基于PCA和优化参数SVM的智能变电站故障诊断方法  

FAULT DIAGNOSIS OF INTELLIGENT SUBSTATION BASED ON PCA AND OPTIMIZED PARAMETER SVM

作　　者：张弛 王广民[1,3] 许会博 佘维 田钊[1] Zhang Chi;Wang Guangmin;Xu Huibo;She Wei;Tian Zhao(School of Software,Zhengzhou University,Zhengzhou 450001,Henan,China;State Grid Tianjin Electric Power Company,Tianjin 300000,China;XJ Electric Co.,Ltd.,Xuchang 461000,Henan,China)

机构地区：[1]郑州大学软件学院,河南郑州450001 [2]国网天津市电力公司电力科学研究院,天津300000 [3]许继电气股份有限公司,河南许昌461000

出　　处：《计算机应用与软件》2022年第7期80-88,共9页Computer Applications and Software

基　　金：国家电网公司总科技项目(5206/2018-19002A)。

摘　　要：针对目前智能变电站故障诊断结构复杂、样本数据量小的问题,构建一种基于主成分分析法和优化参数支持向量机的智能变电站故障诊断模型。通过主成分分析法提取关键故障特征,降低故障诊断的复杂性;结合变电站的运行模式,建立多分类支持向量机分类器,通过帝国竞争算法寻找支持向量机的优化参数;通过真实变电站的故障事件进行实验验证。实验结果表明,该方法能够有效解决训练样本少的问题,同时具备较好的诊断效果。Aiming at the problem of complex fault diagnosis structure and small sample data of intelligent substation, this paper constructs a fault diagnosis model of intelligent substation based on principal component analysis and optimized parameter support vector machine. The key fault features were extracted through principal component analysis to reduce the complexity of fault diagnosis. Based on the operation mode of substation, we established the multi-classification support vector machine classifier. The optimal parameters of support vector machine were found by imperial competition algorithm. The experimental verification were carried out through the fault events of real substation. The results show that the proposed method can effectively solve the problem of few training samples and has a good diagnosis effect.

关 键 词：智能变电站 故障诊断 支持向量机 主成分分析法 帝国竞争算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]