Machine-learning Strategies for Transition/Turbulence Modelling for Low-Pressure Turbines with Unsteady Inflow Conditions

Unsteady flow behaviour induced by wake-blade interactions is crucial for the operational efficiency, aerodynamic stability, and fatigue life of low-pressure turbines, and yet remains challenging to capture with (unsteady) Reynolds-averaged Navier–Stokes calculations. This paper focuses on more reliable estimation of unsteady wake-induced losses, arising primarily from wake mixing and boundary-layer transition under periodic disturbances. To achieve this, the CFD-driven training framework is, for the first time, tailored to LPT flow unsteadiness, enabling revisions of both transition and turbulence models. Firstly, new physics-related features are incorporated into turbulence closure formulations for automated wake-region differentiation, and a new transition-model output is introduced to capture unsteady wake-induced transition. Secondly, model evaluation metrics are supplemented with phase-lock averaged cost functions to ensure consistent improvement throughout the unsteady cycle. The integration of transition and turbulence modelling components is achieved sequentially. A comprehensive assessment is performed using metrics of time-averaged and phase-lock averaged flow features along with secondary statistics, demonstrating solid improvements over the baseline. Detailed model interpretation is also presented to reveal underlying physical insights. Moreover, a-posteriori validation with the machine-learnt transition–turbulence model on different incoming wake frequencies exhibits robust performance, significantly improving the prediction of both wake losses and transition behaviour, not only in a mean sense but also for individual phases. This study highlights the potential of RANS-model development for unsteady multi-stage turbomachinery configurations and provides physical insights into wake-induced unsteadiness in LPTs.