Improved Local Mean Decomposition and Least Squares Support Vector Machines Applied in Fault Selection for Small Current Grounding System
CAO Li-fang,ZHAO Peng-cheng,CHEN Ying,WANG Yu-tian,ZHANG Shu-qing,ZHANG Hang-fei,XU Jian-tao
Institute of Electrical Engineering, the Key Lab of Measurement Technology and Instrumentation of Hebei Province, Yanshan University, Qinhuangdao, Hebei 066004, China
Abstract：A new fault line selection method for the small current grounding system based on improved local mean decomposition(LMD) combined with the least squares support vector machines(LS-SVM) is put forward. Against defects in end effect of LMD, the minimum squared distance-related to improve algorithm is proposed, and end effect is effectively suppressed; LS-SVM is the developed algorithm on the basis of SVM, with a quadratic loss function instead of insensitive loss function, equality constraints instead of inequality constraints, reducing the computational complexity greatly.Compared with classification results of radial basis function neural network(RBF), least squares support vector machine has advantages in nonlinear pattern recognition.The experiment showed that this method could be well selected fault line, provides an effective new method for fault section of small grounding lines.
曹丽芳,赵朋程,陈颖,王玉田,张淑清,张航飞,徐剑涛. 改进LMD和LS-SVM在小电流接地故障选线中的应用[J]. 计量学报, 2016, 37(6): 632-637.
CAO Li-fang,ZHAO Peng-cheng,CHEN Ying,WANG Yu-tian,ZHANG Shu-qing,ZHANG Hang-fei,XU Jian-tao. Improved Local Mean Decomposition and Least Squares Support Vector Machines Applied in Fault Selection for Small Current Grounding System. Acta Metrologica Sinica, 2016, 37(6): 632-637.
［1］Shi J, Zhang N T, Liu X P. A novel fractional wavelet transform and its applications［J］. Science China Information Sciences,2012,55（6）:1270-1279.
［2］俞阿龙. 基于多重小波变换的力微分信号去噪研究［J］. 计量学报, 2008, 29(3): 235-237.
［3］He W P, Zi Y Y, Chen B Q, et al. Tunable Q-factor wavelet transform denoising with neighboring coefficients and its application to rotating machinery fault diagnosis［J］. Science China,2013,56(8):1956-1965.
［4］王晓卫, 田书, 李玉东, 等.基于S变换特征频率序列的小电流接地系统故障区段定位方法［J］.电力系统保护与控制, 2012, 40(14):109-115.
［5］戴桂平, 刘彬. 基于小波去噪和 EMD 的信号瞬时参数提取［J］. 计量学报, 2008, 28(2): 158-162.
［6］刘立君, 王奇, 杨克己, 等. 基于 EMD 和频谱校正的故障诊断方法［J］. 仪器仪表学报, 2011, 32(6): 1278-1283.
［7］Chen Y, Zhang R, Yin D. Multi-polarimetric SAR image compression based on sparse representation［J］. Science China Information Sciences, 2012, 55(8): 1888-1897.
［8］Zheng X, Hao Z. Diagnosis of valve-slap of diesel engine with EEMD-EMD-AGST approach［J］. Transactions of Tianjin University, 2012, 18(1): 26-32.
［9］张淑清, 孙国秀, 李亮, 等. 基于 LMD 近似熵和 FCM 聚类的机械故障诊断研究［J］. 仪器仪表学报, 2013, 34(3): 714-720.
［10］Peng N B, Zhang Y X, Zhao Y H. A SVM-kNN method for quasar-star classification［J］. Science China Physics, Mechanics and Astronomy, 2013, 56(6): 1227-1234.
［11］李志农, 刘卫兵, 易小兵. 基于局域均值分解的机械故障欠定盲源分离方法研究［J］. 机械工程学报, 2011, 47(7): 97-102.
［12］鞠萍华, 秦树人, 赵玲. 基于 LMD 的能量算子解调方法及其在故障特征信号提取中的应用［J］. 振动与冲击, 2011, 30(2): 1-4.
2016, Vol. 37 
