A New Body Force-Based Modelling Approach for Turbomachinery Aerodynamics: Formulation, Verification and Implementation Aspects

Abstract Modelling of turbomachinery flows typically entails singlepassage or full annulus RANS and uRANS simulations. The computational and numerical stability challenges associated primarily with transient simulations of multi-stage and coupled turbo-machinery components may offset their accuracy benefits. Body force modelling provides a reasonable trade-off between computational efficiency and accuracy, replacing the turbomachinery component with a body force field. In the present work, a new body force-based modelling approach is presented, comprising body force-relevant grids, an inviscid compressible solver in the relative frame of reference with metal and aerodynamic blockage and a new body force model comprising a base turning, correction and viscous force component. Distribution coefficients are introduced, mathematically replicating the spatial distribution of key quantities such as loss and flow deviation using custom functions or CFD data. The proposed approach is sufficiently comprehensive to enable its applicability to all types of compressor and potentially turbine blades, while minimising dependence on CFD input. The validity and limitations of the proposed approach are assessed through numerical response test cases. Subsidiary force components and loss mechanisms, associated with abrupt flow turning and blockage effects are captured. Considerations and modelling aspects regarding grid sensitivity, blockage, the solver and the effect of distribution coefficients on compressor performance are analysed and quantified. Limitations associated with the grid topology, body forces and blockage modelling are partially or fully addressed. The paper aims to provide an in-depth analysis in fundamental aspects of body force modelling approaches, while unveiling obscure aspects and limitations underpinning most body force modelling approaches. A new approach is proposed to mitigate the shortcomings of previous methods.