Air-Fuel Ratio Control for Gasoline Engines Based on Physical Model Assisted Extended State Predictive Observer

The air-fuel ratio (AFR) has direct influence on the gasoline engines' power, fuel economy and emission, but suffers from the uncertainty in fresh air and exhaust gas flow dynamics. Therefore, in this paper a AFR controller for gasoline engines is proposed, based on physical model assisted extended state predictive observer (ESPO). The fresh air mass is estimated based on the first principle model considered dual variable valve timing system and low pressure exhaust gas recirculation. For better adaptivity against operating condition variations, two self-leaning factors are embedded in the model and corrected online. Based on the self-learning model, an extended state predictive observer (ESPO) based disturbance rejection controller is designed for model inaccuracy and transport delay compensation. Simulation results show that ESPO reduces the settling time of AFR by 40.2% and 40.8% respectively, compared to the results with the conventional extended state observer (ESO) and Smith predictor in step tests of the target AFR. The deviation of AFR from the target and setting time is reduced by 5.7% and 21% respectively by using the RLS estimator in step tests of throttle position.