A torque-based nonlinear predictive control approach of automotive powertrain by iterative optimization

Abstract

A torque-based nonlinear predictive control approach is developed to manipulate the automotive powertrain in this paper. This approach consists of three components, i.e. torque demand control, nonlinear predictive controller and torque load estimation. A control scheme of torque demand with proportional–derivative (PD) compensator is proposed to meet the torque requirement of drivers. Based on a mean value model of the internal combustion (IC) engine, a torque-based nonlinear predictive controller is designed by use of iterative optimization. A proportional–integral (PI) observer is employed to estimate the torque load of the powertrain. Through experimental validation, it can be concluded that the torque-based nonlinear predictive control approach is a significant candidate for the automotive powertrain to implement dynamic torque control.