Comparison of Different Prediction Methods for Heat Transfer Distribution of Lantern-Shaped Pin-Fin Channel with Different Deep Learning Surrogate Models

GAO Shanghong; ZHANG Weixin; YANG Kefeng; WANG Xiangyu; FENG Zhenping

doi:10.7652/xjtuxb202402004

您当前的位置：

首页 >

文章列表页 >

Comparison of Different Prediction Methods for Heat Transfer Distribution of Lantern-Shaped Pin-Fin Channel with Different Deep Learning Surrogate Models

更新时间：2025-05-13

- Comparison of Different Prediction Methods for Heat Transfer Distribution of Lantern-Shaped Pin-Fin Channel with Different Deep Learning Surrogate Models
- Vol. 58, Issue 2, Pages: 31-42(2024)
- 作者机构：
  
  西安交通大学能源与动力工程学院,西安,710049
- 作者简介：
- 基金信息：
- DOI：10.7652/xjtuxb202402004
  CLC： TK474.7
- Online First：10 February 2024，
  
  Published：2024
- 稿件说明：
移动端阅览
GAO Shanghong, ZHANG Weixin, YANG Kefeng, et al. Comparison of Different Prediction Methods for Heat Transfer Distribution of Lantern-Shaped Pin-Fin Channel with Different Deep Learning Surrogate Models[J]. 2024, 58(2): 31-42.
DOI：

GAO Shanghong, ZHANG Weixin, YANG Kefeng, et al. Comparison of Different Prediction Methods for Heat Transfer Distribution of Lantern-Shaped Pin-Fin Channel with Different Deep Learning Surrogate Models[J]. 2024, 58(2): 31-42. DOI： 10.7652/xjtuxb202402004.

摘要

为提高对灯笼型扰流柱通道端壁面的换热分布预测能力

构建并比较了几种深度学习代理模型的预测性能。应用数据驱动的思想

建立了多项式响应面模型(RSM)、径向基函数模型(RBF)、径向基神经网络模型(RBFNN)、Kriging模型和总体平均近似模型(Ensemble)共5种传统代理模型以及5种基于生成对抗网络的深度学习代理模型

包括有残差网络的pix2pix模型、pix2pixHD模型、CycleGAN模型、StarGAN模型以及单一无残差网络的pix2pix模型

以扰流柱截面参数为设计变量

通过拉丁超立方抽样分别得到了样本数为50、25、12、6和3的训练集

并根据扰流柱尾迹高换热区分布特点

将训练样本数为50的数据集分为宽样本数据集和窄样本数据集

比较了不同代理模型对端壁面换热性能的预测精度、计算成本和泛化能力。结果表明:有残差网络的pix2pix模型相比于无残差网络的pix2pix模型

预测精度得到有效提高

在样本数为50的情况下

面平均值预测误差从0.68%降低到0.32%

平均相对误差从6.89%降低到6.41%

而且当训练样本数减少时

有残差网络的模型预测能力更加突出; 传统代理模型的时间成本可忽略不计

但深度学习模型的单卡训练时间较长

且增加残差网络后的模型计算成本更高; 当训练样本数为50时

传统代理模型和深度学习模型之间预测精度差异不大; 当训练样本数逐渐减少时

深度学习代理模型展示出更高的预测精度和泛化能力; Kriging模型虽然泛化能力强

但是预测结果趋同; RSM模型、RBF模型和Ensemble模型泛化能力最差

训练样本数较少时

预测结果严重失真。可见在换热性能预测方面

深度学习模型在预测精度与泛化能力上均有显著优势

尤其适合于小样本问题

对提高灯笼型扰流柱截面设计效率具有参考价值。

Abstract

In order to improve the prediction ability of heat transfer performance on lantern-shaped pin-fin endwall

several deep learning surrogate models are constructed and their prediction performance compared. Based on data-driven thinking

five traditional surrogate models

namely polynomial response surface model(RSM)

radial basis function model(RBF)

radial basis function neural network(RBFNN)

Kriging model and ensemble of surrogate model(Ensemble)are proposed. Additionally

five generative adversarial network surrogate models based on deep learning

namely pix2pix model

pix2pixHD model

CycleGAN model

StarGAN model all with residual network

and another pix2pix model without residual network are proposed. The training sets with various sample numbers of 50

6 and 3 were obtained by Latin hypercube sampling

respectively

using the pin-fin cross-sectional shape as the design variables. According to the distribution characteristics of high heat transfer zone

the dataset with a training sample number of 50 is divided into a wide sample dataset and a narrow sample dataset. On this basis

the various heat transfer performance of different surrogate models are compared in terms of prediction accuracy

computational cost and generalization ability. The results show that

compared with the pix2pix model without residual network

the prediction accuracy of the pix2pix model with residual network is effectively improved

and the prediction error of the surface mean is reduced from 0.68% to 0.32%

and the averaged relative error is reduced from 6.89% to 6.41% when the sample number is 50. When the number of training samples decreases

the prediction performance of the models with residual network becomes more prominent. The time cost of traditional surrogate models is negligible

but the training time of deep learning models is long especially for models with residual network. When the number of training samples is 50

there is little difference in prediction accuracy between the traditional surrogate models and deep learning models. With gradual decrease in the number of training samples

deep learning surrogate models show higher prediction accuracy and generalization capabilities. Although the Kriging model has strong generalization ability

the prediction results tend to be similar. RSM model

RBF model

and Ensemble model have the lowest generalization ability

and the prediction results are severely distorted when there are fewer training samples. For the heat transfer distribution prediction of the pin-fin endwall

the deep learning models possess significant advantages in terms of both prediction accuracy and generalization ability

which are more suitable for small batch sample problems. This study has important reference value for improving the cross-sectional shape design efficiency of lantern-shaped pin-fin.

关键词

Keywords

references

刘钊, 杨星, 丰镇平. 燃气轮机透平叶片传热和冷却研究: 气膜冷却 [J]. 热力透平, 2014, 43(1): 1-9.

LIU Zhao, YANG Xing, FENG Zhenping. Study on heat transfer and cooling in gas turbine blade: film cooling [J]. Thermal Turbine, 2014, 43(1): 1-9.

METZGER D E, BERRY R A, BRONSON J P. Developing heat transfer in rectangular ducts with staggered arrays of short pin fins [J]. Journal of Heat Transfer, 1982, 104(4): 700-706.

HORBACH T, SCHULZ A, BAUER H J. Trailing edge film cooling of gas turbine airfoils: external cooling performance of various internal pin fin configurations [J]. Journal of Turbomachinery, 2011, 133(4): 041006.

OTTO M, GUPTA G, TRAN P K, et al. Investigation of endwall heat transfer in staggered pin fin arrays [J]. Journal of Turbomachinery, 2021, 143(2): 021009.

周子杰, 王新军, 费昕阳. 燃机透平静叶尾缘柱肋通道内的汽雾/空气冷却流动与换热特性数值研究 [J]. 西安交通大学学报, 2016, 50(11): 21-27.

ZHOU Zijie, WANG Xinjun, FEI Xinyang. Numerical investigation for flow and heat transfer characteristics of air and air/mist cooling in gas turbine stator trail edge path [J]. Journal of Xi'an Jiaotong University, 2016, 50(11): 21-27.

韦宏, 祖迎庆. 冲击+扰流柱双层壁冷却结构的强化换热性能及流阻特性 [J/OL]. 航空动力学报[2023-07-01]. https://doi.org/10.13224/j.cnki.jasp.20220489.

WEI Hong, ZU Yingqing. Enhanced heat transfer performance and flow resistance characteristics of the double-wall cooling structures with jet impingement and pins [J/OL]. Journal of Aerospace Power [2023-07-01]. https://doi.org/10.13224/j.cnki.jasp.20220489.

刘文彪. 基于深度学习的旋转设备故障预测方法研究 [D]. 北京: 中国石油大学(北京), 2021.

穆雪峰, 姚卫星, 余雄庆, 等. 多学科设计优化中常用代理模型的研究 [J]. 计算力学学报, 2005, 22(5): 608-612.

MU Xuefeng, YAO Weixing, YU Xiongqing, et al. A survey of surrogate models used in MDO [J]. Chinese Journal of Computational Mechanics, 2005, 22(5): 608-612.

LI Ping, KIM K Y. Multiobjective optimization of staggered elliptical pin-fin arrays [J]. Numerical Heat Transfer: Part A Applications, 2008, 53(4): 418-431.

MOON M A, KIM K Y. Analysis and optimization of fan-shaped pin-fin in a rectangular cooling channel [J]. International Journal of Heat and Mass Transfer, 2014, 72: 148-162.

MOON M A, HUSAIN A, KIM K Y. Shape optimization of a rotating rectangular channel with pin-fins by Kriging method [C]//2010 14th International Heat Transfer Conference. New York, NY, USA: ASME, 2010: 169-176.

OSTANEK J K. Improving pin-fin heat transfer predictions using artificial neural networks [J]. Journal of Turbomachinery, 2014, 136(5): 051010.

LIU Zhansheng, YANG Kefeng, FENG Zhenping. Aero-thermal coupled design optimization of a turbine vane and the effect on heat load of endwall [C]//ASME Turbo Expo 2020: Turbomachinery Technical Conference and Exposition. New York, NY, USA: ASME, 2020: V02DT38A024.

LIU Zhansheng, YAGN Xing, GAO Chun, et al. Aero-thermal coupled design optimization of the non-axisymmetric endwall for a gas turbine blade [C]//ASME Turbo Expo 2018: Turbomachinery Technical Conference and Exposition. New York, NY, USA: ASME, 2018: V02DT46A018.

LIU Zhansheng, YANG Kefeng, YANG Xing, et al. Optimization of fan-shaped film cooling holes for enhancing cooling effectiveness under variable coolant inlet pressure ratios [C]//ASME Turbo Expo 2020: Turbomachinery Technical Conference and Exposition. New York, NY, USA: ASME, 2020: V07BT12A039.

ZHANG Weixin, LI Feng, JIA Zhe, et al. Experimental study on performance verification test of optimized film cooling holes in a linear cascade [J]. Applied Thermal Engineering, 2022, 212: 118574.

GUI Jie, SUN Zhenan, WEN Yonggang, et al. A review on generative adversarial networks: algorithms, theory, and applications [J]. IEEE Transactions on Knowledge and Data Engineering, 2023, 35(4): 3313-3332.

YANG Li, DAI Wei, RAO Yu, et al. Optimization of the hole distribution of an effusively cooled surface facing non-uniform incoming temperature using deep learning approaches [J]. International Journal of Heat and Mass Transfer, 2019, 145: 118749.

戴维, 杨力, 饶宇. 基于生成对抗网络的涡轮端壁气膜冷却建模方法 [J]. 工程热物理学报, 2020, 41(10): 2420-2424.

DAI Wei, YANG Li, RAO Yu. Modeling method of film cooling of turbine endwall based on generative adversarial networks [J]. Journal of Engineering Thermophysics, 2020, 41(10): 2420-2424.

LI Zuobiao, SU Liangjun, WEN Fengbo, et al. Deep learning method for fast prediction of film cooling performance [J]. Physics of Fluids, 2022, 34(4): 047111.

李左飙, 温风波, 唐晓雷, 等. 基于深度学习的单排孔气膜冷却性能预测 [J]. 航空学报, 2021, 42(4): 307-318.

LI Zuobiao, WEN Fengbo, TANG Xiaolei, et al. Prediction of single-row hole film cooling performance based on deep learning [J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(4): 307-318.

JIANG Chiju, ZHANG Weihao, LI Ya, et al. Multi-scale Pix2Pix network for high-fidelity prediction of adiabatic cooling effectiveness in turbine cascade [J]. Energy, 2023, 265: 126381.

YANG Kefeng, SUI Yongfeng, WANG Xiangyu, et al. Study on flow and heat transfer characteristics in rectangular channels with lantern-shaped pin fin array: part Ⅰ effect of sphere diameter [J/OL]. Numerical Heat Transfer: Part A Applications[2023-07-01]. https://doi.org/10.1080/10407782.2023.2251085.

GOEL T, HAFTKA R T, SHYY W, et al. Ensemble of surrogates [J]. Structural and Multidisciplinary Optimization, 2007, 33(3): 199-216.

ISOLA P, ZHU Junyan, ZHOU Tinghui, et al. Image-to-image translation with conditional adversarial networks [C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2017: 5967-5976.

WANG Tingchun, LIU Mingyu, ZHU Junyan, et al. High-resolution image synthesis and semantic manipu-lation with conditional GANs [C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2018: 8798-8807.

ZHU Junyan, PARK T, ISOLA P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks [C]//2017 IEEE International Conference on Computer Vision(ICCV). Piscataway, NJ, USA: IEEE, 2017: 2242-2251.

CHOI Y, CHOI M, KIM M, et al. StarGAN: unified generative adversarial networks for multi-domain image-to-image translation [C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2018: 8789-8797.

RONNEBERGER O, FISCHER P, BROX T. U-net: convolutional networks for biomedical image segmentation [C]//2015 Medical Image Computing and Computer-Assisted Intervention(MICCAI). Cham, Germany: Springer International Publishing, 2015: 234-241.

HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition [C]//2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). Piscataway, NJ, USA: IEEE, 2016: 770-778.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Catalytic Cracking of Ammonia for Ammonia-Hydrogen Co-Firing in Gas Turbines

Intelligent Quantitative Non-Destructive Evaluation Method for Delamination of Carbon Fiber Composite Tube with Mirror-Reflection Infrared Thermography

Model Predictive Control-Oriented Deep Koopman Global Linear Modeling Method for Wind Turbines

Uncertainty Quantification and Sensitivity Analysis on Aerodynamic Performance of Gas Turbine Exhaust Diffuser

Dynamic Liquid Level Prediction of Steam Generator under Main Steam Pipe Rupture Accidents

Related Author

SI Gengfan

XIAO Tong

ZHANG Meng

WANG Jinhua

HUANG Zuohua

YANG Yang

CHEN Hongen

LIU Bofan

Related Institution

State Key Laboratory of Multiphase Flow in Power Engineering，Xi’an Jiaotong University

School of Aerospace Engineering，Xi’an Jiaotong University

State Key Laboratory for Strength and Vibration of Mechanical Structures，Xi’an Jiaotong University

Shaanxi Engineering Research Center of Nondestructive Testing and Structural Integrity Evaluation，Xi’an Jiaotong University

School of Electrical Engineering, Xi'an Jiaotong University

⁰