Knowledge Distillation for Scalable Nonintrusive Load Monitoring

被引:2
|
作者
Tanoni, Giulia [1 ]
Stankovic, Lina [2 ]
Stankovic, Vladimir [2 ]
Squartini, Stefano [1 ]
Principi, Emanuele [1 ]
机构
[1] Univ Politecn Marche, Dept Informat Engn, I-60131 Ancona, Italy
[2] Univ Strathclyde, Dept Elect & Elect Engn, Glasgow City G1 1XQ, Scotland
关键词
Deep learning (DL); knowledge distillation (KD); multilabel appliance classification; nonintrusive load monitoring (NILM); weak supervision; CLASSIFICATION;
D O I
10.1109/TII.2023.3328436
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Smart meters allow the grid to interface with individual buildings and extract detailed consumption information using nonintrusive load monitoring (NILM) algorithms applied to the acquired data. Deep neural networks, which represent the state of the art for NILM, are affected by scalability issues since they require high computational and memory resources, and by reduced performance when training and target domains mismatched. This article proposes a knowledge distillation approach for NILM, in particular for multilabel appliance classification, to reduce model complexity and improve generalization on unseen data domains. The approach uses weak supervision to reduce labeling effort, which is useful in practical scenarios. Experiments, conducted on U.K.-DALE and REFIT datasets, demonstrated that a low-complexity network can be obtained for deployment on edge devices while maintaining high performance on unseen data domains. The proposed approach outperformed benchmark methods in unseen target domains achieving a F-1-score 0.14 higher than a benchmark model 78 times more complex.
引用
收藏
页码:4710 / 4721
页数:12
相关论文
共 50 条
  • [41] Deep attention and generative neural networks for nonintrusive load monitoring
    Regan, Jacob
    Saffari, Mohsen
    Khodayar, Mahdi
    ELECTRICITY JOURNAL, 2022, 35 (05):
  • [42] Enhancing Electricity Audits in Residential Buildings with Nonintrusive Load Monitoring
    Berges, Mario E.
    Goldman, Ethan
    Matthews, H. Scott
    Soibelman, Lucio
    JOURNAL OF INDUSTRIAL ECOLOGY, 2010, 14 (05) : 844 - +
  • [43] Adversarial Data Anomaly Detection and Calibration for Nonintrusive Load Monitoring
    Yang, Haosen
    Liang, Zipeng
    Shi, Xin
    Cheng, Joseph
    Liang, Jian
    Dong, Hanjiang
    Chung, C. Y.
    IEEE INTERNET OF THINGS JOURNAL, 2025, 12 (01): : 544 - 555
  • [44] Nonintrusive Load Monitoring Based on Complementary Features of Spurious Emissions
    Su, Donglin
    Shi, Qian
    Xu, Hui
    Wang, Wang
    ELECTRONICS, 2019, 8 (09)
  • [45] Nonintrusive Load Monitoring Based on Self-Supervised Learning
    Chen, Shuyi
    Zhao, Bochao
    Zhong, Mingjun
    Luan, Wenpeng
    Yu, Yixin
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2023, 72
  • [46] An Energy Prediction Approach for a Nonintrusive Load Monitoring in Home Appliances
    Buddhahai, Bundit
    Wongseree, Waranyu
    Rakkwamsuk, Pattana
    IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, 2020, 66 (01) : 96 - 105
  • [47] A Novel Feature Extraction Method for Nonintrusive Appliance Load Monitoring
    Chahine, Khaled
    Drissi, Khalil El Khamlichi
    APPLIED COMPUTATIONAL INTELLIGENCE AND SOFT COMPUTING, 2013, 2013
  • [48] Multi-head Attention Networks for Nonintrusive Load Monitoring
    Lin, Nan
    Zhou, Binggui
    Yang, Guanghua
    Ma, Shaodan
    2020 IEEE INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING, COMMUNICATIONS AND COMPUTING (IEEE ICSPCC 2020), 2020,
  • [49] A Smart Socket for Real-Time Nonintrusive Load Monitoring
    Wu, Zhao
    Wang, Chao
    Xiong, Linyun
    Li, Ruiheng
    Wu, Tao
    Zhang, Huaiqing
    IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2023, 70 (10) : 10618 - 10627
  • [50] An Adaptive Two-Stage Load Event Detection Method for Nonintrusive Load Monitoring
    Luan, Wenpeng
    Liu, Zishuai
    Liu, Bo
    Yu, Yixin
    Hou, Yufan
    IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, 2022, 71