Multi-Fidelity Surrogate-Based Optimization of Transonic and Supersonic Axial Turbine Profiles

Automated Fluid-dynamic Shape Optimization plays a key role in the design of  turbomachinery and typically combines Computational Fluid Dynamics (CFD) solvers, parametrization techniques and numerical optimization methods, generally categorized as either direct or surrogate-based (SBO) ones.

Here, a particular focus is given to SBO exploiting surrogate models constructed from low-fidelity models, often referred to as variable- or multi-fidelity optimization.

This paper presents a multifidelity SBO approach for the optimization of the LS89 high pressure axial turbine vane to significantly reduce the computational cost associated to high fidelity CFD simulations while exploiting models of lower fidelity.
A cokriging method is used to simultaneously take into account quantities of interest (QoI) coming from models of different fidelities providing a global surrogate model. A classic bayesian global optimization method permits to iteratively propose desing of interest. It relies on the maximization of the so-called Expected Improvement criterion.
A geometrical parametrization technique based on B-splines is considered to describe the profile geometry. The mass-flow rate and the outlet angle are constrained.
The optimization study reveals significant reduction in computational cost w.r.t. classic optimization frameworks based on a single fidelity, such as, adjoint-based and gradient-free methods, while providing similar improvements in term of fitness functions.

A well-known supersonic turbine nozzle for ORC applications characterized by strong non-ideal gas effects in the flow expansion is also considered to showcase the flexibility of the proposed approach.