Research On Empirical Combustion Model Of Low Speed Two-stroke Marine Diesel Engine

The law of combustion heat realease rate has great influence on the performance of low-speed two-stroke marine diesel engine,and it is an important factor considered in the research and development of low-speed machine performance.In the past,only the combustion experience model established for the MC series models was studied.But the scope of application is narrow and it is difficult to promote.With the application of the electronically controlled fuel injection system on the low speed machine,the control parameter adjustment is independent of the speed and load,and the influence on the heat release law is complex.There is no empirical combustion model built for this type of engine.Therefore,it is of great significance to establish an empirical combustion model for low-speed machines of MC series models and high-pressure common rail fuel injection systems based on the experimental heat release law.Based on the experimental heat release law,the combustion process under various working conditions of various models was analyzed.According to the bore diameter,load and fuel injection system,the heat release rate was divided,and different types of Vibe functions were used for fitting.The heat release law under all conditions of all engines can be simulated by the double-Vibe function with the tail-burning retardation angle as the correction factor.The exothermic process of the large-bore MC series models under high load conditions is concentrated,and the proportion of tail combustion in low-load working conditions and small-bore models is increased.After using the double Vibe function to parameterize the heat release rate,it is found that the diffusion combustion parameters of different models are mainly affected by the cyclic fuel injection mass.The tail-combustion Vibe parameters are affected by speed and the fuel injection mass.A combustion experience model suitable for different MC series models has been established.After verification,it can reflect the true heat release law,and the model effect is good.The influence of control parameters on the heat release law is considered in the empirical model established for the high pressure common rail model.The influence of the thermal effect and dilution effect of EGR on the combustion process is reflected by the O₂ mass fraction in the intake air.The cyclic injection quantity can not fully reflect its influence on the heat release rate.After comprehensive consideration,the O₂ mass fraction,rail pressure,excess air coefficient and cyclic injection quantity are selected as the input of the empirical model,and the empirical model is established.The fitting results are better under medium and high load conditions.After verification,the partial load conditions and the medium and high load EGR conditions are better.The heat release rate of low load EGR conditions is slightly biased.Artificial neural network has strong ability to identify nonlinear relationships between parameters.Based on BP neural network,a combustion model for 6EX340 models is established.The training samples are fitted in low load conditions and superior to empirical relationships.For low load EGR conditions,The predicted results are significantly improved and can be used as a supplement to the empirical combustion model.