Air fuel ratio control in spark injection engines based on neural network and model predictive controller

An application of wavelet neural network or wavenet (WNN) to design model predictive control in control problems of nonlinear systems is investigated in this study. A wavenet is constructed as an alternative architecture to a neural network to approximate a nonlinear system. Based on approximation capability of wavelet network, a suitable adaptive model predictive control law and parameter updating algorithm as applied to nonlinear system uncertainty estimation are developed. It is shown that using wavenets, an effective uncertainty estimation and control strategy can be obtained. This proposed method improves plant performance effectively and provides robustness against variations caused by changes in operating points of the system. By comparing wavenet and neural network, simulation results show the benefits of the proposed method. Defining proper cost function helps saving energy which is necessary these days.