Optimized Control of a Pressure-Wave Supercharger: A Model-Based Feedforward Approach

Abstract

Engine downsizing and boosting is known to substantially improve the fuel economy of passenger cars. Compared to traditional boosting devices, pressure-wave superchargers (PWS) have several advantages. In particular, the substantial delays in the torque dynamics caused by conventional turbochargers can be avoided by a careful tuning of the pressure-wave processes taking place inside the PWS. These processes can be tuned to match the changing thermodynamic boundary conditions of the engine system by changing the rotational speed, by opening or closing a bypass channel, and by adjusting the offset between the two rotor casings. The optimal choice of these control actions is not trivial such that a purely experimental approach is not possible. Accordingly, two models are developed that are able to predict the pressure wave dynamics. They are validated using data measured on an engine test bench. The second model which is based on a lumped-parameter approach, forms the basis for a feedforward controller. The synthesis of such a controller and its validation on the same engine test bench is described as well