Optimization of Turbine Blade Trailing Edge Cooling Using Self-Organized Geometries and Multi-Objective Approaches

Kaibin Hu,Xiaobo Wang,Shengquan Zhong,Cheng Lu,Bocheng Yu,Li Yang,Yu Rao

DOI: https://doi.org/10.1016/j.energy.2023.130013

IF: 9

2024-01-01

Energy

Abstract:Gas turbines are widely utilized as essential power machinery across diverse energy-related industries. Efficient and durable cooling of gas turbine blades is crucial for the performance and life of gas turbine. However, achieving effective cooling for turbine blade trailing edges remained challenging due to the narrow, thin and structurally limited spaces. To address the aforementioned challenge, this study proposed a generative design method by employing self-organization equations to parameterize the topology of wedge-shaped channels within the trailing edge region as a substitute for the conventional pin-fin structures. Additionally, a multi-objective Bayesian approach was utilized to optimize the topological parameters, with comprehensive emphases on improving five design objectives. A comparative analysis of the performance was conducted between self-organized structures and pin-fin structures. Notably, the selected self-organized structure achieved an impressive 43 % increase in the total Nusselt number and a substantial 37 % reduction in local Nusselt number variance under the same pressure loss. The results demonstrated that self-organized structures exhibited superior heat transfer performance with lower pressure drop and improved uniformity. The performance improvement could be attributed to the strategic generation of solids by the proposed generative design algorithm, distributing the flow more reasonable.