基于PointNet和CGAN的锯齿状槽道冷却性能研究

刘金江; 李军现; 黄鑫; 韩泽涛; 张润生; 杜小泽

doi:10.19805/j.cnki.jcspe.2026.250648

欢迎来到能源电力期刊网！

English Version 期刊登录

您当前的位置：

首页 >

文章列表页 >

基于PointNet和CGAN的锯齿状槽道冷却性能研究

更新时间：2026-10-06

- 基于PointNet和CGAN的锯齿状槽道冷却性能研究
- Research on the Cooling Performance of Sawtooth-shaped Channel Based on PointNet and CGAN
- 动力工程学报 2026年46卷第3期页码：158-167
- 作者机构：
  
  1. 南阳师范学院人工智能与软件工程学院,河南,南阳,473000
  2. 河南省数字图像大数据发展创新实验室,河南,南阳,473000
  3. 华北电力大学能源动力与机械工程学院,北京,102206
- 作者简介：
  
  刘金江(1974—),男,河南新乡人,教授,研究方向为数据分析及处理
  张润生(通信作者),男,讲师,博士,E-mail:runshengzhang@ncepu.edu.cn
- 基金信息：
  
  国家自然科学基金资助项目(52306082,62502243);中国博士后科学基金资助项目(2024T170268,GZC20230789);河南省研究生教育改革与质量提升工程项目(YJS2026YBGZZ46)
- DOI：10.19805/j.cnki.jcspe.2026.250648
  中图分类号：
- 网络首发：2026-03-10，
  
  纸质出版：2026-03-10
- 稿件说明：
移动端阅览
刘金江,李军现,黄鑫,韩泽涛,张润生,杜小泽. 基于PointNet和CGAN的锯齿状槽道冷却性能研究动力工程学报, 2026, 46(3): 158-167 https://doi.

org/10.19805/j.cnki.jcspe.2026.250648
刘金江,李军现,黄鑫,韩泽涛,张润生,杜小泽. 基于PointNet和CGAN的锯齿状槽道冷却性能研究动力工程学报, 2026, 46(3): 158-167 https://doi. DOI： 10.19805/j.cnki.jcspe.2026.250648.

org/10.19805/j.cnki.jcspe.2026.250648 DOI：

摘要

针对现有机器学习模型预测气膜冷却效率分布时依赖低维特征输入

且模型不具有通用性的问题

提出了一种基于PointNet与条件生成对抗网络(CGAN)的数据驱动框架。该框架将计算域中的不规则网格节点转换为点云形式

通过PointNet提取全局特征和局部特征

并以此特征作为条件信息

促使CGAN的生成器重构出与计算流体力学(CFD)高度一致的二维流场分布图

然后通过判别器进行对抗训练以提升输出图像的细节。实验中以锯齿状槽道为研究对象

结果表明:训练集和测试集的绝对误差不超过0.05

相对误差不超过6%

计算耗时低于传统CFD方法。所构建的模型在样本充足时能够适配所有气膜冷却结构

并解决了稀疏点云数据无法高精度重构气膜冷却效率分布的问题。

Abstract

To tackle the issues of existing machine learning models

which rely on low-dimensional feature inputs for predicting film-cooling efficiency distributions and lack general applicability

a data-driven framework based on PointNet and conditional generative adversarial network (CGAN) was proposed. In this framework

the irregular mesh nodes within the computational domain were transformed into point cloud data. PointNet was then employed to extract both global and local features from this point cloud. These extracted features serve as conditional information to guide the CGAN generator in reconstructing two-dimensional flow field distributions that closely align with those obtained through computational fluid dynamics (CFD) simulations. Additionally

the adversarial training was conducted by discriminator to improve the detail quality of the generated images. Experiments were carried out through using a sawtooth-shaped channel as the research subject. Results demonstrate that the absolute error between the training and test sets does not surpass 0.05

while the relative error remains below 6%. Moreover

the computational time required is less than that of traditional CFD method. When provided with sufficient training samples

the developed model can be applied to all types of film-cooling structures and effectively address the challenge of accurately reconstructing film-cooling efficiency distributions from sparse point cloud data.

关键词

Keywords

references

张德恒, 贺业光, 赵胜岩, 等. 凹坑造型与深度对气膜冷却效果影响的研究[J]. 动力工程学报, 2025, 45(7): 961-967.ZHANG Deheng, HE Yeguang, ZHAO Shengyan, et al. Influence of crater shape and depth on film cooling effectiveness[J]. Journal of Chinese Society of Power Engineering, 2025, 45(7): 961-967.

GOLDSTEIN R J, ECKERT E R G, TSOU F K, et al. Film cooling with air and helium injection through a rearward-facingslot into a supersonic air flow[J]. AIAA Journal, 1966, 4(6): 981-985.

张振, 苏欣荣, 袁新. 基于物理信息神经网络的气膜冷却湍流模型反演学习[J]. 动力工程学报, 2024, 44(9): 1459-1465.ZHANG Zhen, SU Xinrong, YUAN Xin. PINN-based inversion learning of turbulence model for film cooling[J]. Journal of Chinese Society of Power Engineering, 2024, 44(9): 1459-1465.

BRYNJOLFSSON E, MITCHELL T, ROCK D. What can machines learn, and what does it mean for occupations and the economy?[J] //Proceedings of the AEA Papers and Proceedings, 2018, 108: 43-47.

JORDAN M I, MITCHELL T M. Machine learning: trends, perspectives, and prospects[J].Science, 2015, 349(6245): 255-260.

MILANI P M, LING Julia, SAEZ-MISCHLICH G, et al. A machine learning approach for determining the turbulent diffusivity in film cooling flows[J].Journal of Turbomachinery, 2018, 140(2): 021006.

YANG Li, WANG Qi, HUANG Kang, et al. Establishment of a long-short-term-memory model to predict film cooling effectiveness under superposition conditions[J].International Journal of Heat and Mass Transfer, 2020, 160: 120231.

SELLERS JR J P. Gaseous film cooling with multiple injection stations[J].AIAA Journal, 1963, 1(9): 2154-2156.

WANG Yaning, WANG Wen, TAO Guocheng, et al. Optimization of the semi-sphere vortex generator for film cooling using generative adversarial network[J]. International Journal of Heat and Mass Transfer, 2022, 183: 122026.

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.

袁佳铖, 宗文刚, 曾磊, 等. 基于卷积神经网络的气动热预测方法[J]. 空气动力学学报, 2024, 42(1): 13-25.YUAN Jiacheng, ZONG Wengang, ZENG Lei, et al. CNN-based method for predicting aerodynamic heating[J]. Acta Aerodynamica Sinica, 2024, 42(1): 13-25.

程波, 卢绪祥, 刘雨菲, 等. 基于数据驱动的汽轮机冷端系统建模与运行优化[J]. 动力工程学报, 2025, 45(3): 359-365.CHENG Bo, LU Xuxiang, LIU Yufei, et al. Modeling and operation optimization of steam turbine cold end system based on data driven[J]. Journal of Chinese Society of Power Engineering, 2025, 45(3): 359-365.

刘忠, 周泽华, 邹淑云, 等. 基于迁移学习和CNN-LSTM的水轮机空化状态识别方法[J]. 动力工程学报, 2024, 44(10): 1533-1540.LIU Zhong, ZHOU Zehua, ZOU Shuyun, et al. Cavitation state recognition method of hydraulic turbine based on transfer learning and CNN-LSTM[J]. Journal of Chinese Society of Power Engineering, 2024, 44(10): 1533-1540.

邓志平, 任少君, 翁琪航, 等. 基于物理信息神经网络的生物质气化产物分布预测方法[J]. 动力工程学报, 2024, 44(5): 719-726.DENG Zhiping, REN Shaojun, WENG Qihang, et al. Prediction method for biomass gasification product distribution based on physics-informed neural network[J]. Journal of Chinese Society of Power Engineering, 2024, 44(5): 719-726.

赵亮, 胡杰, 刘汉, 等. 基于语义分割的深度学习激光点云三维目标检测[J]. 中国激光, 2021, 48(17): 1710004.ZHAO Liang, HU Jie, LIU Han, et al. Deep learning based on semantic segmentation for three-dimensional object detection from Point Clouds[J].Chinese Journal of Lasers, 2021, 48(17): 1710004.

李琪琪, 花向红, 赵不钒, 等. 一种室内场景点云平面分割的新方法[J]. 中国激光, 2021, 48(16): 1604002. LI Qiqi, HUA Xianghong, ZHAO Bufan, et al. New method for plane segmentation of indoor scene point cloud[J]. Chinese Journal of Lasers, 2021, 48(16): 1604002.

LU Yiping, DHUNGEL A, EKKAD S V, et al. Effect of trench width and depth on film cooling from cylindrical holes embedded in trenches[J]. Journal of Turbomachinery, 2009, 131(1): 011003.

QI C R, SU Hao, MO Kaichun, et al. PointNet: deep learning on point sets for 3D classification and segmentation[C] //Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition. Honolulu: IEEE, 2017: 77-85.

QI Shizhen, HAN Shaohua, XIANG Zhen, et al. Optimization and mechanistic analysis of the configurational parameters of a serrated trench for improving film cooling performance[J]. International Journal of Thermal Sciences, 2025, 208: 109436.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

预测扇形孔气膜分布的“核”参数模型

基于符号回归与待定系数优化的气膜冷却经验关联式构造方法

凹坑造型与深度对气膜冷却效果影响的研究

计入工业园区电热负荷需求的源网荷储优化调度方法研究

考虑生物质混燃碳捕集与低碳激励机制的虚拟电厂优化调度