Bayesian optimization of in-situ adaptive tabulation for marine engine CFD simulations

This thesis presents a comprehensive optimization study of In-situ Adaptive Tabulation (ISAT) control parameters to enhance computational efficiency in reactive marine diesel engine CFD simulations. Using a multi-objective Bayesian optimization framework integrated with OpenFOAM™, the study systematically tunes ISAT parameters to minimize computational cost while maintaining combustion fidelity. Results demonstrate that poor ISAT configurations can crash or stall simulations, while optimal settings can reduce computational time by up to 50% with controlled impact on accuracy. Guidelines and example configurations are provided for effective ISAT parameter selection.

Recommended citation: D. Virokannas. (2024). "Bayesian optimization of in-situ adaptive tabulation for marine engine CFD simulations." Aalto University.
Download Paper