Model-based control strategies along with an adapted calibration process become more important in the overall vehicle development process. The main drivers for this development trend are increasing numbers of vehicle variants and more complex engine hardware, which is required to fulfill the more and more stringent emission legislation and fuel consumption norms. Upcoming fundamental changes in the homologation process with EU 6c, covering an extended range of different operational and ambient conditions, are suspected to intensify this trend.
One main reason for the increased calibration effort is the use of various complex aftertreatment technologies amongst different vehicle applications, requiring numerous combustion modes. The different combustion modes range from heating strategies for active Diesel Particulate Filter (DPF) regeneration or early SCR light-off and rich combustion modes to purge the NOx storage catalyst (NSC) up to partially premixed combustion modes. A combination of advanced physically oriented control strategies and new model-based calibration procedures seems therefore favorable in order to significantly reduce the overall calibration effort and increase the robustness for the different combustion modes.
Within this publication, a new model-based approach for fuel path control of Diesel engines is presented as an example of such an advanced control strategy. The control concept combines a model-based feed-forward control with a cylinder pressure based feedback control. The models for feed-forward control are derived by splitting the entire thermodynamic process into combustion and energy conversion. For feedback control new fuel path control parameters such as the “centroid-of-heat-release” are introduced and combined with a structure variable controller that changes the control parameters depending on the injection strategy. Besides a detailed description of the novel control algorithm, the benefits in terms of calibration effort are analyzed and discussed. Finally, for experimental validation the control strategy is implemented on a Rapid Control Prototyping (RCP) system. Exemplary results including a rich event during a transient acceleration are shown to illustrate the potential of the control concept.