An application of artificial neural network to diesel engine modelling

The paper presents an application of artificial neural network to modelling of the processes in Diesel engine cylinder. In our study six independent engine control variables have been identified and used These parameters determine the engine behaviour (output parameters). In order to found the relations between input and output parameters the radial artificial neural network has been applied. The ANN has been trained on the base of measurement results in order to approximate measurement results of toxicity of exhaust gas components. Measurements data which have been used to training ANN cover a wide range of steady state engine operating conditions (more then 230). The ANN modelling accuracy has been controlled by optimisation method which has been used in order to find the influence of given centres of Gauss function. Outputs from the ANN for some set of control parameters are compared with measurements. A good correspondence of results has been achieved.