Research On Fault Diagnosis Of Marine Diesel Engine Based On Optimized Probabilistic Neural Network

As the main propulsion of ship,marine diesel engine makes an important impact on the safety of ship and sea crews,so fault diagnosis research on marine diesel engine is essential to ensure navigation safety.Aiming at the typical fault types of marine diesel engine,this paper proposes a marine engine fault diagnosis method based on sparrow search algorithm and probabilistic neural network,and develops the marine diesel engine fault diagnosis software in Qt framework.The main research contents and conclusions of this paper are as follows:1.Taking the marine 4135 diesel engine as the research object,the marine diesel engine simulation model is built and verified in the GT-POWER environment.2.Select 6 typical failure modes and normal operation modes,complete the fault simulation setting of marine diesel engine,calculate and obtain 1400 groups of thermal data including 9 thermal parameters that can characterize the selected typical fault,and expand them to 2800 groups after being processed by piecewise cubic Hermite interpolation method,so as to form a data set for the study of fault diagnosis model.3.Build the fault diagnosis model based on RBF neural network model and probabilistic neural network model,compare the fault diagnosis accuracy of the two models,and select the probabilistic neural network fault diagnosis model with an accuracy of 89.29% as the basic model.4.Establish the fault diagnosis model of marine diesel engine based on sparrow search algorithm optimization probabilistic neural network.Quantum genetic algorithm and sparrow search algorithm are selected to optimize the smoothing parameters of probabilistic neural network in Matlab environment.By comparing the optimization process and optimization effect of the two algorithms,the fault diagnosis accuracy of sparrow search algorithm optimized probabilistic neural network model is 96.67%,and the optimal smoothing parameter is 0.7623.The optimization effect is obvious,which can meet the needs of engineering application.5.Develop marine diesel engine fault diagnosis software based on C + + language under QT framework.Using the mixed programming of QT and MATLAB,QT calls the optimization model in Matlab environment to realize the functions of ship engine fault diagnosis and state evaluation.The research results provide reference for the development of marine engine fault diagnosis software.