Session: 39-01 Adjoint-based and adjoint-enhanced design optimization methods and applications

Paper Number: 82218

82218 - Comparing Gradient-Free and Gradient-Based Multi-Objective Optimization Methodologies on the VKI-LS89 Turbine Vane Test Case 

Numerical optimization is a critical part of the design process for turbomachinery components. The present paper addresses the multi-objective aerodynamic shape optimization of the two-dimensional LS-89 turbine cascade. Subsonic and transonic operating conditions are considered for the multi-objective optimal search. The objective is to minimize the entropy generation at both flow conditions while constraining the mass flow in a certain range and maintaining the blade loading above a threshold value. Nineteen design variables are used to parametrize the geometry. The same problem is optimized using two different optimization strategies. A first optimization strategy uses a gradient-based Sequential Quadratic Programming algorithm for updating the design parameters. This SQP algorithm allows to directly handle the non-linear constraints during the optimization process. An adjoint solver is used for computing the sensitivities of the flow quantities with respect to the design variables, such that the additional gradient computational cost is nearly independent of the number of design variables. In addition, the same optimization problem is performed with a gradient-free - metamodel assisted - evolutionary algorithm. A comparison of the two pareto-fronts obtained with both methods shows that the gradient-based approach allows to find the same optimum at a reduced computational cost. Moreover, the results suggest that the considered optimization problem is uni-modal, in other terms it is characterized by a single optimal solution.

Presenting Author: Romain Hottois von Karman Institute for Fluid Dynamics

Presenting Author Biography: Romain Hottois graduated with a Master's degree in Mechanical Engineering in 2020 at University of Mons (UMons) in Belgium. Within the framework of his Master thesis, he worked on the multi-objective adjoint-based optimization of the VKI-LS89 turbine cascade, in collaboration with von Karman Institute for Fluid Dynamics. The following year, he successfully completed a post-graduate Research Master degree at von Karman Institute within the Turbomachinery and Propulsion Department, where he worked on the optimization of a volute for radial compressors. Romain is now a PhD student at von Karman Institute and UMons, where he will write his thesis on aeroacoustics optimization in Turbomachinery.

Authors:

Romain Hottois von Karman Institute
Arnaud Châtel von Karman Institute
Tom Verstraete von Karman Institute
Grégory Coussement University of Mons
Tom De Bruyn von Karman Institute