With the development of Intelligent Ship 1.0 Research,the fault diagnosis system in a marine power plant is particularly important.The fault diagnosis system can realize the transformation of equipment from the traditionally regular and afterward maintenance to the state-based intelligent use and maintenance,to ensure the efficient operation of ship equipment in the whole life cycle and reduce the operation and maintenance cost.The fault diagnosis system is based on the one-to-one correspondence between fault causes and its phenomena.The research on fault simulation based on simulation technology is not only cost-saving but also an important part of the fault diagnosis system development.In this paper,a low-speed diesel engine steady-state simulation model is established in MATLAB/Simulink environment,including five modules: cylinder working process,intake system,exhaust system,turbocharger,and intercooler,for the fault simulation.The simulation results under four working conditions are compared and analyzed with the test data.Six common faults are set,such as carbon accumulation on the top of piston of low-speed diesel engine,fuel injection advance and lag,air filter blockage,turbine grid blockage and cooling degree decline of the intercooler,to analyze the characteristics of output parameters with different fault sources,thus providing support for fault diagnosis.Based on the fault sample set provided by the fault simulation,84 sets of data are used to train the RBF(Radial Basis Function)neural network model for fault diagnosis.By inputting the main performance parameters of the specified low-speed diesel engine fault into the trained neural network,whether the fault sample set provided by the fault simulation is complete or not can be judged according to the output results.Finally,according to the specified performance parameters input,the fault location is determined,as well as the completeness of the fault sample set provided by the fault simulation.In this paper,a dynamic simulation model of the low-speed diesel engine is also established based on the steady-state model for the fault simulation,with the propeller,rotational balance and governor models added.The simulation results of the dynamic process of the low-speed diesel engine are compared with the experimental values to ensure the accuracy of the dynamic model.And the piston top carbon deposition fault and the oil injection advance and delay fault are simulated to analyze the dynamic variation law of the output parameter with the fault source,which provides support for the fault diagnosis of the low-speed diesel engine dynamic process. |