Study of automobile engine fault diagnosis based on wavelet neural networks

The engine vibration signals characters are extracted using wavelet packet technology. A model of wavelet neural networks is constructed based on wavelet frame theory and neural networks technology. Then multiresolution analysis is used to choose and optimize the wavelet neuron. The model is validated through the testing that simulates the faults of engine valve clearance. The experimental results show that the proposed automobile engine fault diagnostic model based on wavelet neural networks can diagnose the engine fault effectively.

[1]  P. C. Sharma,et al.  Automobile engine fault diagnosis using neural network , 2001, ITSC 2001. 2001 IEEE Intelligent Transportation Systems. Proceedings (Cat. No.01TH8585).

[2]  Eric A. Rying,et al.  Focused local learning with wavelet neural networks , 2002, IEEE Trans. Neural Networks.

[3]  Bin Wu,et al.  Online rotor bar breakage detection of three phase induction motors by wavelet packet decomposition and artificial neural network , 2001, 2001 IEEE 32nd Annual Power Electronics Specialists Conference (IEEE Cat. No.01CH37230).

[4]  Gary G. Yen,et al.  Wavelet packet feature extraction for vibration monitoring , 2000, IEEE Trans. Ind. Electron..

[5]  Xia Limin.  Wavelet packets analysis of rolling bearing vibration signal and fault testing , 2002, Proceedings of the 4th World Congress on Intelligent Control and Automation (Cat. No.02EX527).

[6]  Liu Yong.  Fault Diagnosis Study of the Inlet and Exhaust System Based on Vibration Signal , 2002 .