A GA-SVM Hybrid Classifier for Multiclass Fault Identification of Drivetrain Gearboxes

Authors

Dingguo Lu, University of Nebraska-LincolnFollow
Wei Qiao, University of Nebraska-LincolnFollow

Date of this Version

2014

Citation

Energy Conversion Congress and Exposition (ECCE), 2014 IEEE Year: 2014

Comments

©2014 IEEE

Abstract

This paper presents a genetic algorithm (GA)- support vector machine (SVM) hybrid classifier for multiclass fault identification of drivetrain gearboxes in variable-speed operational conditions. An adaptive feature extraction algorithm is employed to effectively extract the features of gearbox faults from the stator current signal of an AC machine connected to the gearbox. The multiclass GA-SVM classifier is used to identify the faults in the gearbox according to the fault features extracted. A GA is designed to find the optimal parameters of the SVM to obtain the best classification accuracy. The proposed hybrid classifier is validated on a gearbox connected with a permanent-magnet synchronous machine with three different faults. Experimental results show that the multiple types of gearbox faults can be effectively identified and classified by the proposed hybrid classifier with better accuracy than the traditional SVM classifier.