Numerical Simulation and Validation of As-Manufactured NASA High-Efficiency Centrifugal Compressor Using Fully Coupled FSI Approach

Abstract This study presents a numerical validation of the NASA High-Efficiency Centrifugal Compressor (HECC) using as-manufactured geometry under as-measured in situ testing conditions, employing a fully coupled Fluid-Structure Interaction (FSI) approach. The objective is: to further reduce the modelling error within the systematic error hierarchy to improve the accuracy of aerodynamic performance predictions for the test case; and to perform the cold-to-hot transformation using the as-manufactured geometry and share it with the broader community for subsequent analyses. The HECC, experimentally investigated at the NASA Glenn Research Center, is a challenging test case for CFD methods due to its small diameter, high total pressure ratio of 4.7, high work factor of 0.81 and complex flow physics. The HECC vaned diffusor configuration comprises an impeller with 15 main-splitter pairs, a diffuser with 20 main-splitter pairs, and 60 exit guide vanes. Building on previous publications that systematically quantified iteration and discretisation errors, this study focuses on the modelling errors by using the as-manufactured cold impeller geometry confirmed to represent the physical High-Efficiency Centrifugal Compressor from detailed scans accurately. A fully coupled FSI approach integrates the flow and structural solvers, simulating engine hot conditions and accounting for impeller blade displacement within a unified simulation environment. Mapped contact interfaces are utilised between fluid and solid regions to exchange fluid temperature, heat transfer coefficients, pressure, wall shear stresses, and solid displacements via mesh morphing. The fluid domain employs a coupled flow and energy solver, while the solid domain applies a finite element stress and energy solver. The paper explains the methodology of the coupled approach in detail. The impeller aft cavity, which provides cooling, is added to validate and ensure accurate and economical thermal predictions for the impeller, ultimately influencing the hot condition geometry generated from the FSI analysis. Parametric studies on the leakage flow through the cavity, and the labyrinth seal showed good agreement with the experimental data for the impeller temperature. The final baseline model was used to assess the overall compressor characteristics, and the results demonstrated improved agreement with experimental data compared to previous studies using design-intent hot geometry. The results validate the effectiveness of the coupled FSI approach and highlight the accuracy and benefits of using as-manufactured cold scanned geometry, which further minimised the modelling error part of the systematic error hierarchy.