Thiva Albin and Dennis Ritter
Institut für Regelungstechnik, RWTH Aachen University
Monday, February 02, 2015, 11:00 - 12:30
Room 02-012, Georges-Köhler Allee 102, Freiburg 79110, Germany
Significant effort within current developments of powertrains with gasoline engines is set on the reduction of CO2 emissions while maintaining the drivability at the same time. One of the investigated approaches is downsizing, where the engine displacement is reduced and the specific power is increased by influencing the air path of the engine with forced induction devices. Modern air path structures such as the two-stage turbocharging are improving the cylinder filling, but are characterized by high demands on the process control.
In the context of two-stage turbocharging, the Model Predictive Control (MPC) offers great advantages with respect to the demands on the process control. Using MPC, it is possible to systematically consider the complex dynamics of the system and at the same time respect constraints on the process variables. Furthermore the use of MPC in automotive applications offers the possibility to drastically reduce the amount of time needed for controller calibration. In the talk, the modelling of a two-stage turbocharging process, a MPC approach and the evaluation of the control algorithm in a vehicle using on-road tests are demonstrated.