Published by 1. Rafael GIVANILDO, 2. Denis LIMA, 3. Paulo PARIS, 4. Emerson PEDRINO
Department of Computer Science, Federal University of Sao Carlos, Sao Carlos, Brazil
ORCID: 1. 0000-0003-1372-6049, 2. 0000-0002-0457-2562, 3. 0000-0001-8915-8215, 4. 0000-0003-3734-3202
Abstract. This article aims to carry out a brief bibliographical review on the main concepts related to Smart Grid, in addition to the development of a low-cost and open-source smart meter prototype. This research was carried out based on concepts involved and used in developing the CS5463 chip, an embedded Linux system, and several software libraries, which helped with the implementation of the reference algorithm and charging simplification. Furthermore, the prototype had positive results, as it was possible to implement the proposed algorithms with a cost below US $ 50.00 and achieved an accuracy above 90%. Finally, it is concluded that the concept of Smart Grid and everything that permeates it is fundamental, especially given the context of digital transformation in this area. Such a prototype is an initial entry alternative for developing technologies that are trending in this area.
Streszczenie.. Celem tego artykułu jest przeprowadzenie krótkiego przeglądu bibliograficznego głównych pojęć związanych ze Smart Grid, a także opracowanie taniego prototypu inteligentnego licznika o otwartym kodzie źródłowym. Badania te zostały przeprowadzone w oparciu o koncepcję wykorzystana w opracowaniu układu CS5463, wbudowanego systemu Linux oraz kilku bibliotek oprogramowania, które pomogły we wdrożeniu algorytmu referencyjnego i uproszczeniu ładowania. Co więcej, prototyp uzyskał pozytywne wyniki, ponieważ możliwe było wdrożenie proponowanych algorytmów kosztem poniżej 50 USD i osiągnął dokładność powyżej 90%. Na koniec stwierdza się, że koncepcja Smart Grid i wszystko, co się przez nią przenika, ma fundamentalne znaczenie, zwłaszcza w kontekście cyfrowej transformacji w tym obszarze. Taki prototyp jest wstępną alternatywą dla rozwoju technologii, które są trendy w tej dziedzinie. (Tani i dokładny prototyp inteligentnego licznika dla inteligentnych sieci)
Keywords: Smart Meters, Smart Grids, Low-Cost Prototype.
Słowa kluczowe: Smart Grid, miernik inteligentny, pomiar mocy
Introduction
Since the invention of electrical networks, there have been no significant changes in the technology used for the generation, transmission, and distribution of electrical energy, where the technologies used to date back to the end of the 19th-century [1] [2] [3].
As the global demand for energy increases, it is necessary to use techniques to make the network more efficient and actual, giving rise to a new concept to solve this challenge, called Smart Grids [4].
NIST (National Institute of Standards and Technology) defines Smart Grids as an electrical network that uses two-way flow information with secure communication and artificial intelligence technologies to integrate the entire spectrum of the power system, from power generation to final customer [5]. In the Brazilian context, the implementation of this concept is motivated mainly to reduce the non-technical losses of the network, which according to Aneel is around 6.6%. Also, there is a quest to increase the reliability of the system, reduce operating costs – especially those related to measurement, and increased energy efficiency [6] [7].
One of the challenges for the massive implementation of Smart Grid in Brazil is related to the large volume of investment required for its implementation. The key component, which requires most of this investment, is related to the exchange of the meter park for smart meters [8].
In this context, this work develops an initial prototype of an intelligent energy meter. The prototype will be presented as being of low cost concerning the prices of similar equipment (ranging from U$ 50 to U$ 100 for home use), with free software and code compatible with multiple embedded platforms. The decrease in cost of the meter is important, as it decreases the amount of investment needed to update the meter park [9] [10]. Besides, the prototype can be a starting point for future work in the area of Non-intrusive load monitoring (NILM).
This article is organized as follows: Section 2 presents the concept of Smart Grid, section 3 presents the concept of Smart Meter, section 4 presents the developed prototype, and section 5 presents the results obtained. Finally, section 6 presents the conclusions.
Smart Grid Overview
In the current world scenario, the availability of electricity is essential for contemporary societies, as it is closely related to most activities. The unavailability of energy generates several negative impacts.
Unfortunately, classic networks are not resilient and agile, being susceptible to problems with generation (lack of demand) and transmission/distribution (quality problems and issues related to technical and non-technical losses) directly affecting the final consumer.
In 2005, S. Massoud Amin and Bruce F. Wollenberg coined the term Smart Grid in a publication by the IEEE (Institute of Electrical and Electronics Engineers). The authors define the term as large-scale electrical network infrastructure characterized by security, agility, and resilience/robustness facing new threats and unplanned conditions [11]. The term coined by them, meant a major paradigm shift, going beyond the simple implementation of certain technology to something bigger.
There are several motivations to justify investing in Smart Grid. Some of them are improvement in the country’s energy security, reduction of greenhouse gas emissions, and the possibility of reducing operational costs and nontechnical losses.
Brazil, like most developing countries, is in the initial stage of implementing the Smart Grids concept.
Looking from the distribution point of view, where there is an interconnection between the final consumer and the distribution sector, it is necessary to implement an architecture known as AMI – Advanced Metering Infrastructure, which enables bidirectional communication and several new network functionalities.
The key equipment of this architecture is Smart Meters, which collect and send data in a bidirectional way between the customer and the distributor. These devices are the target of this article. The meters can use different communication technologies, such as PLC, RF Mesh, or mobile networks, to establish the connection with the concentrators – or directly with the distributor depending on the case.
Looking at the network architecture in (Fig. 1), AMI is composed of three types of networks, they are Home area network – HAN, Neighborhood area network – NAN, and Wide area network – WAN. HAN is a network generated by smart meters and is responsible for collecting all information on consumption, micro-generation, and household occurrences.
Besides, this network can connect smart devices to the meter, allowing demand control by the distributor if necessary. NAN is the network responsible for concentrating data from existing meters in the neighborhood and sending it to the distributor via WAN. In this stage, concentrators are used, which concentrate data from smart meters and send them via IP network to the distributor’s backbone. After sending data over the WAN, they arrive at the distributor where it will be used for the most diverse services. It is worth mentioning the module known as MDMS (meter data management system) is responsible for managing, storing, and analyzing the data received.
Smart Meter Overview
Before talking about smart energy meters, it is important to talk about the origin and evolution of the meters. Electromechanical meters were the first-meter model and are still widely used today. They are based on the principle of electromagnetic induction, having been invented by Shallenberger in 1888 [13]. This meter measures only the active energy consumption, and manual data collection is required for data collection. The flow of information is unidirectional. The useful life of this type of meter can reach 25 years [10].
With the development of digital systems and their subsequent cheapening, electronic energy meters emerged. This type of meter is based on the use of A/D converters and a microcontroller or microprocessor for sampling techniques to determine the energy consumed by the consumer.
In general, electronic meters are more accurate than electromechanical meters. This type of meter has the following disadvantages: the fact that its useful life is between 13 and 15 years; there is uncertainty about its operation under severe climatic conditions and its cost greater than the electromechanical [13]. Fig. 2 shows some aspects and characteristics of conventional energy meters in relation to Smart Meters.
Over time, electronic meters have evolved into what is now known as smart energy meters. The Edison Electric Institute (EEI) defines Smart Meters as: “electronic metering devices used by utility companies to transmit information for charging the consumer and for operating electrical systems” [14] [15]. There is still no general definition of what features define a smart meter or smart metering system. For this paper, we use the definition by Mohassel, Moahammadi, Fung, and Raahemifar [16]. They are:
• Time-based pricing;
• Providing consumption data for consumers and utility;
• Net metering;
• Failure and outage notification;
• Remote command (turn on / off) operations;
• Load limiting for Demand Response purposes;
• Power quality monitoring including phase, voltage, and current, active and reactive power, power factor;
• Energy theft detection;
• Communication with other intelligent devices;
• Improving environmental conditions by reducing emissions through efficient power consumption.
Looking at the market solutions, we have several meter manufacturers, where we can mention as examples: WEG, General Electric, Itron, Nansen, Siemens, Schneider Electric, among others. In general, these meters are modular, measure active and reactive energy in the four quadrants, active and reactive demand, in addition to items related to network quality. These meters are bidirectional, with the possibility of using multiple forms of communication (PLC, RF, Ethernet, Wifi, Zigbee, GSM / GPRS / CDMA). Also, these meters allow the programming of several charging models, mass memory, and the possibility of fraud detection. Optionally, some meters have tools for use in smart homes.
Smart Meter Prototype
The implemented smart meter prototype proposed is based on non-intrusive, low cost and easy to find components. Based on the design problems presented by Depuru, Wang, and Devabhaktuni [12], we try to address issues related to the technologies used for measurement/charging, meter cost, and communication. Fig. 3 shows the diagram of the mentioned items, which during this section will be better explained.
The measurement and charging technologies involved with the proposed prototype were developed using the Cirrus Logic CS5463 IC, an integrated circuit specialized in measuring electrical parameters, together with the MRAA library – library for embedded Linux systems that easy I/O communication-, implements SPI communication, and allows code portability across multiple embedded systems. The tariff was based on a simplified version of Aneel’s resolution, No. 733 of 2016, which regulates the white tariff (Brazilian seasonal tariff). It was also sought that the prototype minimally could obey the resolution of Aneel nº502, which lists the minimum requirements for energy meters, but maintaining the low-cost prerogative.
Looking at market solutions, the meter features energy measurement, active and reactive demand, bidirectional communication, mass memory, and communication with the Internet via Wi-Fi. Besides, the existing pricing is based on consumption time (Time-based pricing). The final cost of the prototype was US $ 45.00. The basic characteristics of the prototype are:
• Operation: 127 / 220V;
• Measurement of Voltage, Current, Frequency, Active, Reactive and Apparent Power and Power Factor;
• Internet connection via Wi-Fi;
• Ability to save data on the memory card;
• Display with measurement information;
• Non-intrusive, allowing easy installation of the prototype;
• Implementation of a simplified version of the White Tariff;
• Online panel with measurement information;
• Open source code: (https://bitbucket.org/Mud_Owl/ic_mud_owl_v2).
The design and construction of the hardware/software used a bottom-up approach. It started with simplified hardware and software and, after several studies and tests, it evolved to the above characteristics.
Thus, the system architecture shown in Fig. 4 was developed. The sensors gather the information from the network and the meter (IC meter and development board) is responsible for allowing the visualization of the data, making the communication interfaces, and processing the electrical measurements and the tariff.
Hardware
To easy data collection, it was decided to use an integrated circuit dedicated to energy measurement applications, Cirrus Logic CS5463, responsible for obtaining the values of various electrical measurements. This made the work easier, as the measurement algorithms are in hardware. In addition to that, they were tested and had an accuracy established and guaranteed by the manufacturer. The CS5463 is an IC com- posed of two ADC converters – one for voltage and one for current, besides a calculation unit, which calculates several electrical quantities. This IC communicates with the development board using the SPI protocol. For the use of this IC (CS5463) and the modularization of the project, a printed circuit board was developed following the model of the technical sheet found on the Cirrus Logic website [17]. Fig. 5 shows the diagram of the developed circuit board. The JP1 and JP2 components form a socket, which easy the connection of the IC to the board. The TC and TP components represent the connections of the sensors (terminals), connected to the conditioning circuits (voltage dividers that reduce the input voltage to a maximum of 250mV RMS), passive filters, and the IC connections.
The other representations are connecting pins and pins for IC calibration. JP6 and JP7 components are connected to the pins corresponding to the SPI input of the embedded system. The hardware developed and used consists of the following items:
• Cirrus Logic CS5463 integrated circuit, two conditioning circuits for adjusting the sensor voltage to the IC input values, and interface connectors between the sensors and the board;
• 28-pin SMD / DIP adapter;
• Intel Edison with Arduino kit;
• Base shield;
• 16×2 RGB LCD;
• Potential transformer (PT) 127 / 220V to 12 + 12V, as a voltage sensor;
• Current transformer (CT) SCT-013-000, as a current sensor;
• Connection jumpers.
Software
The software developed for the prototype implements the architecture shown above, and it focused on two parts:
• The first part consisted of the implementation of the measurement routines, which took into account the measurement and loading functions. For this, it was needed to use the SPI protocols, for the communication between Intel Edison and CS5463, the I2C protocol, for the communication with the display, and the MQTT protocol – for the WEB communication used to update the dashboard information.
• The second part consisted of implementing user views, where the dashboard was developed. The first part used the Python language in conjunction with several libraries, mainly the MRAA library, which controls the GPIO ports of the card and allows the portability of the software to various hardware on the market and the paho.mqtt library that implements the MQTT client used to send data to the dashboard.
In the flowchart (Fig. 6) on the right, all the necessary steps to measure an electrical quantity are detailed, where it is needed to perform: IC configuration, sending initialization commands and after that, it starts to receive the chip data, which will ultimately need to convert and apply the necessary scales.
In the flowchart (Fig. 6), on the left of the figure, the steps necessary to perform the consumption calculation are detailed according to the white tariff model – power measurement, energy consumption calculation, and application of the white tariff rules (questions related to the day of the week and times ).
The second part consisted of implementing the dashboard. To facilitate the design, a free template built-in Bootstrap [https://getbootstrap.com/] and Highcharts [https://www.highcharts.com/] – used to generate the graphics were used. The functioning of the dashboard consists of receiving the information from the MQTT protocol and presenting it to the user.
Experimental Results
As an initial test, to prove that the developed firmware is functional, it was decided to measure and analyze the error in the current and voltage measurements obtained in different electrical equipment. It was decided this way because current and voltage are the fundamental quantities and the other quantities use these values in the internal calculation of the IC.
To perform the tests, it was needed to perform the calibration steps described in the IC datasheet [17]. This is necessary to obtain the precision described by the manufacturer. Also, there was a need to adjust the conversion scales of the software, once the IC did not present the values directly, but in scale. The test performed was in accordance with the following methodology:
• The prototype smart meter and a set of multimeters configured as ammeter and voltmeter were connected;
• A set of electrical equipment was chosen, in this case: – Amazon Alexa (15 W); – Cellphone Charger (10 W); – Incandescent Lamp (25 W); – Electric Citrus Juicer (250 W); – Led Lamp (9 W); – Computer Monitor (15 W); – Notebook Power Adapter (45 W); – TV (120 W); – Fan ( 80 W).
• One device was turned on at a time and the data obtained from the prototype meter was collected, 5000 samples of each equipment were used (Voltage, Current, Active Power, Apparent Power, Reactive Power, Frequency and Power Factor) and the average was calculated to obtain the values presented in Table 2;
• The data obtained by the prototype were compared with the data obtained by the multimeters and a subsequent calculation of the agreement between the types of data was performed in Table 1.
Table 1. Comparison between multimeter and proposed smart meter.
Table 2. Proposed smart meter collected data
Discussion
By analyzing the table above, it is possible to infer an accuracy above 90% in all the test cases, demonstrating that the data are within an acceptable range. As it is an initial prototype and as there is no need to use test methodologies that are used in commercial products, standardized by INMETRO (Brazilian Metrology Institute), these values are acceptable for this research. The voltage data reached an accuracy close to that described in the datasheet (+/-0.1%), indicating a good fit. Current data achieved an accuracy above 90%, but there is room for improvement during the calibration step aiming at reaching a value close to that indicated in the datasheet. The main objective of this experiment was achieved, as it indicates a good working of the measurement algorithm. In future work, two interesting paths to be followed are: to improve the calibration step – especially in the current part – and carry out measurements at the site’s energy input to have a general value for the installation.
Conclusion
This article presented the scenario of smart grids in Brazil and in the World, contextualizing the importance of the development of the open source smart meter proposed prototype. Concerning the prototype, it was possible to implement several concepts of Smart Meters.
The goal of creating a low-cost prototype was achieved, as equipment costs were below US $50. Another interesting point is that through the implementation of AMI, and consequently the massification of smart meters, a new market will emerge that will be based on the information obtained by smart meters, where there will be many opportunities for the creation of innovative services.
Finally, smart meters can bring new economic and technological opportunities and, mainly, bring investments to modernize electrical systems, bringing more sustainability and awareness about the use of electric energy. For future work, we will extend the research with: first, collect more data about many electrical devices using a low- cost smart meter; second, training a CNN to identify connected electrical devices; third, create a recommendation system for smart home environments.
Acknowledgements This work was supported by funding from FAPESP (Grant 2017/26421-3).
REFERENCES
[1] Y. Wang, Q. Chen, T. Hong, and C. Kang, “Review of smart meter data analytics: Applications, methodologies, and challenges,” IEEE Transactions on Smart Grid, vol. 10, no. 3, pp. 3125–3148, 2018.
[2] B. Yildiz, J. I. Bilbao, J. Dore, and A. B. Sproul, “Recent advances in the analysis of residential electricity consumption and applications of smart meter data,” Applied Energy, vol. 208, pp. 402–427, 2017.
[3] M. R. Asghar, G. Dán, D. Miorandi, and I. Chlamtac, “Smart meter data privacy: A survey,” IEEE Communications Surveys & Tutorials, vol. 19, no. 4, pp. 2820–2835, 2017.
[4] Y. Kabalci, “A survey on smart metering and smart grid communication,” Renewable and Sustainable Energy Reviews, vol. 57, pp. 302–318, 2016.
[5] E. A. Bueno, W. Utubey, and R. R. Hostt, “Evaluating the effect of the white tariff on a distribution expansion project in brazil,” in 2013 IEEE PES Conference on Innovative Smart Grid Technologies (ISGT Latin America). IEEE, 2013, pp. 1–8.
[6] D. Vieira, R. A. Shayani, and M. G. De Oliveira, “Net metering in brazil: Regulation, opportunities and challenges,” IEEE Latin America Transactions, vol. 14, no. 8, pp. 3687–3694, 2016.
[7] M. P. Maceira, L. A. Terry, F. S. Costa, J. M. Damázio, and A. Melo, “Chain of optimization models for setting the energy dispatch and spot price in the brazilian system,” in Proceedings of the power system computation conference-PSCC, vol. 2, 2002, pp. 24–28.
[8] Y. Yan, Y. Qian, H. Sharif, and D. Tipper, “A survey on smart grid communication infrastructures: Motivations, requirements and challenges,” IEEE communications surveys & tutorials, vol. 15, no. 1, pp. 5–20, 2012.
[9] H. Farhangi, “The path of the smart grid,” IEEE power and energy magazine, vol. 8, no. 1, pp. 18–28, 2009.
[10] P. Carvalho, “Smart metering deployment in brazil,” Energy Procedia, vol. 83, pp. 360–369, 2015.
[11] S. M. Amin and B. F. Wollenberg, “Toward a smart grid: power delivery for the 21st century,” IEEE power and energy magazine, vol. 3, no. 5, pp. 34–41, 2005.
[12] S. S. S. R. Depuru, L. Wang, V. Devabhaktuni, and N. Gudi, “Smart meters for power grid—challenges, issues, advantages and status,” in 2011 IEEE/PES Power Systems Conference and Exposition. IEEE, 2011, pp. 1–7.
[13] K. G. Di Santo, E. Kanashiro, S. G. Di Santo, and M. A. Saidel, “A review on smart grids and experiences in brazil,” Renewable and Sustainable Energy Reviews, vol. 52, pp. 1072–1082, 2015.
[14] V. C. Gungor, D. Sahin, T. Kocak, S. Ergut, C. Buccella, C. Cecati, and G. P. Hancke, “Smart grid technologies: Communica- tion technologies and standards,” IEEE transactions on Indus- trial informatics, vol. 7, no. 4, pp. 529–539, 2011.
[15] E.-A.-U. An, “Smart meters and smart meter systems: A metering industry perspective,” Washington, DC, USA, Edison Elect. Inst., White Paper, 2011.
[16] R. R. Mohassel, A. Fung, F. Mohammadi, and K. Raahemifar, “A survey on advanced metering infrastructure,” International Journal of Electrical Power & Energy Systems, vol. 63, pp. 473–484, 2014.
[17] C. Logic, “C55463: Single phase, bi-directional power/energy ic,” Datasheet DS678F2, Apr, 2008
Authors: Rafael Givanildo, M. Sc. Denis Lima, M. Sc. Paulo Paris Prof. Dr. Emerson Pedrino, Computer Science Depart- ment, Federal University of Sao Carlos, SP., Brazil, email: emerson@dc.ufscar.br.
Source & Publisher Item Identifier: PRZEGLĄD ELEKTROTECHNICZNY, ISSN 0033-2097, R. 97 NR 9/2021. doi:10.15199/48.2021.09.16
Power Quality Blog 