Control-Oriented Modeling Of A Thermal Engine Based On A Deep Learning Approach

Abstract The work carried out in this paper focused on "control-oriented modeling of a thermal engine using a deep learning approach". The goal of this work is to develop a neural network model of a thermal engine and to make a prediction of parameters related to engine management and directly impacting pollutant emissions and fuel consumption. For it, data was taken from an experimentation engine and made it possible to make maps of its operation. These maps enabled the calibration of a Simulink model of a thermal engine. Through a systems identification approach, the temporal response of the motor was estimated and made it possible to develop a database which was used for the training the LSTM artificial neural network. The work carried out showed that the learning phase of the neural network proceeded in a consistent way (overall decrease in cost functions) and converged towards a value of RMSE = 1.09 better than those observed in the literature. The resulting neural engine model made it possible to predict several variables (fuel mass flow rate and pollutant mass flow rates) with acceptable residual errors. These results reveal that the neural model obtained correctly predicts the said variables and can therefore be used in closed-loop simulations of the operation of a vehicle or for a context of simulation of the operation of the engine.