微型电脑应用2025，Vol.41Issue(9)：5-8,4.
基于关联分析和深度置信网络的电网智能故障诊断方法
An Intelligent Power Grid Fault Diagnosis Method Based on Correlation Analysis and Deep Belief Network
摘要
Abstract
In this paper,an intelligent fault diagnosis method for power grids based on correlation analysis and deep belief net-work is proposed.Traditional fault diagnosis methods have limitations in dealing with complex systems,which require exten-sive domain knowledge and manual feature extraction.The proposed method aims to overcome these limitations and improve the accuracy and efficiency of fault diagnosis.The feature selection capability of correlation analysis is used for dimensionality re-duction,and deep belief network is used for fault identification.To verify the performance of the proposed fault diagnosis mod-el,experiments are conducted using a personal computer transient analyzer,and the fault diagnosis results of back propagation neural networks and support vector machines are compared with the method in this paper,and the proposed method achieves an average accuracy of 95％in training and 0.01 convergence rate,demonstrating the effectiveness of using deep learning and cor-relation analysis for intelligent fault diagnosis in power grids.关键词
电网/故障诊断/关联分析/深度学习/深度置信网络Key words
power grid/fault diagnosis/correlation analysis/deep learning/deep belief network分类
信息技术与安全科学引用本文复制引用
杨定坤,尹徐珊,孙澄宇,董平,钱冠军..基于关联分析和深度置信网络的电网智能故障诊断方法[J].微型电脑应用,2025,41(9):5-8,4.基金项目
国家自然科学基金(61502229) （61502229）
