Minimizing Downstream Heat Transfer for Turbomachinery Design via Bayesian Optimization

When a turbine engine is manufactured, an inter-platform gap is created between the combustor annulus and the turbine annulus due to them having to be manufactured as separate parts.  This gap has to be supplied with leakage air to prevent ingestion of the extremely hot combustion gases into the interior of the engine. The two parameters that affect the effectiveness of this cooling system are the blowing ratio of the air inlet and the radius of curvature of the outlet of the gap. This work utilizes Bayesian optimization to find the ideal configuration of these parameters such that downstream cooling is maximized. Here, Bayesian optimization is a sequential design strategy that utilizes Bayes Theorem to direct the search for a global maximum or minimum of a black box function, and  results for every iteration were obtained by large-eddy simulation.