A Multi-Fidelity Approach for Improving Modeling of eVTOL Rotor Noise

Abstract:
A methodology was developed and validated to predict total noise from a 1/5th scale eVTOL rotor in hover by coupling 2D-RANS airfoil simulations with the comprehensive code, CHARM, for rotor loading, the acoustic code, PSU-WOPWOP, for tonal noise, and the broadband noise code, UCD-QuietFly, for broadband noise prediction. Improved sectional 2D aerodynamic inputs obtained from 2D-RANS simulations were used throughout the framework, replacing XFOIL-derived inputs to enhance the prediction accuracy of low Reynolds-number effects such as transition and laminar separation bubbles. Predictions were validated against measurements at Virginia Tech. Results indicate that turbulence model selection influences local boundary-layer development and sectional aerodynamic loading, producing modest differences in tonal noise at 4000 RPM (first BPF) and spectral differences of approximately 4 dB in broadband noise, while integrated OASPL shows slight sensitivity to turbulence model choice. At 2000 RPM, broadband noise dominates the total SPL, with transitional models predicting thinner boundary layers and reduced high-frequency trailing-edge noise. In addition, high-fidelity 3D-DDES simulations performed in OpenFOAM capture the mid-frequency noise content that is underpredicted by CHARM. Overall, the combined mid-fidelity broadband noise results and high-fidelity mid-frequency range noise results provide improved agreement with experimental spectra and demonstrate an effective approach for eVTOL rotor noise prediction.

Did you attend Forum 82? Click the preview below to access the full paper.