Nonlinear GDI rail pressure control: Design, analysis and experimental implementation

Abstract

Rail pressure tracking control is a landmark application of feedback control in gasoline direct injection (GDI) engines, since improved control performance and robustness translate into better combustion stability, fuel economy and emissions. This paper develops a nonlinear, parameter-varying, input-output affine model for fuel rail system and then designs a rail pressure controller using the triple step procedure. In order to meet the engineering requirement for low computation load, a process is proposed for the map-based implementation of the triple step controller. The steady state control, the reference variation feedforward control and the error feedback gains are realized as maps, while keeping the controller structure and the dynamic error feedback unchanged. Moreover, input-to-state stability (ISS) is achieved for the closed-loop tracking error system, where disturbances and uncertainties are lumped into an additive disturbance. Finally, the map-based controller is tested on an engine control HiL (Hardware in the Loop) platform as well as on a GDI engine test-bench. Experimental results demonstrate the good control performance of the proposed controller.