An implementable multi-calibration strategy to address real driving emissions of CI engines

In the last years emissions related to transportation have been in the focus of various legislation changes. The soon to be introduced Worldwide Harmonized Light-Duty Vehicles Test Procedure (WLTP) is one example for that. However, as shown by recent studies a substantial, often overlooked, factor is the driver itself and fixed procedures for testing within the legislation will not help in suppressing this influence. Classical calibration based engine control does not allow to counteract the drivers behavior or account for varying demands. Against this background we propose an engine control strategy based on multiple calibrations, where the selection of a calibration is the degree of freedom that can be used to counteract external influences. The calibration selection problem is mathematically formulated as a Boolean Linear Program (BLP) and an implementable approximate solution, which utilizes route specific information, is introduced. The method and its assumptions are validated by applying it in an experimental setup based on real road data and an engine testbed. The results show that the method is able to select a calibration in a sensible way and that the approximations lead to small constraint violations only.