Fault diagnosis based on wavelet package and Elman neural network for a hydraulic pump

Considering the low signal-to-noise, faint failure characteristics of hydraulic pump, and slow convergence speed and the instability of BP network, a new fault diagnosis method based on the wavelet package analysis and Elman neural network is presented. The wavelet package analysis is adopted to eliminate the noise in the actual signals and to extract the fault characteristics. Through signal decomposition and single reconfiguration with wavelet package, the noise can be eliminated from signals to strengthen the failure signal and to extract fault feature in every frequency bands effectively. Energy of various frequency bands acting as the fault feature vector is input into the improved Elman neural network to realize the mapping between the feature vector and the fault mode. The experiment results verified the effectiveness of the proposed method in the hydraulic pump fault diagnosis.