Entropy generation rate optimisation for profiled endwall design for axial turbines

Martinez-Castro, Pedro A; Williams, Richard J; Ingram, Grant L

Abstract

This paper investigates the formulation of the optimal objective function for aerodynamic design of turbomachinery using genetic algorithms (GA) supported by Computational Fluid Dynamics (CFD). The aim is to outline entropy generation rate minimisation as the most suitable objective function for improving turbomachinery. To this end, the calculation of this variable in RANS-CFD simulations is explained, followed by a sensitivity analysis to mesh size, turbulence model and calculation method in a turbine passage of a low-speed linear cascade. Entropy generation rate minimisation was then used as the objective function to improve profiled endwall (PEW) design for the same cascade. The results of entropy generation rate per unit surface area of optimised designs were compared with previous studies' endwall designs. This confirmed that the PEWs reduce secondary loss by weakening the pressure side of the horseshoe vortex and its interaction with the suction surface boundary layer. Comparison of loss coefficient CFD results with measurements from the reference cascade with original planar and the optimised PEW showed good agreement between simulations and experiments, demonstrating the proposed design approach's effectiveness. This paper recommends adopting entropy generation rate as the objective function because it yields superior designs and provides deeper insights into loss mechanisms.

Citation

Martinez-Castro, P. A., Williams, R. J., & Ingram, G. L. (2024, September). Entropy generation rate optimisation for profiled endwall design for axial turbines. Presented at GPPS Chania 2024, Chania

• Presentation Conference Type: Conference Paper (published)
• Conference Name: GPPS Chania 2024
• Start Date: Sep 4, 2024
• End Date: Sep 6, 2024
• Acceptance Date: Sep 4, 2024
• Online Publication Date: Sep 4, 2024
• Publication Date: Sep 4, 2024
• Deposit Date: Oct 23, 2024
• Publicly Available Date: Oct 24, 2024
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.33737/gpps24-tc-081
• Public URL: https://durham-repository.worktribe.com/output/2981505
• Publisher URL: https://gpps.global/gpps-chania24/