Study on Marine Diesel Engine Fault Identification Based on Neural Network

Abstract

In order to further improve the accuracy and real-time of Marine diesel engine fault identification, an intelligent identification method based on Shffled Frog Leaping algorithm and Harmonic search algorithm and optimized RBF neural network was proposed to diagnose Marine diesel engine fault. This method optimizes the hidden node, center vector and width parameters of RBF neural network, and carries out simulation experiment on Marine diesel engine fault identification under MATLAB environment. In the experimental process, the RBF neural network was built, and the HS algorithm was used to optimize the hyperparameters of the RBF network, and the SFLA algorithm was used to optimize the harmony memory library to further improve the accuracy of fault identification. Experimental results show that the RBF neural network trained by this method has good convergence effect and high diagnostic accuracy, which verifies the validity and rationality of the proposed method.