机电工程技术2025，Vol.54Issue(10)：145-149,5.DOI:10.3969/j.issn.1009-9492.2025.10.030
基于CNN-TCN的电能质量扰动信号识别模型
Power Quality Disturbance Signal Identification Model Based on CNN-TCN
于红 1任行 2俞瑞龙2
作者信息
- 1. 湖南经研电力设计有限公司,长沙 410116
- 2. 昆明理工大学电力工程学院,昆明 650500
- 折叠
摘要
Abstract
The identification and classification of power quality disturbance signals is a key scientific issue in power system monitoring.Traditional methods rely on manual feature extraction,which is inefficient and susceptible to subjective factors.A hybrid model based on convolutional neural network(CNN)and temporal convolutional network(TCN)is proposed,combined with a self-attention mechanism,aiming to solve the automatic and efficient classification of power quality disturbance signals.By using CNN to accurately extract spatial features and TCN to enhance the analysis and processing of time series data,this model has significantly improved the accuracy and efficiency of identifying power quality disturbance signals.Experimental results show that the model exhibits excellent performance in processing a variety of power quality disturbance signals,with a test accuracy of 98.7%and excellent generalization ability.The hybrid model has not only successfully verified its effectiveness in classifying power quality disturbance signal,but also demonstrated the huge application potential of hybrid deep learning technology in the field of power system monitoring.关键词
电能质量扰动信号识别/分类模型/卷积神经网络/时序卷积网络/卷自注意力机制Key words
power quality disturbance signal identification/classification model/convolutional neural network/sequential convolutional network/self-attention mechanism分类
信息技术与安全科学引用本文复制引用
于红,任行,俞瑞龙..基于CNN-TCN的电能质量扰动信号识别模型[J].机电工程技术,2025,54(10):145-149,5.
