Desagregação de Energia Baseada em Deep Learning e Transformação de Wavelet
Resumo
A desagregação de energia é uma área que busca identificar o consumo individual de diferentes aparelhos usando apenas o sinal agregado medido a partir de um único ponto. Este trabalho propõe uma rede neural treinada com dados reduzidos Wavelets para realizar a desagregação de energia. Além da desagregação, que geralmente apenas obtemos uma resposta binária identificando o momento de ativação do aparelho, também estamos interessados em estimar o valor de consumo do aparelho. Consideramos o conjunto de dados UK-DALE para realizar nossos experimentos. Usando nossa abordagem, em comparação com outro trabalho bem estabelecido, alcançamos melhorias por aparelho de 27,8% (F1-score) no processo de desagregação e 11,4% (acurácia estimada) no valor de consumo do aparelho.
Downloads
Referências
Batra, N., Kelly, J., Parson, O., Dutta, H., Knottenbelt, W., Rogers, A., Singh, A., and Srivastava, M. (2014). Nilmtk: an open source toolkit for non-intrusive load monitoring. In Proceedings of the 5th international conference on Future energy systems, pages 265–276. ACM.
Chang, H.-H., Lian, K.-L., Su, Y.-C., and Lee, W.-J. (2013). Power-spectrum-based wavelet transform for nonintrusive demand monitoring and load identification. IEEE Transactions on Industry Applications, 50(3):2081–2089.
Clevert, D., Unterthiner, T., and Hochreiter, S. (2015). Fast and accurate deep network learning by exponential linear units (elus). In Bengio, Y. and LeCun, Y., editors, 4th Int. Conf. on Learn. Represent., ICLR 2016.
Eddy, S. R. (2004). What is a hidden Markov model? Nature Biotechnology, 22(10):1315–1316. Gillis, J. M. and Morsi, W. G. (2016). Non-intrusive load monitoring using semisupervised machine learning and wavelet design. IEEE Transactions on Smart Grid, 8(6):2648–2655.
Hecht-Nielsen, R. (1992). Theory of the backpropagation neural network. In Neural networks for perception, pages 65–93. Elsevier.
Hochreiter, S. (1998). The Vanishing Gradient Problem During Learning Recurrent Neural Nets and Problem Solutions. International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems, 6(2):107–116.
Kelly, J. and Knottenbelt, W. (2015a). Neural nilm: Deep neural networks applied to energy disaggregation. In Proceedings of the 2nd ACM International Conference on Embedded Systems for Energy-Efficient Built Environments, pages 55–64. ACM.
Kelly, J. and Knottenbelt, W. (2015b). The uk-dale dataset, domestic appliance-level electricity demand and whole-house demand from five uk homes. Scientific data, 2:150007.
Kingma, D. P. and Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
Kong, W., Dong, Z. Y., Wang, B., Zhao, J., and Huang, J. (2019). A practical solution for non-intrusive type ii load monitoring based on deep learning and post-processing. IEEE Transactions on Smart Grid, 11(1):148–160.
Muller, M. (2007). Dynamic time warping. ¨ Information retrieval for music and motion, pages 69–84. Roser, M., Ritchie, H., and Ortiz-Ospina, E. (2013). World population growth. Most recent substantial revision in May 2019.
Santos, E., Freitas, C., and Aquino, A. (2019a). Uma arquitetura de aprendizagem profunda para desagregação de energia. Anais da XIX Escola Regional de Computação Bahia, Alagoas e Sergipe, pages 314–322, Porto Alegre, RS, Brasil. SBC.
Santos, E., Ramos, G., and Aquino, A. (2021). An energy disaggregation approach based on deep neural network and wavelet transform. IEEE Transactions on Industrial Informatics, pages 1–1.
Santos, E. G., Freitas, C. G. S., and Aquino, A. L. L. (2019b). A deep learning approach for energy disaggregation considering embedded devices. In 2019 IX Brazilian Symposium on Computing Systems Engineering (SBESC), pages 1–8.
Shin, C., Joo, S., Yim, J., Lee, H., Moon, T., and Rhee, W. (2019). Subtask gated networks for non-intrusive load monitoring. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 33, pages 1150–1157.
Zhang, C., Zhong, M., Wang, Z., Goddard, N., and Sutton, C. (2018). Sequence-to-point learning with neural networks for non-intrusive load monitoring. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32.
Zhao, B., Stankovic, L., and Stankovic, V. (2015). Blind non-intrusive appliance load monitoring using graph-based signal processing. In 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP), pages 68–72. IEEE.
