Date & Time: October 17th, 13:15-14:00
Location: Seminar Room M 3170-73 in the M-building, LTH
Author: Lukas Christensson
Title: Machine Learning for Air Charge Estimation: A residual Error Approach
Supervisor:  Alba Gurpegui Ramon
Examiner:  Bo Bernhardsson
 

Abstract: 
This thesis examines the possibility of complementing classical formulas for engine calibration with supporting machine learning models to improve the precision. Both a Gaussian Process Regressor (GPR) and an Artificial Neural Network (ANN) are trained for the task and Bayesian optimization is performed to optimize the structure of the models. The study shows promise for utilizing machine learning models to improve combustion engine efficiency, and that further studies for hardware modification and implementation are needed.