Kim, Jihoon (The University of Tokyo), Yamasaki, Yudai (The University of Tokyo)

Prediction of IMEP Fluctuation Trends in a Gasoline Engine Using Random Forest

Scheduled for presentation during the Invited session "Learning-based Optimization and Control for Intelligent Electrified Vehicles" (ThB1), Thursday, October 31, 2024, 13:10−13:30, Room T1

7th IFAC Conference on Engine and Powertrain Control, Simulation and Modeling, Oct 30 - Nov 1, 2024, Dalian, China

This information is tentative and subject to change. Compiled on May 16, 2025

Abstract

This study focuses on predicting cyclic variations in a gasoline engine using random forest classification. Engine cyclic variation, characterized by fluctuations in combustion parameters such as indicated mean effective pressure (IMEP), significantly impacts engine performance. These variations are influenced by factors such as air-fuel mixture uniformity, residual gas concentration, and in-cylinder turbulence, making precise quantitative prediction challenging. To address this, a model was constructed by using a random forest and evaluated to predict whether IMEP will increase or decrease in the next cycle. The results demonstrate the potential of the model in predicting IMEP fluctuations.