Multi-objective 3d optimization of film cooling configurations in a high pressure vane based on genetic algorithm -artificial neural network

In this paper, the optimum parameters of a row of cylindrical film cooling holes has been investigated using a multi-objective evolutionary approach so as to achieve a compromise between film cooling effectiveness and coolant massflow rate which are in opposite directions and compete with each other. For this purpose, chordwise position of film holes, as well as diameter and injection angles were chosen as design parameters. Thirty samples were generated as database through CFD runs and artificial neural network (ANN) method was used to construct the surrogate model to approximate the optimization targets as functions of design parameters and genetic algorithm (GA) was used as optimizer. Design iterations were repeated 6 times and optimum configuration resulted in objective function values was found.