Intelligent Fault Diagnosis Of Gasoline Engine Based On Neural Network

Engine is the common power source of automobiles and other machines. Developing automobile engine fault diagnosis system, which was applied while the automobile kept assembled, had significant meaning with its structure becoming more and more complex, working condition becoming more and more severe, fault diagnosis becoming more and more difficult. This paper gave the diagnostic method of engine fault based on vibration signal and devised the experiment. Vibration signal of engine was tested at the order and some frequent failure. Some traditional and recent methods are brought out to the signal disposal and fault diagnosis. It was verified that the wavelet packet should effectively process the vibration signals and abstract the fault character. ANN(Artificial Neural Network)had a powerful nonlinear mapping ability, so fault diagnosis was accomplished by Artificial Neural Network. In the paper, it was also discussed the problem of the fault diagnosis at different speeds, optimizing the structure and power values of Artificial Neural Network by Genetic Algorithm, connecting VB and Matiab smoothly and so on. At last, the method given by this paper was tested by experiment. The result indicated the Artificial Neural Network optimized by Genetic Algorithm was an effective way to classify the fault of engine. The method had wide applied foreground and was straightforward and facile.