Predictive Control Of Transient Air-Fuel Ratio for Gasoline Engine Based on Wavelet Network Inversion

Abstract

To achieve accurate control of the transient air-fuel ratio for gasoline engines, a composite predictive control method based on wavelet network is proposed. The inverse system of transient air-fuel ratio system was to identify with wavelet network, which is able to dynamically compensate intake flow with feedforward control, and a pseudo-linear system is formed by cascading the inverse system with the original system. Then the dynamic matrix control(DMC) was used as a compensation for both internal and external disturbance in real time, and the effective control for transient air-fuel ratio system of nonlinear, time-delay and time-variation is realized. Finally simulated with experiment data at transient conditions, the results show that the inverse model based on wavelet network can accurately approximate the air fuel ratio transient process, and this method has stronger robustness and anti-interference. Therefore, the composite predictive control strategy has good control performances and can improve the control precision of air-fuel ratio in transient conditions.