The application and research of the intelligent fault diagnosis for marine diesel engine

The marine diesel engine is a complex system, which has the important function to guarantee the marine security. In this paper a novel approach of optimizing and training fuzzy neural network based on the ant colony algorithm is proposed for the intelligent fault diagnosis of this kind of diesel engine. The structure and the parameter of fuzzy neural network for fault diagnosis system are introduced. Its weight and the threshold value are trained by the ant colony optimization algorithm. This method may effectively avoid the question that the BP algorithm usually chosen to train network easily to sink into the partial extreme value and also has the characteristics of quick convergence. Finally this fuzzy neural network system optimized by ant colony algorithm training is applied in the fault diagnosis of the marine diesel engine. The comparison of simulation results shows good performance and validity of the proposed method.