Non-intrusive load identification based on improved non-local attention module

被引：0
|
作者：
Wu H. [1 ]
Zhang Y. [1 ]
Wang Y. [2 ]
Ma Q. [1 ]
机构：
[1] School of Control Science and Engineering, Shandong University, Jinan
[2] Northern Suburbs Branch, Jinan Thermo Electron Group Co. Ltd., Jinan
来源：
Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition) | 2023年 / 51卷 / 05期
关键词：
convolutional neural network; deep learning; non-intrusive load monitoring; non-local neural network; residual learning;
D O I：
10.13245/j.hust.229112
中图分类号：
学科分类号：
摘要：
Aiming at the problems of high model complexity，large number of parameters，and weak ability to obtain long-range dependencies between features in the field of deep learning-based non-intrusive load identification，a lightweight load identification model based on attention mechanism was proposed．The model took the device current information of low time dimension as input，and built a lightweight time residual convolutional neural network by introducing an improved non-local attention module to model the feature relationship of currents in different times．Experiments on the public PLAID (plug-level appliance identification data set) and WHITED (worldwide household and industry transient energy data set) show that the model computational complexity is as low as 4×105 and the number of parameters is less than 5.2×104，when the recognition rate of electrical appliances reaches 97.32% and 99.32%，respectively. © 2023 Huazhong University of Science and Technology. All rights reserved.
引用
收藏
页码：19 / 25
页数：6
相关论文
共 25 条
  • [1] HART G W．, Nonintrusive appliance load monitoring[J], Proceedings of the IEEE, 80, 12, pp. 1870-1891, (1992)
  • [2] FAUSTINE A, PEREIRA L．, Improved appliance classification in non-intrusive load monitoring using weighted recurrence graph and convolutional neural networks[J], Energies, 13, 13, pp. 3374-3374, (2020)
  • [3] CIANCETTA F，, BUCCI G，, FIORUCCI E, A new convolutional neural network-based system for NILM applications[J], IEEE Transactions on Instrumentation and Measurement, 70, pp. 1-12, (2020)
  • [4] GURBUZ F B，, BAYINDIR R, BULBUL H I．, A brief review of non-intrusive load monitoring and its impact on social life[C], Proc of 2021 9th International Conference on Smart Grid, pp. 289-294, (2021)
  • [5] BERGES M, GOLDMAN E, MATTHEWS H S, User-centered nonintrusive electricity load monitoring for residential buildings[J], Journal of Computing in Civil Engineering, 25, 6, pp. 471-480, (2011)
  • [6] KHAN M M R，, SIDDIQUE M, SAKIB S．, Non-intrusive electrical appliances monitoring and classification using k-nearest neighbors[C], Proc of 2019 2nd International Conference on Innovation in Engineering and Technology (ICIET), pp. 1-5, (2019)
  • [7] LU C, MA L, XU T, Non-intrusive load monitoring method based on improved differential evolution algorithm[C], Proc of 2019 11th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA), pp. 279-283, (2019)
  • [8] DUARTE C，, DELMAR P，, GOOSSEN K W, Non-intrusive load monitoring based on switching voltage transients and wavelet transforms[C], Proc of 2012 Future of Instrumentation International Workshop (FIIW), pp. 1-4, (2012)
  • [9] JARAMILLO A, LAVERTY D M, DEL RINCON J M, Supervised non-intrusive load monitoring algorithm for electric vehicle identification[C], Proc of 2020 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), pp. 1-6, (2020)
  • [10] VAYGAN E K，, RAJABI R，, ESTEBSARI A．, Short-term load forecasting using time pooling deep recurrent neural network[C], Proc of 2021 IEEE International Conference on Environment and Electrical Engineering and 2021 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS Europe), pp. 1-5, (2021)
123