Air Fuel Ratio Control for Gasoline Engine Using Neural Network Multi-step Predictive Model

Abstract

Air fuel ratio is a key index affecting the emission of gasoline engine, and its accurate control is the foundation of enhancing the three-way catalytic converting efficiency and improving the emission. In order to overcome the existed transmission delay of air fuel ratio signal, which affects the control accuracy of air fuel ratio if using directive air fuel ratio sensor signal., and a multi-step predictive control method of air fuel ratio based on neural network was provided in the paper. A multi-step predictive model of air fuel ratio based on back propagation neural network was set up firstly, and then a fuzzy controller was designed using the error of predictive values and expected values and its derivative. The simulation was accomplished using experiment data of HL495 gasoline engine, and the results show the air fuel ratio error is less than 3% in the faster throttle movement and it is less than 1.5% in the slower throttle movement.