邢海青,郭瑞峰,杨浙川,熊小雨,施永涛
XING Haiqing,GUO Ruifeng,YANG Zhechuan,XIONG Xiaoyu,SHI Yongtao

• 1:国网浙江杭州市余杭区供电公司
• 2:杭州电子科技大学自动化学院

摘要(Abstract)：

非侵入式负荷监测技术不仅能将电能流向透明化，还能简化智能电表安装流程，从而有效降低负荷监测成本。为提高非侵入式负荷监测中的负荷识别准确性，提出了基于数据扩充与无阈值递归图的非侵入式负荷识别方法。采用去噪扩散概率模型对小样本负荷数据进行数据扩充，以提升负荷识别方法的鲁棒性；通过去除递归图的Heaviside函数实现无阈值递归图以高效表征负荷特征，并结合Transformer深度学习网络构建负荷识别框架。将所提出的方法应用到3个实测数据集中，实验结果表明，所提方法能有效提高负荷识别准确度，改善分类效果。
Non-intrusive load monitoring(NILM) not only makes the flow of electric energy transparent but also simplifies the installation process of smart meters, effectively reducing the cost of load monitoring. To enhance the accuracy of load recognition in NILM, a method for load recognition based on data augmentation and threshold-free recurrence plot(RP) is proposed. a denoising diffusion probability model(DDPM) is utilized to augment the load data of small samples to enhance the robustness of the load recognition method. Furthermore, a threshold-free RP, achieved by removing the Heaviside function of the recurrence graph, efficiently represents load characteristics.This is combined with a Transformer deep learning network to construct a load recognition framework. The proposed method is applied to three real-world datasets, and experimental results demonstrate its effectiveness in improving load recognition accuracy and enhancing classification performance.

关键词(KeyWords)： 非侵入式负荷监测;数据扩充;负荷识别;深度学习;递归图
NILM;data augmentation;load identification;deep learning;RP

Abstract:

Keywords:

基金项目(Foundation): 浙江省重点研发计划（2021C01144）;; 浙江大有集团有限公司科技项目（2021-DY16）

作者(Author): 邢海青,郭瑞峰,杨浙川,熊小雨,施永涛
XING Haiqing,GUO Ruifeng,YANG Zhechuan,XIONG Xiaoyu,SHI Yongtao

DOI: 10.19585/j.zjdl.202406010

参考文献(References)：

• [1] CHANG H H,CHEN K L,TSAI Y P,et al.A new measurement method for power signatures of nonintrusive demand monitoring and load identification[J].IEEE Transactions on Industry Applications,2012,48(2):764-771.
• [2]董哲，陈玉梁，薛同来，等.基于全局与滑动窗口结合的Attention机制的非侵入式负荷分解算法[J].电测与仪表，2023,60(11):74-80.DONG Zhe,CHEN Yuliang,XUE Tonglai,et al. Nonintrusive load monitoring algorithm based on Attention mechanism combined with global and sliding window[J].Electrical Measurement&Instrumentation,2023,60(11):74-80.
• [3]严萌，于雅雯，王玲静，等.基于多特征联合稀疏表达的SOM-K-means非侵入负荷辨识[J].电力建设，2023,44(5):61-71.YAN Meng,YU Yawen,WANG Lingjing,et al.SOM-Kmeans Non-intrusive load identification based on multi feature joint sparse expression[J].Electric Power Construction,2023,44(5):61-71.
• [4]余贻鑫，刘博，栾文鹏.非侵入式居民电力负荷监测与分解技术[J].南方电网技术，2013,7(4):1-5.YU Yixin,LIU Bo,LUAN Wenpeng. Nonintrusive residential load monitoring and decomposition technology[J].Southern Power System Technology,2013,7(4):1-5.
• [5]邱一昊.基于暂态过程的非侵入式家用电器负荷监测方法的研究[D].苏州：苏州大学，2019.QIU Yihao.Research on load monitoring method of non intrusive household appliances based on transient process[D].Suzhou:Soochow University,2019.
• [6]刘睿迪，汪震.基于低频功率数据处理的负荷分解方法[J].能源工程，2021(6):79-84.LIU Ruidi,WANG Zhen. Load decomposition method based on low frequency power data processing[J].Energy Engineering,2021(6):79-84.
• [7] PAN Y G,LIU K,SHEN Z Y,et al. Sequence-tosubsequence learning with conditional Gan for power disaggregation[C]//ICASSP 2020-2020 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).May 4-8,2020.Barcelona,Spain.IEEE,2020:3202-3206.
• [8] JIA D N,LI Y X,DU Z H,et al.Non-intrusive load identification using reconstructed voltage-current images[J].IEEE Access,2021,9:77349-77358.
• [9] WENNINGER M,BAYERL S P,MAIER A,et al.Recurrence plot spacial pyramid pooling network for appliance identification in non-intrusive load monitoring[C]//2021 20th IEEE International Conference on Machine Learning and Applications(ICMLA). December 13-16,2021.Pasadena,CA,USA.IEEE,2021:108-115.
• [10] YAN L,SHEIKHOLESLAMI M,GONG W L,et al.Challenges for real-world applications of nonintrusive load monitoring and opportunities for machine learning approaches[J].The Electricity Journal,2022,35(5):107136.
• [11]王毓琦，高嵩，万校宏，等.电网负荷分类评价反馈算法研究[J].山东电力技术，2022,49(3):20-24.WANG Yuqi,GAO Song,WAN Xiaohong,et al. Research on feedback algorithm of power load classification evaluation[J]. Shandong Electric Power,2022,49(3):20-24.
• [12]汪繁荣，向堃，吴铁洲.基于聚类特征及seq2seq深度CNN的家电负荷识别方法研究[J].电测与仪表，2023,60(10):79-86.WANG Fanrong,XIANG Kun,WU Tiezhou. Research on household appliance load identification method based on clustering features and seq2seq depth CNN[J].Electrical Measurement&Instrumentation,2023,60(10):79-86.
• [13]朱海南，李丰硕，孙华忠，等.基于改进AlexNet-GRU深度学习网络的配电网短期负荷预测方法[J].电力电容器与无功补偿，2023,44(4):48-54.ZHU Hainan,LI Fengshuo,SUN Huazhong,et al.Shortterm load prediction method of distribution network based on improved AlexNet-GRU deep learning network[J].Power Capacitor&Reactive Power Compensation,2023,44(4):48-54.
• [14]封钰，宋佑斌，金晟，等.基于随机森林算法和粗糙集理论的改进型深度学习短期负荷预测模型[J].发电技术，2023,44(6):889-895.FENG Yu,SONG Youbin,JIN Sheng,et al. Improved deep learning model for forecasting short-term load based on random forest algorithm and rough set theory[J].Power Generation Technology,2023,44(6):889-895.
• [15]刘晓东，常飞，王璇，等.基于GBRT和LGBM的多能负荷组合预测方法[J].电力电容器与无功补偿，2023,44(3):97-102.LIU Xiaodong,CHANG Fei,WANG Xuan,et al.Multienergy load combined forecasting method based on GBRT and LGBM[J].Power Capacitor&Reactive Power Compensation,2023,44(3):97-102.
• [16]黄河，王燕，姜念，等.考虑用户诉求差异的居民可控负荷资源优化控制[J].发电技术，2023,44(6):896-908.HUANG He,WANG Yan,JIANG Nian,et al. Optimal control of residents′controllable load resources considering different demands of users[J].Power Generation Technology,2023,44(6):896-908.
• [17] BERRETTONI G,BOURELLY C,CAPRIGLIONE D,et al.Preliminary sensitivity analysis of combinatorial optimization(CO)for NILM applications:effect of the meter accuracy[C]//2021 IEEE 6th International Forum on Research and Technology for Society and Industry(RTSI).September 6-9,2021.Naples,Italy.IEEE,2021:486-490.
• [18] GURBUZ F B,BAYINDIR R,VADI S.Comprehensive non-intrusive load monitoring process:device event detection,device feature extraction and device identification using KNN,random forest and decision tree[C]//2021 10th International Conference on Renewable Energy Research and Application(ICRERA).September 26-29,2021.Istanbul,Turkey.IEEE,2021:447-452.
• [19] LIU H,WU H P,YU C M.A hybrid model for appliance classification based on time series features[J].Energy and Buildings,2019,196:112-123.
• [20] DAVIES P,DENNIS J,HANSOM J,et al.Deep neural networks for appliance transient classification[C]//ICASSP 2019-2019 IEEE International Conference on Acoustics,Speech and Signal Processing(ICASSP).May12-17,2019.Brighton,United Kingdom.IEEE,2019:8320-8324.
• [21] WU Q,WANG F.Concatenate convolutional neural networks for non-intrusive load monitoring across complex background[J].Energies,2019,12(8):1572.
• [22] ZHENG Z,CHEN H N,LUO X W.A supervised eventbased non-intrusive load monitoring for non-linear appliances[J].Sustainability,2018,10(4):1001.
• [23] SOHL-DICKSTEIN J,WEISS E A,MAHESWARANATHAN N,et al.Deep unsupervised learning using nonequilibrium thermodynamics[C]//Proceedings of the 32nd International Conference on International Conference on Machine Learning-Volume 37.July 6-11,2015,Lille,France.ACM,2015:2256-2265.
• [24] HO J,JAIN A,ABBEEL P.Denoising diffusion probabilistic models[C]//Proceedings of the 34th International Conference on Neural Information Processing Systems.December 6-12,2020,Vancouver,BC,Canada. ACM,2020:6840-6851.
• [25] NICHOL A Q,DHARIWAL P.Improved denoising diffusion probabilistic models[C]//International Conference on Machine Learning,2021:8162-8171.
• [26] KINGMA D P,WELLING M.Auto-encoding variational bayes[J].arXiv preprint arXiv:1312.6114,2013.
• [27] KULLBACK S,LEIBLER R A.On information and sufficiency[J].The Annals of Mathematical Statistics,1951,22(1):79-86.
• [28] SONG Y,ERMON S.Generative modeling by estimating gradients of the data distribution[J]. Advances in Neural Information Processing Systems,2019,32.
• [29] ZAGORUYKO S,KOMODAKIS N. Wide residual networks[J].arXiv preprint arXiv:1605.07146,2016.
• [30] ECKMANN J P,KAMPHORST S O,RUELLE D.Recurrence plots of dynamical systems[M]//Turbulence,Strange Attractors and Chaos. World Scientific,1995:441-445.
• [31] VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.December 4-9,2017,Long Beach,California,USA.ACM,2017:6000-6010.
• [32] KAHL M,KRAUSE V,HACKENBERG R,et al.Measurement system and dataset for in-depth analysis of appliance energy consumption in industrial environment[J].Tm-Technisches Messen,2019,86(1):1-13.
• [33] MEDICO R,DE BAETS L,GAO J K,et al.A voltage and current measurement dataset for plug load appliance identification in households[J].Scientific Data,2020,7:49.
• [34] KAHL M,HAQ A U,KRIECHBAUMER T,et al.Whited-a worldwide household and industry transient energy data set[C]//3rd International Workshop on NonIntrusive Load Monitoring,2016:1-4.
• [35] KARL P.LIII.On lines and planes of closest fit to systems of points in space[J]. Philosophical Magazine Series 6,1901,2(11):559-572.
• [36] VAN DER MAATEN L,HINTON G. Visualizing data using t-SNE[J]. Journal of Machine Learning Research,2008,9:2579-2625.
• [37] GULRAJANI I,AHMED F,ARJOVSKY M,et al.Improved training of Wasserstein GANs[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems. December 4-9,2017,Long Beach,California,USA.ACM,2017:5769-5779.
• [38] MATASSINI L,KANTZ H,HO?YST J,et al.Optimizing of recurrence plots for noise reduction[J].Phys Rev E Stat Nonlin Soft Matter Phys,2002,65(2):021102.