Surrogate-based Model Optimization of Intake Manifold for an Engine

This is the result of individual study of Mr. Potsawat BOONJAIPETCH Student ID 650631110: This study develops and tests an automated Bayesian Optimization (BO) framework for intake-manifold shape exploration using a low-fidelity computational fluid dynamic (CFD). The framework was designed to automatically generate geometry candidates, perform CFD simulations, extract the objective value, update the surrogate model, and propose new designs through an acquisition function. This study served as a practical test problem to confirm that the complete optimization loop could operate successfully under a limited computational budget.