Intelligent Fault Diagnosis of Rotating Machinery Using Support vector Machine and Improved ABC

Abstract

An intelligent fault diagnosis method by means of local mean decomposition (LMD), support vector machine (SVM) and improved artificial swarm (IABC) is proposed in this paper. Firstly, the vibration signals are decomposed by means of LMD method and frequency-domain and time-domain statistical characteristics of fault information are extracted. Then, a classifier model is present, which combines IABC and SVM to improve classification accuracy. Finally, SVM model identifies different fault situations adopting the optimal features and model parameters. The experiment's result shows the effectiveness of the proposed method in fault feature extraction and fault diagnosis of rolling bearings.