Tuneable model predictive control of a turbocharged diesel engine with dual loop exhaust gas recirculation

The air path of a turbocharged diesel engine is a multi-input multi-output (MIMO) system with strong nonlinearity, coupling effect, delay and actuator constraints. This makes the design and tuning of the controller complex. In this paper, a tuneable model predictive control (TMPC) controller for a diesel engine's air path with dual loop exhaust gas recirculation (DLEGR) is presented. The objective is to regulate the intake manifold pressure and exhaust gas recirculation (EGR) mass flow in each loop to meet the time-varying setpoints through coordinated control of the variable geometry turbocharger (VGT) and EGR valves. The TMPC controller adopts the design framework of an MPC controller. This controller is also able to provide a map-based switching scheme for the local controller and the controller's weightings. A comparison between the TMPC controller and a conventional PID controller is conducted on a validated real-time engine model. The simulation results show that the TMPC controller achieves lower overshoot, faster response and a shorter settling time on the manipulated objects. These improvements are beneficial for obtaining lower fuel consumption. In order to test the capability of the TMPC controller, it is validated on a hardware in the loop (HIL) platform. The results show that the agreement between the simulation and the actual ECU's response is good.