Combustion simulation of turbocharger HSDI Diesel engines during transient operation using neural networks

The purpose of this investigation is to develop a procedure to simulate, with a fast and simple empirical model, the rate of combustion in Diesel engines during transient operation, with the further goal of including this simulation in a given one-dimensional gas dynamic model for global transient behaviour prediction. Due to the nature of transient phenomena, this simulation has to be performed for a wide range of engine conditions and that is why the route followed in this study was to apply the self-adjusting capability of neural networks (NN) to relate engine operating conditions with the parameters of combustion heat release laws. Finally, for a given turbocharged Diesel engine, the rates of heat release obtained with the NN methodology have been compared to the rates obtained experimentally, the temporal evolution of some combustion parameters have been analysed and the influence of NN combustion simulation in global transient modelling has been presented.