Aeroacoustics — Airfoil Shape Optimization

Shape optimization of airfoils with multi-fidelity CFD and noise prediction models.

Implemented differentiable boundary layer data extraction from the CFD solvers ADflow and CMPLXFOIL.
Developed components of semi-empirical noise models for constructing coupled analyses with MPhys.
Currently researching multi-objective gradient-based constrained optimization.

Boundary layer quantities are extracted differentiably from the flow field, validated against reference CFD and experiment, then fed into semi-empirical wall-pressure-spectrum models that predict trailing-edge noise.

Boundary layer profile evaluation against reference data

Yuan Lyu, Arjit Seth, and Rhea P. Liem. “Fuel consumption and trailing-edge noise tradeoff studies via mission-based airfoil shape optimization”. Aerospace Science and Technology 151 (2024), p. 109331. doi:10.1016/j.ast.2024.109331
Stéphane Redonnet, Turzo Bose, Arjit Seth, and Larry K. B. Li. “Airfoil Self-Noise Prediction Using Deep Neural Networks”. Engineering Analysis with Boundary Elements 159 (2024), pp. 180–191. doi:10.1016/j.enganabound.2023.11.024
Dajung Kim, Arjit Seth, and Rhea P. Liem. “Geometric Programming for Airfoil Shape Optimization With Geometric Constraints”. AIAA SCITECH 2025 Forum. doi:10.2514/6.2025-0653

Related publications