Simulation-based Engine Control for an Ignition-Assisted Diesel Engine with Varying Cetane Number Fuels

Compression ignition engines are complex propulsion/power-generating devices. For highaltitude aviation propulsion applications, a wide range of challenging conditions necessitate the use of ignition assistance methods such as a glow plug, for the combustion of low cetane number (CN) fuels such as sustainable aviation fuels (SAF). Control model development and training is thus vital for maintaining optimal combustion under these conditions. Due to challenges associated with experimental data generation, such as time delays with manufacturing & setup of experiments, engine malfunctions, etc., developing a control strategy based on only experimental data can be costly and time consuming. To accelerate the development of the control model, well-validated CFD simulations can be used to generate data for control model training and explore the design space. This study demonstrates a control strategy developed for optimal operation of an energy assisted compression ignition (EACI) engine for low cetane number fuels, built using CFD data. Moreover, the viability of using CFD simulations as a virtual engine (i.e., digital twin) to replace the real engine for control model testing is also demonstrated. The convergence of the current engine CA50 (i.e., the crank angle at which 50% fuel energy has been release) to the desired engine CA50 via the cetane estimation algorithm in the control model is shown. Finally, the CFD simulation-based control model demonstration is compared to the corresponding experimental control model demonstration.