Research on Thruster Fault Diagnosis of Underwater Vehicle

摘要

During the underwater vehicles (UVs) operation, the thruster can malfunction due to foreign body entanglement and damaged blade. Therefore, there is need for research and development on data driven methodologies and condition monitoring techniques which are able to achieve fast, reliable and high-quality fault diagnosis. In this paper, a novel method combining Back Propagation Neural Network (BPNN) and Support Vector Machine (SVM) is proposed towards fast and accurate fault diagnosis of UVs. Firstly, wavelet packet transform is used to decompose the thruster vibration signal, and then wavelet packet frequency-band energy analysis (WPFE) method is used to calculate the decomposed vibration signal, and extract characteristic vectors. Finally, the BPNN-SVM method is used to identify the fault. The experimental results show that the performance of the proposed method is verified by the comparison with other widely used methods.