基于机器学习的篦齿封严流动传热性能预测

高峰 ,  张敏慧 ,  王春华 ,  张居兵

南京师范大学学报(工程技术版) ›› 2026, Vol. 26 ›› Issue (2) : 9 -20.

PDF (4352KB)
南京师范大学学报(工程技术版) ›› 2026, Vol. 26 ›› Issue (2) : 9 -20. DOI: 10.3969/j.issn.1672-1292.2026.02.002
动力工程及工程热物理

基于机器学习的篦齿封严流动传热性能预测

作者信息 +

Prediction of Flow and Heat Transfer Performance of Labyrinth Seals Based on Machine Learning

Author information +
文章历史 +
PDF (4456K)

摘要

以光滑衬套直通式篦齿为研究对象,基于机器学习方法实现了篦齿封严流动传热性能的精细化和快速化预测. 遴选篦齿封严间隙、齿距、篦齿前倾角、入口总压和转速为机器学习模型输入,基于拉丁超立方抽样和计算流体力学(computation fluid dynamic,CFD)仿真方法建立机器学习所需的学习样本. 针对计算域划分独立的机器学习网格节点(其数量远小于 CFD 网格),把径向基函数神经网络独立建立在每个节点上并直接预测对应位置的总压、总温及速度. 将径向基函数神经网络与 CFD 计算结果对比分析,结果表明:对于总压、总温及速度,径向基函数神经网络预测的平均绝对百分比误差分别为 0.07%、0.17%和 0.14%,均方根误差分别为 1.18kPa、1.8K 和 6.45m/s. 此外,对机器学习模型输入和输出参数之间的数值关联进行了详细分析. 研究成果为篦齿封严流动传热高效、低成本建模提供了一种新的技术途径.

Abstract

The study focuses on the fine-grained and rapid prediction of the flow and heat transfer performance of labyrinth seals with smooth bushings using machine learning methods. Input parameters are selected for the machine learning model include the labyrinth seal clearance, tooth spacing, forward angle of the labyrinth tooth, inlet total pressure, and rotational speed. Learning samples required for the machine learning model are established based on Latin hypercube sampling and computation fluid dynamic (CFD) simulation methods. Radial basis function neural networks are independently constructed at each node of the machine learning grid (with a number of nodes significantly fewer than the CFD grid) to directly predict the total pressure, total temperature, and velocity at the corresponding locations. Comparison between the radial basis function neural network and CFD calculation results reveals that the mean absolute percentage errors for total pressure, total temperature, and velocity are 0.07%, 0.17%, and 0.14%, with root mean square errors of 1.18 kPa, 1.8 K, and 6.45 m/s, respectively. A detailed analysis of the numerical relationships between the input and output parameters of the machine learning model is conducted. The research findings provide a new technical approach for efficient and low-cost modeling of the heat transfer performance of labyrinth seals.

关键词

航空发动机 / 篦齿封严 / 机器学习 / 径向基函数神经网络 / 泄漏系数

Key words

aircraft engine / labyrinth seals / machine learning / radial basis function neural network / leakage coefficient

引用本文

引用格式 ▾
高峰,张敏慧,王春华,张居兵. 基于机器学习的篦齿封严流动传热性能预测[J]. 南京师范大学学报(工程技术版), 2026, 26(2): 9-20 DOI:10.3969/j.issn.1672-1292.2026.02.002

登录浏览全文

4963

注册一个新账户 忘记密码

参考文献

[1]

聂顺鹏, 刘高文, 雷昭, 等. 有无蜂窝的直通斜齿流动特性旋转实验研究[J]. 推进技术, 2020, 41(5): 1015-1022.

[2]

任国哲, 卢德正, 孙丹, 等. 结构参数对篦齿封严流动特性与最佳封严齿数影响研究[J]. 推进技术, 2025, 46(2): 293-305.

[3]

陈礼顺, 王彦岭, 卢建红, 等. 航空发动机封严技术的研究和应用进展[J]. 航空制造技术, 2008(8): 82-84, 95.

[4]

Denecke J, Dullenkopf K, Bauer H J. Dimensional analysis and scaling of rotating seals[C]// ASME Turbo Expo 2005:Power for Land,Sea,and Air. Reno,Nevada,USA, 2005: 1149-1160.

[5]

Denecke J, Dullenkopf K, Wittig S. Experimental investigation of the total temperature increase and swirl development in rotating labyrinth seals[C]// ASME Turbo Expo 2005:Power for Land,Sea,and Air. Reno,Nevada,USA, 2005: 1161-1171.

[6]

Millward J A, Edwards M F. Windage heating of air passing through labyrinth seals[J]. Journal of Turbomachinery, 1996, 118(2): 414-419.

[7]

吴丁毅, 刘振侠. 交错式篦齿封严特性的实验研究[J]. 推进技术, 1997, 18(2): 87-90.

[8]

Kulkarni D Y, Mare L D. Development of a new loss model for turbomachinery labyrinth seals[J]. Journal of Engineering for Gas Turbines and Power, 2023, 145(6): 061016.

[9]

Bermejo O,Gallardo M J,Sotillo A, et al. Numerical and experimental study of flutter in a realistic labyrinth seal[J]. International Journal of Turbomachinery,Propulsion and Power, 2024, 9(2): 13.

[10]

Xie L, Du Q, Liu G, et al.. Investigation of flow characteristics in a rotor-stator cavity under crossflow using wall-modelled large-eddy simulation[J]. Journal of Zhejiang University:Science A, 2023, 24(6): 473-497.

[11]

李驰, 张勃, 马壮, 等. 台阶式斜篦齿封严压降与风阻温升数值分析[J]. 润滑与密封, 2024, 49(6): 29-35.

[12]

李昶威, 孙丹, 赵欢, 等. 静态篦齿封严温变效应产生机理数值研究[J]. 航空动力学报, 2024, 39(2): 120-128.

[13]

Yan X, Li J, Song L M, et al.. Investigations on the discharge and total temperature increase characteristics of the labyrinth seals with honeycomb and smooth lands[J]. Journal of Turbomachinery, 2009, 131(4): 041009.

[14]

唐萌, 赵晨旭, 翟世杰. 射流对不同封严篦齿特性的影响分析研究[J]. 机械制造与自动化, 2022, 51(3): 130-134.

[15]

王鹏飞, 郭文, 张靖周. 典型阶梯篦齿高转速密封性能试验[J]. 航空动力学报, 2017, 32(12): 3057-3063.

[16]

孔晓治, 刘高文, 雷昭, 等. 齿型对压气机级间封严特性影响的实验研究[J]. 推进技术, 2018, 39(9): 2085-2093.

[17]

陈志英, 汪宇峰, 刘勇, 等. 基于封严间隙的涡轮盘篦齿综合优化设计[J]. 航空发动机, 2019, 45(5): 19-25.

[18]

Chun Y H, Ahn J. Effects of geometric parameters of a staggered labyrinth seal on leakage flow[J]. Journal of Mechanical Science and Technology, 2023, 37(6): 2959-2968.

[19]

Lee I S, Kang J Y, Kim J W, et al.. Effects of tip clearance,number of teeth,and tooth front angle on the sealing performance of straight and stepped labyrinth seals[J]. Journal of Mechanical Science and Technology, 2021, 35(4): 1539-1547.

[20]

Enache M, Carlanescu R, Mangra A, et al.. Investigation of flow through a labyrinth seal[J]. INCAS Bulletin, 2021, 13(2): 51-58.

[21]

Nayak C K. Effect of rotation on leakage and windage heating in labyrinth seals with honeycomb lands[J]. Journal of Engineering for Gas Turbines and Power, 2020, 142(8): 081001.

[22]

刘一帆, 王春华, 张卓, 等. 基于机器学习的全覆盖气膜冷却性能预测[J]. 中南大学学报(自然科学版), 2024, 55(6): 2334-2347.

[23]

王振宇, 王春华, 张靖周. 涡轮集气腔结构参数优化研究[J]. 中南大学学报(自然科学版), 2022, 53(7): 2756-2765.

[24]

Li Y, Wei D, Yu R, 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.

[25]

Pazireh S, Jeffrey J, Defoe A. New loss generation body force model for fan/compressor bladerows:An artificial-neural-network based methodology[J]. International Journal of Turbomachinery,Propulsion and Power, 2021, 6(1): 5.

[26]

Lee S, You D. Data-driven prediction of unsteady flow over a circular cylinder using deep learning[J]. Journal of Fluid Mechanics, 2019, 879(1): 217-254.

[27]

Wang Y, Wang W, Tao G, et al.. Two-dimensional film-cooling effectiveness prediction based on deconvolution neural network[J]. International Communications in Heat and Mass Transfer, 2021, 129: 105621.

[28]

张霞, 张效伟, 朱惠人, 等. 直通型篦齿封严特性的试验[J]. 航空动力学报, 2010, 25(8): 1753-1757.

[29]

Ma Z, Gao J, Hu W, et al.. Risk-adjustable stochastic schedule based on Sobol augmented Latin hypercube sampling considering correlation of wind power uncertainties[J]. IET Renewable Power Generation, 2021, 15(11): 2356-2367.

[30]

Huy K D, Duc K T, Seung E K. Stochastic analysis of semi-rigid steel frames using a refined plastic-hinge model and Latin hypercube sampling[J]. Engineering Structures, 2023, 291: 116313.

[31]

Hector C E, Erik C, Karla A, et al.. An advanced initialization technique for metaheuristic optimization:A fusion of Latin hypercube sampling and evolutionary behaviors[J]. Computational and Applied Mathematics, 2024, 43: 234.

[32]

姜峰, 洪林雄, 李华聪. 基于自适应径向基函数模型的区间不确定性分析方法[J]. 航空动力学报, 2024, 39(11): 73-82.

AI Summary AI Mindmap
PDF (4352KB)

0

访问

0

被引

详细

导航
相关文章

AI思维导图

/

〈 〉