Classification of fault location and performance degradation of a roller bearing

• Published in: Measurement : journal of the International Measurement Confederation Vol. 46; no. 3; pp. 1178 - 1189
• Main Authors: Zhang, Ying, Zuo, Hongfu, Bai, Fang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.2013
• Subjects: Classification; cost effectiveness; Ensemble empirical mode decomposition; Fault diagnosis; Fault location; Feature extraction; Kernel principal component analysis; Kernels; Mathematical analysis; Mathematical models; Performance degradation; principal component analysis; Roller bearings; Support vector machine; Vectors (mathematics); vibration; Fault diagnosis; Feature extraction; Ensemble empirical mode decomposition; Support vector machine; Kernel principal component analysis

► EEMD signal processing method can overcome the mode mixing problem of EMD. ► We propose two types of features based on EEMD. ► We use KPCA to feature extraction and eliminate redundancy. ► We use PSO-SVM for the intelligent pattern classification. ► We compare the identification ability of several...