Non-intrusive Load Monitoring Method Based on Multi-task Learning Convolutional Network

DENG Xuhui; CHEN Zhong; YANG Kai; LIU Bo

doi:10.7500/AEPS20220517001

您当前的位置：

首页 >

文章列表页 >

Non-intrusive Load Monitoring Method Based on Multi-task Learning Convolutional Network

Basic Research | 更新时间：2025-06-16

- Non-intrusive Load Monitoring Method Based on Multi-task Learning Convolutional Network
- Automation of Electric Power Systems Vol. 47, Issue 8, Pages: 189-197(2023)
- 作者机构：
  
  东南大学电气工程学院，江苏省南京市 210096
- 作者简介：
- 基金信息：
  
  Natural Natural Science Foundation of China(52077035)
- DOI：10.7500/AEPS20220517001
  CLC：
- Received：17 May 2022，
  
  Revised：2022-11-10，
  
  Published Online：22 December 2022，
  
  Published：25 April 2023
- 稿件说明：
移动端阅览
邓旭晖,陈中,杨凯等.基于多任务学习卷积网络的非侵入式负荷监测方法[J].电力系统自动化,2023,47(8):189-197.

DENG Xuhui,CHEN Zhong,YANG Kai,et al.Non-intrusive Load Monitoring Method Based on Multi-task Learning Convolutional Network[J].Automation of Electric Power Systems,2023,47(8):189-197.
邓旭晖,陈中,杨凯等.基于多任务学习卷积网络的非侵入式负荷监测方法[J].电力系统自动化,2023,47(8):189-197. DOI： 10.7500/AEPS20220517001.

DENG Xuhui,CHEN Zhong,YANG Kai,et al.Non-intrusive Load Monitoring Method Based on Multi-task Learning Convolutional Network[J].Automation of Electric Power Systems,2023,47(8):189-197. DOI： 10.7500/AEPS20220517001.

摘要

随着居民用户对设备耗能管理需求的增长，低硬件成本的非侵入式负荷监测技术具有巨大的工程应用价值。目前，负荷监测深度网络存在输出假阳性功率片段问题，造成对电器开关运行状态的误判，影响负荷分解精度。为此，提出一种基于多任务学习的非侵入式负荷监测方法。首先，建立基于多分支卷积网络及改进后处理的负荷监测推理框架，在负荷功率分解网络主分支的基础上，设立开关状态判定分支网络，引入电器开关序列监督信息，通过多任务聚合损失函数调整网络训练时梯度反向传播过程，降低了运行状态误判率。然后，采用加权均值滤波对网络输出的开关状态评估值和功率输出值进行后处理，进一步降低负荷分解误差。最后，在UK-DALE数据集上开展对比实验，实验结果表明所提方法能够较好地分离出包含复杂功率特性的电器负荷，验证了方法的有效性。

Abstract

With the increasing demand for energy management of devices by residential users

the non-intrusive load monitoring (NILM) technology with low hardware cost has great engineering application value. At present

the load monitoring deep network has the problem of outputting false positive power segments

which leads to the misjudgment of on/off operation status of the electric appliance and affects the accuracy of load decomposition. Therefore

an NILM method based on multi-task learning is proposed. First

a load monitoring inference framework based on the multi-branch convolutional network and improved post-processing is established. Based on the main branch of the load power decomposition network

an on/off status evaluation subbranch network is set up

and the supervision information of on/off sequence of the electric appliance is introduced. The gradient backpropagation process during network training is adjusted by the multi-task aggregation loss function

which reduces the misjudgment rate of the operation status. Then

the weighted mean filtering is used to post-process the on/off status evaluation value and power output value of the network to further reduce the load decomposition error. Finally

the comparative experiments are carried out on the UK-DALE data set. The experimental results show that the proposed method can well separate the electric appliance load containing complex power characteristics

and the effectiveness of the method is verified.

关键词

Keywords

references

武昕，严萌，郭一凡，等 . 基于结构化特征图谱的组合支持向量机非侵入式负荷辨识［J］. 电力系统自动化， 2022 ， 46 （ 12 ）： 210 - 219 .

WU Xin ， YAN Meng ， GUO Yifan ， et al . Non-intrusive load identification by combined support vector machine based on structured characteristic spectrum ［J］. Automation of Electric Power Systems ， 2022 ， 46 （ 12 ）： 210 - 219 .

PAMULAPATI T ， MALLIPEDDI R ， LEE M . Multi-objective home appliance scheduling with implicit and interactive user satisfaction modelling ［J］. Applied Energy ， 2020 ， 267 ： 114690 .

HART G W . Nonintrusive appliance load monitoring ［J］. Proceedings of the IEEE ， 1992 ， 80 （ 12 ）： 1870 - 1891 .

刘国龙，赵俊华，文福拴，等 . 面向配用电侧负荷数据的深度端到端超分辨率感知方法［J］. 电力系统自动化， 2020 ， 44 （ 24 ）： 28 - 35 .

LIU Guolong ， ZHAO Junhua ， WEN Fushuan ， et al . Deep end-to-end super-resolution perception method for load data at distribution side ［J］. Automation of Electric Power Systems ， 2020 ， 44 （ 24 ）： 28 - 35 .

KONG W C ， DONG Z Y ， MA J ， et al . An extensible approach for non-intrusive load disaggregation with smart meter data ［J］. IEEE Transactions on Smart Grid ， 2018 ， 9 （ 4 ）： 3362 - 3372 .

冯人海，袁万琦，葛磊蛟 . 基于图信号交替优化的居民用户非侵入式负荷监测方法［J］. 中国电机工程学报， 2022 ， 42 （ 4 ）： 1355 - 1365 .

FENG Renhai ， YUAN Wanqi ， GE Leijiao . Non-intrusive load monitoring method for resident users based on alternating optimization in graph signal ［J］. Proceedings of the CSEE ， 2022 ， 42 （ 4 ）： 1355 - 1365 .

KHODAYAR M ， LIU G Y ， WANG J H ， et al . Spatiotemporal behind-the-meter load and PV power forecasting via deep graph dictionary learning ［J］. IEEE Transactions on Neural Networks and Learning Systems ， 2021 ， 32 （ 10 ）： 4713 - 4727 .

罗平，樊星驰，章坚民，等 . 基于电器运行状态和深度学习的非侵入式负荷分解［J］. 电力系统自动化， 2021 ， 45 （ 12 ）： 49 - 56 .

LUO Ping ， FAN Xingchi ， ZHANG Jianmin ， et al . Non-intrusive load decomposition based on operation state of electrical appliances and deep learning ［J］. Automation of Electric Power Systems ， 2021 ， 45 （ 12 ）： 49 - 56 .

KELLY J ， KNOTTENBELT W . Neural NILM： deep neural networks applied to energy disaggregation ［C］// 2nd ACM International Conference on Embedded Systems for Energy-Efficient Built Environments ， November 4-5 ， 2015 ， Seoul， South Korea .

BATRA N ， KUKUNURI R ， PANDEY A ， et al . Towards reproducible state-of-the-art energy disaggregation ［C］// 6th ACM International Conference on Systems for Energy-Efficient Buildings， Cities， and Transportation ， November 13-14 ， 2019 ， New York， USA .

KHODAYAR M ， WANG J H ， WANG Z Y . Energy disaggregation via deep temporal dictionary learning ［J］. IEEE Transactions on Neural Networks and Learning Systems ， 2020 ， 31 （ 5 ）： 1696 - 1709 .

廖荣文，刘刚，肖刚 . 基于时间模糊化长短时记忆的非侵入式负荷分解方法［J］. 电力系统自动化， 2021 ， 45 （ 24 ）： 73 - 80 .

LIAO Rongwen ， LIU Gang ， XIAO Gang . Non-intrusive load decomposition method based on time-fuzzified long short-term memory ［J］. Automation of Electric Power Systems ， 2021 ， 45 （ 24 ）： 73 - 80 .

CUI G C ， LIU B ， LUAN W P ， et al . Estimation of target appliance electricity consumption using background filtering ［J］. IEEE Transactions on Smart Grid ， 2019 ， 10 （ 6 ）： 5920 - 5929 .

CHEN K J ， ZHANG Y ， WANG Q ， et al . Scale- and context-aware convolutional non-intrusive load monitoring ［J］. IEEE Transactions on Power Systems ， 2020 ， 35 （ 3 ）： 2362 - 2373 .

KONG W C ， DONG Z Y ， WANG B ， et al . A practical solution for non-intrusive type Ⅱ load monitoring based on deep learning and post-processing ［J］. IEEE Transactions on Smart Grid ， 2020 ， 11 （ 1 ）： 148 - 160 .

HU J ， SHEN L ， ALBANIE S ， et al . Squeeze-and-excitation networks ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence ， 2020 ， 42 （ 8 ）： 2011 - 2023 .

KELLY J ， KNOTTENBELT W . The UK-DALE dataset， domestic appliance-level electricity demand and whole-house demand from five UK homes ［J］. Scientific Data ， 2015 ， 2 ： 150007 .

杨劲男，刘博，栾文鹏，等 . 多电器混叠场景NILM负荷事件优化匹配方法［J］. 电网技术， 2022 ， 46 （ 2 ）： 691 - 698 .

YANG Jinnan ， LIU Bo ， LUAN Wenpeng ， et al . Optimal load events matching of NILM in multi-appliance operation scenario ［J］. Power System Technology ， 2022 ， 46 （ 2 ）： 691 - 698 .

SHIN C ， RHO S ， LEE H ， et al . Data requirements for applying machine learning to energy disaggregation ［J］. Energies ， 2019 ， 12 （ 9 ）： 1696 .

BONFIGLI R ， FELICETTI A ， PRINCIPI E ， et al . Denoising autoencoders for Non-Intrusive Load Monitoring： improvements and comparative evaluation ［J］. Energy and Buildings ， 2018 ， 158 ： 1461 - 1474 .

ZHOU Z J ， XIANG Y M ， XU H ， et al . A novel transfer learning-based intelligent nonintrusive load-monitoring with limited measurements ［J］. IEEE Transactions on Instrumentation and Measurement ， 2021 ， 70 ： 1 - 8 .

王毅星 . 基于深度学习和迁移学习的电力数据挖掘技术研究［D］. 杭州：浙江大学， 2019 .

WANG Yixing . Power data mining technology based on deep learning and transfer learning ［D］. Hangzhou ： Zhejiang University ， 2019 .

ZHANG Y ， YANG Q . An overview of multi-task learning ［J］. National Science Review ， 2017 ， 5 （ 1 ）： 30 - 43 .

SHIN C ， JOO S ， YIM J ， et al . Subtask gated networks for non-intrusive load monitoring ［J］. Proceedings of the AAAI Conference on Artificial Intelligence ， 2019 ， 33 （ 1 ）： 1150 - 1157 .

ZHOU G ， LI Z ， FU M ， et al . Sequence-to-sequence load disaggregation using multiscale residual neural network ［J］. IEEE Transactions on Instrumentation and Measurement ， 2021 ， 70 ： 1 - 10 .

KELLY J . UK domestic appliance level electricity ［EB/OL］. ［ 2021-12-02 ］. https：//data.ukedc.rl.ac.uk/browse/edc/efficiency/residential/EnergyConsumption/Domestic/UK-DALE-2017 https://data.ukedc.rl.ac.uk/browse/edc/efficiency/residential/EnergyConsumption/Domestic/UK-DALE-2017 .

PICCIALLI V ， SUDOSO A M . Improving non-intrusive load disaggregation through an attention-based deep neural network ［J］. Energies ， 2021 ， 14 （ 4 ）： 847 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Transient Frequency Safety Assessment Based on Independent Spatial-temporal Feature Extraction and Grouped Enhancement

Short-term Load Probabilistic Forecasting Based on Conditional Enhanced Diffusion Model

Ultra-short-term Regional Load Forecasting Based on Spatio-Temporal Graph Attention Network

Electric Vehicle Charging Load Prediction Based on Dynamic Adaptive Graph Neural Network

Related Author

YANG Wenxian

SHI Fang

SONG Xuemeng

TIAN Shuoshuo

NIE Liqiang

CHENG Dingyi

LIU Jinxiang

ZHANG Jiangfeng

Related Institution

State Grid Shandong Electric Power Research Institute

School of Electrical Engineering, Shandong University

School of Computer Science and Technology, Shandong University

National Joint Engineering Laboratory of Power Grid with Electric Vehicles (Shandong University)

Jinhua Institute of Zhejiang University

⁰