Non-intrusive Load Monitoring Method based on Improved Temporal Convolutional Network

LÜ li; LEI Shaobo; FAN Haoyan; Niu Hong

English Version Journal Login

您当前的位置：

首页 >

文章列表页 >

Non-intrusive Load Monitoring Method based on Improved Temporal Convolutional Network

更新时间：2026-05-21

- Non-intrusive Load Monitoring Method based on Improved Temporal Convolutional Network
- Vol. 63, Issue 5, (2026)
- 作者机构：
  
  内蒙古电力(集团)有限责任公司
- 作者简介：
- 基金信息：
- DOI：
  CLC：
- Published：2026
- 稿件说明：
移动端阅览
LÜ li, LEI Shaobo, FAN Haoyan, et al. Non-intrusive Load Monitoring Method based on Improved Temporal Convolutional Network[J]. 2026, 63(5).
DOI：

LÜ li, LEI Shaobo, FAN Haoyan, et al. Non-intrusive Load Monitoring Method based on Improved Temporal Convolutional Network[J]. 2026, 63(5). DOI：

摘要

针对当前NILM技术在处理复杂电器负荷和长时间序列时准确性不足及计算效率低的问题，文中提出了一种TCN-1DEMA-MTL模型。首先，模型利用TCN模块提取数据的上下文特征并实现多层次特征融合；然后，引入1DEMA注意力机制以动态加权优化特征提取效果；并且，采用MTL框架将负荷分解与状态识别任务集成以进一步提升预测精度。实验阶段与现有的双向LSTM、S2P模型和LDwA模型进行了比较，所提模型在各项性能指标上均表现出显著优势。实验结果验证了该方案的有效性和实用性，为居民电力负荷监测的发展提供了有力的借鉴。

Abstract

In view of the problems of insufficient accuracy and low computational efficiency of the current NILM technology in processing complex electrical loads and long time series

a TCN-1DEMA-MTL model is proposed. First

the model uses the TCN module to extract the contextual features of the data and realize multi-level feature fusion; then

the 1DEMA attention mechanism is introduced to dynamically weight and optimize the feature extraction effect; and the MTL framework is used to integrate the load decomposition and state recognition tasks to further improve the prediction accuracy. In the experimental stage

the proposed model is compared with the existing bidirectional LSTM

S2P model and LDwA model

and all performance indicators show significant advantages. The experimental results verify the effectiveness and practicality of the scheme

which provides a powerful reference for the development of residential power load monitoring.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

CUSUM Transient Event Detection Algorithm Based on Compound Sliding Window

An approach for non-intrusive load event detection based on Bayesian iteration

Nonintrusive household load identification method based on random forest

Insulator detection model based on feature Pyramid and multi-task learning

Power quality disturbance recognition method based on multi-task learning and one-dimensional convolutional neural network

Related Author

Hu Wenshan

Zhou Hong

Zhou Dongguo

Deng Shicong

Zhang Heng

Shi Shuaibin

Zhao Jiaqing

Fang Guoquan

Related Institution

武汉大学动力与机械学院

深圳供电局有限公司

国网江苏省电力有限公司苏州供电公司

东南大学电气工程学院

武汉大学电气与自动化学院

AI问答

⁰