Machine Learning Analysis of Factors Impacting Cycle-to-Cycle Variation in a Gasoline Spark-Ignited Engine

In this work, we have applied a machine learning (ML) technique to provide insights into the underlying causes of cycle-to-cycle variation (CCV) in a gasoline spark-ignited (SI) engine. The analysis was performed on a set of large eddy simulation (LES) calculations of a single cylinder of a four-cylinder port-fueled SI engine. The operating condition studied was stoichiometric, without significant knock, and represents a load of 16 bar brake mean effective pressure (BMEP), at an engine speed of 2500 revolutions per minute. A total of 123 cycles was simulated. Of these, 49 were run in sequence, while 74 were run in a parallel manner. For the parallel approach, each cycle is initialized with its own synthetic turbulent field (through perturbation of the base field) to generate CCV, as part of another work performed by us. In the current work, we post-processed three-dimensional information from all 123 cycles to compute various flame topology and pre-ignition flow-field metrics. We then evaluated correlations between these computed metrics, and peak cylinder pressure (PCP) employing an ML technique called random forest which was used to learn the correlation between PCP, and these flame topology and pre-ignition flow-field metrics. The computed metrics form the inputs to the random forest model developed, and PCP is the predicted output. The random forest model inherently captures the effect of all inputs, as well as interactions between them owing to its decision-tree structure. The goal of this work is to demonstrate (as a first step) that ML models can implicitly learn complex relationships between pre-ignition flow-fields, flame shapes, and the eventual outcome of the cycle (whether a cycle will be a high or a low cycle).