ROLLING-BEARINGS FAULT DIAGNOSIS

BASED-ON EMPIRICAL MODE DECOMPOSITION

AND LEAST SQUARE SUPPORT VECTOR MACHINE

WANG Taiyong  HE Huilong  WANG Guofeng  LENG Yonggang  XU Yonggang  LI Qiang

(School of Mechanical Engineering, Tianjin University, Tianjin 300072)

Abstract: In order to extract equipment condition information effectively, a new fault diagnosis method is proposed based-on EMD(empirical mode decomposition) and LS-SVM(least square support vector machine), which takes vibration signals’ Renyi entropy, a complexity measure, as measure criterion. Firstly, vibration signals are decomposed into several IMFs (intrinsic mode functions), then the Renyi entropy of each IMF is computed and regarded as the input characteristic vectors of LS-SVM for fault classification. The rolling-bearings fault diagnosis examples prove the practicability of the method.

Key words: Empirical mode decomposition Least square support vector machine Renyi-entropy Fault diagnosis Rolling-bearings

CLC No: TH165

国家自然科学基金资助项目(50475117, 50675153). Received 20060704, received in revised form 20061126