Data Characterization for Electrical Load Disaggregation using Supervised Learning

  • Wesley Souza Universidade Federal de São Carlos
  • Tiago Almeida Universidade Federal de São Carlos

Resumo


A energia elétrica é um serviço essencial e o seu uso otimizado é uma demanda cada vez maior nas redes elétricas modernas. A recente evolução dos microcontroladores permitiu a imersão de inteligência artificial em uma nova geração de medidores, chamados de medidores cognitivos. A principal característica do medidor cognitivo é a possibilidade de realizar a desagregação do consumo de energia elétrica por cargas. Para que isso seja possível, é necessário utilizar uma boa representação para os dados, de tal forma que leituras provenientes de medições elétricas possam ser usadas para identificar cada equipamento elétrico. Neste contexto, este artigo apresenta uma proposta de caracterização de dados para a desagregação do consumo por meio de 140 atributos. Para validar essa proposta, foi construído um conjunto de dados reais coletados a partir de 28 equipamentos elétricos, os quais foram usados para treinar três métodos tradicionais de classificação. Com base nos resultados obtidos, conclui-se que é possível realizar a efetiva desagregação de cargas em cenários reais e em tempo quase real, abrindo um grande conjunto de possibilidades para explorações de novos recursos e empregos de técnicas de inteligência computacional em medidores cognitivos.

Palavras-chave: desagregação de cargas, medidor cognitivo, engenharia de atributos, aprendizado de máquina, nilm

Referências

Abubakar, I., Khalid, S., Mustafa, M., Shareef, H., and Mustapha, M. (2017). Application of load monitoring in appliances’ energy management - a review. Renewable and Sustainable Energy Reviews, 67:235–245.

Azzini, H. A. D., Cypriano, J. G. I., de Souza, W. A., Monzani, R. C., da Silva, L. C. P., and Venerando, A. C. (2018). Comparação das técnicas de monitoramento de cargas para a desagregação do consumo de energia elétrica. In Congresso Brasileiro de Automática, pages 2525–8311.

Barai, G. R., Krishnan, S., and Venkatesh, B. (2015). Smart metering and functionalities of smart meters in smart grid - a review. In IEEE Electrical Power and Energy Conference, pages 138–145.

Baranski, M. and Voss, J. (2004). Genetic algorithm for pattern detection in nialm systems. In IEEE International Conference on Systems, Man and Cybernetics, volume 4, pages 3462–3468 vol.4.

Basu, K., Debusschere, V., Bacha, S., Maulik, U., and Bondyopadhyay, S. (2015). Nonintrusive load monitoring: A temporal multilabel classification approach. IEEE Transactions on Industrial Informatics, 11(1):262–270.

Blum, A. L. and Langley, P. (1997). Selection of relevant features and examples in machine learning. Artificial Intelligence, 97(1):245 – 271. Relevance.

Bouhouras, A. S., Gkaidatzis, P. A., Panagiotou, E., Poulakis, N., and Christoforidis, G. C. (2019). A nilm algorithm with enhanced disaggregation scheme under harmonic current vectors. Energy and Buildings, 183:392 – 407.

Chang, H.-H. (2012). Non-intrusive demand monitoring and load identification for energy management systems based on transient feature analyses. Energies, 5(11):4569–4589.

Dong, M., Meira, P. C. M., Xu, W., and Freitas, W. (2012). An event window based load monitoring technique for smart meters. IEEE Transactions on Smart Grid, 3(2):787– 796.

Drenker, S. and Kader, A. (1999). Nonintrusive monitoring of electric loads. IEEE Computer Applications in Power, 12(4):47–51.

Fernandes, R. A. S., da Silva, I. N., and Oleskovicz, M. (2013). Load profile identification interface for consumer online monitoring purposes in smart grids. IEEE Transactions on Industrial Informatics, 9(3):1507–1517.

Figueiredo, M. B., de Almeida, A., and Ribeiro, B. (2011). An experimental study on electrical signature identification of non-intrusive load monitoring (nilm) systems. In Adaptive and Natural Computing Algorithms, pages 31–40, Berlin, Heidelberg. Springer Berlin Heidelberg.

Garcia, F. D., Marafao, F. P., d. Souza, W. A., and d. Silva, L. C. P. (2017). Power metering: History and future trends. In IEEE Green Technologies Conference, pages 26–33.

Gillis, J. M., Alshareef, S. M., and Morsi, W. G. (2016). Nonintrusive load monitoring using wavelet design and machine learning. IEEE Transactions on Smart Grid, 7(1):320–328.

Hart, G. W. (1992). Nonintrusive appliance load monitoring. Proceedings of the IEEE, 80(12):1870–1891.

Huang, T. D., Wang, W., and Lian, K. (2015). A new power signature for nonintrusive appliance load monitoring. IEEE Transactions on Smart Grid, 6(4):1994–1995.

Kelly, J. and Knottenbelt, W. (2015). Neural nilm: Deep neural networks applied to energy disaggregation. In ACM International Conference on Embedded Systems for Energy-Efficient Built Environments, pages 55–64.

Lin, S., Zhao, L., Li, F., Liu, Q., Li, D., and Fu, Y. (2016). A nonintrusive load identification method for residential applications based on quadratic programming. Electric Power Systems Research, 133:241 – 248.

Makonin, S., Popowich, F., and Gill, B. (2013). The cognitive power meter: Looking beyond the smart meter. In IEEE Canadian Conference on Electrical and Computer Engineering, pages 1–5.

Marceau, M. and Zmeureanu, R. (2000). Nonintrusive load disaggregation computer program to estimate the energy consumption of major end uses in residential buildings. Energy Conversion and Management, 41(13):1389 – 1403.

Opris, I. and Caracasian, L. (2013a). On the implementation of the functionalities of smart metering systems. In International Symposium on Advanced Topics in Electrical Engineering, pages 1–6.

Opris, I. and Caracasian, L. (2013b). The relation between smart meters and electricity consumers. In International Conference on Environment and Electrical Engineering, pages 325–329.

Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12:2825–2830.

Platt, G., West, S., and Moore, T. (2015). The real-world challenges and opportunities of distributed generation. In IEEE Energy Conversion Congress and Exposition, pages 1112–1116.

Rathmair, M. and Haase, J. (2013). Load identification and management framework for private households. In Annual Conference of the IEEE Industrial Electronics Society, pages 5729–5734.

Ruzzelli, A. G., Nicolas, C., Schoofs, A., and O’Hare, G. M. P. (2010). Real-time recognition and profiling of appliances through a single electricity sensor. In IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks, pages 1–9.

Sadeghianpourhamami, N., Ruyssinck, J., Deschrijver, D., Dhaene, T., and Develder, C. (2017). Comprehensive feature selection for appliance classification in nilm. Energy and Buildings, 151:98–106.

Schirmer, P. A. and Mporas, I. (2019). Statistical and electrical features evaluation for electrical appliances energy disaggregation. Sustainability, 11(11).

Shaw, S. R., Leeb, S. B., Norford, L. K., and Cox, R. W. (2008). Nonintrusive load monitoring and diagnostics in power systems. IEEE Transactions on Instrumentation and Measurement, 57(7):1445–1454.

Souza, W. A., Marafao, F. P., Liberado, E. V., Diniz, I. S., and Serni, P. J. A. (2015). Power quality, smart meters and additional information from different power terms. IEEE Latin America Transactions, 13(1):158–165.

Souza, W. A., Marafão, F. P., Liberado, E. V., Simões, M. G., and Silva, L. C. P. (2018). A nilm dataset for cognitive meters based on conservative power theory and pattern recognition techniques. Journal of Control, Automation and Electrical Systems, 29(6):742–755.

Sultanem, F. (1991). Using appliance signatures for monitoring residential loads at meter panel level. IEEE Transactions on Power Delivery, 6(4):1380–1385.

Teshome, D. F., Huang, T. D., and Lian, K. (2016). Distinctive load feature extraction based on fryze’s time-domain power theory. IEEE Power and Energy Technology Systems Journal, 3(2):60–70.

Xiao, P. and Cheng, S. (2019). Neural network for NILM based on operational state change classification. arXiv.

Zhang, X., Shen, J., Yang, T., Tang, L., Wang, L., Liu, Y., and Xu, P. (2019). Smart meter and in-home display for energy savings in residential buildings: a pilot investigation in shanghai, china. Intelligent Buildings International, 11(1):4–26.
Publicado
15/10/2019
Como Citar

Selecione um Formato
SOUZA, Wesley; ALMEIDA, Tiago. Data Characterization for Electrical Load Disaggregation using Supervised Learning. In: ENCONTRO NACIONAL DE INTELIGÊNCIA ARTIFICIAL E COMPUTACIONAL (ENIAC), 16. , 2019, Salvador. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2019 . p. 226-237. DOI: https://doi.org/10.5753/eniac.2019.9286.