Comparison of Two Fourier-Based Methods for Simulating Inlet Distortion Unsteady Flows in Transonic Compressors

The aerodynamic performance of transonic compressors, particularly the stall margin, is significantly influenced by inlet distortion. While time-marching methods accurately simulate such unsteady flows, they can be time-consuming. To enhance the computational efficiency, two Fourier-based methods are proposed in this paper: the time-accurate method with interface filtering and the time-space collocation (TSC) method. The time-accurate method with interface filtering ignores the rotor-stator interaction effects, enabling a larger time step and faster convergence. In contrast, the TSC method accounts for harmonics of conservative variables and transforms the unsteady simulation into multiple steady-state calculations, thereby reducing computational costs. The two Fourier-based methods are validated using NASA Stage 67 and a two-stage transonic fan. Near the peak efficiency point, the results from both methods closely match that of URANS simulation and experimental data. The time-accurate method with interface filtering demonstrates a speed enhancement of 4 to 5 times as a result of a reduction in the iteration steps. In contrast, the TSC method exhibits a speed improvement of at least 20 times in two specific cases, attributable to the significantly smaller mesh size and iteration steps employed in the TSC method compared to the URANS method. Near the stall point, more harmonics for inlet distortion are necessary in TSC simulation to accurately capture flow separation. In the two-stage transonic fan simulations, the strong rotor-stator interaction effects lead to deviations from the URANS simulation; nevertheless, the Fourier-based simulations accurately reflect the trend of the stall margin under total pressure distortion. Overall, the Fourier-based methods show promising potential for engineering applications in estimating the performance degradation of compressors subjected to inlet distortion.