Sparse representation-based classification for the planetary gearbox with improved KPCA and dictionary learning

Abstract

A fault diagnosis method for the planetary gearbox according to sparse representation-based classification (SRC) has been presented in this paper. Considering the real-time performance and accuracy rate of the fault diagnosis, the proposed method has introduced the improved kernel principal component analysis (KPCA) and dictionary learning. First, some time domain and frequency domain features are combined into a feature vector to represent a sample, which can reduce the computational burden and enhance the real-time performance of fault classification. Second, the feature sets are transformed into a new feature space through the improved KPCA, which can improve the precision of fault classification. Then, the training samples are used to implement dictionary learning, and the testing samples are taken as the input of the SRC for classifying. Finally, a planetary gearbox fault diagnosis experiment is designed to verify the effectiveness of the proposed method.