A novel method of intelligent fault diagnosis for diesel engine

On the point view of complementary strategies, a new hybrid algorithm to optimize the RBF network based on artificial immunology was proposed. A dynamic clustering algorithm based on clonal selection algorithm was used to specify the amount and initial position of the RBF centers; then RBF network was trained by the immune evolutionary algorithm. Combining with the rough sets-based attribute reduction algorithm, a novel hybrid system of rough sets and Immune-RBF network for intelligent fault diagnosis were put forward. The diagnosis of diesel demonstrates that the method can effectively simplify the structure of network and increase the efficiency and precision of diagnosis.