Fault Diagnosis of Diesel Fuel Ejection System Based on Improved WNN

To remedy the disadvantages of conventional diesel engine fuel ejection system´s fault diagnosis method, which couldn´t get exact results and dispel noise effectively, a new way based on wavelet neural network (WNN) which combines merits of wavelet transform (WT) and RBF neural network (RBFNN) was put forward. Moreover, after picking up fault characteristic parameters, we trained it with genetic algorithm (GA) and simulated annealing (SA). Finally we applied the improved WNN to fault diagnosis of diesel engine fuel ejection system. The results show that the algorithm is good at dispelling noise, stable, and effective in high precise fault diagnosis