摘要
交流接触器的状态识别是确保其正常工作的重要前提,针对交流接触器的运行状态具有多元、多属性交互影响和动态变化的特点,提出一种基于灰狼算法优化的支持向量机(GWO-SVM)的交流接触器运行状态识别方法。从接触器触头分合闸阶段的振动信号提取时域以及频域特征,采用皮尔逊相关系数对参数进行特征选择,得到最优特征子集;通过GWO-SVM方法进行接触器运行状态识别,与其他识别算法对比,所提方法在对接触器运行状态的识别准确率方面更优异。
The state recognition of AC contactor is an important prerequisite to ensure its normal operation.In view of the characteristics of multivariate,multi-attribute interaction and dynamic change of AC contactor operation state,an AC contactor operation state recognition method based on Grey Wolf optimizer support vector machine(GWO-SVM)is proposed.The time domain and frequency domain features are extracted from the vibration signal of the contactor contact opening and closing stage,and the Pearson correlation coefficient is used to select the parameters to obtain the optimal feature subset.Finally,the contactor operation state is identified by GWO-SVM method.By comparing other identification algorithms,it shows that the proposed method is better in the recognition accuracy of the contactor operation state.
作者
李瑞新
刘树鑫
曹云东
LI Ruixin;LIU Shuxin;CAO Yundong(Institute of Electrical Apparatus New Technology and Application,Shenyang University of Technology,Shenyang 110870,China)
出处
《电器与能效管理技术》
2022年第6期39-44,68,共7页
Electrical & Energy Management Technology
基金
辽宁省科技重大专项(2020JH1/10100012)
辽宁省教育厅项目(LTGD2020001)
沈阳中青年科技创新人才计划(RC210354)。
关键词
交流接触器
特征选择
灰狼算法
支持向量机
AC contactor
feature selection
Gray Wolf algorithm
support vector machine(SVM)
