Data-driven augmentation of a RANS turbulence model for Transonic Flow Prediction

Kurzfassung

Purpose: The paper aims to improve Reynolds-Averaged Navier Stokes (RANS) turbulence models using a datadriven approach based on machine learning. A special focus is put on determining the optimal input features used for the machine learning model. Methodology: The Field Inversion and Machine Learning (FIML) approach is applied to the negative Spalart-Allmaras turbulence model for transonic flows over an airfoil where shock-induced separation occurs. Feature selection methods are applied to the results of the field inversion to determine the optimal input features for the machine learning model. Findings: Optimal input features and a machine learning model are developed which improve the existing negative Spalart-Allmaras turbulence model with respect to shock-induced flow separation. Originality: A comprehensive workflow is demonstrated that yields insights on which input features and which machine learning model should be used in the context of the FIML approach