基于PCA特征重构的油浸式变压器故障诊断方法研究  

PCA-based Feature Reconstruction Method of Oil Immersed Transformer Trouble Diagnosis

作　　者：高翔 靳智嵩 赵雪静 程鹏 GAO Xiang;JIN Zhisong;ZHAO Xuejing;CHENG Peng(State Grid Jibei Electric Power Co.,Ltd.Langfang Branch,Langfang 065000,China)

机构地区：[1]国网冀北电力有限公司廊坊供电公司,廊坊065000

出　　处：《自动化与仪表》2025年第4期61-66,共6页Automation & Instrumentation

摘　　要：油浸式电力变压器的故障诊断对保障电网安全稳定运行至关重要。为此,研究采用主成分分析法对油浸式变压器的故障数据进行特征重构。并利用灰狼优化算法来优化支持向量机的参数,搭建基于改进支持向量机的故障诊断模型。结果表明,在特征重构后的数据测试中,故障算法的准确率约高出重构前2%~4%。所提故障诊断模型的诊断准确率为95.68%,比传统支持向量机算法高出5.94%,且运行时间也更短。经实验测试,所提特征重构方法和故障诊断模型满足了变压器故障诊断上的应用需求。研究结果有助于提高电力系统的可靠性和安全性,优化维护资源的分配。The fault diagnosis of oil immersed power transformers is crucial for ensuring the safe and stable operation of the power grid.Therefore,the principal component analysis method was used to reconstruct the fault data of oil immersed transformers.And use grey wolf optimization algorithm to optimize the parameters of support vector machine,and build a fault diagnosis model based on improved support vector machine.The results show that in the data testing after feature reconstruction,the accuracy of the fault algorithm is about 2%~4%higher than before reconstruction.The diagnostic accuracy of the proposed fault diagnosis model is 95.68%,which is 5.94%higher than traditional support vector machine algorithms,and the running time is also shorter.The feature reconstruction method and fault diagnosis model proposed through experimental testing meet the application requirements of transformer fault diagnosis.The research results contribute to improving the reliability and safety of the power system,optimizing the allocation of maintenance resources.

关 键 词：故障诊断 主成分分析法 变压器 支持向量机 

分 类 号：TM411[电气工程—电器] TP277[自动化与计算机技术—检测技术与自动化装置]