Enhancing Generalization in Endwall Film Cooling Prediction: Incorporating the Superposition Principle into Transformer-Based Neural Operators

The paper addresses fluid and thermal analysis for turbine endwall film cooling. It introduces a superposition-based deep neural operator that divides endwall temperature-field prediction into two stages and incorporates the film-cooling superposition principle into a transformer-based neural operator framework. The method is designed to improve generalization across cooling layouts with different numbers of film holes.