东北电力技术2026，Vol.47Issue(1)：43-48,6.
基于改进EPO-BP神经网络的变压器故障诊断方法
Transformer Fault Diagnosis Method Based on Improved EPO-BP Neural Network
王帆 1王茜雯 2柯渊 1宁鑫淼 1安睿1
作者信息
- 1. 国网宁夏电力有限公司超高压公司,宁夏 银川 750011
- 2. 国网北京市电力公司检修分公司,北京 100069
- 折叠
摘要
Abstract
The safe and stable operation of transformers is a basic requirement for ensuring the quality of electric power.According to the existing transformer fault diagnosis methods suffering from poor adaptability and low accuracy,it proposes a transformer fault diag-nosis method based on an improved(emperor penguin optimizer)EPO-BP(back propagation)neural network.Firstly,in response to the problems of slow convergence speed and tendency to fall into local optimum during the iterative process of EPO,it introduces a driving training mechanism.It introduces a driving training mechanism to enhance the accuracy of emperor penguins movement trajec-tories towards clusters and their operational efficiency.Secondly,based on the improved EPO algorithm,it optimizes the weights and thresholds of the BP neural network to enhance the performance and classification accuracy of the model.It collectes and divides the normal operation and fault operation data of the transformer into the training set and the test set.Finally,it carries out fault diagnosis of the transformer based on the improved EPO-BP neural network model.The results show that this fault diagnosis model has stronger adaptability and higher classification accuracy.关键词
变压器/故障诊断/帝企鹅优化器/驾驶训练机制/BP神经网络Key words
transformer/fault diagnosis/EPO/driving training mechanism/BP neural network分类
信息技术与安全科学引用本文复制引用
王帆,王茜雯,柯渊,宁鑫淼,安睿..基于改进EPO-BP神经网络的变压器故障诊断方法[J].东北电力技术,2026,47(1):43-48,6.
