Feature Extraction of Kernel Regress Reconstruction for Fault Diagnosis Based on Self-organizing Manifold Learning

作     者：CHEN Xiaoguang LIANG Lin XU Guanghua LIU Dan 

作者机构：School of Mechanical EngineeringXi’an Jiaotong University State Key Laboratory for Manufacturing Systems EngineeringXi’an Jiaotong University 

出 版 物：《Chinese Journal of Mechanical Engineering》 (中国机械工程学报（英文版）)

年 卷 期：2013年第26卷第5期

页      面：1041-1049页

核心收录：

学科分类：0711[理学-系统科学] 07[理学] 0817[工学-化学工程与技术] 080202[工学-机械电子工程] 08[工学] 0807[工学-动力工程及工程热物理] 080401[工学-精密仪器及机械] 0804[工学-仪器科学与技术] 080402[工学-测试计量技术及仪器] 0802[工学-机械工程] 0811[工学-控制科学与工程] 0801[工学-力学（可授工学、理学学位）] 

基　　金：supported by National Natural Science Foundation of China(Grant No.51075323) 

主　　题：feature extraction manifold learning self-organize mapping kernel regression local tangent space alignment 

摘      要：The feature space extracted from vibration signals with various faults is often nonlinear and of high ***,nonlinear dimensionality reduction methods are available for extracting low-dimensional embeddings,such as manifold ***,these methods are all based on manual intervention,which have some shortages in stability,and suppressing the disturbance *** extract features automatically,a manifold learning method with self-organization mapping is introduced for the first *** the non-uniform sample distribution reconstructed by the phase space,the expectation maximization(EM) iteration algorithm is used to divide the local neighborhoods adaptively without manual *** that,the local tangent space alignment(LTSA) algorithm is adopted to compress the high-dimensional phase space into a more truthful low-dimensional ***,the signal is reconstructed by the kernel *** typical states include the Lorenz system,engine fault with piston pin defect,and bearing fault with outer-race defect are *** with the LTSA and continuous wavelet transform,the results show that the background noise can be fully restrained and the entire periodic repetition of impact components is well separated and identified.A new way to automatically and precisely extract the impulsive components from mechanical signals is proposed.