Turbocharging technology is one of the main ways to improve the power performance and economy of engine, while traditional exhaust gas turbocharging has hysteresis, especially in the low-speed range of high-power density diesel engines. As an auxiliary and optimization of exhaust gas turbocharging technology, the electric supercharging technology can effectively make up for the shortcomings of turbocharging technology. However, due to the increased degree of control freedom of the electric supercharger, the coordinated control of electric composite supercharging system becomes the key to guaranteeing its performance. In this paper, an electric composite diesel engine intake control strategy based on model predictive control (MPC) is proposed. Firstly, a discrete predictive control model and an exhaust gas turbine observer and flow observer are established based on the mathematical and physical equations of electric supercharger, and then the control strategy and control parameters are verified and calibrated through a co-simulation platform,. Finally, the effects of MPC and traditional PID control under different working conditions as well as the ability to resist power supply voltage disturbance are compared. The results show that the control algorithm based on the observers can realize the coordinated control of the electric composite supercharging system, with maximum steady-state error less than or equal to 1%, without adding turbine speed sensors. It can improve the low-speed response, reduce the transient lag and reach the target inlet pressure ratio within 5.2s, as adapting the response characteristics of exhaust gas turbine under different normal operating conditions. Compared with the traditional PID control algorithm, it has better anti-interference and control accuracy, improving 42% dynamic error.
