Search Results

Flame lift-off length (FLOL), ignition delay time (IDT), liquid length (LL), and Soot are essential parameters defining spray combustion characteristics. They help understand the combustion dynamics and validate the spray and combustion models for numerical simulations. However, obtaining extensive data from experiments is costlier and time-consuming. Machine learning (ML) models have advanced to the point where they could create efficient models that could be used as surrogates for experiments. In this study, five different ML algorithms have been trained using the experimental dataset available through the engine combustion network (ECN) community. A novel genetic algorithm-based hyperparameter optimization code has been used to optimize the models to improve prediction accuracy. The model performances were compared, and the better model was chosen as an experimental surrogate to predict FLOL, IDT, LL, and Soot.

The study showcases the strength of machine learning (ML) models in imitating the operation of an advanced engine concept - the gasoline compression ignition (GCI) - at low loads. The GCI engine is prone to exceeding the limits of criteria emissions at such loads, especially at the cold start when the catalyst is not activated. One proposition to accelerate catalyst light-off is using spark-ignition. This, however, adds an extra level of complexity in identifying an optimum operation point. The ML models can be a useful tool in guiding the engine calibration process. In this study, the ML models are trained on GCI engine experiments, covering different intake conditions, injection strategies, and spark settings. The models can predict seven engine performance parameters: fuel consumption, four engine-out emissions, exhaust temperature, and coefficient of variation (COV) in indicated mean effective pressure (IMEP).