基于LS-TSVM与降维ESPRIT谐波检测的研究  

Harmonics Detection Method Based on LS-TSVM and Dimension Reduction ESPRIT

作　　者：柏林[1] 唐智 刘小峰[1] 刘子军 BO Lin;TANG Zhi;LIU Xiaofeng;LIU Zijun(The State Key Laboratory of Mechanical Transmission,Chongqing University,Chongqing 400044;VI Service Network Co.,Ltd.,Chongqing 400044,China)

机构地区：[1]重庆大学机械传动国家重点实验室,重庆400044 [2]重庆聚星仪器有限公司,重庆400044

出　　处：《铁道学报》2018年第6期60-66,共7页Journal of the China Railway Society

基　　金：国家自然科学基金(51475052);中国博士后科学基金(2015M582519);中央高校基本科研业务费专项资金(CDJZR14110004)

摘　　要：本文针对电网中大量谐波威胁电力系统正常运行的问题,提出基于最小二乘双支持向量机(LS-TSVM)结合多级维纳滤波器降维ESPRIT谐波检测算法。以高速列车作为研究对象,在SVM基础上发展LS-TSVM谐波检测算法,为进一步提升间谐波检测精度与检测效率,采用多级维纳滤波器ESPRIT降维技术。研究结果表明:在同等采样数据量的情况下以及不失精度的同时,该方法与支持向量机结合ESPRIT方法以及矩阵束法比较,具有更好的检测效率。In this paper,to deal with the problem of a large number of harmonics existing in the power grid that threatens the safe operation of the power system,a harmonic detection method was proposed based on the least square support vector machine combined with multi stage Wiener filter to reduce the dimension of ESPRIT.The power supply system of high speed trains was studied to develop the harmonic detection method based on the least square method on the basis of support vector machine.In order to further improve the accuracy and efficiency of the inter harmonics detection,a multi stage Wiener filter was used to reduce the dimension of ESPRIT.The research results show that in the case of the same sampling data and detection accuracy,the method has better detection efficiency,compared with the method combining the support vector machine with the ESPRIT and the matrix pencil method.

关 键 词：谐波 最小二乘双子支持向量机 多级维纳滤波器 旋转不变子空间技术 

分 类 号：U223.6[交通运输工程—道路与铁道工程]