Cooperative optimization of cooling units arrangement on gas turbine endwall with generating adversarial network-based surrogate models

Yu Song, Zhao Liu, Yuhui Han, Weixin Zhang and Zhenping Feng

Energy, 2025, vol. 323, issue C

Abstract: The optimization of film cooling arrangement significantly enhances the operational reliability of gas turbine blades, making it an essential design process for advanced gas turbines. Generating adversarial network-based surrogate models are considered highly advantageous for image generation. To demonstrate their application potential in improving the prediction of film cooling effectiveness over the turbine endwall, this study utilizes U-Net framework as the architecture for image processing. The genetic algorithm was selected as the global optimization method. After training the surrogate model, the mean absolute error of the validation set was 0.98 %, and the correlation index was 0.999. The results showed that the optimal film hole arrangement significantly improved the endwall film cooling performance while reducing passage total pressure losses. Specifically, film cooling performance increased by 119.73 % to 158.26 %, and the passage total pressure loss was reduced by 12.23 % to 13.59 % for optimized balance samples. The increase in the total heat transfer coefficient mainly resulted from the mixing of the leakage with the mainstream downstream of the mid-passage gap. Optimization achieves significant improvements in endwall cooling, with only a slight reduction in phantom cooling.

Keywords: Surrogate model; Generating adversarial network; Multi-objectives optimization; Cooling arrangement; Turbine endwall (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225014513
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:323:y:2025:i:c:s0360544225014513

DOI: 10.1016/j.energy.2025.135809

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().