Published by Andrzej POPENDA, Czestochowa University of Technology
Abstract. In the paper the uncomplicated structure of the active limiter of current reference for a BLDC motor control system is proposed. The limiter allows for uninterruptible operation of a speed controller due to the automatically adjusted rate of change of angular velocity reference depending on actual moment of inertia. The results of investigations, which confirm the effectiveness of the proposed structure, are presented.
Streszczenie. W artykule zaproponowano nieskomplikowaną strukturę aktywnego ogranicznika prądu zadanego w układzie regulacji prędkości kątowej silnika BLDC, która umożliwia bezprzerwowe działanie regulatora prędkości dzięki automatycznemu dostosowaniu szybkości narastania prędkości zadanej do dowolnego momentu bezwładności. Zaprezentowano wyniki badań, które potwierdzają skuteczność działania proponowanej struktury. (Tłumienie drgań w elektrycznym układzie napędowym z długim elementem sprężystym).
Keywords: electric drive system, speed controller, transmission shaft, vibration. Słowa kluczowe: elektryczny układ napędowy, regulator prędkości, wał transmisyjny, drgania.
Introduction
Rapid development and diversity of electric drive systems require extensive theoretical and practical knowledge as well as the use of a wide range of theoretical and constructional solutions from the contemporary designers and users. Electric drive systems are used in all branches of industry, therefore, the trouble-free operation of these systems is of crucial importance. An analysis of the states of electric drive operation is often related to ensuring its safety. A key issue is to detect the mechanical resonance phenomenon or the phenomena close to resonance as the most unsafe ones for the system. Ignoring the analysis and the ongoing diagnostics of vibrations in electromechanical systems results in many failures [1, 2].
Electric motor, being a part of an electric drive, is coupled with a working mechanism via a driving shaft that is an element of mechanical power transmission. Mechanical power transmissions can be single-path or multi-path and can also include gear trains and clutches [3, 4]. The long driving shafts, defined as transmission shafts, are used first of all in the drive systems for the steel industry, mainly in the drive systems for rolling mills – the transmission shafts are over 10 meters long and their diameters are of 0.5 to 0.8 m [1-8]. Transmission shafts are also used in drive systems for polymerization reactors [5]. The length of these shafts is from 4 to 7 meters. Moreover, transmission shafts are used in hydro generator sets, ship drive systems, submarine drive systems, etc. [1].
Control systems for electric motors are usually equipped with the controllers of position, speed and current or torque, e.g. [10]. Step change or rapid change of angular velocity reference result in the temporary lock-down of the speed controller as a consequence of the applied limiter on the controller output (current reference limiter). Particularly negative consequences of such lock-down can be observed in drive systems, in which there are long elastic couplings (transmission shafts) between an electric motor and a working machine. As a consequence, transmission shafts are being twisted and moments of torsion of significant magnitude occur. The amplitude of these moments can be much higher than the rated torque of motor.
In the paper the uncomplicated structure of the active limiter of current reference for a brushless dc (BLDC) motor control system is proposed. The limiter allows for uninterruptible operation of speed controller due to the automatically adjusted rate of change of angular velocity reference depending on actual moment of inertia.
The electric drive system with a long elastic coupling
The investigated drive system includes the BLDC motor of 4 kW with a control system, steel transmission shaft of length 0.66 m and diameter 0.008 m, the additional rotating mass with moment of inertia JL and dc generator (Fig. 1) [1, 2]. Two rotary incremental encoders with a resolution of 3600 pulse/rev, installed on the transmission shaft ends, are used to measure angular displacement and angular velocity. Hall effect transducers were used to measure motor phase currents. The measuring signals from the transducers are sent to a laboratory computer equipped with two multifunctional I / O devices.
In Fig. 2 the block diagram of a standard structure of BLDC motor (BLDCM) control system is shown. Closed-loop controllers of rotor angular velocity and armature current are included in the system, e.g. [9]. Similar solutions are used to control both brushless and brushed dc motors, but the armature current of brushed dc motor is obtained as a result of a direct measurement. It should be noted that also control systems for ac motors include speed controllers and current controllers or alternatively torque controllers, e.g. [11].
In Figs. 3 to 6 the exemplary waves of angular velocities at shat input and output, difference in angular velocity and angle of shaft twist for the investigated drive system during starting a motor, an idle run and a run under load, respectively, are shown [1]. The waves were obtained by using both computer simulation and measurements on the test stand (Fig. 2). The long driveshaft modelling algorithm based on an electric transmission line [1] was used in the simulation.
Fig.1. A photo of the investigated electric drive system
Fig.2. A standard BLDC motor control system, where SC is speed controller (angular velocity controller), ST is Schmitt trigger, RPS are rotor position sensors (Hall effect sensors), HC is Hall code, ACC is armature current calculator, ESS is electronic signal select
Fig.3. Angular velocity at shaft input vs. time
Fig.4. Angular velocity at shaft output vs. time
Fig.5. Difference in angular velocity vs. time
Fig.6. Angle of twist vs. time
The parameters of 4 kW BLDC motor and steel transmission shaft of length 0.66 m and diameter 0.008 m, according to the real laboratory setup elements [1], were taken into simulation. The end of the transmission shaft was loaded by a rotating mass JL = 0.11 kg·m2 and a mechanical torque of rated value.
A lock-down of speed controller is clearly visible within the first second of starting a motor i.e. from the instant when a step change of motor angular velocity reference occurs at the system input (Figs. 3 to 6). However, as far as the speed controller exits from the limiter zone, the vibrations are damped quickly as a consequence of unlocking the speed controller operation. In addition, a standard control system (Fig. 2) reduces vibrations effectively without the additional damping circuits or devices.
The active limiter of current reference for a BLDC motor control system
The solution for vibration damping at each operational state of drive system, also during starting a motor, is a slowdown of the angular velocity reference rate of rise. However, the intensity of such slowdown should correspond with actual moment of inertia of motor rotor and rotating masses connected to it. In more complex working mechanisms this task requires determination of an equivalent moment of inertia expressed in motor shaft terms. Instead, this paper proposes the uncomplicated structure of the active limiter of current reference (Fig. 7) which allows for uninterruptible operation of speed controller due to the automatically adjusted rate of change of angular velocity reference depending on actual moment of inertia. This structure may be connected to the standard BLDC motor control system (Fig. 2).
Fig.7. Active limiter of armature current reference, where C is speed comparator, I is integrator, NF is nonlinear function
In Figs. 8 to 19 the exemplary waves of angular velocity reference, difference in angular velocity, angular velocities at shat input and output, electromagnetic torque of motor and angle of shaft twist in the investigated drive system equipped with the proposed active limiter of current reference (Fig. 7) are shown.
Fig.8. Angular velocity reference vs. time, JL = 0.11 kg·m2
Fig.9. Difference in angular velocity vs. time, JL = 0.11 kg·m2
Fig.10. Angular velocity at shaft input vs. time, JL = 0.11 kg·m2
Fig.11. Angular velocity at shaft output vs. time, JL = 0.11 kg·m2
Fig.12. Electromagnetic torque of motor vs. time, JL = 0.11 kg·m2
Fig.13. Angle of twist vs. time, JL = 0.11 kg·m2
Fig.14. Angular velocity reference vs. time, JL = 0.94 kg·m2
Fig.15. Difference in angular velocity vs. time, JL = 0.94 kg·m2
Fig.16. Angular velocity at shaft input vs. time, JL = 0.94 kg·m2
Fig.17. Angular velocity at shaft output vs. time, JL = 0.94 kg·m2
Fig.18. Electromagnetic torque of motor vs. time, JL = 0.94 kg·m2
Fig.19. Angle of twist vs. time, JL = 0.94 kg·m2
The same operational conditions were taken into account like for the described previously system with a static current limiter (Fig. 2) i.e. starting a motor, an idle run and a run under load, respectively. The end of the transmission shaft was loaded by an additional rotating mass JL = 0.11 kg·m2 and 0.94 kg·m2, respectively, and a mechanical torque of rated value.
Conclusions
In the paper the uncomplicated structure of the active limiter of current reference for a brushless dc (BLDC) motor control system is proposed. The limiter allows for uninterruptible operation of speed controller due to the automatically adjusted rate of change of angular velocity reference depending on actual moment of inertia. The proposed solution allows for damping of vibration at each operational state of drive system i.e. during starting a motor, an idle run and a run under load, etc. This is particularly important in drive systems, in which there are long elastic couplings (transmission shafts) leading to the significant level of vibration in mechanical system.
REFERENCES
[1] Popenda A. , Lis M., Nowak M., Blecharz K., Mathematical Modelling of Drive System with an Elastic Coupling Based on Formal Analogy between the Transmission Shaft and the Electric Transmission Line, Energies, 1181(2020), No. 13, 1-14 [2] Lis M. , Modelowanie matematyczne procesów nieustalonych w elektrycznych układach napędowych o złożonej transmisji ruchu, Wydawnictwo Politechniki Częstochowskiej, Częstochowa 2013 [3] Popenda A., Mathematical modelling of transmission shafts based on electrical and mechanical similarities, Przegląd Elektrotechniczny, 95(2019), No. 12, 196-199. [4] Rusek A., Stany dynamiczne układów napędowych z silnikami indukcyjnymi specjalnego wykonania, Wydawnictwo Politechniki Częstochowskiej, Częstochowa 2012 [5] Popenda A., Modelowanie i symulacja dynamicznych stanów pracy układów napędowych do reaktorów polimeryzacji z silnikami indukcyjnymi specjalnego wykonania, Wydawnictwo Politechniki Częstochowskiej, Częstochowa 2011 [6] Czaban A. , Lis M. , Mathematical Modelling of Transient States in a Drive System with a Long Elastic Element, Przegląd Elektrotechniczny, 88(2012), No. 12b, 167–170. [7] Lis M. , Szaf raniec A. , Model matematyczny synchronicznego układu pompowego o podatnej transmisji ruchu, Maszyny Elektryczne – Zeszyty Problemowe, 118(2018), nr 2, 165–170. [8] Szaf raniec A. , Modelowanie matematyczne procesów oscylacyjnych w napędzie elektrohydraulicznym o podatnej transmisji ruchu, Przegląd Elektrotechniczny, 93(2017), nr 12, 167-170 [9] Andr zejewski A. , Time-Optimal Position Control of DC Motor Servo Drive, Przegląd Elektrotechniczny, 95(2019), No. 12, 85-88 [10] Jakubiec B. , Napęd bezszczotkowego silnika prądu stałego z rozmytym regulatorem prędkości, Przegląd Elektrotechniczny, 90(2014), No. 12, 211-213 [11] Olesiak K. , Application of a fuzzy logic controller for a permanent magnet synchronous machine drive, Przegląd Elektrotechniczny, 92(2016) No. 12
Author: dr hab. inż. Andrzej Popenda, profesor uczelni, Politechnika Częstochowska, Wydział Elektryczny, Katedra Elektroenergetyki, al. Armii Krajowej 17, 42-200 Częstochowa, E-mail: andrzej.popenda@pcz.pl.
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 12/2020. doi:10.15199/48.2020.12.49
Published by Abdelkader BOUDERBALLA1, Boubakeur ZEGNINI2, Tahar SEGHIER2, Abderrahmane TALEB Higher Normal School (ENS) of Laghouat (1), Amar TELIDJI University of Laghouat (2)
Abstract. The protection of the equipment, the safety of persons and the continuity of the electrical supply are the main objectives of the grounding system. For its precise design, the possible potential distribution on the ground and the equivalent resistance of the system must be determined. Grounding systems are considered to be rod electrodes and flat electrodes buried vertically in the ground. A simulation grounding model was established through FEM method on relevant constraints, such as soil resistivity, number, position and length of electrodes. Through this study, the grounding resistance values obtained with COMSOL Multiphysics are compared with the proven analytical formulas. The results indicate that this simulation can be used as fast solution for industrial applications and has acceptable results for calculating the resistance of a vertical grounding.
Streszczenie. Ochrona urządzeń, bezpieczeństwo osób i ciągłość zasilania elektrycznego są głównymi celami systemu uziemienia. W celu dokładnego zaprojektowania systemu należy określić ewentualny rozkład potencjału w ziemi i równoważną rezystancję systemu. Za systemy uziemienia uważa się elektrody prętowe i płaskie zakopane pionowo w ziemi. Symulacyjny model uziemienia został ustalony metodą FEM na podstawie odpowiednich ograniczeń, takich jak rezystywność gruntu, liczba, pozycja i długość elektrod. Dzięki temu badaniu, wartości rezystancji uziemienia uzyskane za pomocą COMSOL Multiphysics są porównywane ze sprawdzonym wzorami analitycznymi. Wyniki wskazują, że symulacja ta może być wykorzystana jako szybkie rozwiązanie do zastosowań przemysłowych i daje akceptowalne wyniki do obliczania oporności uziemienia pionowego. (Określanie rezystancji uziemienia z zastosowaniem metody FEM)
Keywords: grounding resistance, finite element method, soil resistivity. Słowa kluczowe: rezystancja uziemienia, metoda elementów skończonych, rezystywność gruntu gruntu.
Introduction
For several decades, research has been intensifying in the field of the ground of electrical installations. The vast majority of this research was aimed at behaviors of these groundings at industrial frequency and under steady state conditions. In addition, the resistivity of the soil considered was generally close to 100 Ω.m (value often in temperate regions), which is not the case in the tropics in certain types of soil. Soil resistivity can reach several thousands of ohms meters. The study of the behaviour of the earth network requires a previous analysis of the distribution of the electric potential in the surrounding soil. This distribution is a function of the electrical characteristics of the ground, i.e. its resistivity (to a lesser extent its permittivity too), the geometrical characteristics of the ground network and the source. The design of an earth network must therefore be preceded by an investigation.
The safety, reliability and correct operation of the power system depend on the design and construction quality of your grounding system [1]. The construction project aims to protect the equipment and personnel of the electrical system from the danger of electric shock. The grounding system can include only one grounding electrode or the entire electrode group. In many of the applications, low grounding resistance is essential to meet electrical safety standards [2].
The grounding resistance is often called the earth resistance. Grounding resistance is defined as the relationship between the voltage on the grounding device and the current flowing to the earth through the grounding device. The value of the earthing resistance is an important technical parameter related to the safety of personnel and equipment. If the earth resistance is high and earth current occurs, it can result in death or personal injury and equipment damage. The value of the earthing resistance at a certain value of the grid current determines the value of the dangerous voltage inside or around the substation.
The main purpose of designing the grounding system of a substation or power plant is to provide safety to personnel in the event of a ground fault. The primary purpose of the electrical system substation is to maintain reliable operation and provide protection to personnel and equipment in the event of failure. The basic configuration of the grounding grid is composed of round steel that forms a two-dimensional grid (usually square or rectangular) and is buried close to the earth surface. The ground rods are only effective if a significant portion of their length is in contact with a low resistivity soil. From the point of view of the safety of people and equipment, there is no doubt that the ground door is the most important part of the electrical system. The reliability and availability of the electrical system depends on the quality of the design and construction of the grounding grids.
The main objectives of the grounding system are:
• Protect personnel from electrical hazards by limiting overvoltage that can be found in the event of a failure in the electrical system.
• The safety and continuity of electrical equipment by limiting the overvoltage that can occur in extreme conditions or accidents.
• The correct operation of the equipment and electrical protection equipment allows detecting faults and selecting all the measures destined to disconnect areas of equipment due to ground fault current.
The calculation method used to simulate the grounding grid considers the following simplified assumptions:
• The soil is an infinite, flat, isotropic and multilayer environment.
• The electromagnetic method is suitable for calculating the grounding resistance and the lattice potential.
• The ground network conductors are considered linear, connected to each other and buried near the surface of the earth.
• The behavior of the earthing grid at the estimated operating frequency can be determined by electromagnetic field analysis techniques for fixed fields (the propagation time can be ignored).
The different methods of calculating the grounding resistance are based on determining the potential or capacitance of the grounding electrode. The method of calculating the grounding grid uses different mathematical techniques to determine the grounding resistance and the step voltage and contact voltage, applying the hypotheses that allow us to model the real system in other theoretical of comparable results. These studies are generally performed on the ground mesh with symmetry and uniformity [2] or two or more layers [3] – [7]. The initial method was based on calculating the earth fault current (amps) from the random grid potential and obtaining the earth resistance value as the ratio between voltage and current. After calculating the earth resistance, the actual grid potential (calculated as the product of the actual fault current and the calculated earth resistance value) is again used to analyze the finite element model to obtain the potential distribution in the model and the touch voltage and the step voltage order.
Recent studies based on the finite element method (FEM) have been used to calculate the ground resistance of earth systems [8],[9] FEM allows to obtain the grounding resistance as a function of the resistivity of soil . Therefore it is possible to justify the use of FEM in the dimensioning of the grounding systems. The FEM method is used to evaluate the solution of the partial differential equation that governs the behavior of the system. The grounding system is considered to be a rod electrode buried vertically in the ground. Earth resistance is also determined using FEM and with calculating the dissipated power or from the stored energy. Then, by integrating the potential distribution on ground surface, current density distribution inside soil, we finally calculate the amplitude of the current passing through the earth rod or region, the ground resistance is determined as the ratio between the voltage and the calculated current. In this paper , we will propose a simple model of calculation by FEM to evaluate the impulse characteristics of the soils, coming as close as possible physical realities. This method will allow us to study the effect of certain parameters (soil resistivity, geometric positions of the electrodes) on performance of grounding systems in order to propose concrete solutions for the best realization of earthing systems, respecting the standards for the safety of equipment and people.
Material and Method
The latest research on fundamental analysis is based on the finite element method. The design of the earthing system using the FEM method aims to determine the earth resistance. Compared to the traditional soil method, the finite element method provides more accurate results [8]. The oldest FEM method consists of current analysis using the electrode potential. Once the current is calculated, the ground resistance can be found by dividing the voltage by the current. In this method, the main disadvantage is that the size of the model is chosen as a considerable ground distance from the ground electrode. Because the earth electrode of the grounding point considers the analysis of each potential in the grounding of the selected point.
Where: From tests of various designs, d1 can be determined by:
.
where: D is the diagonal distance of grounding electrode, d1 is the distance from electrode to the points where semispherical model of equipotent surface distribution.
Resistance of grounding electrode can be derived from Fig.(1):
.
where R1 is the resistance inside the semispherical surface, R2 is the resistance outside the semi-spherical surface.
Resistance R2, ground resistance between d1 and infinite, which is small compared to, can be calculated by applying expression (3) to calculate the resistance of a semi-spherical resistor of internal radius and infinite external radius see Fig( 1). R2 is computed from the following equation
.
where: ρs the soil resistivity
Fig.1. The resistance R2 (d2 is the distance from electrode to the points where electrical potential goes to zero)
The R1 ground resistance of a semi-sphere with radius d1 (see Fig. (1)).Being d1 the earth distance for which the distribution of potentials can be supposed spherical.
This is where finite element analysis exactly takes its place. In general, R1 can be calculated from dissipated power given in the following equation:
.
R1 can be detailed by replacing the terms as in Equation (4)
.
where: VG is the potential in the grounding electrode; VB is the potential in the boundary d1; E is the electric field; dV is the volume element; σ is the electrical conductivity.
At the same time, dissipated power is determined by :
.
Dissipated power (6) is calculated by applying current flow analysis to the finite-element model. The finite-element model is comprises d1 of a semi-spherical volume of radius Fig. (1) and is defined according to the geometry of the grounding grid and the soil structure.
.
where: V is the electrical potential; σ is the electric conductivity, je is the external current density and Qj is the current source density.
In the proposed method, the ground around the rod is divided into three cylindrical shaped parts as shown in Fig. (2). It is believed that the first cylinder, referred to as the grounding system, has a radius of 2.5 times the length of the rod to maintain a sufficient volume of the beam to guarantee efficient current discharge.
The second cylinder, called the effective soil, is usually considered to have a radius and a height of 10 m. The next layer or the third cylinder with a height and a radius of more than 10 meters, which does not have a significant effect on the soil resistivity, is considered to be infinite.
The cylindrical electrode, although it is not really used, has great importance in the study of grounding systems since equipotential surfaces of any grounding rods become hemispheric at a sufficiently long distance form the rods themselves. Therefore, it is possible to approximate any shaped rod with a cylindrical electrode having an equivalent radius.
Fig.2. Grounding model
FEM is known with its unique triangles. Initializing the mesh allows the designer to see the triangles made by the COMSOL multiphysics solver, which is shown in Fig. (3).
Fig.3. Initialized mesh model
Fig.4. Subdomain and physics selection
Table 1.Subdomain for the rod electrode and Soil
.
The boundary conditions define the interface between the geometry of the model and its environment. You can also set the interface conditions on the internal limits of the model geometry. Different boundary conditions can be defined for each boundary is shown in Fig. (5).
Fig.5. The rod and the soil grounding boundary conditions
You can also define different boundary conditions for each limit. For the bar, the potential limit is defined as V0 = 100 V. It should be noted that the surrounding soil actually has infinite dimensions. Therefore, it is possible to model with a domain whose size is much larger than the size of an earthing system with boundary conditions set to zero potential.
For the soil, the boundaries are selected as follows:
• From conditions shown in Fig. (2), the four sides, and the bottom of the soil cylinder were set to ground boundary condition.
• The top of the soil cylinder was set to electrical insulation boundary. This model is the model governed by Laplace equations, which is the governing equation for the earthing system under design. Dependent variable is by default set to 100 V, which is the behavior parameter to be studied and analyzed. At industrial frequencies, the ground connection is modeled by resistance.
To obtain the Laplace equation final solution, simply clicking the Solve button on the Main toolbar yields the solved model.
Fig.6. Solved model
Fig.7. Post-processing from Menu Bar
Fig.8. Subdomain Integration Dialog Box
Fig.9. Resistive heating selection
Fig.10. Results of subdomain integration, resistive
To calculate the power loss, the following steps from Fig (6) to Fig (9) have been performed:
1-Select Finishing in the menu bar;
2-From selection after processing to subdomain integration, the Subdomain Integration dialog box opens;
3-Selection of resistive heating from the predefined quantities with remaining values Select subdomain 1 in the subdomain integration dialog
4-By pressing OK, the resistive heating in the bottom medium is calculated, which represents the power loss between the pole edge and the bottom limit;
5-Review the results while viewing the message log to see the results in the drawing area at the bottom of the message log.
The dissipated power is calculated using this software by integrating the appropriate domain for the predefined quantity of resistive heating. The domain of the soil has been chosen to calculate the dissipated power between the rod edge and the outer boundary of the soil. Resistive loss is chosen by clicking the volume integration button in the software window, and the value of the resistive loss is found. To obtain the ground resistance of the earthing system from COMSOL Multiphysics we use the resistive heating integration in the Subdomain of soil, to get the dissipated power from the rod in the soil around it, and the resistance R1 is equal to the square of the voltage of rod over this power, a detailed calculation can be found in the reference [12].
Simulation Results and Discussion
The results obtained by the COMSOL program package based on FEM are given. COMSOL is software for various physics and engineering applications (especially related phenomena or multiphysics). For comparative analysis, the AC / DC module is used in conjunction with the “current” interface. The geometry of the ground system is modelled based on the actual geometry of the grounding system.
The rod, is designed as a cylindrical shaped element with radius r, and length l, and made of copper with conductivity of copper that is σ = 5.99 107 (Ω. m)−1 or resistivity of copper which is ρ = 1.66 10-8 (Ω .m). The rod is driven vertically into the soil.
For the evaluation, we will start with a simple hemispherical electrode. The soil is represented by a hemisphere with a large radius (towards infinity). First, to simplify the calculation, let’s consider homogeneous soil. The calculation will be done in asymmetric situations and in 3D.
Next, we will consider several layers of soil with different resistivity’s. The model used is identical to the one that the soil has a single layer; the analytical relations depend on the length of the electrode in relation to the depth of the layer Fig (11).
A two-layer soil model is generally used for nonhomogeneous soil characterizations. The first layer thickness is h = 5.5 m with a specific resistivity of 1000 Ω. m, where the second layer with resistivity is 100 Ω. m, the calculations were carries out in 3D and axi-symmetric.
Fig.11. Cylindrical grounding geometry in 3D, and axi-symmetric
In all simulations, a 10 A current has been injected into the ground rod in order to evaluate the potential on the earth surface from the rod itself to a sufficient long distance characterized by a null the ground potential.
The computations are conducted in axi-symmetric Fig. (12) shows the possible potential distribution of a 10 A current flowing through a 4 m long electrode.
Fig.12. Soil potential in two layers
For a cylindrical copper electrode of radius 0.0125 m, in a homogeneous soil of resistivity 100 Ω.m, the grounding resistance is given by the Fig (13), as a function of the length of the electrode, in axi-symmetric and 3D. Also represented on this graph are the resistances calculated with the analytical relationships of Rudenberg, DwightSunde and Liew-Darveniza (see Table 2).
Table 2. Analytic equation of the grounding resistance
.
With ρs the resistivity of the soil, l the length of the electrode and d the diameter of the electrode, r the radius of the electrode.
Fig.13. Resistance according to the length of the electrode
The relative errors considering that the analytical relation Dwight–Sunde is exact are indicated below on the Fig. (13), these curves show us that the modeling which we have chosen is adapted to a cylindrical electrode. The errors are less than 10% compared to the analytical calculations Fig. (14).
Fig.14. Relative error FEM Axi and FEM 3D
We also note that analytical relationships of DwightSunde and Liew-Darveniza give more accurate results, close to those obtained by finite elements, the axisymmetric calculation being more precise than the 3D.
As for the cylindrical electrode, the 3D calculation is very expensive in term of computing time and resources. For copper electrodes of radius r = 0.0125 m, the distance between electrodes being equal to twice their length, and for a resistivity of 100 Ω.m, the resistance of the set of two stakes in line and the relative error by considering analytical equation of Dwight-Sunde gives results closer to those calculated by finite elements, with relative errors of less than 5%.
According to the standard, improving the grounding system means that the resistance of the grounding system is reduced. For this reason, an effective way to reduce resistance must be found. We will list these different resources and assess the impact of improving their authorized grounding. In this evaluation, we will always calculate the ground potential after current injection, then we can deduce the resistance value from it and compare it with the literature [13],[14].
Table 3. Resistance as a function of the diameter of the vertical cylindrical electrode
.
To achieve low resistance, a single vertical electrode is usually not sufficient. It is important to use several electrodes in parallel. Fig. (15) and Fig. (16) show the ground potential and the earth resistance depending on the number and distance between vertical cylindrical electrodes 2 m length, and r = 0.0125 m with a current of 10 A with soil resistivity of 100 Ω.m. Tables 3 and 4 show the effect of increasing the number and distance of electrodes on the grounding resistance.
Table 4. Resistance as a function of the distance between the 4 cylindrical electrodes
.
Fig.15. The potential distribution at the ground surface as a function of the diameter of the vertical cylindrical electrode. ( l =2 m length, I = 10 A current , soil resistivity of 100 Ω.m).
Fig.16. The potential distribution at the ground surface as a function of the distance between the vertical cylindrical electrodes. ( l =2 m length, and r = 0.0125 m, I = 10 A current , soil resistivity of 100 Ω.m).
For example, when changing from one electrode to 4 electrodes, the resistance is divided almost by four. We realize that when changing from one electrode to two electrodes, the resistance can be minimized, which means that it is efficient, but takes up space, and increase installation costs.
According to the results in Table 4, we noticed that the grounding resistance decreases as the distance between the four earth stakes increases. For example, if the distance goes from 2 m to 4 m, the ground resistance decreases by 12. 36%, while that from 2 m to 10 m, the decrease is only 20%. This shows once again that for cost and space considerations there is a limit to the distance between the electrodes. The literature reports that the 6 m distance between the electrodes is economically a limit to the cost of grounding [13].The impact of resistivity on the grounding resistance for an electrode of l = 2m, r = 0.0125m through which a current of 10A flows in a soil of different resistivity (10 Ω.m, 100 Ω.m, 1300 Ω.m) are represented by Fig. (17). Differences in results of calculations and simulations were caused by simplifications adopted in calculations.
Fig.17. The grounding resistance [Ω] as a function of soil resistivity
In high resistivity soils, the conductivity of the earth conductor is less than that of low resistivity soil, making it difficult for the earthing system to dissipate the current injected into soil. Resistivity can be reduced by treating the soil with charcoal and other products such as wood bentonite and salt [5].The IEC 62305-3 standard deals with the protection, inside a structure, against physical damage and against injury to living beings due to touch and step voltages. The essential and most reliable protective measure for the protection of structures against physical damage is considered to be the lightning protection system 18]. Main protection measures against injury to living beings due to touch and step voltages are intended to: reduce the dangerous current flowing through bodies by insulating exposed conductive parts, and/or by increasing the surface soil resistivity. Using durable materials to ensure long, continuous operation. In dry soil, but in corrosive acid, salty, or humid environments require regular maintenance. Soil resistivity depends on the content of electrolytes in the soil, geological and chemical considerations and seasonal variations in a site. In different seasons, the resistivity of the surface soil layer will change, which will affect the safety of grounding system, and the grounding resistance, step and touch voltage will move to the safe side or to the hazard side. In this evaluation we will always calculate the potential in the earth after the injection of the current, then we can deduce the value of the resistance and we compare it with the analytic expressions.
Conclusion
The purpose of this work was to define an alternative method for assessing earth resistance with FEM methods. This method makes it possible to take into account more specific cases than what the standards formulas (calculations according to fundamental formulas) for calculating the earth resistance of different ground electrodes. The model was validated by comparing the well-known equation of ground resistance for a vertical cylindrical electrode with one simulated by FEM analysis.
In this study, in order to evaluate the grounding resistances (also the electric field and potential distribution and current density) in industrial frequency, the finite element calculation models were presented in 2D axisymmetric and 3D, for some electrode configurations. These calculation models, which have the advantage of being closer to the actual physical situation, and which apply to more complex configurations have been validated against analytical calculations, with differences in relative error low and often in the order of 10%. These results showed that axi-symmetric calculations are much cheaper in terms of calculation time and resources because they allow much finer meshing. The ground resistance should be as low as possible to ensure the safety of personnel and equipment. The presented FEM method is very useful for calculating grounding resistance as a function of the resistivity of soil. Then, it will be possible to justify the use of this method in dimensioning of the grounding systems.
A future development of this work may be the study of the frequency variation, especially the behavior of the earth for high-frequency components that characterize power electronic devices or telecommunication systems.
Acknowledgments: This work was financially supported by Directorate General for Scientific Research and Technological Development (DGRST) Ministry of Higher Education and Scientific Research Algeria, under the Scientific Program/ PRFU. Contract/, No A01L07UN030120180002.
REFERENCES
[1] Hyung-Soo L., Jung-Hoon K., Dawalibi F. P., and Jinxi,,M. “Efficient ground grids designs in layered solids,” IEEE Trans. Power Del., vol. 13, no. 3, pp. 745–751, Jul. 1998. [2] Thapar B., Gerez V., Balakrishnan A. and Blank D. A. “Evaluation of a grounding grid of anyshape”, IEEE Trans. on Power Delivery, Vol. 6,No. 2, pp. 640-645, April 1991. [3] Dawalibi F. P. and Mukhedkar D., “Optimum design of substation grounding in two-layer earth structure”, IEEE Trans. Power Apparatus and Systems, Vol. PAS-94, No, 2, pp. 252-272, April 1975. [4] Seedher,H.R. Arora J.K. and Thapar B., “Finite expressions for computation of potential in two layer solid”, IEEE Trans. on Power Delivery, pp.1098-1102, 1987. [5] Lee H.S., . Kim J.H, Dawalibi F.P. and Ma J., “Efficient ground grid designs in layered soils”, IEEE Trans. On Power Delivery, Vol. 13, No. 3, pp.745-751, July 1998. [6] Thapar B.. and Goyal.S.L., “Scale model studies of grounding grids in Non-uniform soils”. IEEE Trans. on Power Delivery, Vol. 2, pp. 1060-1066, October 1987. [7] Takahashi T. and Kawase T., “Calculation of earth resistance for a deep-driven rod in a multi-layer earth structure”, IEEE Trans. On Power Delivery, Vol. 6, No. 2, pp. 608-614, April 1991. [8] Nenad N. Cvetković, Dejan B. Jovanović, Aleksa T. Ristić, Miodrag S. Stojanović, Dejan D. Krstić, Comparison of different models for determining the grounding rod resistance, ELECTROTECHNICA & ELECTRONICA, (E+E), Volume 50, Issue 9-10, 2015, p p.35-39. 2015. [9] Sajad Samadinasab, Farhad Namdari,Mohammad Bakhshipour, A Novel Approach for Earthing System Design Using FiniteElement Method, Journal of Intelligent Procedures in Electrical Technology, Vol. 8 – No.29 – pp54-63 Spring 2017. [10] AbuBakar,A. Dow R.S., “Simulation of ship grounding damage using the finite element method”, International Journal of Solids and Structures, Vol. 50, No. 5, pp. 623-636, 2013. [11] Boudreballa Abdelkader, Zegnini.Boubakeur, Seghier Tahar, Gueffaf Hamza, implementation of magneto dynamic to evaluate grounding performance, 5th international conference on advances in mechanical engineering Istanbul 2019, 17-19 december 2019, Istunbul –Turkey.2019. [12] Boudreballa Abdelkader, Zegnini.Boubakeur, Seghier Tahar, Saadaoui Yahia, ‘’implentation and design of grounding systems using COMSOL multiphysics’’, conference:2020,1st International Conference on Communications Control and Signal Processing (CCSSP), Procceding EEE, pp 513 – 517. 2020. [13] Vijayaraghavan G., Mark Brown, Malcolm Barnes,”Practical Grounding, Bonding, Shielding and Surge Protection, ” Elsevier, 2004. [14] Yaqing Liu, “Transient Response of Grounding Systems Caused by Lightning: Modeling and Experiments”, PhD thesis Department of Engineering Sciences, University of Uppsala, 2004. [15] Standard for ENA ER/S.34, “A guide for assessing the rise of earth potential at substation sites”, Energy Network Association, Issue 1, 1986. [16] Tagg G.F. “Earth Resistance”, George Newnes Ltd, England, 1964. [17] Sunde E. D. “Earth conduction Effects in Transmission line Systems”, Dover Publications Inc., 1968. [18] Grzegorz KARNAS, Stanislaw WYDERKA, Robert ZIEMBA, Kamil FILIK, Grzegorz MASLOWSKI, Analysis of lightning current distribution in lightning protection system and connected installation, Przeglad Elektrotechniczny, ISSN 0033-2097, R. 90 NR 2/2014, pp 174-178, 2014.
Authors: PhD student Abdelkader Boudreballa, Laboratoire d’études et développement des matériaux semi-conducteurs et diélectriques (LeDMaScD), Amar Telidji University of Laghouat., Abderrahmane Taleb Higher Normal School (ENS) of Laghouat, Ghardaia road Laghouat 03000, Algeria,E-mail: bouderbalalagh@gmail.com. prof dr Boubakeur Zegnini , Laboratoire d’études et développement des matériaux semiconducteurs et diélectriques (LeDMaScD), Amar Telidji University of Laghouat. ,PoBox 37 G, Mkam Laghouat 03000,Algeria,E-mail: b.zegnini@lagh-univ.dz, prof dr Seghier Tahar , Laboratoire d’études et développement des matériaux semi-conducteurs et diélectriques (LeDMaScD), Amar Telidji University of Laghouat. ,PoBox 37 G, Mkam Laghouat 03000,Algeria,E-mail: t.seghier@lagh-univ.dz
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 5/202. doi:10.15199/48.2021.05.02
Published by Gideon UMOH1, Chinedu OBE2, Cosmas OGBUKA2, George EKPO3, Emeka OBE2, Department of Electrical Engineering, Maritime Academy of Nigeria, Oron, Akwa-Ibom State, Nigeria(1), Department of Electrical Engineering, University of Nigeria, Nsukka, Enugu State, Nigeria(2) Department of Electrical and Electronic Engineering, Akwa Ibom State Polytechnic, Ikot Osurua, Nigeria(3)
Abstract. The direct-phase variable (DPV) model of a five-phase synchronous reluctance motor (SRM) for direct-on-line starting is presented. The model eliminates the dependence of inductances on rotor position. Machine performance characteristics namely vector potential, speed, flux linkage and current were monitored under different conditions. The DPV model was simulated in MATLAB/Simulink while a finite element model (FEM) was simulated using ANSYS Maxwell FEA software for comparison and validation. Very close similarities in obtained results justify the DPV model.
Streszczenie. Przedstawiono moddel pięciofazowego reluktancyjnego silnika synchronicznego wykorzystywanego do startu on-line. Zbadano potencjał wektorowy, szybkość, strumień rozproszony i prąd w różnych warunkach pracy. (DPV model i analiza pracy pięciofazowego synchronicznego silnika reluktancyjnego do bezpośredniego startu on-line) .
Keywords: direct-on-line (DOL), direct-phase variable (DPV), finite element model (FEM), five-phase synchronous reluctance motor (SRM) Słowa kluczowe: silnik reluktancyjny, silnik pięciofazowy, silnik synchroniczny.
Introduction
The conventional starting method in most industrial processes utilizing the induction motor are the direct-on-line starting, star-delta starting and the auto-transformer starting, which have served the industries for several years [1,2]. Owing to the high cost of utilization of the power electronics drives systems, advancement in the area of electric machine design can minimize the use of the power electronics drives, with minimal or no compromise to the machine performance especially where variable speed operation is not desired. Considering the most common and easily used starting method; the direct-on-line staring, the machine windings can be modelled to have a similar waveform as its supply voltage. The squirrel cage placed on the rotor has been the backbone of direct-on-line starting methods. The synchronous reluctance motor (SRM) can also be modelled to harness these advantages by using a cage rotor. The SRM has a conventional stator similar to that of the induction motor (IM), but utilizes reluctance torque for energy conversion, and can be modelled as a simplest form of a salient pole synchronous motor without any field winding [3,4]. Earlier design improvements of the SRM seeks to improve saliency ratio of the rotor [5,6,7,8].
For the synchronous machine, the stator inductances depend on the rotor position, as the transformation of the machine variables to the d-q frame to remove these dependencies is adopted by many authors especially for the multi-phase machines [9]. The unnecessary encumbrances associated with the d-q modelling are readily eliminated using the DPV model [10]. The phase variable model has remained an essential model in predicting the machine performance during faults and unbalanced conditions [4, 11,12]. The phase variable model takes into account the stator phase windings and the rotor d-q windings of the machine, representing the machine equations in terms of the machine variables, as compared to the direct-quadrature (d-q) axis transformation of which the stator phases are transformed to a fictitious axis of references. The combination of the Winding Function (WF) model and the direct-phase variable model has been proven to give accurate result that can be compared with the Finite Element Method (FEM) [4]. The finite element model [4, 13,14,15,16,17,18,19,20,21], has helped in electric machine design optimization, and in reducing design error to a minimum. The use of the multi-phase machines has assisted in torque production without an increase in the copper loss of the machine by the utilization of additional harmonics with the fundamental component, and helps in the optimization of the air-gap magneto motive force (MMF) [11,22]. The advantage of the 5-phase over 3-phase machine is that torque pulsation especially for SRM is highly reduced [23,24,25].
The present study is aimed at developing a direct-phase variable (DPV) model of a five-phase synchronous reluctance motor for direct-on-line starting. The developed model will be subjected to performance test during normal, loaded, fault and unbalanced conditions considering the following performance characteristics; vector potential, speed, flux linkage and current. The suitability of the developed DPV model will be validated by comparing the results with results obtained using the FEM which is widely considered as near ideal.
Phase variable model of five-phase synchronous reluctance motor
A five-phase (5-ph), cage rotor synchronous reluctance motor having 4-pole, 40 slots is modelled for direct-on-line starting. The supply is a 5ph voltage from a converter, without any form of control, modelled to have a similar waveform as in Fig. 1 (i.e. a sinusoidal waveform of the (first and the third harmonics) to reduce torque ripples.
Fig.1. Supply voltage waveform
The voltage equation for a five-phase SRM is given in [4] as.
.
.
where θr is the rotor position. The rotor inductances are function of the rotor position. and V are the machine current and voltage matrices and are given in (3) and (4) respectively. The inductance matrix is given in (5).
.
where,
.
where Lss is the stator inductance matrix, L’r is the rotor inductance matrix referred to the stator, L’sris the mutual inductance matrix between the stator and the rotor referred to the stator, L’lkq is the rotor q-axis leakage inductance referred to the stator and L’lkd is the rotor d-axis leakage inductance referred to the stator.
.
where R is the resistance matrix. The expression for the electromagnetic torque is given in (10).
.
Where Is , Ir, J , p and Tlare the stator current matrix, rotor current matrix, inertia constant, number of machine poles and load torque respectively.
The Stator resistance can be calculated from (14),
.
Where: σc, vxl and Acare the conductivity, total volume of conductor per phase and the conductor cross-sectional area.
Inductance Calculation
The motor structure showing the cross-section of the five-phase SRM is shown in Fig. 2. On the basis of this cross-section, the configuration of the full-pitched double layer winding is tabulated in Table 1.
Fig.2. Cross Sectional Area of 5ph SRM showing winding layout
The winding function can also be approximated using the fundamental and the third harmonic components to represent the different phases with regard to their respective phase shifts.
The different phase windings for any of the phases will be of the form shown in (15).
.
where, ∝ is 0, – 2π/5, – 4π/5, 4π/5, 2π/5 for A, B, C, D and E phases respectively, while Φs is the stator circumferential position.
Fig. 3 shows the winding functions plot against the stator circumferential position of phase A of the machine using the winding function method and its approximation from utilizing only the fundamental and the third harmonic components to represent the actual winding function. Similar plot can be made with the necessary phase shift.
The expression for the calculation of stator inductances is presented in (16).
.
where Nx(φ) nd Ny(φ) are the winding functions of phase X and Y respectively and φ is the stator circumferential position, g-1(φ,θr) is the inverse air gap function which is a function of the stator circumferential position (φ) and the rotor position (θr), while ݈ is the axial length of the air gap of the machine, r is the radius to the mean of the air gap and µ0 is the permittivity of free space. The inverse air-gap function including the third harmonic component is given in (17).
Fig.3. A-phase winding function
.
where,
.
where ga the main air gap length, gb is the inter-polar slot space and β is the ratio of pole arc to pole pitch.
From a generalized equation of inductances, (16) yields:
.
where the expression for and are given in (21) and (22) as:
.
.
where A = 2θr; B = 2θr–θ1; C = 2θr – θ2; D = 2θr + θ2; E = 2θr + θ1;
The mutual inductances between the stator and the rotor are given in (8). If desired, the q-axis and the d-axis magnetizing inductances can be obtained from the stator inductance values, and are given in (23) and (24) respectively.
.
where,
.
Simulation of the Dynamic Process
The design parameters of the 5-ph SRM are given in Table 2. The developed machine was monitored for performance characteristics at synchronism, loading and during fault and loss of synchronism.
The machine speed characteristics are simulated using MATLAB/SIMULINK for the DPV and ANSYS Maxwell software for the FEM. A load torque value of 50N-m introduced at 0.98 seconds, representing a continuous running duty cycle for the motor, and subsequently a loss of e-phase fault while on load, lasting for 0.5 seconds was created at 1.7 seconds in both simulations. A ramp load torque spanning from 50 Nm to 110 Nm within 0.5 seconds was introduced at 3.0 seconds to determine the load carrying capacity of the models. The combined load torque is shown in Fig. 4.
Table 2. 5-Phase SRM machine dimensions and circuit parameters
.
Table 3. Settling time and loss of synchronism for speed characteristics
.
Table 4. Speed performance characteristics
.
Fig.4. Load torque against time
The speed characteristics for the DPV and FEM are presented in Fig. 5. Greater settling time is observed in the FEM model. At synchronism, a settling time of 0.887 seconds is recorded for DPV as compared to a settling time of 0.887 seconds for FEM. The settling time for the speed during different conditions of synchronism, loading, fault and subsequent rectification of fault are tabulated in Table 3.
The speed transient characteristics at start showed the least percentage value rise of 8.9 % about the synchronous speed for FEM as compared to a percentage value rise of 15.27 % for the DPV. The speed transient characteristics at loading showed a least percentage rise of 2.27% and 2.13% for FEM and DPV respectively. At fault a least percentage transient rise values of 4.47 % and 6.07% is observed for DPV and FEM respectively. When the loss of e-phase fault was cleared, the least transient percentage rise value of 2.93% and 6.27% are recorded for DPV and FEM respectively. A lower value of difference in percentage rise were observed on loading, fault and when the fault was cleared. The Speed transient performance characteristics at start, loading, fault and when the fault is cleared are tabulated in Table 4.
The Speed characteristics plots for FEM showing transient at start, loss of phase fault and loss of synchronism are presented in Fig. 6, Fig. 7, and Fig. 8 respectively.
Fig.5. Speed characteristics from start to loss of synchronism (FEM and DPV)
The flux linkage at synchronism, loading fault and when the fault was cleared was monitored with the FEM and the DPV and are tabulated in Table 5. The plot of the a-phase flux linkage during synchronism for FEM and DPV are presented in Fig. 9. A 1.95% drop is observed in the FEM as compared to 0.69% drop as recorded in the DPV. A 22.7% drop is recorded for FEM during fault as compared to 2.53 % drop for the DPV. The current characteristics for the five phases are similar since a symmetrical winding and loading is employed, the A-phase stator current characteristics are presented for both models in Fig. 10 for FEM and DPV, and tabulated in Table 7.
Fig.6. Speed characteristics (showing transient at start to synchronism)
Fig.7. Speed characteristics (showing transient at loss and subsequent restoration of e-phase fault)
Fig.8. Speed characteristics (showing loss of synchronism on loading)
Fig.9. Flux linkage characteristics at synchronism for FEM and DPV
Fig.10. Current characteristics for DPV and FEM
A high transient rise is observed in the FEM at start, with a value of 111.3 A as compared to a value rise of 87.73 A for DPV as shown in Table 6.
At synchronism, a value of 16.11A and 16.56 A is recorded for FEM and DPV respectively. On loading, a current rise of 19.16A and 16.56A was observed for FEM and DPV respectively. The highest variation is observed during fault with a value of 4.02% as compared to at synchronism, loading and when the fault was cleared.
The torque performance of both models at loss of synchronism are tabulated in Table 7. At loss of synchronism due to the introduction of ramp load at 3.0 seconds, spanning between 50Nm and 110Nm a carrying capacity of 63.74Nm and 70.16 Nm are recorded for FEM and DPV respectively.
Table 5. Flux linkage performance characteristics
.
Table 6. Stator A-Phase Current performance characteristics
.
Table 7. Maximum load torque
.
Conclusion
The developed DPV model of five-phase synchronous reluctance motor (SRM) for direct-on-line starting has been analysed using MATLAB/Simulink and validated using ANSYS Maxwell. The performance characteristics of the vector potential, speed, flux linkage, a-phase current and the load carrying capacity were considered. The machine was modelled with full-pitched stator winding configuration with the developed model accounting for the third harmonics of the Air-gap MMF.
The developed DPV model and the FEM were monitored and compared for performance characteristics at start, on loading, during loss of e-phase fault, and loss of synchronism due to loading. Apart from a higher transient rise at start, observed with the FEM, both models show similar performance characteristics of the parameters for the considered conditions. A highest variance of 4.02 % rise in current is observed during fault between the models while the load accommodating capacity of 70.16 N-m at 3.168 seconds for the DPV and 63.74 at 3.115 seconds for the FEM were recorded.
With these competitive results when compared to the FEM which is nearly ideal, it is sufficient to conclude that the developed DPV model can satisfactorily give adequate information on the analysis of 5-phase SRM and can as well be extended to other multi-phase machines.
REFERENCES
[1] Aree P., Precise Analytical Formula for Starting Time Calculation of Medium- and High-Voltage Induction Motors under Conventional Starter Methods, Electrical Engineering, 100 (2018),1195–1203 [2] Kocman S., Orsag P., Pecinka P., Simulation of Start-Up Behaviour of Induction Motor with Direct Online Connection, Advances in Electrical and Electronic Engineering, 15 (2017), No.5, 754-762 [3] Ogbuka C., Nwosu C., Umoh G., A New Cross-Saturated Torque Model of Highly Utilized Synchronous Reluctance Machine, Archives of Electrical Engineering, 67 (2018), No. 1 109–121 [4] Obe E., Binder A., Direct-Phase-Variable Model of a Synchronous Reluctance Motor Including All Slot and Winding Harmonics, Energy Conversion and Management, 52 (2011), No. 1, 284-291 [5] Stipetic S., Zarko D., Cavar N., Adjustment of Rated Current and Power Factor in a Synchronous Reluctance Motor Optimally Designed for Maximum Saliency Ratio, IEEE Transactions on Industry Applications, 56(2020), No. 3, 2481 – 2490 [6] Jurca F., Inte R., Martis C., Optimal Rotor Design of Novel Outer Rotor Reluctance Synchronous Machine, Electrical Engineering, 102 (2020), 107–116 [7] Chai W., Zhao W., Kwon B., Optimal Design of Wound Field Synchronous Reluctance Machines to Improve Torque by Increasing the Saliency Ratio, IEEE Transactions on Magnetics, 53 (2017), No. 11 [8] V. Hrabovcová V, Makyš P, Rafajdus P, Šebest M., Improved Barriers Rotor of the Reluctance Synchronous Motor, Electrical Engineering, 99 (2017), 1325–1335 [9] Alves de Souza S., Suemitsu W., Five-Phase PermanentMagnet Synchronous Motor, IEEE Latin America Transactions, 15 (2017), No. 4, 639 – 645 [10] Umoh G., Ogbuka C., Obe E., Modelling and Analysis of FivePhase Permanent Magnet Synchronous Motor in Machine Variables, Przegląd Elektrotechniczny, 96 (2020), No.1, 87-92 [11] Umoh G., Obe E., Five-Phase Synchronous Reluctance Motor: a Better Alternative to The Three-Phase Synchronous Reluctance Motor, ICEPENG 2015 International Conference Nsukka, (2015), 56- 61 [12] Umoh G., Obe E., Analysis of Five-Phase Synchronous Reluctance Motor under Loaded and Faulted Conditions, ICEPENG 2015 International Conference Nsukka, (2005), 37-43 [13] Mahmoudi A., Kahourzade S., Rahim N., Ping H., Improvement to Performance of Solid-Rotor Ringed Line-Start Axial-Flux Permanent-Magnet Motor, Progress in Electromagnetics Research, 124 (2012), 383-404 [14] Torkaman H., Afjei E., Gorgani A., Faraji N., Karim H., Arbab N., External Rotor SRM with High Torque per Volume: Design, Analysis, and Experiments, Electrical Engineering, 95 (2013), 393–401 [15] Duan Y., Ionel I., A Review of Recent Developments in Electrical Machine Design Optimization Methods with a Permanent-Magnet Synchronous Motor Benchmark Study, IEEE Transactions on Industry Applications, 49 (2013), No.3,1268 – 1275 [16] Tahi S., Ibtiouen R., Bounekhla M., Design Optimization of Two Synchronous Reluctance Machine Structure with Maximized Torque and Power Factor, Progress in Electromagnetics Research, 35 (2011), 369-387 [17] Krüttgen C., Steentjes S., Glehn G., Hameyer K., Parametric Homogenized Model for Inclusion of Eddy Currents and Hysteresis in 2-D Finite-Element Simulation of Electrical Machines, IEEE Transactions on Magnetics. 53 (2017), No. 6 [18] Zhao W., Xing F., Wang X., Lipo T., Kwon B., Design and Analysis of a Novel PM-Assisted Synchronous Reluctance Machine with Axially Integrated Magnets by the Finite-Element Method, IEEE Transactions on Magnetics, 53 (2017), No. 6 [19] Tiegna H., Bellara A., Amara Y., Barakat G., Analytical modeling of the open-circuit magnetic field in axial flux permanent-magnet machines with semi-closed slots, IEEE Transactions on Magnetics, 48 (2012), No. 3, 1212-1226 [20] Azizi H., Vahedi A., Sensitivity Analysis and Optimum Design for the Stator of Synchronous Reluctance Machine using the Coupled Finite and Taguchi Method, Turkish Journal of Electrical and Computer Sciences, 23 (2015), 17 – 27 [21] Nekoubin A, Soltani J, Dowlatshahi M (2020) Comparative Analysis of Three-Phase and Five-Phase Permanent-Magnet Motor Based on Finite Element Method, Journal of Electrical Engineering & Technology, 15 (2020), 1705–1712 [22] Perreira L., Scharlau C., Pereira L., Haffner S., Influence of Saturation on the Airgap Induction Waveform of Five-Phase Induction Machines, IEEE Transactions on Energy Conversion, 27 (2012), No. 1, 29-41 [23] Parsa L., Toliyat H., Goodarzi A., Five-phase Interior Permanent-Magnet Motors with Low Torque Pulsation, IEEE Transactions on Industry Applications, 43 (2007), No. 1 40–46 [24] Baek J., Bonthu S., Choi S., Design of Five-Phase Permanent Magnet Assisted Synchronous Reluctance Motor for Low Output Torque Ripple Applications, IET Electric Power Applications, 10 (2016), No. 5, 339 – 346 [25] Lin C., Hwang C., High Performances Design of a Six-Phase Synchronous Reluctance Motor Using Multi-Objective Optimization with Altered Bee Colony Optimization and Taguchi Method, Energies, 11 (2018), 1-14
Authors: Engr. Dr. Gideon Umoh, Department of Electrical Engineering, Maritime Academy of Nigeria, Oron, Akwa-Ibom State, Nigeria, E-mail: umoh.gideon@gmail.com; Engr. Chinedu Obe, Department of Electrical Engineering, University of Nigeria, Nsukka, Enugu State, Nigeria, E-mail: chinedu.obe@unn.edu.ng, Engr. Dr. Cosmas Ogbuka (Corresponding Author), Department of Electrical Engineering, University of Nigeria, Nsukka, Enugu State, Nigeria, E-mail: cosmas.ogbuka@unn.edu.ng; George Ekpo, Email: georgeekpo@gmail.com, Department of Electrical and Electronic Engineering, Akwa Ibom State Polytechnic, Ikot Osurua, Nigeria; Engr. Prof. Emeka Obe, Department of Electrical Engineering, University of Nigeria, Nsukka, Enugu State, Nigeria, E-mail: simon.obe@unn.edu.ng
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 1/2021. doi:10.15199/48.2021.01.05
Published by Ľubomír BEŇA1, Paweł KUT2, Rzeszow University of Technology, Faculty of Electrical and Computer Engineering (1) Rzeszow University of Technology, Faculty of Civil and Environmental Engineering and Architecture (2)
Abstract. The large number of wind farms in the power system makes it possible to use them in the process of voltage regulation in the nodes to which they were connected. The regulation possibilities depend on the generators in which the wind farm has been equipped. Currently, Doubly-Fed Induction Generators are the most commonly used ones, which have wide possibilities of reactive power and voltage control at the wind farm connection point. The article presents an analysis of the connection of a wind farm consisting of wind turbines equipped with DFIG generators to the power system for the possibility of voltage regulation. Simulations were carried out using PowerWorld Simulation software.
Streszczenie. Duża liczba farm wiatrowych w systemie elektroenergetycznym stwarza możliwość wykorzystania ich w procesie regulacji napięcia w węzłach do których zostały przyłączone. Możliwości regulacyjne zależą od generatorów w jakie zostały wyposażone elektrownie wiatrowe. Obecnie najczęściej znajdują zastosowanie generatory asynchroniczne dwustronnie zasilane, które posiadają szerokie możliwości regulacji mocy biernej, a co za tym idzie napięcia w punkcie przyłączenia farmy wiatrowej. W artykule przedstawiono analizę przyłączenia farmy wiatrowej, składającej się z elektrowni wiatrowych wyposażonych w generatory asynchroniczne dwustronnie zasilane do systemu elektroenergetycznego pod kątem możliwości regulacji napięcia. Symulacje zostały przeprowadzone z wykorzystaniem oprogramowania PowerWorld Simulator. Farmy wiatrowe w procesie regulacji napicia w systemie energetycznym
Keywords: wind farm, voltage, reactive power, power system Słowa kluczowe: farma wiatrowa, napięcie, moc bierna, system elektroenergetyczny
Introduction
Renewable energy sources are now considered to be the most prospective energy sector. Solar energy and its derivatives are a free and inexhaustible source of energy [1]. Thanks to the rapid development of technologies in the field of renewable energy, the efficiency of generating units increase year by year, while the investment cost decrease. Increasing the share of renewable energy sources in the energy systems of the Member States is now a priority in the European Union energy policy. Funds earmarked for this purpose are to accelerate the development of renewable energy sources and enable diversification of fuels and gradual independence from conventional fuels. The current climate package assumes an increase in renewable energy sources share in the European Union to 20% by 2020. In 2016, the share of renewable energy in final energy consumption was 17% for the European Union [2]. Another target will be to increase energy production in renewable energy sources to 27% in 2030 [3].
According to the data of the Polish Energy Regulatory Office [4], in the Polish national power system, the total installed capacity in renewable energy sources at 30.05.2018 amounted to 8 584,552 MW, and the power in installations using wind energy was 5 874,778, which is 68,43% of the total installed power in renewable energy. One can notice the slowdown in the development of wind energy in Poland as a result of the Wind Farm Investment Act [5,6]. In 2016, the installed capacity in wind sources was 5 807,416 MW, so within 2 years the power increased by only 67,362 MW.
The high installed capacity in wind farms makes it possible to use them for the process of voltage regulation in the power system nodes. Thanks to the use of wind farms with large reactive power control options, the transmission system operator can use wind farm to maintain the required voltage level at the connection point [7]. The use of wind farms in the reactive power control process also allows limiting voltage fluctuations resulting from the stochastic nature of wind and reducing power losses in the internal network of the wind farm and in the network to which it is connected. The connection of wind farms to the power system does not increase the voltage distortion at the connection point [8].
Voltage in the power system
To ensure correct operation of the power system, it is necessary to balance the active and reactive power. One of the basic parameters affecting the quality of electricity is frequency and voltage. Maintaining a constant frequency requires balancing of active power, while a proper balancing of reactive power is associated with maintaining the correct voltage in the system nodes.
The demand of receiving nodes for reactive power is around 42%. The remaining 58% are own needs of the network, which include: longitudinal losses in lines – 21%, longitudinal losses in transformers – 20%, generators demand – 9% and losses in network transformers – 8% [9].
Constant changes in the system load make it necessary to adjust the voltage levels in the power system nodes. Voltage deviations below the rated value are caused by [10]:
• voltage drops in medium and low voltage lines and in transformers;
• too low voltage on the medium voltage side in station 110/MV, resulting from fault conditions. Voltage deviations above the rated value are caused by:
• positive value of longitudinal voltage loss induced by capacitive reactive power flows;
• too high voltage on medium voltage substation bus and MV/LV transformers in abnormal operating conditions.
The value of voltage drop in overhead lines and transformers depends primarily on the part of the longitudinal voltage loss, which is dependent on the reactive component of the current:
.
where: Ib – reactive component of the current.
The longitudinal part of the voltage loss depends on the active current component is much smaller:
.
where: Ic– active component of the current.
This is due to the fact that the value of the overhead line reactance is much higher compared to the resistance. The longitudinal voltage loss increases the value of the voltage at the end of the line when capacitive and less at inductive. For this reason, voltage regulation is closely related to the reactive power regulation.
Wind turbines equipped with an Doubly-Fed Induction Generator
Wind turbines equipped with DFIG (Doubly-Fed Induction Generator) generators have a wide range of active and reactive power regulation, which is why they currently belong to the most commonly used generators in wind energy sector. DFIG generators allow to obtain better quality of electricity compared to other generators used in wind farms. They provide active suppression of voltage and power oscillations as well as current and voltage harmonics.
Doubly-Fed Induction Generators are equipped with an energy electronic converter connected to the rotor circuit, thanks to which it is possible to transmit energy in both directions: from and to the rotor [11]. DFIG generators enable operation at both super-synchronous and sub-synchronous speeds. In the case of super-synchronous operation, the energy flows from the rotor to the grid, while during the work with the sub-synchronous speed, the energy flows from the stator to the rotor.
Figure 1 shows the characteristics of the Vestas V90 – 3 MW wind turbine equipped with Doubly-Fed Induction Generator, on which the area of permissible operating conditions is marked.
Fig.1. Area of permissible operating conditions of the DFIG generator at Vestas V90 – 3 MW wind turbine [12]
Wind turbines Vestas V90 – 3 MW enable operation in the constant power factor mode in the range of 0,98cap – 0,96ind. It is possible to work with a different power factor, but with a reduction in the value of active power generated. When generator is connected in a triangle, the maximum reactive power generated is 1500 kvar, while for star connection 750 kvar.
Analysis of the possibility of using a wind farm in voltage regulation
Simulations were carried out using the PowerWorld Simulator program. PowerWorld Simulator enables analysis of active and reactive power distribution as well as voltage level analysis in power system nodes.
The analysis of the wind farm’s impact on the voltage level in the power network was carried out for a wind farm consisting of 10 wind turbines Vestas V90 – 3 MW (Fig.2, EW1-EW10). The 30 MW wind farm was connected to the 110 kV network.
Figure 2 shows the diagram of the internal network. The wind farm has a three radial lines connected to the main supply point located on the wind farm. In the case of a radial structure, damage to the cable stops the transmission of energy from wind turbines located behind the damaged part of the internal network. The ring topology is characterized by greater reliability, but requires higher investment costs.
Fig.2. Diagram of the internal network of the analyzed wind farm
The diagram of the analyzed fragment of the power system, modelled in the PowerWorld Simulator is shown in figure 3. The wind farm was connected at node 13.
Fig.3. Diagram of the analyzed fragment of the power system
Table 1 presents the values of active and reactive power of a wind farm for which simulations have been carried out. Reactive power was determined based on the simulation of the internal network of the wind farm.
Table 1. The values of active and reactive power in simulations
.
Figure 4 shows the results of the simulation depending on the power factor of wind farms.
Fig.4. Relation between the voltage in the nodes and generated active power by wind farm for: cosφ = 1(a), cosφ = 0,98cap (b), cosφ = 0,96ind (c)
The regulations possibilities of a wind farm consisting of wind turbines equipped with Doubly-Fed Induction Generators, create the possibility of improving voltage conditions in the node to which the wind farm is connected and in neighbouring nodes. In case the voltage at the connection point is lower than required, the wind farm may became a source of reactive power, which will increase the voltage in the node. In the analyzed case, when wind turbines operate with a power factor of cosφ = 0,98cap, the voltage at the connection point (node 13), compared to work with the power factor cosφ = 1, increased by 0,9% for a wind farm working with 30 MW. In the case of wind turbines operating with a power factor cosφ = 0,96ind, it is possible to obtain a constant voltage in the nodes, despite increasing the active power generation. Figure 5 shows relation between voltage and reactive power.
Fig.5. Relation between voltage and reactive power for: P = 30 MW (a), P = 1,76 MW (b)
Figure 6 shows the power losses in the analyzed fragment of the power system depending on the active power for the analyzed values of the power factor cosφ.
Fig.6. Relation between active power losses in the network and generated active power
In the case of wind turbines with cosφ = 0,98cap, the losses of active power in the network to which the farm is connected are reduced as compared to the work with the power factor cosφ = 1.
Table 2 shows the calculated coefficient kU,P, which is a measure of how the voltage in the network nodes changes depending on the active power generated. The higher value of this coefficient means the greater difference between the voltage in the node depending on the generated active power. The coefficient kU,P is defined by the formula:
.
where: ∆U – voltage difference, ∆P – difference between maximum and minimum active power generated by wind farm.
.
Table 2. Values of the coefficient kU,P
.
As can be seen from the above results, increasing the reactive power output increases the voltage at all nodes of the network.
Summary
Wind farms, thanks to their regulation capabilities, can be used by the transmission system operator to regulate the voltage in the node to which it was connected and in neighbouring nodes.
The ability to work with reactive power capacitive and inductive by wind turbines with Doubly-Fed Induction Generators allows to maintain a constant voltage at the connection point despite increasing the active power generation and increasing or decreasing the voltage in the node.
The connection of wind farms close to the recipients also positively influences the level of active power losses in the network, which decreases when the wind farm work with higher active power.
REFERENCES
[1] Proszak-Miąsik D., Bukowska M., Nowak K., Rabczak S., Astronomical and meteorological conditions of a solar system operation, Iop Conf Ser-Mat Sci., 245, (2017) [2] Statistical data on energy from renewable sources, Eurostat [3] http://www.cire.pl/item,96778,1,0,0,0,0,0,ke-do-2030-r-wzrostefektywnoscienergetycznej-o-30-proc.html, Dostęp. 01.12.2018 [4] http://www.ure.gov.pl/pl/rynki-energii/energiaelektryczna/odnawialne-zrodla-ener/potencjal-krajowyoze/5753,Moc-zainstalowana -MW.html, dostęp 01.12.2018 [5] Ustawa z dnia 20 maja 2016 r. o inwestycjach w zakresie elektrowni wiatrowych [6] Ustawa z dnia 7 czerwca 2018 roku. o zmianie ustawy o odnawialnych źródłach energii oraz niektórych innych ustaw [7] Pijarski P., Kacejko P., Wancerz M., Gryniewicz-Jaworska M., Układ sterowania mocą bierną farmy wiatrowej wykorzystujący możliwości regulacyjne przekształtników, dławika zaczepowego oraz pojemność kabla zasilającego farmę, Przegląd Elektrotechniczny, 92 (2016), nr. 8, 44-47 [8] Gała M., Praca turbin wiatrowych w systemie elektroenergetycznym oraz ich wpływ na jakość energii elektrycznej, Przegląd Elektrotechniczny, 93 (2017), nr.6, 37-40 [9] Kot A., Bilans I zapotrzebowanie mocy biernej w Krajowym Systemie Elektroenergetycznym, Acta Energetica, 1 (2013), nr.14, 68-71 [10] Praca zbiorowa, Poradnik inżyniera elektryka, WNT, 2011 [11] Klucznik J, Udział farm wiatrowych w regulacji napięcia w sieci dystrybucyjnej, Acta Energetica, 1 (2010), nr. 3, 39-39 [12] Grządzielski I., Sposoby kompensacji mocy biernej, prezentacja Międzynarodowe Targi Energetyki Expopower, Poznań 2010.
Authors: dr hab. inż. Ľubomír Beňa, prof. PRz, Rzeszow University of Technology, Faculty of Electrical and Computer Engineering, Department of Power Electronics and Power Engineering, ul. Wincentego Pola 2, 35-959 Rzeszów, E-mail: lbena@prz.edu.pl; mgr inż. Paweł Kut, Rzeszow University of Technology, Faculty of Civil and Environmental Engineering and Architecture, Department of Heat Engineering and Air Conditioning, Al. Powstańców Warszawy 6, 35-959 Rzeszów, E-mail: p.kut@prz.edu.pl.
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 95 NR 8/2019. doi:10.15199/48.2019.08.33
Published by Joanna KOZIEŁ, Grzegorz KOMARZYNIEC, Andrzej WAC-WŁODARCZYK, Ryszard GOLEMAN, Department of Electrical Engineering and Electrotechnologies, Lublin University of Technology
Abstract. The article presents the characteristics of the net power of a wind farm as a function of wind speed, and lists the factors determining the selection of a wind farm. Classification of large wind turbine generators is illustrated. The work areas of a synchronous, 3-phase generator with permanent magnets are depicted. Responses of a given wind farm to the demand for capacitive and inductive reactive power are presented.
Streszczenie. W artykule przedstawiono charakterystykę mocy netto farmy wiatrowej w funkcji prędkości wiatru, oraz wymieniono czynniki decydujące o wyborze farmy wiatrowej. Zilustrowano klasyfikację generatorów dużych turbin wiatrowych. Zobrazowano obszary pracy generator synchronicznego, 3-fazowego z magnesami trwałymi. Przedstawiono odpowiedzi określonej farmy wiatrowej na zapotrzebowanie na moc bierną o charakterze pojemnościowym, jak i indukcyjnym. (Analiza wpływu farmy wiatrowej na jakość energii elektrycznej w sieci dystrybucyjnej).
Keywords: wind farm, farm location, permanent magnet 3-phase synchronous generator, reactive power. Słowa kluczowe: farma wiatrowa, lokalizacja farmy, generator synchroniczny 3 fazowy z magnesami trwałymi, moc bierna.
Introduction
A specific feature of electricity is that, in terms of certain parameters, its quality is more influenced by the user than the producer or system operator. In these cases, the system user is the basic partner of the system operator in the efforts to maintain the proper quality of electricity.
It should be noted that these problems are dealt with in other standards already published or in development. The emission standards define the levels of electromagnetic disturbances which can be caused by the devices of the system users. Immunity standards set levels of disturbance that devices can tolerate without undesirable damage or loss of function. The third set of standards for electromagnetic compatibility [1],[2] levels allows the emission standards and immunity standards to be coordinated and coherent in such a way as to achieve the overarching goal of electromagnetic compatibility [3].
Fig.1. Characteristics of the net power of a wind farm as a function of wind speed
Modern wind turbines usually have a horizontal axis of rotation, and the turbine itself usually has three blades. The vast majority of turbines are equipped with asynchronous generators [4].
The wind pressure on the blades creates a pressure difference in front of and behind the blades. The turbine begins to rotate and the rotor drives a generator located in the nacelle. It is in the generator that the mechanical energy is converted into electricity. The rotational speed of a typical asynchronous generator oscillates around 1500 rpm, which requires the use of a gear located between the rotor and the generator [4].
A wind farm – also known as a wind power plant, is a group of wind turbines equipped with generators, generating electricity, powered by wind power. The energy obtained in this way is referred to as “clean” energy, because during its production the turbine does not emit pollutants related to fuel combustion.
The actual net power recorded at the transformer terminals, depending on the wind speed, is shown in Fig. 1.
Factors determining the location of the wind farm
Choosing the right location of the facility is crucial for the success of a given investment, therefore it is necessary to thoroughly analyse all factors [5]. In the case of wind farms one should:
• ensure the access of the wind farm to the National Power System, • keep the required distances between the turbines and arrange them in such a way as to make the most efficient use of production capacity, • exclude protected areas, areas of national parks and nature reserves, • exclude areas adjacent to airports, railways, expressways, • keep appropriate distances from human clusters, • keep the statutory distances from transmission lines, e.g. high voltage and oil and gas pipelines, • ensure compliance with applicable legal acts, including environmental guidelines.
However, the most important determinant of the location is the average wind speed for a given area during the year. It depends on the wind speed whether the wind farm will be able to produce enough energy to make the investment profitable. And so, on the basis of data collected at over 60 stations throughout the country, average wind speeds were determined for given areas [5], [6].
It should be noted that winter is the windiest period, therefore the most desirable and generating the largest amounts of energy produced by wind farms.
Wind speed is a decisive factor in the ability to produce and regulate energy in the distribution system. For an exemplary Vestas V112 turbine is use in this farm. The limiting wind speed at which the Vestas V112 turbine starts producing electricity is 3 m/s. According to the characteristics, it is the equivalent of the production of 22 kW of active power. The critical speed can also be read from the characteristics of the turbine. For this wind turbine model it is 25 m/s. Above this wind speed, the turbine will stop. As it is easy to read from the diagram, the turbine already reaches the rated power of 3.3 MW at a speed of about 13 m/s. In the event of winds blowing at speeds below 3 m/s, as recommended by both the wind farm operators and the turbine manufacturer, the turbine should be stopped. Restarting it is often problematic, e.g. in winter, when ice blocks can form on the blades. In order to prevent this, turbines are usually not stopped in such cases, but instead take power from the power grid and drive the rotor (they work by drawing power from the National Power System).
Table 1. Summary of wind speed for various time periods
.
Fig.2. A photo of the panel visible to the wind farm worker
Fig.3. Area of achievable states of a synchronous generator operating in a power plant, connected directly to the grid (blue, red and black lines) and wind farms (brown line)
On a day when low wind speed prevails, the wind farm draws power from the grid to drive the wind turbines. The appearance of the wind farm SCADA system panel is shown in Figure 2. Despite the use of automation, it is a simpler method of control, because the start-up of wind turbines is associated with a certain risk, resulting from a power outage for the electronics in the nacelle. As can be seen, the WTG04 turbine is stopped due to some error. The farm automation has the ability to constantly interfere with the operation of wind turbines (load change, regulation change, setting angle in relation to the wind direction, starting or stopping the turbine operation). Probably due to an error or in time to repair the operation of the turbine was stopped. The remaining turbines draw their active power from the grid. For example, the WTG07 turbine draws 69.5 kW of power, and the wind speed measured at the top of its nacelle is 2.3 m/s. These are instantaneous values, refreshed by the system at regular intervals.
Each of the turbines operates at wind speeds below 3 m/s. For this reason, turbines need power from the grid to operate. It seems that only the WTG04 turbine does not draw power from the grid, but it is the turbine whose operation in the wind farm system has been suspended.
Regulation of voltage and reactive power of a wind farm in the power grid
The participation of wind turbines in the process of voltage regulation in the power grid is strictly dependent on their ability to generate reactive power. It results directly from the areas of turbine operating states. Modern wind farms are adapted not only to the production (generation) of reactive power, but also to its consumption from the grid. The common ones include wind farms with characteristics similar to a triangle, or more often to a polygon. The reactive power generation capabilities of such farms are large and reach up to 50% of the rated active power. Unfortunately, for small values of wind farm power, there is a close dependence of the generated active power on reactive power (Fig. 3) [7], [8], [9]. Nevertheless, the reactive power regulation function is very important from the point of view of the voltage stability of the system. In order to take a closer look at the above issue, the division of generators used in large wind turbines should be considered. Classification of large wind turbine generators : Large wind turbine generators,
• Induction (asynchronous) generators, • Synchronous generators, • Double-fed induction generators, • Cage induction generators, • Generators with wound rotors, • Generators with permanent magnets, • Generators with slip ring-powered rotors, • Generators with brushless rotors, • Generators with embedded magnets, • Generators with surface-mounted magnets.
The generators most often used in wind turbines include a synchronous three-phase generator with permanent magnets. The areas of its work are presented in Figure 3.
The characteristics of a classic generator block, defined at the level of the generator busbars, are marked in red. The blue colour is the characteristic of the upper voltage of the block transformer. The black lines, in turn, limit the minimum and maximum value of the active power generated by the steam turbine.
This is due to the generation limitations of this turbine and boiler. Wind turbines are an excellent source of energy, as shown in Figure 3. The working area is wide. Wind farms can operate with a higher reactive power consumption while simultaneously having small active power generations. Unfortunately, for low values of the generated active power, this range is very limited and so, most of the time during the year, there is uncertainty as to the possibility of obtaining reactive power at all [10],[11]. Wind farm responses to the demand for capacitive and inductive reactive power are presented in Figure 4.
Fig.4. The regulation system working in the reactive power regulation mode
Fig.5. Characteristics of Q = f (t) of a wind farm
Conclusions
Wind farms have a significant impact on the regulation of voltage parameters in the power grid. The appropriate location of the wind farm has a decisive influence on the regulatory possibilities of the wind farm. The potential of wind farms has been noticed and they are undoubtedly becoming an important element supporting the regulation of the National Power System. Regulatory capacity of wind farms is not applicable to less developed energy systems, e.g. island systems. The wind farms can regulate the frequency and active power, voltage and reactive power, resulting in a constant power factor cosφ. Wind farms are able to react quickly to grid disruptions thanks to their power reserve.
REFERENCES
[1] Michałowska J., Mazurek P.A., Gad R., Chudy A., Kozieł J., Identification of the electromagnetic field strength in public spaces and during travel, 2019 Applications of Electromagnetics on Modern Engineering and Medicine PTZE 2019, pp.121-124, 8781737. [2] Mazurek P.A., Michałowska J., Kozieł J., Gad R., Wdowiak A., The intensity of the electromagnetic field in the coverage of GSM 900, GSM 1800, DECT, UMTS, WLAN in build – up areas, 2018 Applications of Electromagnetics on Modern Engineering and Medicine PTZE 2018, pp. 159-162, 8503156. [3] Norma EN 50160:2010 [4] Wolańczyk F,. Elektrownie wiatrowe, Wydawnictwo KaBe, Krosno 2013 [5] Dygulska A., Perlańska E., Mapa wietrzności Polski, project Czysta Energia, Słupsk, 2015 [6] Montusiewicz J., Gryniewicz-Jaworska M., Pijarski P., Looking for the optimal location for wind farms, Advances in Science and Technology Research Journal, vol. 9, s. 135-142, 2015 [7] Kacejko P., Pijarski P., Gałązka K., Prosument – krajobraz po bitwie, Rynek Energii, vol. 117, nr 2, s. 40-44, 2015 [8] Pijarski P., Wydra M., Kacejko P., Optimal control of wind power generation, Advances in Science and Technology Research Journal, vol. 12, nr 1, s. 9-18, 2018 [9] Zmarzły D., Badania jakości energii w wybranej farmie wiatrowej, Politechnika Opolska, Opole 2014 [10] Kacejko P., Pijarski P., Gałązka K., Prosument – przyjaciel, wróg czy tylko hobbysta, Rynek Energii, vol. 114, nr 5, s. 83- 89, 2014 [11] Pijarski P., Rzepecki A., Wydra M., Efektywne zarządzanie mocą farm wiatrowych, Rynek Energii, vol. 111, nr 2, s. 69-74, 2014
Authors: dr inż. Joanna Kozieł, Department of Electrical Engineering and Electrotechnologies, Faculty of Electrical Engineering and Computer Science, Lublin University of Technology, ul. Nadbystrzycka 38A, 20-618 Lublin, e-mail: j.koziel@pollub.pl; dr hab. inż. Grzegorz Komarzyniec, prof. uczelni, Department of Electrical Engineering and Electrotechnologies, Faculty of Electrical Engineering and Computer Science, Lublin University of Technology, ul. Nadbystrzycka 38A, 20-618 Lublin, e-mail: g.komarzyniec@pollub.pl; prof. dr hab. inż. Andrzej WacWłodarczyk, Department of Electrical Engineering and Electrotechnologies, Faculty of Electrical Engineering and Computer Science, Lublin University of Technology, ul. Nadbystrzycka 38A, 20-618 Lublin, e-mail: a.wacwlodarczyk@pollub.pl; dr hab. inż. Ryszard Goleman, prof. uczelni, Department of Electrical Engineering and Electrotechnologies, Faculty of Electrical Engineering and Computer Science, Lublin University of Technology, ul. Nadbystrzycka 38A, 20-618 Lublin, e-mail: r.goleman@pollub.pl.
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 12/2020. doi:10.15199/48.2020.12.43
Published by Justyna HERLENDER, Jan IŻYKOWSKI, Eugeniusz ROSOŁOWSKI, Wroclaw University of Science and Technology, Department of Electrical Power Engineering
Abstract. This paper deals with analysis of impedance-differential protection applied to locating faults on power transmission line. Based on the voltage and current measurements at both line ends, the differential impedance is calculated. It enables to formulate efficient protective algorithm. Moreover, the presented impedance-differential protection has ability to determine the fault location for an inspection-repair purpose. The fault signals from ATP-EMTP simulations of faults on the sample transmission line was applied for evaluating the fault location accuracy and to compare with the other fault location methods.
Streszczenie. Artykuł prezentuje analizę impedancyjnego zabezpieczenia różnicowego w zastosowaniu do lokalizacji zwarć w linii przesyłowej. Stosując pomiary napięć i prądów na obu końcach linii wyznaczana jest impedancja różnicowa. Pozwala ona na sformułowanie efektywnego algorytmu zabezpieczeniowego. Ponadto takie zabezpieczenie pozwala na lokalizowanie zwarć do celów inspekcyjno-remontowych. Sygnały zwarciowe z symulacji zwarć w przykładowej linii przesyłowej z użyciem programu ATP-EMTP zastosowano do oceny dokładności lokalizacji i porównania z innymi metodami lokalizacji (Impedancyjne zabezpieczenie różnicowe jako lokalizator zwarć w linii przesyłowej).
Keywords: impedance-differential protection, transmission line, fault location, fault simulation. Słowa kluczowe: impedancyjne zabezpieczenie różnicowe, linia przesyłowa, lokalizacja zwarć, symulacja zwarć.
Introduction
Published fault statistics [1]-[2] unambiguously indicate that majority of total number of power system faults occur on overhead power lines. Such faults have to be detected and then located by protective relays as well as by fault locators [2]. In order to prevent spreading out the fault effects, the identified fault has to be cleared by a circuit breaker tripped by a protective relay as quickly as possible. Improvement of protective relays operation is of concern in many researches performed all over the world. Application of synchronized measurements [3]-[5] appears as one of the means for that purpose. In particular, such measurements allow to get modern differential protection systems of overhead power transmission lines [6]-[8].
This paper deals with impedance-differential relay providing effective protection of transmission lines [7]. The traditional current differential relays [6] apply measurements of three-phase currents at the line ends, while the impedance-differential relay under consideration [7] utilizes the measurements of both currents and voltages from the line ends. Thus, more information on the fault is provided in the case of impedance-differential relay. As a result, effective protection of transmission line is achieved [7]. Moreover, a distance to fault can be determined [9]-[12] which can be utilized for an inspection-repair purpose, i.e., for sending the repair crew to remove the fault and thus allowing the line to be switched on into operation. This paper is analyzing the fault location feature of the impedance-differential relay. In particular, a comprehensive evaluation of fault location accuracy with use of the simulation data is presented.
Impedance-differential protection – formulation for single phase system
Figure 1 presents a simplified single phase model of the transmission line utilized to present the impedance differential protection principle [7]. After deriving the algorithm for such a case then it will be extended to three-phase system.
The line is represented with the lumped impedance (ZL) and the line shunt admittances uniformly distributed:
.
where CLis the line shunt capacitance.
It is assumed that the fault (F) is on the line S-R, at the relative distance d [p.u.], counted from the bus S.
Fig.1. Single phase model of faulted transmission line
The following expressions can be written for the circuit of Figure 1:
.
The coefficients at the measured voltage Vs, VR in (2) are different and depend on the distance to fault d. However replacing them by their average value results in:
.
The differential impedance and the compensated differential impedance are introduced as follows:
.
Now the locational differential impedance is defined as:
.
Taking into account (4)-(6) one can obtain that the locational differential impedance is the following function of the sought distance to fault:
.
Therefore the fault location can be performed using
.
At the right-hand side of (8) the real part is taken to reject some imaginary part which can appear due to the calculation errors.
Impedance-differential protection – formulation for three phase system
For the purpose of paper conciseness only calculations of asymmetrical faults are demonstrated here, while the symmetrical faults consideration are presents in [7].
Differential impedance regarding asymmetrical faults can be determined using the symmetrical components. For calculation, in this paper, single-phase-to-earth fault is applied. Figure 2 represents the positive-, negative-, and zero-sequence network for a single phase-to-earth fault.
Fig.2. Interconnection of equivalent networks for positive-, negative- and zero-sequence components under L1-E fault
From Figure 2 following relationships can be observed:
.
Considering that the fault occurs in phase L1, and after implementation of symmetrical component properties, it can be obtained from (9):
.
In view of zero sequence circuit presented in Figure 2, the voltage drop can be formulated as:
.
After substituting (11) to (10), the following formula is obtained:
.
The formula (12) is analogous to (3) which was obtained for a single phase system. Therefore, taking this analogy, one can extend usage of the set of equations (4)-(8) to the single phase fault (L1–E) in three-phase system by taking:
.
Analogous substitutions one has to apply for the remaining single phase faults (L2–E, L3–E). Moreover, this can be applied for phase1-phase2 and phase1-phase2-earth faults as well.
Data of simulated system
For evaluation of the presented protection algorithm, the model of the 400 kV double fed transmission line (Tab. 1) has been tested. The simulation was performed using the ATPDraw [13], while protection algorithm was implemented in MATLAB software. The phasors of measured currents and voltages were determined by the full-cycle Fourier filtering.
Table 1. Parameters of the modeled transmission line
.
Fault resistance influence
Length of the investigated line varied, and was equal to the following values: 80 km, 200 km and 300 km. However, for the sake of briefness, only results for 300 km line are shown in this paper. In order to test the proposed protection algorithm, short-circuit simulations were conducted inside the line as well as beyond it. The inner faults were simulated, referring to the S side at distances of d = 0; 0.1; 0.2; 0.3;…1. The faults applied outside the protected line, were located behind the terminals S and R, respectively. The studies included symmetrical faults (three-phase-to-earth faults) and different asymmetrical faults (phase-to-earth (L1-E), phase-to-phase (L1-L2), and phase-to-phase-to-earth (L1-L2-E) faults).
The presented fault location algorithm was compared with the two-end synchronized fault location algorithms (14) presented in [2]
.
where i – kind of processed signals.
The error of studied impedance-differential algorithm was defined as
.
The presented results in Table 2 and 3 concern phase-to-earth (L1-E) faults inside the protected line, regarding to the fault resistance. The results included phase-to-phase faults are presented in Table 4. The fault location errors were determined as follows:
• errorDIFF – use of signals of impedance-differential protection • errorDIFF0 – as for errorDIFF but without shunt capacitances compensation • error1 – use of positive sequence component • error2 – use of negative sequence component • errorL1-E – use of signals applied in distance protection for L1-E fault • errorL1-L2 – use of signals applied in distance protection for L1-L2 fault.
It is visible that fault location computation concerning compensation were more accurate than neglecting it. Maximal error obtained by the presented algorithm (with compensation) did not exceed 0.8% while without compensation this value was insignificantly higher than 1%. In contrast, the result calculated in case of positive sequence component based location algorithm was even greater than 4%.
Additionally, for phase-to-phase faults, the average error computed for considered protection method was equaled to 0.0904% (with compensation) and it means that from all used methods this calculated faults location the most accurately. On the contrary, the average error calculated for negative sequence based algorithm was the highest in case of the faults simulated for fault resistance amount to 2 Ω.
Generally, the differential impedance algorithm enabled to locate faults with maximal average error equaled to 0.5086%. Only algorithm based on negative sequence components worked more accurate, and the maximal average error did not exceed 0.3418%. However, taken into consideration all average error results, differential impedance protection method was the most precise from all of compared algorithms in except of the simulation made for L1-E fault with RF=10 Ω.
The sample example is presented in Figure 3 – 7. The specifications of it are as follows: phase-to-earth (L1-E) fault at the midpoint of 300 km line, RF=10 Ω. The computed fault location is depicted in Figure 7 where d(41÷60)ms was obtained by averaging within the interval (41÷60) ms after the fault inception.
Fig.3. The example – voltage at terminal R
Fig.4. The example – voltage at terminal S
While simulating short-circuits in the middle of the protected line, it was observed that independently of the applied fault location algorithm, the computed distances were characterized by the smallest error. This situation was observed for all fault types.
As presented in Tables 2 – 4, the considered protection algorithm allows to detect faults in all conditions, regardless of different fault types and fault resistance.
Fig.5. The example – current at terminal R
Fig.6. The example – current at terminal S
Fig.7. The example – computed distance
Conclusions
The aim of this paper is to present the concept of impedance-differential protection for long transmission lines. The demonstrated protection algorithm enables not only for internal fault detection, but can be applied also as a fault locator.
Based on simulation results it can be concluded that presented method can be used for transmission lines with different lengths as well as is not influenced by the fault resistance changes.
Moreover, capacitive charging current which constitutes the main drawback of current differential protection is eliminated in presented protection method and does not influence on the fault location determination.
In addition, the accuracy of fault location achieved in case of impedance-differential relay allows to improve the faults location calculation obtained from the use of two-end synchronized fault location algorithms. The precision of presented algorithm is on the same level as for method using negative sequence components and even has superiority over it, in case of phase-to-phase faults (L1-L2 faults, Tab. 4).
The obtained results approve the high reliability of the impedance-differential protection.
For the further studies of demonstrated protection algorithm as a transmission line fault locator, the impact of source strength or shunt reactors application could be evaluated.
REFERENCES
[1] Kacejko P., Machowski J., Zwarcia w systemach elektroenergetycznych, WNT Warszawa (2002) [2] Saha M.M., Izykowski J., Rosolowski E., Fault Location on Power Networks, Springer, London (2010) [3] Halinka A., Szewczyk M., Talaga M., Metodyka pomiarów synchronicznych (PMU) oraz przykłady zastosowania. Wiadomości Elektrotechniczne, 82 (2014), no 8, 21-25 [4] Iżykowski J., Rosołowski E., Synchroniczne pomiary rozproszone w zastosowaniu do lokalizacji zwarć w liniach napowietrznych, Przegląd Eektrotechniczny, 85 (2009), no. 11, 21-25 [5] Szewczyk M., Time synchronization for synchronous measurements in Electric Power Systems with reference to the IEEE C37.118TM Standard – selected tests and recommendations, Przegląd Elektrotechniczny, 91 (2015), no. 4, 144–148 [6] Altuve Ferrer H.J., Kasztenny B., Fischer N., Line current differential protection, A collection of technical papers representing modern solutions, Schweitzer Engineering Laboratories, (2014) [7] Ghanizade Bolandi T., Seyedi H., Hasemi S.M., Soleiman Nezhad P., Impedance-differential protection: A new approach to transmission-line pilot protection, IEEE Transaction on power delivery, 30 (2015), no. 6, 2510-2518 [8] Suonan J.L., Deng X.Y., Liu K., Transmission line pilot protection principle based on integrated impedance, IET Trans. Distrib. Gen., 5 (2011), no. 10, 1003–1010 [9] Kowalik R., Rasolomampionona D., Glik K., Detection, classification and fault location in HV lines using travelling waves, Przegląd Elektrotechninczy, 88 (2012), no. 1a, 269-275 [10] Wiszniewski A., Dokładna lokalizacja miejsca zwarcia w liniach napowietrznych elektroenergetycznych, Przegląd Elektrotechniczny, 60 (1984), no. 2, 41-44 [11] Smolarczyk A., Szweicer W., Porównanie wybranych metod lokalizacji miejsca zwarcia, Przegląd Elektrotechniczny, 79 (2003), no.2, 59-64 [12] Iżykowski J., Saha M., Rosołowski E., Wykorzystanie prądów wejściowych zabezpieczeniowych przekaźników różnicowych do lokalizacji zwarć, Przegląd Elektrotechniczny, 84 (2008), no.5, 9-13 [13] Dommel H., ElectroMagnetic Transients Program, BPA, Portland, Oregon, (1986)
Authors: MSc. Justyna Herlender, prof. dr. Jan Iżykowski, prof. dr. Eugeniusz Rosołowski, Wroclaw University of Science and Technology, Department of Electrical Power Engineering, 27 Wybrzeże Wyspiańskiego St., 50-370 Wroclaw, Poland; E-mails: justyna.herlender@pwr.edu.pl, jan.izykowski@pwr.edu.pl; eugeniusz.rosolowski@pwr.edu.pl
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 93 NR 11/2017. doi:10.15199/48.2017.11.40
Published by Maciej CIUBA1, Michał WOJCIECHOWSKI1, Maciej OWSIŃSKI2, Michał BORECKI1, Warsaw University of Technology, Koszykowa 75, 00-662 Warszawa, Poland (1) Institute of Power Engineering – Research Institute, Mory 8, 01-330 Warszawa, Poland (2)
Abstract. The presented article discusses the differences in the results of the electric field simulation in a medium voltage heat-shrinkable cable termination with the most probable assembly faults. Two types of voltage excitation were set as the boundary condition for a model of a real object. The first was a typical electrostatic excitation, and the second was the AC voltage with mains frequency. Both were used for cable accessories with selected assembly omissions. Consideration of the effect of the excitation type suggests that for cable accessories, field simulation using only electrostatics leads to unreal results and incorrect inference about the location of zones with the highest electrical stresses.
Streszczenie. Prezentowany artykuł omawia wpływ zadanego wymuszenia na różnice w wynikach symulacji pola elektrycznego w termokurczliwej głowicy kablowej średniego napięcia z najbardziej prawdopodobnymi błędami montażu. Przyjęto dwa rodzaje wymuszenia napięciowego jako warunek brzegowy w modelu bazującym na konstrukcji rzeczywistej głowicy. Pierwszym było typowe wymuszenie elektrostatyczne, a drugiem wymuszenie napięcia przemiennego. Oba zostały zastosowane dla osprzętu kablowego z wybranymi błędami montażu. Uwzględnienie wpływu typu wymuszenia sugeruje, że w przypadku osprzętu kablowego symulacja pola z zastosowaniem wyłącznie elektrostatyki prowadzi do nierzeczywistych wyników oraz błędnego wnioskowania co do lokalizacji stref o największych naprężeniach elektrycznych. Wybór rodzaju wymuszenia a wyniki symulacji pola elektrycznego w głowicy kablowej SN
Keywords: Cable termination, numerical simulation, electric field distribution, assembly fault Słowa kluczowe: głowica kablowa, symulacja numeryczna, rozkład pola elektrycznego, błąd montażu
I. Introduction
The need to use cable accessories causes discontinuity of cable insulation and which in turn causes a nonuniformity of electric field distribution. Damages of cable accessories are one of the main reasons for the failure of cable power systems in general [1]-[3] even with sophisticated types of protecting devices [11, 12]. Possible sources of local increase of electric field strength may include assembly faults during the preparation of cable termination like gaseous cavities, contaminants on the insulation surface, the omission of assembling of the stress control element or just incorrect cable accessory choice. Further parts of this paper describe the results of two possible ways to simulate electric field distribution of heat shrink cable termination which is based on its common use.
II. Compared Computing Methods
Electrostatic Analysis
The dependent variable for this analysis is the electric potential. Considered electrostatic analysis based on two Maxwell equations for linear media
.
where (1) is Gauss’s law with D – electric flux density and ρ – charge density, (2) is Faraday’s law with E – electric field strength
Conductivity does not occur in formulas mentioned above so all materials between conductors were considered as perfect insulators (σ = 0 (S⋅m-1)). Therefore electric field distribution depends only on the permittivity of materials and possible occurrence of uncompensated electric charges.
AC Analysis
This type of analysis also uses electric potential as a dependent variable to estimate electric field strength (3) implemented in the time-harmonic equation of continuity (4).
.
with: V – electric potential, E – electric field strength, J – current density, D – electric flux density, Je– external current density, σ– conductivity, ω – angular frequency and j – imaginary unit. Equations (3) and (4) describe electric field in lossy materials where both permittivity and resistivity affect the distribution of electric field between conductors. Both methods described above were implemented to simulations of two different cases:
• cable termination correctly assembled according to the installation instructions; • the omission of assembly semiconducting mastic; • assembly of semiconducting mastic and stress control tube has been ignored.
III. Model preparation
The created model was based on real heat-shrinkable termination (Fig.1 and Fig.2). The terminated cable is XRUHAKXS 120/50RMC 12/20 kV type of XLPE extruded MV cable [4].
Fig.1. General view of examined cable termination
Fig.2. Part of sectioned modeled termination with visible semiconducting mastic (yellow) and semiconducting tube (black).
The cross-section of a real object allowed measuring the dimensions of each part of the accessory. Dimensions were measured with a caliper of 0.02 mm precision. Materials properties mentioned in [5]-[9] were used in the considered model.
Table I. Values of material properties used in model
.
Different cases of modeled cable termination were computed with the use of the AC/DC Module in the Comsol Multiphysics environment. Figure 3. shows a general view of the axisymmetric model of correctly assembled heatshrinkable MV cable termination and magnified part with a cutting point of cable screening covered by stress control mass.
To fulfill the requirements of the FEM algorithm some boundary conditions should be applied. In both, electrostatic and AC analysis, Dirichlet boundary conditions were chosen, which means selected conductors surfaces obtain electric potentials. The next step for any solved problem is to build the mesh. Discretization (Fig.4.) of examined areas was made by automatic built-in algorithm with manually slightly changed parameters like the ratio between the size of two adjacent grid elements to dense mesh in important areas with expected electric field intensification.
To solve partial differential equations MUMPS (MUltifrontal Massively Parallel Sparse direct Solver) direct solver was used.
Fig.3. Axisymmetric model of heat-shrinkable cable termination with magnified cable screen cutting point area
Fig.4. Discretization of considered cable termination with magnified cable screen cutting point area
IV. Results of simulation
Correctly assembled cable termination
This section aims to investigate the electric field distribution of clean and properly assembled termination and compare results of simulation on the same model but under two different physics mentioned in section II.
Fig.5. Potential distribution as a result of a) electrostatic and b) AC analysis for correctly assembled cable termination
The first results of electric field simulation for both possibilities show (Fig.5) substantial difference in its distribution. Area of discontinuity of uniformity of electric field has moved from cable screen ending closer to cable semiconducting screen layer ends. Graph (Fig.6) present electric field strength for both possible simulation types along the straight line in cable termination at radius r = 11 (mm), where the green line is for electrostatic analysis and the blue one is for AC analysis.
Fig.6. Electric field strength inside cable termination at radius 11(mm) from the axis in properly assembled object
Cable termination without semiconducting mastic
This section consist investigation of electric field distribution of termination assembled with omission of semiconducting mastic and compare results of “electrostatics” and “electric current” simulation on the same model (section II).
Fig.7. Potential distribution as a result of a) electrostatic and b) AC analysis for incorrectly assembled cable termination (without semiconducting mastic).
As in the first case also two quantities distribution was considered: potential distribution (Fig.7) on the surface of the cross-sectional area of termination and part of its surrounding, and electric field strength (Fig.8) longwise line on radius r = 11(mm) from the axis of a 2d-axisymmetric model of cable termination.
Fig.8. Electric field strength inside cable termination without mastic at radius 11(mm) from the axis.
Cable termination without semiconducting mastic and stress control tube
This section aims to investigate the electric field distribution of improperly assembled termination and compare results of simulation on the same model but under two different types of physics (section II).
Moreover, maximal electric field potential value from AC simulation as before is higher than in electrostatic solution.
Once more electric field distribution significantly changes (Fig. 9) only because of use different type of simulation conditions. Both analyzed quantities: potential (Fig.5, 9) and electric field strength (Fig. 6, 10) show that difference.
Fig.9. Potential distribution as a result of a) electrostatic and b) AC analysis for incorrectly assembled cable termination (without semiconducting mastic and stress control tube).
Fig.10. Electric field strength inside cable termination without semiconducting mastic and tube at radius 11mm from the axis.
V. Conclusions
That means the results of modeling could affect the design of a final product and its reliability and also position on the market. As it has been shown in the previous part of the article, incorrect choice of physics type could easily lead to erroneous conclusions. The possibility of an incorrect assembly of cable accessory constrain the designer to change the project and also the production process, therefore, it is very important to carefully build a numerical model with the correct parameters and correctly chosen physics. The best way to obtain it is to set model parameters possibly closest to the designed work conditions of the real object.
In the case of cable termination, it is very risky to simulate it with the electrostatic field. Regardless of modeled mistakes in the assembly process, this study shows that for electrostatics the most electrically stressed area appears always at the end of the deflected cable screen.
The omission of semiconducting termination parts assembly lead to an increase of potential gradients close to cable screen end.
REFERENCES
[1] Gulski, “Knowledge rules for partial discharge diagnosis in service”, Cigre TF 15.11/33.03.02, 2002, pp. 1-88. [2] K.P. Meena, B.N. Rao, T. Thirumurthy, G.K. Raja, “Failure analysis of medium voltage cable accessories during qualification tests”, 10th IEEE Int. Conf. Properties and Applications of Dielectric Materials, Bangalore, India, 2012, pp. 1-4 [3] G. Mazzanti; M. Marzinotto, “Combination of probabilistic electro-thermal life model and discrete enlargement law for power cable accessories”, 2016 IEEE International Conference on Dielectrics (ICD), vol.2, 2016, pp.780-783 [4] NKT Cables, XRUHAKXS 12/20 kV Data Sheet, 2016 [5] Aarnio, Anssi, “Characterization of non-metallic materials for medium voltage, cable accessories”, Master’s thesis, Tampere University of Technology, Material Science, 2010 [6] “SCO Silicone Grease Compound”, Electrolube company catalogue, Revision: 3 April 2016 [7] Illias, H. A.; Lee, Z. H.; Bakar, A. H. A. & Mokhlis, H. 2012. Distribution of electric field in medium voltage cable joint geometry. 2012 IEEE International Conference on Condition Monitoring and Diagnosis 23–27 September 2012, Bali, Indonesia. [8] Olli Kuusisto, “The Effects of Installation-Based Defects in Medium Voltage Cable Joints”, Bachelor thesis, Helsinki Metropolia University of Applied Sciences, Electrical Power Engineering, 2016 [9] I.A. Metwally, A.H. Al-Badi, A.S. Al-Hinai, F. Al Kamali and H. Al-Ghaithi, “Influence of Design Parameters and Defects on Electric Field Distributions inside MV Cable Joints”, IEEE 2016, Proceedings of the 18th Mediterranean Electrotechnical Conference (MELECON) [10] Ciuba M., Owsiński M., Sul P.: Analiza rozkładu pola elektrycznego w termokurczliwej głowicy kablowej z wadą montażu, in: Elektronika – konstrukcje, technologie, zastosowania, vol. 1, no. 11/2016, 2016, pp. 64-66, DOI:10.15199/13.2016.11.13 [11] M. Borecki, J. Starzyński, Z. Krawczyk, “The comparative analysis of selected overvoltage protection measures for medium voltage overhead lines with covered conductors”, Conference on Progress in Applied Electrical Engineering, ISSN 978-1-5386-1528-7, pp. 1-4, 2017, DOI: 10.1109/PAEE.2017.8009014 [12] M. Borecki, S. Wincenciak, “Simulation of Electric Field on the Surface of a Long Flashover Arrester”, 17th International Conference on Computational Problems of Electrical Engineering (CPEE 2016), ISSN 978-1-5090-2800-9, pp. 1-4, 2016, DOI: 10.1109/CPEE.2016.7738763
Authors: Maciej Ciuba1, Michał Wojciechowski1 , Maciej Owsiński2 , Michał Borecki1, Warsaw University of Technology (1), Koszykowa 75, 00-662 Warszawa, Poland e-mail: maciej.ciuba@ee.pw.edu.pl, michal.wojciechowski@ee.pw.edu.pl Institute of Power Engineering – Research Institute (2), Mory 8, 01- 330 Warszawa, Poland e-mail: maciej.owsinski@ien.com.pl
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 96 NR 4/2020. doi:10.15199/48.2020.04.25
Published by Juliya MALOGULKO, Svyatoslav VYSHNEVSKY , Iryna KOTYLKO, Natalia SOBCHUK, Vinnitsa National Technical University
Abstract. The article analyzes the pace of increase in the generation of photovoltaic stations in the context of the combined electricity system of Ukraine and the energy supply company of PJSC “Vinnytsyaoblenergo”. An analysis of existing regulatory documents regulating the work of photovoltaic stations has been carried out. Within the framework of the considered documents, the criteria for assessing the reliability of the operation of electric networks, namely, the length of long breaks in electric power supply of consumers of electric energy SAIDI are defined. The interconnection of the change of reliability indices of electric networks operation with the increase in the number and installed capacity of renewable energy sources, in particular photovoltaic stations (PV), is shown. In order to increase the reliability of power supply, a method of its renewal has been proposed, due to the joint use of generation of small hydroelectric power stations and PV stations.
Streszczenie. W artykule analizowano kombinowany system energetyczny na Ukrainie zawierający ogniwa fotowoltaiczne. Na podstawie dokumentacji analizowano niezawodność systemu biorąc pod uwagę przerswy w dostawie energii. Zaproponowano poprawę przez dołączenie małych elektrowni wodnych. Wpływ rozproszonych źródeł energii na niezawodność sieci elektrycznej
Keywords: dispersed energy sources, photovoltaic stations, local electric systems, small hydroelectric power stations. Słowa kluczowe: rozproszone źródła energii, ogniwa fotowoltaiczne, niezawodność
Introduction
In 2015, our state was one of the first to ratify the Paris Climate Agreements, thus confirming its intentions and commitments to integrate into the EU energy system and carry out energy reforms in the framework of the requirements of the Third Energy Package, which includes, among other things, the creation of favorable conditions for the introduction of new energy generating capacities renewable energy sources (RES).
The National Renewable Energy Action Plan for the period up to 2020 stipulates that the share of renewable energy generation in final energy consumption should reach 11% [1].
Nowadays, Ukraine demonstrates the highest pace of signing agreements for future accession of RES, but it poses great risks to the outdated energy network [2]. The key is that, according to official information of SRFEU, in I quarter. In 2018, 159.4 MW were put into operation, generating capacities – 54 objects of the electric power industry (2.4 times exceeding the capacity put into operation for the same period in 2017). At the same time, the objects of wind electric station (WES) and photovoltaic (PV) make up 92% of the installed capacities, and the average unit capacity of the electric energy objects introduced at that time is 3 MW. The installed capacity of WES and PV in Ukraine as of mid-2018 is 1353 MW (512 and 841 MW respectively), which has almost no effect on the balance of power. Their deviations from the planned generation are offset by maneuvering power.
In 2017, the number of technical specifications and contracts signed with Ukrenergo for joining high-voltage networks of green energy facilities, compared with 2016, increased by more than 30 times the power indicator. This is a crazy pace and this tendency persists.
According to Ukrenergo, contracts for joining by 2025 to networks of green energy installations with a capacity of 7426 MW (WES – 4200 MW, and PV – 3226 MW, excluding large hydroelectric power stations) have already been signed. However, the united energy system (UES) can only accept up to 3,000 MW of solar and wind power plants without the risk of unbalance and serious changes in its structure.
The system operator in his research stresses that PV and WES in terms of stability [3-6] of electricity supply are unreliable. Deviations from scheduled charts over the course of the day amount to more than 450 MW at a set power of 1217 MW. One more specific feature of the installation of renewable energy sources is their uneven distribution throughout Ukraine. Thus, the presence of one powerful source of up to 3 MW or several less powerful ones up to 0.5 MW connected to one substation of the distribution electric network (DEN), makes it possible to consider DEN as a local electrical system (LES). And for the local electrical system, there are not yet clear legislative acts that will require the operation of renewable energy sources.
Particularly acute for distributive electrical networks is the question of reliability and uninterrupted power supply. According to the Resolution of the National Commission for State Regulation in the Fields of Energy and Utilities (SRFEU) dated June 12, 2018 “On Approving the Procedure for Ensuring Quality Standards for Electricity Supply and Providing Compensation to Consumers for Failure”, [7] determined indicative, qualitatively characterizing the level of reliability of work DEN But, taking into account the pace of increasing the generation capacity of RES, in particular PV, it is expedient to carry out an analysis of changes in the determined indicators in terms of its growth.
Aim of the research – The purpose of the article is to assess the influence of photovoltaic generation on the reliability of the operation of electric networks and develop a method for its improvement.
Main materials of the research
In accordance with the IEEE 1366-2012 [8] and the SRFEU Resolution “On Approval of the Targets of Reliability (Continuity) Electricity Supply for 2018” [9], the main indicators of the reliability of operation of electric networks, including those with renewable energy sources, are quantified and qualitative breaks in electricity supply.
Classification of interruptions in electrical supply according to DSTU EN 50160: 2014:
a) scheduled, when the consumer is informed in advance about them;
b) Emergencies caused by long-term or short-term short circuits that are most often the result of external events, equipment failure or third-party interference in its operation Random breaks are classified as:
1) long interruptions (longer than three minutes); 2) short-term interruptions (including up to three minutes).
– System Average Interruption Duration Index
.
Where ri – the time of restoration of electricity, Ni – the number of breaks in the electricity supply of consumers in the reporting period, NT – the total number of consumers in the electrical network.
Increasing the share of generating renewable energy sources, for example, only those with the largest increase in power, namely wind and photovoltaic stations, are analyzed (Fig. 1).
At the beginning of 2015, the total installed capacity of the PV was 315 MW. Over the past four years, their capacity has increased more than 3 times and at the end of 2018 – 1100 MW. It should be noted that the PV are unevenly located on the territory of Ukraine, and in their turn, it is quite difficult to assess their impact on the reliability of electricity supply networks.
Fig.1. Dynamics of power generation of renewable energy sources in the UES of Ukraine. The total installed capacity of wind power stations in UES, 1 – wind electric station, 2 – photovoltaic stations, 3 – total renewable energy sources.
In fig. 2 shows the SAIDI change for 2011, 2015-2018, the average for the UES for urban and rural electric networks.
Fig.2. Change of target indicator SAIDI (red curve) and actual (blue curve) for (a) urban electric networks; (b) rural electricity networks of the UES of Ukraine
Based on statistical data, an increase in the generation capacity of renewable energy sources, the active introduction of which into electricity networks began to increase in 2015, may be the reason for prolonged electric power interruptions (SAIDI) electricity networks. The pace of increase in the generation of RES in the section of each power supply company is analyzed, among others Vinnytsyaoblenergo company has been allocated (Fig. 3), since here, starting from 2015, the growth of power generation capacity of the PV was the largest. Only the generation of PV is analyzed, because the wind potential for this region is insignificant. Consequently, the generation capacity in early 2015 was 41.3 MW and increased almost four times in the next three years – at the end of 2018, the power plant is 180 MW. However, the effect of the PV on the reliability of the networks here is significantly different from the impact on the network of the IEC as a whole (see Figure 4).
Fig.3. The pace of increase in the generation of PV in company ” Vinnytsyaoblenergo” (1) and the UES of Ukraine (2)
Simultaneous improvement of the level of technical equipment of the networks, as observed in Vinnytsyaoblenergo, together with the development of the PV, allows us to reveal their potential in view of the possibility of ensuring compliance with the indicator of the length of long interruptions in electricity supply in urban and rural electric networks (Fig. 4).
Fig.4. Change of target indicator SAIDI (red curve) and actual (blue curve) for a) city electric networks, b) rural electric networks of company “Vinnytsyaoblenergo”
The analysis of dependencies in Figures 2 and 4 allows us to conclude that it is possible to estimate and achieve the maximum effect from the implementation of renewable energy sources in view of the possibility of providing normative indicators of reliability (continuity) of electricity supply, taking into account the technical condition of the electricity network to which they are joining.
Thus, there is a whole range of tasks facing the industry experts to improve these indicators. Taking into account the rapid growth of electricity generation from distributed energy sources, it is expedient to use them jointly, in the context of a partial restoration of power supply to network demands. However, the use of such an approach is impossible without the coordination with relay protection systems and emergency automation of distribution networks 6-10kV. The connection of distributed generation objects to the DEN leads to a change in the main characteristics of the power system, on the basis of which the generally accepted concept of building relay protection (RP) was formed. At the level of the distribution network, multilateral power of the site of damage becomes possible, there are new, previously not typical types of disturbances and accidents, characteristics of electromagnetic and electromechanical transients are changing.
The problem of constructing RP significantly expands and complicates and, in the integrated approach, includes the solution of two groups of tasks:
– connected with the provision of the necessary technical perfection of RH of distribution electrical networks in which such stations are operated;
– associated with the creation of relay protection and automatics (RPA), which is installed at the point joining RES to the electrical network.
One of the ways to reduce the number and duration of failures in electricity supply is the possibility of joint use of different types of RES, namely, photovoltaic and small hydroelectric power stations in the tasks of partial recovery of consumers in the absence of electricity supply from centralized electricity supply networks. Another advantage of the joint use of such RES is to reduce the losses of the PV owners from the lack of electric power in the absence of tension on bus stations. These include the assurance of reliability of electric power supply to consumers maintenance of voltage levels within permissible limits, optimization of power flows in order to reduce losses, as well as maintenance of balance reliability in LES with combined electric supply from local and centralized sources of energy [11]. Determining the priority of solving problems arising in LES, note the balance reliability as the reliability of LES when its calculation model is determined by the balance of consumption and generation of electric energy, with the external supply being taken into account. The successful solution of other problems depends on the methods and means used to ensure the balance reliability. Technical and economic indices of LES depend on the balance of its active and reactive power [12]. The optimization of LES and EPS’s joint operation is considered in a number of research works [13]. Therefore, scheduling of the generation of photovoltaic stations is very important for maintenance of normal operating modes of the power system.
In the new economic conditions, photovoltaic power stations of direct transformation of energy are more and more widely used. Their use, in addition to making a profit from electricity sales, allows, under certain conditions, to unload the electricity networks and improve the quality of electricity [14].
Modeling Of Electrical Supply Restoration In Local Electrical Systems After Loss Of Centralized Power
For the calculation of the regime parameters of the fragment of the scheme of the Yampil DEN, the source information about the Galzhbievska photovoltaic plant was analyzed:
– put into operation – 2013; – Installed power – 1381 kW; – Estimated annual generation – 1515 MWh; – type of Multi-Si modules.
Generation at this power plant takes place using SMA circular inverters, as well as multicrystalline silicon modules. This configuration ensures the optimal operation of the PV. The solar panels generate DC power, then through the inverters it enters 3 KTPs: KTP 0.4/10-630 kVA, 0.4/10- 1000kVA and 0.4 / 10-250 kVA.
The normal scheme of power output by the power plant implies that the electricity generated by the PV through the KTP of 0.4 / 10 – 630 kVA and 0.4 / 10 – 1000 by the line 27- 23 is supplied to the network connected to the substation Yampil 110 / 10kV. The model contains PV, consisting of 380 parallel connected strings of 15 photovoltaic modules per each type of panels – monocrystal (YL235P-29b). At the output of the inverter, a 0.4 / 10 kV transformer and a transformer 10 kV three-phase switch is installed. Another possibility of this model is that the generation of a photovoltaic station changes during simulation time, since the main parameters influencing the PV output, namely the level of solar insolation and the ambient temperature, therefore, the simulation of these parameters is presented in the form of a schedule that changes during the day.
The length of the feeder lines 15 of the 110/10 “Yampil” is 18 km. This feeder contains: 37 knots, 16 transformer substations, Galzhbievskaya PV and Galjbievska sHPP. Synchronous hydrogenerator with a power of 667 kVA, at the output of which is a voltage of 10 kV. The power network is represented in the form of four feeders and a centralized power source in the form of a 110/10 kV Yampil substation.
The total power of the transformer substations from which consumers feed is 2 149 kW. (Fig. 5), were considered.
Fig.5. Fragment of electric network where dispersed generation located
Fig.6. Simulink model of Galjbievska PV station which situated on the 998 bus
Fig. 7. Simulink model of Galjbievska small hydropowerplant which situated on the 999 bus
After analyzing the technical condition of the equipment of the electric networks, it was established that the most damaged element is the power lines. The computer model which showed in Figure 5, we divided for two computer mode (fig.6) – is a model of Galjbievska photovoltaic station and Figure 7 – is a model of small hydro power plant, which can supplying power to the photoelectric station bus. Due to the loss of centralized power supply from a centralized substation, consider two cases of restoration of electricity supply:
1. The combined use of a photoelectric and small hydroelectric power plant to restore consumers electricity supply.
2. Duration of the transition process, which will take place at the voltage supply to the PV bus from sHPP. In the framework of this work, the following modes of work are investigated and modeled:
1. The sinusoidal analysis of the voltage curve in paralleloperated photovoltaic and hydroelectric power plants at the power output of the common bus and in the absence of power supply from the 110 / 10kV Yampil 110 substation, it should be noted that the part of the load did not lose electricity due to the combined generation of the PV plant and sHPP (Fig. 8)
2. The next possible version of power supply for LES users, it is advisable to consider the mode of operation of the electrical grid, in which the running PV plant voltage is applied after the launch of a small hydroelectric power plant (Fig. 9).
Fig.8. Changing the voltage curve as a result of the parallel operation of the PV plant and sHPP with the electric network
Fig.9. Changing the voltage curve as a result of voltage supply to the PV plant bus in sHPP
The results of simulation, on restoring the electricity supply of LES demands, by means of water drainage by a small hydroelectric power plant in order to supply voltage to the bus of a photovoltaic station, shows a slight distortion of the voltage curve.
The main idea of the work, is the implementation of mathematical modeling of the possibility of restoring the power supply of consumers of the electrical network that lost centralized power. Given the dependence of the photovoltaic power station’s operation on the electrical network, it can not function independently without supplying voltage. Proceeding from this, as a voltage source, hydroelectric power stations of low power are considered.
Conclusions
To increase the technical and economic efficiency of joint operation of distributed power sources and distribution electric networks, it is necessary to solve a number of tasks, which will allow to increase electricity generation of RES, reduce electricity losses in distribution electric networks, improve the quality and reliability of electricity supply to consumers.
In order to efficiently exploit distributed energy sources and their integrated use in power grids, especially in the sense of improving the reliability of power supply, it is necessary to develop a method for restoring consumer power supply, with the loss of centralized power supply. In the context of the transition from the wholesale electricity market of a single buyer to balancing and to electricity supply under bilateral agreements, in recent years and in perspective in Ukraine, there is a tendency of transition from purely centralized electricity to combined, when the number of local power sources is increasing. Moreover, the share of the latter in the energy balance of power systems is increasing. Local power sources, operating directly in the grids of 10-6-0.38 kV, include both traditional sources of low power and alternatives.
The results of simulation of the possibility of restoring electricity supply to LES users in the event of a loss of centralized power show high potential for the use of RES in this direction. In particular, RES may be a significant time to maintain the power of consumers while ensuring a standard for the quality of electric energy and the cost- effectiveness of the operation of electricity grids
REFERENCES
[1] Derzhavna tsilʹova ekonomichna prohrama enerhoefektyvnosti i rozvytku sfery vyrobnytstva enerhonosiyiv z vidnovlyuvanykh dzherel enerhiyi ta alʹternatyvnykh vydiv palyva na 2010-2020 roky// Zatverdzhena postanovoyu Kabinetu Ministriv Ukrayiny vid 01.03.2010 r. № 243 [Elektronnyy resurs] // Rezhym dostupu: http://zakon2.rada.gov.ua/laws/show/243-2010-%D0%BF [2] Zvit pro rezulʹtaty diyalʹnosti u 2017 rotsi [Elektronnyy resurs] // Zatverdzheno postanovoyu SRFEU vid 23 bereznya 2018 roku № 360. Rezhym dostupu do resursu – URL: http://www.nerc.gov.ua/data/filearch/Catalog3/Richnyi_zvit_SRFEU_2017.pdf [3] Lezhniuk P. D. Otsinyuvannya vplyvu na yakistʹ funktsionuvannya lokalʹnoyi elektrychnoyi systemy vidnovlyuvanykh dzherel elektroenerhiyi / P. D. Lezhniuk, V. O. Komar, D. S. Sobchuk // Visnyk Kharkivsʹkoho natsionalʹnoho tekhnichnoho universytetu silʹsʹkoho hospodarstva imeni Petra Vasylenka. Problemy enerhozabespechennya ta enerhozberezhennya APK Ukrayiny. – Kharkiv : KHNTUS·H, 2013.Vypusk 141. S. 8–10. [4] Kulyk V. V. Optymalʹne keruvannya rozoseredzhenymy dzherelamy elektroenerhiyi z asynkhronnymy heneratoramy zasobamy Smart Grid [Elektronnyy resurs] / V. V. Kulyk, T. YE. Mahas, YU. V. Malohulko // Naukovi pratsi VNTU. – 2011. – №4.– S. 1–6. – Rezhym dostupu do resursu : http://praci.vntu.edu.ua/article/view/1404/999. [5] Kyrylenko O. V. Power Engineering of Sustainable Development: Challenges and Construction Paths / O. V. Kirilenko, AV Prachevnik // Proceedings of the Institute of Electrodynamics of the National Academy of Sciences of Ukraine. Special issue. 2010. 10-16. [6] Kyrylenko O. V. Tekhnichni aspekty vprovadzhennya dzherel rozpodilenoyi heneratsiyi v elektrychnykh merezhakh / O. V. Kyrylenko, V. V. Pavlovsʹkyy, L. M. Lukʺyanenko // Tekhnichna elektrodynamika. 2011. № 1. S. 46–53. [7] Pro zatverdzhennya Poryadku zabezpechennya standartiv yakosti elektropostachannya ta nadannya kompensatsiy spozhyvacham za yikh nedotrymannya// Zatverdzhena postanovoyu Kabinetu Ministriv Ukrayiny vid 12.06.2018 r. № 375 [Elektronnyy resurs] // Rezhym dostupu: http://zakon.rada.gov.ua/laws/show/v0375874-18 [8] IEEE Guide for Electric Power Distribution Reliability Indices,” in IEEE Std 1366-2012 (Revision of IEEE Std 1366-2003) , vol., no., pp.1-43, 31 May 2012 doi: 10.1109/IEEESTD.2012.6209381 [9] Pro zatverdzhennya tsilʹovykh pokaznykiv nadiynosti (bezperervnosti) elektropostachannya na 2018 rik // Zatverdzhena postanovoyu Kabinetu Ministriv Ukrayiny vid 14.06.2018 r. № 392 [Elektronnyy resurs] // Rezhym dostupu: http://www.nerc.gov.ua/index.php?id=32667 [10] Lezhnyuk P.D. Vozobnovlyayemyye istochniki energii v raspredelitel’nykh elektricheskikh setyakh: Monografiya / P.D. Lezhnyuk., A.A. Koval’chuk, A.V. Nikitorovich, V.V. Kulik. – Vinnitsa, 2014. – 204 s. [11] Q. Jiang, G. Geng, C. Guo, and Y. Cao, “An efficient implementation of automatic differentiation in interior point optimal power flow,” IEEE Trans. Power Syst., vol. 25, no. 1, pp. 147–155, Feb. 2010. [12] Kulyk V.V. Operational control of the flow of reactive power in the distribution of electrical dispersed generation // Proceedings of the Institute of Electrodynamics National Academy of Sciences. Collected Works. Special-start you. – 2013. – P. 151-158. [13] Seyed Mohammad Hossein Nabavi, Somayeh Hajforoosh, Mohammad A.S. Masoum, “Placement and Sizing of Distributed Generation Units for Congestion Management and Improvement of Voltage Profile using Particle Swarm Optimization”, IEEE Transactions on Power Systems, 2011. [14] C. L. T. Borges and D. M. Falcao, “Optimal distributed generation allocation for reliability, losses, and voltage improvement”, Int. J. Elect Power Energy Syst., vol. 28, no. 6, pp. 413–420, Jul. 2006.
Published by Jan NEDOMA, Marcel FAJKUS, Radek MARTINEK, VSB – Technical university of Ostrava
Abstract. This article deals with the measurement of electric current in the energy via optical fibers. Nowadays, the measurement of the electrical current by using optical fiber most commonly based on the principle of Faraday effects, thus the magneto-optic effect. FOCS (Fiber-Optic Current Sensor) is very accurate, modular and easy to install. Another advantage is the isolation of the measuring part from the primary technology, which is sensed. Optical fibers can also be used to measure the inside of the transformer. It also offers the possibility of measuring the temperature of winding. The main contribution of the paper is to summarize interesting published results to date, approaches and basic principles leading to the analysis and to defining the electrical values such as electrical current using fiber optic technology.
Streszczenie. W artykule opisano możliwości pomiaru prądu przy wykorzystaniu światłowodów. Czujnik FOCS (fiber optic current sensor) jest dokładny I łatwy do instalacji. Inną zaletą jest oddzielenie galwaniczne od toru prądowego. W artkule przedstawiono przegląd prac na ten temat, przegląd rozwiązań konstrukcyjnych I zastosowań oraz analizę właściwości metody. Pomiar pradu elektrycznego przy wykorzystaniu światłowodów.
Keywords: fiber-optic technology, electrical current, current sensor, magneto-optic effect. Słowa kluczowe: światłowody, pomiar prądu, zjawisko Faradaya.
Introduction
The fiber-optic sensor is based on the use of Faraday’s magneto-optical effect (year of discovery is 1845), in which there is a magnetic rotation of plain of the polarized light, see Figure 1. However, the use of this phenomenon for the measurement of electrical quantities has only occurred thanks to the development of fiber optic technology. The rate of twisting of the polarization plane is directly proportional to the path d, after which the light in the given environment spreads and size of the components of the vector of magnetic induction B in the direction of light propagation. The orientation of the vector determines the meaning of the polarization plane. The size of the angle of rotation β, by which the polarization plane is rotated (see Fig. 1) can easily be calculated by a simple relationship (1):
(1) β = V * B * d,
where: V – Verdet constant, B – density of magnetic induction, d – length of path The Verdet constant depends on the wavelength of the light and it is an optical “constant” that describes the strength of the Faraday-effect for a particular material.
Fig.1. Basic scheme of the Faraday-effect.
Fiber-optic current sensors are referred to as FOCS (Fiber-Optic Current Sensor). Figure 2 shows the principle of the fiber-optic current sensor FOCS. FOCS benefits are high accuracy, high bandwidth: detection of current ripple and transients, wide temperature range, full digital processing, uni- or bi-directional current measurement, analogue and digital outputs, easy to install, adaptable shape of sensing head, small size and weight, no magnetic centering necessary, no magnetic overload problem, immunity to electromagnetic interference and many others.
Fig.2. Principle of the fiber-optic current sensor FOCS
An important application area of FOCS is the metallurgical industry, where an electrolytic process is used to obtain precious metals. Typical electrolyzers work with DC (Direct Current) in the order of several hundred thousand amperes (up to 500 kA). Fiber-optic sensors can be used as measuring transducers and provide a number of benefits, such as easier installation, smaller dimensions and high accuracy of measurement. Nowadays, the target application area for the use of the fiber optic FOCS sensors is to measure, control and protect the substations. Also, the advantage is that due to small dimensions and weight, FOCS can be integrated into an existing device, such as switches and bushings. The application is also in the DC lines of very high voltage (HVDC) used for the transmission of electricity over long distances. From the point of view of the security function of the protection there is a significant advantage that even in the case of high short-circuit currents, there is no overloading and distortion of the output signal, which we encounter with conventional current transformers. Another advantage is the isolation of the measuring part from the primary technology, which is sensed [1-6].
The main contribution of the paper is to summarize interesting published results to date, approaches and basic principles leading to the analysis and to defining the electrical values such as electrical current using fiber optic technology.
Approaches and results
An advanced fiber-optic current sensor is based on the recirculating architecture of fiber loop for significantly enhancing the current sensitivity. The recirculating loop is created by a 2×2 optical switch and the single mode fiber is used as the sensing head, see Fig. 3. Authors experimentally obtained a sensitivity of 11.5 degrees/A for1000 meters fiber and a sensitivity of 21.2 degrees/A for 500-meters fiber [7].
Fig.3. Basic scheme of Sagnac fiber optic current sensor
A smaller fiber sensing coil is required in order to that the FOCS meet the needs of the high voltage watt-hour meter for measuring smaller current with the better performance. The FOCS sensing coils (<10 cm in diameter) is achieved by the special spun highly birefringent fiber. The ability to resist bending is analyzed in the comparison between the fiber sensing coils by the special spun highly birefringent fiber and the fiber sensing coils by the low birefringent fiber. The measurement error of the FOCS is ± 0.2 % from -40 °C to 70 °C under considering temperature compensation, and the FOCS shows the excellent performance of temperature characteristics, linearity and accuracy in a current range from 1 A to 120 A [8].
Next article studies the using an FOCS for the measurement of plasma current in the fusion reactor (ITER). The sensor is based on a classical FOCS interrogator involving the measurement of the state of polarization rotation when the light in presence of a magnetic field (Faraday effect) in an optical fiber surrounding the current and terminated by a Faraday mirror. The objective of the simulations is to quantify the ratio (beat length over the spun period of the spun fiber) enabling a measurement error in agreement with the ITER specifications. The simulation work showed that a L-B/S-P ratio is 19.2 [9].
The paper [10] researched on the error ratio of FOCS induced by temperature drift. The reason of FOCS error drift was described to explain the relationship among temperature, linear birefringence and error of FOCS ratio. In the range of testing temperature, both the ratio error of FOCS and linear birefringence had a linear relationship with temperature. The FOCS ratio error had a linear relationship with experiment temperature similar to 60 °C.
Complex transforming processes of polarization state induce illegal linear polarization state and illegal circular polarization state in Sagnac fiber-optic current sensor, which, decrease the performance of S-FOCS. Based on polarization state error models of Sagnac fiber-optic current sensor, authors made experiments to evaluate performances of several key components, including polarizer, quarter-wave retarder and sensing head, then, investigate the influence of several main polarization error factors on S-FOCS’s performance. The result shows that the changing degree of polarization state causes bias instability, nonlinear of scale ratio, and random noises of SFOCS, and influents S-FOCS’s performance [11].
Paper [12] describes the structure and principle of the FOCS created by Faraday mirror. The performance of the FOCSs is limited by the linear birefringence (LB). Faraday mirror can be employed to compensate the LB using the non-reciprocity of Faraday effect and the reciprocity of LB. The results indicate if that the influence of LB disappears then the current is null. However, if the current is not zero, the LB is not removed and the effect extent is different for different LB. The Faraday rotation can be deduced from the detected signals and the LB need not be compensated physically by employing this technique.
A special spun linear birefringent fiber was designed and created for FOCS based on polarization-rotated reflection interferometers. In contrast with conventional sensing fibers used in FOCS, the fiber uses a function of a quarter wave plate. The output of sensor has a good linearity performance in a wide range of current (10 to 5000 A/rms). The ratio error is ± 0.1 % and the phase error is ± 2 min at AC current of 1000 A [13].
Article [14] investigate the design principle exploiting the geometric rotation effect for the sensing coil of the fiberoptic current sensor (FOCS) on the basis of the polarization-rotated reflection interferometer. The sensing coil is formed by winding the low birefringence single-mode optical fiber in a toroidal spiral. If the rated current is 1200 A/rms, the sensing coil can ensure the scale factor error of the sensor of 0.2 S over a temperature range from -40 °C to 60 °C.
A prototype fiber-optic current sensor (FOCS) created by Sagnac interferometer is designed and tested for monitoring current up to 4000 A. Sensor is tested for nominal current 1 A up to 800 A. The output of sensor has nonlinearity of ± 0.5 % [15].
A highly accurate fiber-optic current sensor for AC and DC up to 100 kA was investigated. Reciprocal optical circuit and method of signal processing based on the relationship of modulation and demodulation of this sensor were analyzed. The sensor achieved accuracy to within ± 0.2 % at -40 °C to 80 °C with inherent temperature compensation, resolving power for small AC current was less than 0.5 A, angle difference was less than ± 2 min [16].
Authors [17] propose a new technique to reduce the bending-induced linear birefringence (LB) by fiber polarization rotator (FPR) in the reflective fiber-optic current sensor (FOCS), see Fig. 4. A comparison was made between the proposed FPR scheme and the conventional scheme on measurement errors. Simulation shows that the proposed scheme has a large improvement on such LB reduction by an order of magnitude.
Fig.4. Basic scheme of FOCS without FPR.
A new method for measurements of lightning on wind turbines by fiber-optic current sensors (FOCS) was developed. FOCS are resistive with respect to electromagnetic interference (EMI), because the magnetic field produced by the current of lightning is directly converted into an optical signal in this device. The sensor cannot be damaged by over-current coming from an unexpected surge caused by a lightning. However, the accuracy of current measurements with FOCS is affected by the environmental perturbations, such as mechanical vibration and temperature changes [18].
A fiber-optic current sensor employs phase shifting algorithms to process the optical signal is described. Here, the sensing element consists of a coil low birefringence fiber which placed between one polarizer and four analyzers. In the polarimeter, the output light from the sensing element of current is divided into 4 beams through 3 non-polarizing beam splitters, and an analyzer and a detector are placed in each beam path. The design of the sensor, results and shift algorithms will be presented [19].
Authors presented fiber-optic current sensor for DC up to 500 kA. The sensor offers significant advantages with regard to performance and ease of installation compared to state-of-the-art Hall-effect-based current transducers. The sensor exploits the Faraday effect in an optical fiber and measures the path integral of the magnetic field along a closed loop around the current-carrying bus bars. The differential magneto-optic phase shift of left and right circular light waves propagating in the fiber is detected by a polarization-rotated reflection interferometer. The sensor achieves accuracy ± 0.1 % over a wide range of currents and temperatures [20].
Next paper presents a fiber-optic current sensor (FOCS), customized for measurement of harmonic current in high-voltage electric power systems. The practical application of this device has been verified experimentally at a thermal power plant. The measurements have been made for the operation of a 6kV induction motor with different load conditions and the thyristor excitation of a 15 kV AC generator [21].
A fiber-optic current sensor (FOCS) created by the Faraday effect is presented. The sensor of current is realized using the all-fiber low-coherent reflection interferometer. An Erbium-doped fiber, a super-fluorescent radiation source, and a sensing spun high-birefringence fiber coil are applied in the interferometer. The sensitivity of the sensor is about 70 mA/root Hz, the scale factor error is about 0.5 %, the range of measured currents is 0.1 A~3000 A, and the bandwidth is up to 10 kHz [22].
Authors [23] present FOCS based on Faraday effects with the magnetic concentrator. According to the measured values of AC up to 1 kA, a calibration procedure was performed. A well-known temperature dependence of the Faraday current sensor and its influence on the measurement accuracy are tested using a special doublelayer thermal insulated chamber.
A fiber-optic current sensor (FOCS) based on a Sagnac interferometer is presented to measure high-voltage AC current in electric power systems from 5 A to 3200 A, see Fig. 5. A simple analytical expression for the sensor has been derived, the input-output performance and the temperature dependence of FOCS have been experimentally investigated. The simple geometry of sensor gives high accuracy and sensitivity, wide dynamic range, and immunity to slow-variance temperature and other environmental fluctuations [24].
Fig.5. Basic scheme of FOCS using Differentiating Sagnac Interferometer.
Next, it is discussed a new configuration of FOCS which increases the linearity range of the device without decreasing its sensitivity. In this way, the whole system behaves as a ”null-detector”. However, the effect of bending-induced linear birefringence is taken into account in the design and optimization of the sensor. It is shown that the response of the experimental apparatus is temperature independent and that its bandwidth gets values higher than 1 MHz [25].
Fiber-optic current sensor based on the Faraday effect of magneto-optic materials, in the isotropic optical transparent medium, and the magnetic field can make the plane polarized light propagating polarization plane rotated along the magnetic field. By mathematical optimization method, paper [26] discusses the end fitting, the average points fitting, least square method, and the optimal linear least absolute deviation method which is applicated to the linear characteristics of the fiber-optic current sensor. The results show that different linear fitting methods have different results. Using the least square method and best linear fitting method (regression line) obtained FOCS’s linearity fitting degree is ± 0.069 % and ± 0.071 % respectively.
The error drift caused temperature shift reduces the accuracy of FOCT. Using two models, authors quantitatively researched the error ratio drift, and obtained temperature characteristics of FOCT: within the tested range of temperature (10~60 °C), both the FOCT ratio error and linear birefringence phase difference changes linearly with temperature [27].
The residual linear birefringent of sensing fiber, temperature and vibration sensitivity severely influence the accuracy of Sagnac fiber-optic current sensor. A sensing fiber can be used in FOCS with spun high-birefringent fiber. This S Hi-Bi fiber includes three sections: two terminal sections with variable spin-rate along fiber were utilized to substitute the fiber quarter-wave plates, respectively converting the light polarization state from the linear one to the circle one, and vice versa; and the middle section with a uniform spin-rate was utilized as the current sensing fiber which maintains the circular state polarization and compresses the residual linear birefringent during the light propagation [28].
The fiber-optic current sensors utilize the effects of magnetic-field imposed on the change of polarization state of light in the fibers. This sensor has a lot of advantages over conventional current sensors, see Fig. 6. To eliminate the vibration sensitivity, an improved light path of the reflective fiber-optic current sensor is proposed, which makes the discharge of Sagnac effect with Sagnac effect itself and do not disturb Faraday effects [29].
Fig.6. Basic scheme of a novel fiber-optic current sensor.
To improve the performance, a fiber-optic current sensor was presented and using Jones matrix, its polarization error was studied. By establishing the Jones matrix expression of the main optical devices, an interference expression of this sensor was introduced and the influence of imperfection of optical devices on the measurement accuracy was analyzed [30].
High-precision methods and devices are one of the most important modern engineering development lines in optimal energy consumption sphere. However, there are a few small-sized instruments which allow conducting precision measurements without breaking the circuit. Paper [31] presents information about conducted Faraday Effect research and general construction of the simple fiber-optic current sensor (FOCS).
In 2005, ABB introduced a high-performance fiber-optic current sensor for the measurement of DC up to 600 kA. Recently, ABB has developed the sensor further with a view to implementing it in high-voltage substations. It can be integrated into primary high-voltage equipment such as circuit breakers. By appropriately selecting the number of fiber loops, the measurement range can be optimized for specific applications [32].
According to the polarization coupling model of the polarization-maintaining delay fiber coil, considering the reciprocal parasitic waves with large amplitude, the interference intensity of the optical system in the fiber-optic current sensor (FOCS) is calculated, and the theoretical relationship between the scale factor and the polarization crosstalk of the delay fiber coil is obtained. The effect mechanism of the temperature dependence of the polarization crosstalk on the scale factor is revealed, and the corresponding suppression methods are proposed. The experimental results show that the temperature dependence of the FOCS can be improved greatly when lowering the tension of winding fiber, reducing the quantity of the glue, and utilizing the frame with low-temperature coefficient. The variation of the scale factor is decreased from 0.63 % to 0.07 % over the temperature range from -40 °C to 70 °C [33].
In Sagnac interferometer fiber-optic current sensor, errors from the residual linear birefringence and environmental temperature and vibration sensitivity severely influence the accuracy of this sensor. The spun highly linearly birefringent fiber was designed and the vibration insensitive Sagnac interferometer fiber-optic current sensor scheme was set up. In this scheme, the compensation fiber coil was designed to compensate the error caused by Sagnac effect in sensing fiber coil [34].
The fiber-optic current sensor was presented. According to a characteristic of interference signal, square wave modulation technique was applied to enhance the sensitivity of FOCS, and correlative demodulation scheme was proposed to show phase difference information. The sensor achieved accuracy within ± 0.25 % at -40~60 °C and the bandwidth exceeded 6 kHz [35].
ABB developed a new configuration of air-insulated switchgear (AIS) substations by using new circuit breaker (CB) technologies and optical sensors. The new disconnecting circuit breaker (DCB) ‘Combined’ placed inside the breaking chamber without any other components. The fiber-optic current sensors based on the Faraday effect is able to determine the current with a fiber-optic loop integrated with the conductors. FOCS also provide an interface with process-level devices for a substation automation system (SAS). The use of new DCB and optical sensors can improve the performance, efficiency, and reduce the footprint of a substation [36].
The nonlinearities of the response of an interferometric fiber-optic current sensor, which is associated with deviations of the light waves from perfect circular polarization, are investigated. The consequences of inherent temperature compensation of the Faraday effect using a non-90°-retarder are investigated for currents up to several 100 kA and temperatures between – 40 °C and 80 °C. The results are of particular interest to sensors for high DC in the electro-winning industry where measurement accuracy of ± 0.1 % is required up to 500 kA [37].
Fiber-optic current sensors (FOCS) based on Faraday magneto-optical effect have plenty of advantages in comparison to the traditional current sensors. However, the residual linear birefringence and environmental vibration sensitive problem are the fatal drawback in Sagnac fiberoptic current sensors. A vibration immunity sensing loop with a function of passive fiber-optic polarization control is proposed, which consist of a spun or twisted fiber with highly birefringence. A spun or twisted fiber possesses two crucial functions: Remove the residual linear birefringence and control polarization of light. It has been demonstrated that this FOCS has high sensitivity, considerable wide dynamic range, resistance to electromagnetic interference, and immunity of vibration [38].
Fig.7. Basic scheme of principle of S-FOCS.
The intrinsic birefringence of the sensing fiber is the main cause which affects the precision of measurement, and elimination of this cause is the key research topic. In paper [39] an interferometric fiber-optic current sensor configuration with a conjugate reflector is presented. The results show that the configuration with a conjugate reflector is more insensitive to linear birefringence than the configuration with an ordinary reflector. The residual linear birefringence of the sensing coil is about 6 times than that in ordinary mirror configuration for ± 0.2 % measuring accuracy.
Firms require highly accurate DC current sensors to control their processes and operations. FOCS has suitable performance and functionality and is smaller and lighter. Using the FOCS accuracy increased by 10 times, specified accuracy is maintained over a wide temperature range, and large bandwidth to enable rapid response to current ripple and transients. FOCS is also responsible for handling uni- and bi-directional DC to ± 500 kA with negligible power consumption [40].
ABB developed a sensor that represents a quantum leap in high DC measurements. This sensor offers outstanding precision and is smaller, lighter and much less complex than traditional transducers and it is about to change the future of high DC measurements. The new fiber-optic current sensor (FOCS) for high DC makes use of the Faraday effects [41].
The problem of polarimetric sensors is not discussed in this paper, but currently polarimetric sensors are also commonly used. Information on polarimetric sensors is available in articles published in the Electrical Review or in article [42].
Conclusion
This article is to summarize the interesting practical and theoretical published results to date, approaches and basic principles based on Faraday effect leading to the analysis and to defining the electrical values such as electrical current using the fiber-optic technology. Fiber-optic current sensor benefits are obvious and very useful for current applications The high accuracy, high bandwidth: detection of current ripple and transients, wide temperature range, no magnetic centering necessary, no magnetic overload problem, immunity to electromagnetic interference and many others.
Acknowledgements – This article was supported by the Technology Agency of the Czech Republic TA04021263 Project and by the Ministry of Education of the Czech Republic (Projects Nos. SP2017/128 and SP2017/79). This research was partially supported by the Ministry of Education, Youth and Sports of the Czech Republic through Grant Project no. CZ. 1.07/2.3.00/20.0217 within the framework of the Operation Programme Education for Competitiveness financed by the European Structural Funds and from the state budget of the Czech Republic. The Ministry of the Interior of the Czech Republic (Projects Nos. VI20152020008 and VI2VS/444) also provided support for this article. The Ministry of Industry and Trade of the Czech Republic (Project No. FV 10396) also provided support for this article.
References
[1] P. Siddons, N. C. Bell, Y. Cai, C. S. Adams, and I. G. Hughes, “A gigahertz-bandwidth atomic probe based on the slow-light Faraday effect,” Nat. Photonics, vol. 3, no. 4, 2009. [2] L. Weller, K. S. Kleinbach, M. A. Zentile, S. Knappe, I. G. Hughes, and C. S. Adams, “Optical isolator using an atomic vapor in the hyperfine paschen-back regime,” Opt. Lett., vol. 37, no. 16, 2012. [3] E. Prati, “Propagation in gyroelectromagnetic guiding systems,” J. Electromagn. Waves Appl., vol. 17, no. 8, 2003. [4] M. L. Kales, “Modes in wave guides containing ferrites,” J. Appl. Phys., vol. 24, no. 5, 1953. [5] S. Mancuso and S. R. Spangler, “Faraday rotation and models for the plasma structure of the solar corona,” Astrophys. J., vol. 539, no. 1 PART 1, 2000. [6] W. F. Magie, “Faraday’s Diary,” Science (80-. )., vol. 77, no. 1995, 1933. [7] J. Du, Y. Tao, Y. Liu, L. Ma, W. Zhang, and Z. He, “Highly sensitive and reconfigurable fiber optic current sensor by optical recirculating in a fiber loop,” Opt. Express, vol. 24, no. 16, 2016. [8] F. Yang, X. Hou, Y. Huang, and N. Peng, “Optical fiber current sensor with small size for the high voltage watt-hour meter,” in Proceedings of SPIE – The International Society for Optical Engineering, 2016, vol. 10158. [9] M. Wuilpart, B. Vanus, A. Andrasan, A. Gusarov, P. Moreau, and P. Mégret, “Study of a fibre optics current sensor for the measurement of plasma current in ITER,” in Proceedings of SPIE – The International Society for Optical Engineering, 2016, vol. 9916. [10] S. Cheng, Z. Guo, G. Zhang, W. Yu, Y. Shen, and S. Jin, “The analysis on temperature characteristic of fiber optic current sensor ratio error,” in Proceedings – 5th International Conference on Instrumentation and Measurement, Computer, Communication, and Control, IMCCC 2015, 2016. [11] L. Wang, M. Cao, Q. Liu, G. Wei, B. Li, and C. Lin, “Modeling and experimental verification of polarization errors in Sagnac fiber optic current sensor,” Optik (Stuttg)., vol. 126, no. 20, 2015. [12] R. Zhang, X. S. Yao, T. Liu, and L. Li, “The effect of linear birefringence on fiber optic current sensor based on Faraday mirror,” in Proceedings of SPIE – The International Society for Optical Engineering, 2014, vol. 9274. [13] P. Nai, S. Wang, and W. Tao, “A special spun birefringent fiber optic current sensor,” in Proceedings – 2013 12th International Conference on Optical Communications and Networks, ICOCN 2013, 2013. [14] C. Zhang, C. Li, X. Wang, L. Li, J. Yu, and X. Feng, “Design principle for sensing coil of fiber-optic current sensor based on geometric rotation effect,” Appl. Opt., vol. 51, no. 18, 2012. [15] A. Prakash et al., “Development of fiber optic current sensor for electric power industry application,” in 2012 International Conference on Fiber Optics and Photonics, PHOTONICS 2012, 2012. [16] C. Zhang, C. Wang, and S. Chen, “Theory and experimental research of a Y-phase-modulator based optical fiber current sensor,” in Proceedings of SPIE – The International Society for Optical Engineering, 2011, vol. 8199. [17] S. Zhou and X. Zhang, “Simulation of Linear Birefringence Reduction in Fiber-Optical Current Sensor,” IEEE Photonics Technol. Lett., vol. 19, no. 19, 2007. [18] S. G. M. Krämer, F. P. León, Y. N. M. Hernández, and B. Lewke, “Integration of a distributed fiber optic current sensor setup for lightning detection in wind turbines,” in Conference Record – IEEE Instrumentation and Measurement Technology Conference, 2007. [19] J. L. Flores, G. García-Torales, and V. H. Ortiz, “Optical current sensor using phase shift algorithms,” in Proceedings of SPIE – The International Society for Optical Engineering, 2007, vol. 6770. [20] K. Bohnert, H. Brändle, M. G. Brunzel, P. Gabus, and P. Guggenbach, “Highly accurate fiber-optic dc current sensor for the electrowinning industry,” IEEE Trans. Ind. Appl., vol. 43, no. 1, 2007. [21] S. J. Petricevic, Z. Stojkovic, and J. B. Radunovic, “Practical application of fiber-optic current sensor in power system harmonic measurement,” IEEE Trans. Instrum. Meas., vol. 55, no. 3, 2006. [22] V. P. Gubin et al., “All-fiber optical sensor of electrical current with a SPUN fiber sensing element,” in Proceedings of SPIE – The International Society for Optical Engineering, 2006, vol. 6251. [23] P. Mihailovic, S. Petricevic, Z. Stojkovic, and J. B. Radunovic, “Development of a portable fiber-optic current sensor for power systems monitoring,” IEEE Trans. Instrum. Meas., vol. 53, no. 1, 2004. [24] T. Wang, C. Luo, and S. Zheng, “A fiber-optic current sensor based on a differentiating Sagnac interferometer,” IEEE Trans. Instrum. Meas., vol. 50, no. 3, 2001. [25] A. Irace, G. Breglio, A. Luciano, L. Zeni, and A. Cutolo, “A novel configuration for a wide-range high-power fiber-optic current sensor,” Eur. Trans. Electr. Power, vol. 7, no. 5, 1997. [26] Z. Yue, W. Xiaofei, and M. Hao, “Study on the characteristics of all fibre optical current sensor linear fitting algorithm,” in 2015 IEEE 12th International Conference on Electronic Measurement and Instruments, ICEMI 2015, 2016, vol. 3. [27] S. Cheng, Z. Guo, G. Zhang, Y. Shen, P. Song, and H. Huang, “Temperature characteristic of fiber optic current sensor,” Gaodianya Jishu/High Volt. Eng., vol. 41, no. 11, 2015. [28] Y. Wang, M. Kang, L. Ren, and K. Ren, “Design of spun highbirefringent fiber for fiber optic current sensor,” Hongwai yu Jiguang Gongcheng/Infrared Laser Eng., vol. 44, no. 1, 2015. [29] S. Tao, F. Yue, L. Yipu, Z. Mengxue, D. Hailong, and W. Xinlao, “Simulation of polarization errors for all-fiber optic current sensors,” in Proceedings – 8th International Conference on Future Generation Communication and Networking, FGCN 2014, 2014. [31] P. Liu, K. Chen, S. Chen, and Z. Liu, “Study on polarization errors of re-entrant fiber-optic current sensor,” Bandaoti Guangdian/Semiconductor Optoelectron., vol. 35, no. 2, 2014. [31] M. G. Grigoriev and N. V. Turushev, Development of fiber optic current sensor, vol. 1040. 2014. [32] K. Bohnert, R. Thomas, and M. Mendik, “Light measures current: A fiber-optic current sensor integrated into a highvoltage circuit breaker | Strom messen mit Licht: Faseroptischer Stromsensor in einem HochspannungsLeistungsschalter,” ABB Rev., no. 1, 2014. [33] C. Li, C. Zhang, H. Sun, and H. Cui, “Effect of polarization crosstalk of polarization-maintaining delay optical fiber coil on the fiber-optic current sensor,” Zhongguo Jiguang/Chinese J. Lasers, vol. 41, no. 11, 2014. [34] Y. Wang, M. Kang, L. Ren, K. Ren, J. Xu, and N. Shi, “A novel compensation fiber coil for eliminating Sagnac Effect in fiber optic current sensor,” in Chinese Control Conference, CCC, 2013. [35] C.-Y. Zhang, C.-X. Zhang, X.-X. Wang, Z.-F. Ma, and Q.-Q. Liu, “Signal processing system for digital closed-loop fiber optic current sensor,” Zhongguo Dianji Gongcheng Xuebao/Proceedings Chinese Soc. Electr. Eng., vol. 29, no. 30, 2009. [36] L. Jing, H.-E. Olovsson, J. Fan, and R. Thomas, “Small footprint high performance,” ABB Rev., no. SPEC. REP., 2008. [37] K. Bohnert, P. Gabus, J. Nehring, S. Wiesendanger, A. Frank, and H. Brändle, “Nonlinearities in the high-current response of interferometric fiberoptic current sensors,” in Proceedings of SPIE – The International Society for Optical Engineering, 2008, vol. 7004. [38] M. Kang, Y. Wang, J. Xu, K. Ren, and S. Liu, “Vibration immunity fiber optic current sensor employing a spun or twisted highly linear birefringence fiber,” in Chinese Control Conference, CCC, 2013. [39] B. Jiao, Z. Wang, F. Liu, and W. Bi, “Interferometric fiber-optic current sensor with phase conjugate reflector,” in Proceedings of IEEE ICIA 2006 – 2006 IEEE International Conference on Information Acquisition, 2006. [40] K. Bohnert and P. Guggenbach, “Innovation in high DC current measurement,” Elektron, vol. 22, no. 7, 2005. [41] K. Bohnert and P. Guggenbach, “A revolution in high dc current measurement,” ABB Rev., no. 1, 2005. [42]S. Torbus, “Frequency analysis of work of polarimetric current sensor with measurement coil made with the single mode optical fiber not resistant to bending,” in Przegląd Elektrotechniczny, vol. 2017, no. 6, 2017.
Authors: Ing. Jan Nedoma, VSB – Technical university of Ostrava, Faculty of Electrical Engineering and Computer Science, Department of Telecommunications, 17. listopadu 15, 708 33 Ostrava-Poruba Czech Republic, E-mail: jan.nedoma@vsb.cz; Ing. Marcel Fajkus, VSB – Technical university of Ostrava, Faculty of Electrical Engineering and Computer Science, Department of Telecommunications, 17. listopadu 15, 708 33 Ostrava-Poruba Czech Republic, E-mail: marcel.fajkus@vsb.cz; doc. Ing. Radek Martinek, VSB – Technical university of Ostrava, Faculty of Electrical Engineering and Computer Science, Department of Cybernetics and Biomedical Engineering, 17. listopadu 15, 708 33 Ostrava-Poruba Czech Republic, E-mail: radek.martinek@vsb.cz.
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 93 NR 11/2017. doi:10.15199/48.2017.11.30
Published by Amina DJABOREBBI, Boubakeur ZEGNINI, Djillali MAHI, Amar TELIDJI University of Laghouat, PoBox 37 G, Mkam Laghouat 03000, Algeria.
Abstract. The paper deals with transient analysis of grounding systems wind turbines. To improve the accuracy of lightning impact studies on wind power grid generation, it is necessary to develop faster, more accurate simulation tools and to use increasingly sophisticated models .First, we identify and characterize the different parameters that influence the behaviour of grounding systems, particularly when they broadcast a lightning current. To do this, an electromagnetic model from the theory of antennas equation’s by – Euler method with incorporating soil ionization allows to represent the behavior of an earthing system in the frequency domain.. Different configurations with several complexity degrees have been simulated. To validate the obtained results, we compare our TLM results to the measurement results and FDTD simulation. A comparison between two different configurations of wind turbine grounding systems with comparing the transient potential, impulse impedance, and DC component of transient impedance between the two configurations when buried in soil. A number of illustrative computational examples are presented in the paper.
Streszczenie. W celu poprawy dokładności badań wpływu wyładowań atmosferycznych na produkcję energii wiatrowej konieczne jest opracowanie szybszych i dokładniejszych narzędzi symulacyjnych oraz wykorzystanie coraz bardziej wyrafinowanych modeli. W pierwszej kolejności identyfikujemy i opisujemy różne parametry wpływające na zachowanie się systemów uziemienia, zwłaszcza gdy przekazują one prąd piorunowy. W tym celu opracowany został model elektromagnetyczny z teorii równań Eulera z wykorzystaniem jonizacji gruntu, który pozwala na przedstawienie zachowania się systemu uziemienia w dziedzinie częstotliwości. Symulowane są różne konfiguracje o kilku stopniach złożoności. W celu walidacji uzyskanych wyników, porównujemy nasze wyniki TLM z wynikami pomiarów i symulacji FDTD.Porównanie dwóch różnych konfiguracji uziemienia turbiny wiatrowej z porównaniem potencjału przemijającego, impedancji impulsowej i składowej stałej impedancji przemijającej pomiędzy tymi dwiema konfiguracjami, gdy są one zakopane w gruncie. (Modelowanie odpowiedzi na prąd piounowy w uziemionym systemie turbiny wiatrowej)
Keywords: Grounding system, Transmission line method, Finite difference time domain, Soil ionization, Transient behaviour, Słowa kluczowe: System uziemienia, metoda linii transmisyjnej, domena czasowa różnic skończonych, jonizacja gleby.
Introduction
The grounding systems transient behavior was the object of several investigations, specifically their transient response when subjected to lightning current, either by experiments [1] or by numerical simulations. The numerical simulations are based principally on three main theories: Field calculation using Finite Element Method [2], Field calculation using Antenna Theory Methods [3,4,5] or by using Transmission Line Modeling (TLM) [6, 7, 8, 9]. Some electrical elements are characterized by their high length like high voltage towers and wind turbines. Since the lightning strokes are attracted by high-rise buildings [10], many recent studies have been presented to the show the transient behaviour of high voltage line towers [5, 11, 12] and Wind turbine grounding systems [4, 13] when subjected to lightning current.
In our case, we are interested to the protection of wind turbines against the lightning strokes, and this by evaluating the transient response of their grounding systems, because that numerous wind turbines have been placed in regions characterized by high activity of lightning [14], and that turbines contains a very sensitive electronic components which control several systems like the convertors and pitch angles controller. So, their grounding systems are designated to avoid the lightning current to the ground without damaging its components [15]. In many investigations, the grounding systems are modeled using TLM theory because of its simplicity and effectiveness with giving results in good accordance with the other theories and the experimental results [6, 7, 16, 17, 18]. And this method has been implemented in several simulation programs like MATLAB and EMTP.
Zalhaf et al. [19, 20] have simulated the transient behavior of the whole wind turbine connected to grounding system using MATLAB. The investigation has been continued by another one [21], in which the simulations by MATLAB have been compared with some experimental results and other simulations results obtained by PSCAD/EMTDC software. A good accordance has been obtained. Several studies in the same subject have been published which using ATP-EMTP code. Sekioka et al. [22] have used ATP-EMTP to study the perturbations generated by the grounding systems subjected to lightning current, where they have studied the induced overvoltages in the surrounding of wind turbines grounding systems. Using the same code, Pastromas et al. [23] have simulated the wind turbine grounding system to present the current distribution in different segments. The grounding system of offshore wind turbine farms have been studied too by ATP-EMTP code, Heng et al. [24] have simulated the proposed tower model to evaluate the transient voltage obtained when injecting lightning current.
The TLM has been solved by FDTD method in [15]. In their investigation Raju et al. have studied the impact of soil parameters and distance between wind turbines on the transient response of grounding systems installed in homogeneous soil. In their simulation, the soil ionization haven’t been considered. For another configuration of wind turbine grounding system, the frequency variation of impedance has been studied by Sunjerga et al [13] with Antenna Theory by using NEC-4 Code. In this investigation, the transient response and soil ionization haven’t been evoked.
Senthilkumar et al [25] have used CDEGS software to study the grounding potential rise in ground grid buried in multilayer soil when subjected to power frequency current. In precedent investigation [17] we have used TLM to simulate several grounding system configurations and to study the optimal configuration for grounding system. The results have shown that the grid configuration always gives the lowest transient impedance. In another investigation [18] we have used the same model based on TLM to study the transient behavior of grounding grids buried in homogeneous and in heterogeneous soil. In the investigation we have studied the influence of the grid dimensions, the kind of the ground and the current injection point on the grounding system response (transient potential and impedance). The obtained results show that the grid containing the greater number of conductors permits to obtain the lowest transient potential and impedance.
This investigation treats the modeling of grounding system using TLM equation solved by – Euler method with incorporating soil ionization. Different configurations with several complexity degrees have been simulated. To validate the obtained results, we compare our TLM results to the measurement results and FDTD simulation ones. The novelty of our paper consists in a comparison between different wind turbine grounding configurations, and this comparison is made for grounding systems buried in homogenous and heterogeneous soil. After choose the best configuration, we study the current and the potential along the chosen grounding configuration when buried in homogeneous and stratified soil, to evaluate the components participation in the transient response, and this to authors’ knowledge haven’t been studied before. A comparison between two different configurations of wind turbine grounding systems with comparing the transient potential, impulse impedance, and DC component of transient impedance between the two configurations when buried in homogeneous and stratified soil. Finally, we choose the best configuration and we study the transient current and potential along this configuration when buried in homogeneous and stratified soil, and we give some discussions in the conclusion.
Transmission line method and soil ionization
The transient currents are characterized by high frequency components, for this the study of the systems subjected to lightning currents requires the use of some methods takes into configuration the electromagnetic wave propagation in the studied systems [8, 18]. Since it has improved its effectiveness for the modeling of such phenomena, the transmission line has been adopted to be used in this investigation for simulating the transient response of grounding system [17, 18]. It consists of divide every grounding conductor into several segments. Each segment is modeled by transmission line as shown in Fig. 1. The longitudinal part represents heat losses and magnetic effect of the conductor, and the transversal parts are modeling the heat losses and the polarization in the soil surrounding the electrode.
For a grounding conductor buried in soil of resistivity ρs and permittivity εs and permeability μsand subjected to current of frequency f, the current wavelength λ can be determined by the formula [6]:
.
Fig.1. Transmission Line Model of grounding conductor segment.
These parameters are calculated for each segment of grounding grid according to [26] by the formulas for the horizontal conductors:
.
.
And for the vertical conductor the parameters become [26]:
.
where ρc is the soil resistivity, ρs is the soil resistivity, l is the length of the electrode, h is the radius of the electrode, h is the burial depth, r is conductor radius, μ is the soil permeability (4πx10-7 (H/m)), εris relative permittivity, and ε0 is the vacuum permittivity (8.859×10-12 F/m).
When the current Ie is injected, we determine in each segment of grounding system: The input voltage Ui, the branch current Iij, the output voltage Uj and the output current Is. These parameters are shown in Fig.2.
Fig.2. The voltages and currents in each studied segment
By applying Kirchhoff laws, we determine the next differential equations:
.
These equations are valid only for one segment, and the output current and voltage of this segment will be the inputs of the next segment.
These equations and the other ones of rest of grounding electrode segments are solved by using iterative methods. In our simulation we have used Euler one.
To incorporate soil ionization, the transversal conductance is considered time dependent parameter G(t) determined from G and calculated by [7]:
.
With I(t) is the injected current and Ig is current is the current from which the soil ionization initiates in the soil which determined from soil electrical critical field. This later has been determined by [27]:
.
For this; the equations (10) and (12) become:
.
The TLM simulation is validated if every conductor segment length will be smaller than one tenth of wavelength (∆l≪λ /10) [6].
Simulation Results and validation
The proposed model has been already validated by comparing with ATP-EMTP results in previous investigation [17, 18].
In [28], Visacro et al. have make some experiments on horizontal electrode buried in homogeneous soil and subjected to impulse current. The chosen electrode is of 12 m length and 7 mm radius buried at 0.5 m depth. The electrode is buried in two different kinds of soil: the first is high resistivity soil (4 kΩ.m) and the second is low resistivity soil (300 Ω.m).
The soil permittivity has been considered 20. The current injected is of 2 A peak value. The authors of [28] have simulated the same configurations using their developed Hybrid Electromagnetic Model based on Antenna Theory. Their obtained results are presented on the Fig. 3.
Fig.3. The simulation and measured results obtained by [ICLP] a) Result for 300 Ω.m soil, b) Result for 4 kΩ.m soil
We have estimated the injected current, and we have used this current to simulate the same configuration with same soil and conductor parameters. Our obtained results for this simulation are shown in Fig. 4.
When comparing our TLM results and those obtained by Hybrid Electromagnetic Model [28] with the experimental results, we observe that the voltages computed using our TLM are clearly show better agreement with the experimental voltages compared to those obtained using the proposed [28]. We note the existence of a difference doesn’t exceed 10%.
Fig.4. Our simulation results obtained TLM solved by Euler a) Result for 300 Ω.m soil, b) Result for 4 kΩ.m soil
After validating our TLM calculation with comparing with experimental result, we validate our method for more complicated configuration of grounding system. We study the response of wind turbine grounding system.
We study the transient behavior of the grounding system composed of two squares of 12m x12 m (ring earth) and 6m x 6m (foundation reinforcing) co-centered buried at 2 m depth. The squares are related by two conductors placed as cross (bonding bar). The extremities of extern square are related to 10 m vertical electrodes (points 1, 2, 3 and 4) as shown in figure 5. The conductor radius has been considered 10 mm.
The current injected in the center has been considered of 50 kA magnitude and 0,25 μs front time.
Fig.5. The wind turbine grounding electrode (Configuration A) [15]
This configuration has been studied by [15] using FDTD method.
For our study, we will simulate this configuration in two cases: the first one with ignoring soil ionization, and the second with considering the soil ionization. The configuration has been evaluated for two soil resistivity values namely 1000 Ωm and 2000 Ωm and the relative permittivity has been considered 9. Our obtained results for TLM with ignoring soil ionization phenomenon are presented in figure 6 while the potentials obtained with incorporating of soil ionization are shown in Fig. 7.
When confronting our TLM potentials presented in Fig. 6. to those obtained by [15], we note a good accordance with the potentials obtained using FDTD method [15], and this accordance validates our TLM simulation. When incorporating soil ionization phenomena, we note a great decrease on the transient potential (about 25% for 2000 Ω.m and 10% for 1000 Ω.m), while the waveform has been kept for the two cases. So, the incorporation of soil ionization in TLM simulation causes a significant potential decrease. When the soil is more resistive, the potential will be significantly reduced which means that the soil ionization phenomena must be considered for the grounding systems buried in resistive soil.
Fig.6. Transient potential at injection point without incorporating soil ionization
Fig.7. Transient potential at injection point with incorporating of soil ionization.
After validation of the simulation of grounding system presented in Fig.5., we simulate here another configuration of wind turbine grounding system. We name the configuration presented in Fig. 5. by Configuration A, and the second configuration that we will study by configuration B which is presented in the Fig. 8.
Fig.8. Configuration B studied by Sunjerga et al [13]
The grounding system of configuration B consists of several rings connected with several rods. The geometrical parameters are presented in Table 1 where the rings (such like r2 shown in figure 8) and the depth of each ring are noted.
Table 1. The Geometry of the wind turbine grounding system B
.
In this part, we will inject the same impulse current into the two configurations. The current formula is given by i(t)= 10 (e-27000t – e-5600000t ). Three soil parameters have been considered of resistivity/relative permittivity values: 10Ωm/80, 100Ωm/40 and 1000Ωm/9.
Fig.9. Comparison between the transient potentials obtained for the configurations A and B for homogeneous soil of: (a) 10Ωm/80, (b) 100Ωm/40, (c) 1000Ωm/9
We note that for both configurations A and B, the conductor radius value has been considered 7 mm. The obtained results for the configurations A and B are shown in Fig. 9: (a) for grounding systems buried in soil of resistivity 10Ωm and relative permittivity 80, (b) for soil of resistivity 100Ωm and permittivity 40 and (c) for the grounding systems buried in soil of resistivity 1000Ωm and permittivity 9. We note that for all of the configurations and for any soil parameters, the principal variations are observed between 0 and 4 µs. After that, the potential becomes approximately constant, that means that after 4 µs the impedance becomes constant.
Discussions
In the three tests, we observe on the figures 9 (a), (b) and (c) that the configuration B gives the lower values of transient potential all of the soil resistivity values:
– For the soil resistivity/permittivity 10Ωm/80 the peak value of configuration B present 58% comparing to the one obtained for the configuration A.
– When the soil resistivity/permittivity are 100Ωm/40 the peak value of configuration B present 62.5% comparing to the peak value of the potential obtained for the configuration A.
– When the grounding systems are buried in soil of resistivity/permittivity 1000Ωm/9, the transient potential waveshape has been changed, and many oscillations have appeared for the both potentials.
In this case the configuration B gives lower transient potential when comparing to the configuration A with a difference of 70% noted between the potential peaks.
We note that for the configuration A, when increasing soil resistivity, the wave shape of the potential becomes different to the current one. For high resistivity values (1000Ωm/9), the potential contains significant oscillations.
We define the impulse impedance Zp by the ratio of peak voltage Vp to peak current Ip.
.
We define the DC component of the transient impedance by the ratio of the constant value of potential (In this case, the potential value after 4 µs) to the current peak.
We present the impulse impedance of each configuration (A and B) and for the all of soil resistivity on the Table 2, and the DC component of the transient impedance on the Table 3.
Table 2. The impulse impedance values for each configuration
.
The Table 2 presents show the configuration B gives the lowest impulse impedance when comparing to the values obtained for the configuration A. So, we can adopt that the configuration B which contains more rings distributed on several levels can reduce the transient impedance.
Table 3. The DC component of transient impedance values for each configuration
.
From the Table 3, we can see that the DC component of each configuration increase linearly with the increasing of soil resistivity, so this DC component is a constant depend to the grounding system geometry multiplied by the soil resistivity.
When comparing DC components of configurations A with B, we observe that the DC component of A is the triple of the one of B. So, the configuration B gives the lowest DC component of transient impedance. We note that for high resistivity soil, the impulse impedance becomes equal to the DC component of transient impedance.
Conclusion
We first identified and characterized the different parameters that influence the behaviour of earthing systems, especially when they diffuse a lightning current. The resistivity is the most important element in the design of grounding systems. When its value is very high, the potential generated by an atmospheric discharge becomes important. Next, a model was described to represent the transient behaviour of the grounding systems in the frequency domain. To do this, we are based on the electromagnetic model. This model based on the antennas, uses the numerical method known as Euler by using two applications .The first application was the incorporating soil ionization in the model of wind turbine grounding systems. A potential decrease is noted when incorporating soil ionization phenomena, however the waveform has been kept. The second application was a comparison between two wind turbine grounding systems; the first is the one which has been validated with FDTD simulation composed by two squares co-centered buried at same depth, and related to long vertical electrodes, and the second is another configuration composed of five rings disposed on several depths and connected to short vertical electrodes.
We note that for high resistivity soil, a remarkable voltage oscillation appears on the transient potential, and its waveform becomes different to the current one.
We have evaluated the DC component of the transient impedance; it consists of ratio between voltage stabilized value and the current peak one. This component may be determined by a constant which depend to the system geometry multiplied by soil resistivity.
Finally, our model also allowed us to estimate the temporal responses of the wave electromagnetic created by the grounding system on its material environment and propose optimization solutions for more resistive flooring.
REFERENCES
[1] H. Rochereau,: Response of earth electrodes when fast fronted currents areflowing out. EDF: Buletin de la Direction des Etudes et Recherches B, (2), 1988, pp. 13–22 [2] M. Akbari, K. Sheshyekani and M. R. Alemi,”The Effect of Frequency Dependence of Soil Electrical Parameters on the Lightning Performance of Grounding Systems,” in IEEE Transactions on Electromagnetic Comptibility,vol.55,no.4, pp.739-746,Aug.2013. [3] D. Poljak, S. Sesnic, S. V. Tkachenko, K. El KhamlichiDrissi, K. Kerroum, “Time Domain Analysis of the Horizontal Grounding Electrode: Antenna theory approach versus transmission line approximation,” Proc. of the 2014 International Symposium on Electromagnetic Compatibility (EMC Europe 2014), Gothenburg, Sweden, 181-185, September 1-4, 2014. [4] R. Araneo, G. Lovat, S. Celozzi and P. Burghignoli. Numerical Analysis of Mutual Transient Voltages in Grounding Systems of Offshore Wind Farms. International Applied Computational Electromagnetics Society Symposium (ACES). 2018. [5] Javad Gholinezhad, and Reza Shariatinasab .Time Domain Modeling of Tower-Footing Grounding Systems based on Impedance Matrix. IEEE Transactions on Power Delivery. Volume: 34, Issue: 3.2019 PP : 910 – 918. [6] Yang L, Wu G N, Cao X B, “An optimized transmission line model of grounding electrodes under lightning currents,” Sci China Tech Sci, Vol. 56, 335-341, 2013. [7] Daniel S. Gazzana, Arturo S.Bretas, Guilherme A. D. Dias,Marcos Telló, Dave W. P. Thomas and Christos Christopoulos.The Transmission Line Modeling Method to Represent the Soil Ionization Phenomenon in Grounding Systems. IEEE transactions on magnetics, vol. 50, no. 2, February 2014. [8] Daniel S. Gazzana, Guilherme A. D. Dias, Roberto C. Leborgne, Arturo S. Bretas,Marcos Telló, DaveW.P. Thomas, Christos Christopoulos. Novel Formulation to Determine the Potential on the Soil Surface Generated by a Lightning Surge. ieee transactions on magnetics, vol. 52, no. 3. march 2016 [9] M. Mokhtari, Z. Abdul-Malek and Z. Salam. An Improved Circuit-Based Model of a Grounding Electrode by Considering the Current Rate of Rise and Soil Ionization Factors. IEEE Transactions On Power delivery, vol.30, no. 1, pp:211-219. February 2015 [10] IEC. Protection Against Lightning Part 1: General Principles. IEC 62305-1, Ed. 1.0b, 2006 [11] W. R. SI, C. Z. FU, T. Y. WU, S. J. Wang, P. YUAN. Measurement and Analysis of Impulse Grounding Impedance for UHV Transmission Tower. IEEE International Conference on High Voltage Engineering and Application (ICHVE) 2016 [12] Amauri G. Martins-Britto, Sébastien R. M. J. Rondineau and Felipe V. Lopes. Transient Response of the Grounding Grid of a Power Line Tower Subject to a Lightning Discharge. Workshop on Communication Networks and Power Systems (WCNPS). 2018 [13] Antonio Sunjerga , Quanxin Li , Dragan Poljak , Marcos Rubinstein , Farhad Rachidi. Transient Impedance of Interconnected Wind Turbine Grounding Systems 26th International Conference on Software, Telecommunications and Computer Networks (SoftCOM) 2018 [14] D. Cavka, D. Poljak, V. Doric, and R. Goic, “Transient analysis of grounding systems for wind turbines,” Renew. Energy, vol. 43, pp. 284–291, 2012. [15] Md. Raju Ahmed and Masaru Ishii. Electromagnetic Analysis of Lightning Surge Response of Interconnected Wind Turbine Grounding System. 011 International Symposium on Lightning Protection (XI SIPDA), Fortaleza, Brazil, October 3-7, 2011. [16] Yaqing Liu, Mihael Zitnik, and Rajeev Thottappillil, “An Improved Transmission-Line Modelof Grounding System,” IEEE Transactions on Electromagnetic Compatibility, VOL. 43, No. 3, 348-355, AUGUST 2001. [17] Djaborebbi Amina, Boubakeur Zegnini, Tahar Seghier ,Hamza Gueffaf, Djillali Mahi. Economic optimization of copper conductor for transient response of grounding system. 5TH International Conference On Advances In Mechanical Engineering Istanbul 2019, 17-19 DECEMBER 2019. [18] DJABOREBBI Amina, ZEGNINI Boubakeur, SEGHIER Tahar, Mahi Djillali. Evaluation and Performance of Grounding Grids Buried in Soil Under Impulse lightning Current. The First International Conference on Communications, Control Systems and Signal Processing.16-17 March 2020 El-Oued, Algeria [19] A. S. Zalhaf, Mahmoud Ahmed, S. Ookawara and Mazen Abdel-Salam. Computation of Transient Induced Voltages along a Wind Turbine Struck by Lightning. 2nd International Conference on Power and Renewable Energy 2017. [20] A. S. Zalhaf, Mahmoud Ahmed, S. Ookawara and Mazen Abdel-Salam. A Simplified Model of Wind Turbine for Lightning Transient Analysis as Influenced by Structure of Grounding System. 5th International Conference on Electric Power and Energy Conversion Systems (EPECS) 2018. [21] Amr S. Zalhaf, Mazen Abdel-Salam, Diaa-Eldin A. Mansour, Shinichi Ookawara, Mahmoud Ahmed. Assessment of wind turbine transient overvoltages when struck by lightning: experimental and analytical study. IET Renewable Power Generation Volume: 13, Issue: 8 .Year: 2019 . PP 1360 – 1368 [22] Shozo Sekioka, Hitomi Otoguro, and Toshihisa Funabashi. A Study on Overvoltages in Windfarm Caused by Direct Lightning Stroke. IEEE Transactions on Power Delivery ( Volume: 34 , Issue: 2 , April 2019 )pp : 671 – 679. [23] S. A. Pastromas, K. Maimaris, I. K. Stasinos, I. A. Naxakis, E. C. Pyrgioti. Assessment of Wind Turbine Grounding System. IEEE International Conference on High Voltage Engineering and Application (ICHVE) 2018 [24] Lu Heng, Xia Nenghong, Qian Chao and Tian Jianlin. Modeling of Offshore Wind Turbine Under Lightning Stroke and Analysis of Impact Factors on Transient Overvoltage. 11th Asia-Pacific International Conference on Lightning (APL) 2019. [25] R.T.Senthilkumar, K.Selvakumar and K.Nagarajan. Optimization of Multilayer Earth Structure by using Steepest Descent Method and Estimation of Transient Ground Potential Rise in Substation. 2018 IEEE PES Asia-Pacific Power and Energy Engineering Conference (APPEEC) [26] Sunde ED, “Earth Conducting Effects in Transmission Systems,” New York, N. Y.: Dover publications, Inc.; 1968. [27] T.K. Manna, P. Chowdhury. Generalized equation of soil critical electric field Ec based on impulse tests and measured soil electrical parameters, IET Generation, Transmission and Distribution. 1 (5) (2007) 811-817. [28] Silvério Visacro, Wyllian L.F. Pinto, Fuad S. Almeida, M.H. Murta Vale, Gláucio Rosado. Experimental Evaluation Of Soil Parameter Behavior In The Frequency Range Associated To Lightning Currents. 29th International Conference on Lightning Protection 23rd – 26th June 2008 – Uppsala, Sweden
Authors: Amina Djaborebbi, Laboratoire d’études et développement des matériaux semi-conducteurs etdiélectriques(LeDMaScD), Amar Telidji University of Laghouat. ,PoBox 37 G, Mkam Laghouat 03000,Algeria E-mail:a.djaborebbi@lagh-univ.dz, prof dr Boubakeur Zegnini , LeDMaScD, Amar Telidji University of Laghouat. ,PoBox 37 G, Mkam Laghouat 03000,Algeria,E-mail: b.zegnini@lagh-univ.dz, prof dr Djillali Mahi , LeDMaScD,Amar Telidji University of Laghouat. ,PoBox 37 G, Mkam Laghouat 03000,Algeria,E-mail: d.mahi@lagh-univ.dz
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 2/2021. doi:10.15199/48.2021.02.33
Power Quality Blog 