Multi-objective optimization of fan-shaped film hole for balancing heat transfer and thermal stress based on ANN and NSGA-II

Modern aerospace engines demand turbine blade designs that achieve an optimal balance between efficient cooling and structural strength under complex working conditions. Traditional serial design methods, which typically involve a step-by-step approach to addressing each design objective separately, can be time-consuming and insufficient to meet the demands of multi-objective optimization. This study presents a multi-objective optimization methodology for fan-shaped film cooling holes, integrating high-precision artificial neural networks (ANN) with the non-dominated sorting genetic algorithm II (NSGA-II) to efficiently balance heat transfer and thermal stress in design. Meanwhile, the optimization approach can automatically extract features, significantly reducing the reliance on manual interventions. The results demonstrate the optimization improves film cooling effectiveness by 55.8% while reducing the maximum equivalent thermal stress by approximately 14.9%. Parametric analysis reveals that increasing the blowing ratio (0.5 to 2.0), length–diameter ratio (2.8 to 4.6), and streamwise torsion angle (3.0°to 14.0°) can effectively enhance cooling efficiency while concurrently reducing thermal stress. This method identifies optimal geometric parameters for real operating conditions, providing an efficient and practical pathway for the multidisciplinary optimization design of complex film structures.