A Framework for Heat Release Predictions in Compression Ignition Engines with Multiple Injection Events

Abstract

This paper details the development of a zero-dimensional combustion model for compression ignition engines with the main focus on incorporating physics-dependent features. Physical processes, such as injection, evaporation and mixing of fuel and air were given special attention and were linked to physical parameters such as injector geometry and fuel properties to allow for a realistic representation of the in-cylinder combustion process. In its current form, the model aids offline optimization and will further serve as the basis for a model predictive control strategy. Modeling parameters were calibrated against experimental steady-state data of a heavy-duty multi-cylinder engine, mostly at 1038rpm and 14.1bar brake mean effective pressure. Calibration coefficients were fitted to various parameter sweeps, including intake valve closure, injection pressure, combustion phasing and EGR. The modeling framework is capable of predicting premixed and diffusion combustion and was calibrated with up to three injection events. The predicted and experimentally measured combustion phasings are within 1 crank angle degree at high loads, while increasing discrepancies were found for lower loads.