Adjoint-based trailing edge shape optimization of a transonic turbine vane using large eddy simulations

TL;DR

This paper develops an adjoint-based shape optimization method using large eddy simulations to improve turbine vane trailing edge design, effectively reducing pressure loss and heat transfer despite turbulence complexities.

Contribution

It introduces a viscosity stabilized unsteady adjoint method combined with Bayesian optimization for LES-based turbomachinery shape optimization.

Findings

Optimized vane designs show reduced pressure loss.

Bayesian optimization effectively handles noise in LES data.

Method demonstrates feasibility on supercomputing resources.

Abstract

The shape of the trailing edge of a gas turbine nozzle guide vane has a significant effect on the downstream stagnation pressure loss and heat transfer over the surface of the vane. Traditionally, adjoint-based design optimization methods for turbomachinery components have used low-fidelity simulations like Reynolds averaged Navier-Stokes. To reliably capture the complex flow phenomena involved in turbulent flow over a turbine vane, high-fidelity simulations like large eddy simulation (LES) are required. In this paper, an adjoint-based trailing edge shape optimization using LES is performed to reduce pressure loss and heat transfer over the surface of the vane. The chaotic dynamics of turbulence limits the effectiveness of the adjoint method for long-time averaged objective functions computed from LES. A viscosity stabilized unsteady adjoint method is used to obtain gradients of the design objective function with reasonable accuracy. A gradient utilizing Bayesian optimization is used to robustly handle noise in the objective function and gradient evaluations. The trailing edge shape is parameterized using a linear combination of $5$ convex designs. Results from the optimization, performed on the supercomputer Mira, are compared with optimal designs generated using derivative-free design optimization of the same problem.