基于LSTM的航空发动机涡轮盘局部服役温度-应力快速预测方法

程穆威; 杨晓光; 范永升; 姜仁杰; 石多奇

doi:10.13224/j.cnki.jasp.20240547

基于LSTM的航空发动机涡轮盘局部服役温度-应力快速预测方法

doi: 10.13224/j.cnki.jasp.20240547

北京航空航天大学能源与动力工程学院，北京 100191

基金项目: 国家科技重大专项（J2019-Ⅳ-0017-0085）；国家自然科学基金（52205139，12172021）

详细信息

作者简介:
程穆威（1997-），男，博士生，主要从事航空发动机高温结构强度与寿命预测的研究。E-mail：muwei_cheng@buaa.edu.cn

通讯作者:
范永升（1991-），男，副教授，博士，主要从事航空发动机结构完整性的研究。E-mail：fanys@buaa.edu.cn

中图分类号: V232.3
计量
- 文章访问数: 662
- HTML浏览量: 298
- PDF量: 73
- 被引次数: 0
出版历程
- 收稿日期: 2024-08-05
- 网络出版日期: 2025-07-31

Rapid prediction method for local service temperature and stress on aeroengine turbine disks based on LSTM network

School of Energy and Power Engineering，Beihang University，Beijing 100191，China

摘要

摘要:
建立基于发动机服役历程数据的涡轮盘关键位置温度-应力预测模型，旨在为热端部件的服役损伤评估和寿命分析提供温度和应力历程输入。通过发动机热力循环过程推算涡轮盘载荷相关的气路参数，基于数值模拟结果，构建并训练了基于长短期记忆（LSTM）神经网络的涡轮盘温度和应力预测模型。通过交叉验证和噪声植入，评估了模型的泛化能力。结果表明：所建立的模型在测试集上达到了5.4×10⁻⁴的归一化均方误差。利用所建立的方法分析了典型飞行动作下的关键位置温度、应力演化历程，为发动机的任务设计和涡轮盘服役损伤评估提供了重要输入。
- 涡轮盘 /
- LSTM网络 /
- 服役历程 /
- 载荷谱 /
- 数据驱动
Abstract:
A predictive model for temperature and stress at critical locations of turbine disks was developed using engine service data, with its aim to provide essential inputs for service damage assessment and life analysis of hot-section components. The research derived gas path parameters related to turbine disk loading through engine thermodynamic cycle analysis. Based on numerical simulation results, a Long Short-Term Memory (LSTM) neural network model was constructed and trained to predict turbine disk temperature and stress. The generalization of proposed model capability was evaluated through cross-validation and noise injection. Result demonstrated that the established model achieved a normalized mean square error of 5.4×10⁻⁴ on the test set. The developed methodology allowed to analyze temperature and stress evolution patterns during typical flight maneuvers, providing crucial inputs for engine mission design and turbine disk service damage assessment.
- turbine disk /
- LSTM network /
- operating history /
- load spectrum /
- data-driven

HTML

图 1 涡轮盘服役应力和温度预测方法流程图

Figure 1. Method flowchart for predicting turbine disc service stress and temperature

下载: 全尺寸图片幻灯片

图 2 双转子涡扇发动机结构示意图

Figure 2. Structural diagram of the two-spool turbofan engine

下载: 全尺寸图片幻灯片

图 3 LSTM单元和网络结构

Figure 3. Structure of LSTM cell and network

下载: 全尺寸图片幻灯片

图 4 涡轮盘几何模型和边界条件

Figure 4. Geometry model and boundary conditions of turbine disk

下载: 全尺寸图片幻灯片

图 5 起飞状态下的涡轮盘云图

Figure 5. Contour of turbine disk under take-off state

下载: 全尺寸图片幻灯片

图 6 发动机典型飞行任务下涡轮盘热边界条件剖面

Figure 6. Spectra of disk thermal boundary conditions under typical engine flight mission

下载: 全尺寸图片幻灯片

图 7 网络训练过程中损失函数

Figure 7. Loss function during network training

下载: 全尺寸图片幻灯片

图 8 模型预测误差带图

Figure 8. Model prediction error bands graph

下载: 全尺寸图片幻灯片

图 9 模型预测结果与测试集数值模拟结果对比

Figure 9. Comparison of model predictions with numerical simulation results from the test set

下载: 全尺寸图片幻灯片

图 10 叠加噪声对模型预测效果的影响

Figure 10. Impact of noise on the model prediction accuracy

下载: 全尺寸图片幻灯片

图 11 典型机动动作下的高压转速谱

Figure 11. Rotational speed spectra of high-pressure spool during typical maneuver

下载: 全尺寸图片幻灯片

图 12 起飞阶段关键位置的温度和Mises应力谱

Figure 12. Temperature and Mises stress spectra at critical location during takeoff stage

下载: 全尺寸图片幻灯片

图 13 起飞阶段轮心处2方向的热应力分析

Figure 13. Thermal stress analysis in direction 2 at critical location during takeoff

下载: 全尺寸图片幻灯片

图 14 不同三类次循环下关键位置的温度和应力谱

Figure 14. Temperature and stress spectra at critical location under different type Ⅲ sub-cycle

下载: 全尺寸图片幻灯片

图 15 不同三类次循环轮心处2方向的热应力分析

Figure 15. Thermal stress spectra in direction 2 at the disk hole under different type Ⅲ sub-cycle

下载: 全尺寸图片幻灯片

表 1 交叉验证时的模型误差（E_MSE）

Table 1. Model error （E_MSE） in cross-validation

任务误差（E_MSE）/10⁻⁴

1 5.2
2 8.5
3 6
4 2.8
5 6
6 8.3
7 6.2
8 4.9
9 5.4
平均 8.7

下载: 导出CSV

参考文献(27)

[1]	苏清友. 航空涡喷、涡扇发动机主要零部件定寿指南[M]. 北京: 航空工业出版社, 2004. SU Qingyou. Guide for life determination of main components of aviation turbojet and turbofan engines[M]. Beijing: Aviation Industry Press, 2004. (in Chinese SU Qingyou. Guide for life determination of main components of aviation turbojet and turbofan engines[M]. Beijing: Aviation Industry Press, 2004. (in Chinese)
[2]	杨兴宇, 郑小梅, 孙燕涛, 等. 航空发动机使用寿命控制技术[M]. 北京: 科学出版社, 2018. YANG Xingyu, ZHENG Xiaomei, SUN Yantao, et al. Service life control technology of aero-engine[M]. Beijing: Science Press, 2018. (in Chinese YANG Xingyu, ZHENG Xiaomei, SUN Yantao, et al. Service life control technology of aero-engine[M]. Beijing: Science Press, 2018. (in Chinese)
[3]	TINGA T, VISSER W P J, DE WOLF W B, et al. Integrated lifing analysis tool for gas turbine components[C]// ASME Turbo Expo 2000: Power for Land, Sea, and Air. Munich, Germany: American Society of Mechanical Engineers, 2000: V001T01A020.
[4]	ABDUL GHAFIR M F, LI Y G, SINGH R, et al. Impact of operating and health conditions on aero gas turbine: hot section creep life using a creep factor approach[C]// ASME Turbo Expo 2010: Power for Land, Sea, and Air. Glasgow, UK: American Society of Mechanical Engineers Digital Collection, 2010: 533-545.
[5]	ESHATI S, LASKARIDIS P, HASLAM A, et al. The influence of humidity on the creep life of a high pressure gas turbine blade: Part Ⅱ case study[C]// ASME Turbo Expo 2012: Turbine Technical Conference and Exposition. Copenhagen, Denmark: American Society of Mechanical Engineers, 2012: 281-288.
[6]	NAEEM M. Impacts of low-pressure (LP) compressor’s deterioration upon an aero-engine’s high-pressure (HP) turbine blade’s life consumption[J]. The Aeronautical Journal, 2006, 110(1106): 227-238. doi: 10.1017/S0001924000001202
[7]	NAEEM M. Implications of turbine erosion for an aero-engine’s high-pressure-turbine blade’s low-cycle-fatigue life-consumption[J]. Journal of Engineering for Gas Turbines and Power, 2009, 131(5): 052501. doi: 10.1115/1.3078383
[8]	ABU A O. Integrated approach for stress based lifing of aero gas turbine blades[D]. Cranfield, East of England, UK: Cranfield University, 2013.
[9]	ESHATI S, ABDUL GHAFIR M F, LASKARIDIS P, et al. Impact of operating conditions and design parameters on gas turbine hot section creep life[C]// ASME Turbo Expo 2010: Power for Land, Sea, and Air. Glasgow, UK: American Society of Mechanical Engineers, 2010: 547-558.
[10]	徐敬沛, 王学民, 卿华, 等. 基于降维和随机森林的航空发动机涡轮盘应力预测[J]. 推进技术, 2023, 44(12): 2208072. XU Jingpei, WANG Xuemin, QING Hua, et al. Stress prediction of aero-engine turbine disk based on dimension reduction and random forest[J]. Journal of Propulsion Technology, 2023, 44(12): 2208072. (in Chinese XU Jingpei, WANG Xuemin, QING Hua, et al. Stress prediction of aero-engine turbine disk based on dimension reduction and random forest[J]. Journal of Propulsion Technology, 2023, 44(12): 2208072. (in Chinese)
[11]	BREITKOPF G E, SPEER T M. 飞行中受高瞬态热应力作用的关键转动构件低循环疲劳寿命损耗的估算[R]. 齐日春译. 沈阳: 中国航发沈阳发动机研究所, 1992. BREITKOPF G E, SPEER T M. Estimation of low cycle fatigue life loss of key rotary clock components subjected to high transient thermal stress in flight[R]. Translator: QI Richun. Shenyang: China Aero Engine Corporation Shenyang Engine Research Institute, 1992. (in Chinese BREITKOPF G E, SPEER T M. Estimation of low cycle fatigue life loss of key rotary clock components subjected to high transient thermal stress in flight[R]. Translator: QI Richun. Shenyang: China Aero Engine Corporation Shenyang Engine Research Institute, 1992. (in Chinese)
[12]	GABRIELSSON R, LUNDBERG R, AVRAN P. Status of the European gas turbine program: AGATA[C]// ASME 1998 International Gas Turbine and Aeroengine Congress and Exhibition. Stockholm, Sweden: American Society of Mechanical Engineers, 1998: V002T04A006.
[13]	王林山. 时滞递归神经网络[M]. 北京: 科学出版社, 2008. WANG Linshan. Delayed recurrent neural network[M]. Beijing: Science Press, 2008. (in Chinese WANG Linshan. Delayed recurrent neural network[M]. Beijing: Science Press, 2008. (in Chinese)
[14]	SHERSTINSKY A. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network[J]. Physica D: Nonlinear Phenomena, 2020, 404: 132306. doi: 10.1016/j.physd.2019.132306
[15]	皮骏, 马圣, 张奇奇, 等. 基于改进果蝇算法优化的GRNN航空发动机排气温度预测模型[J]. 航空动力学报, 2019, 34(1): 8-17. PI Jun, MA Sheng, ZHANG Qiqi, et al. Aero-engine exhaust gas temperature prediction model based on IFOA-GRNN[J]. Journal of Aerospace Power, 2019, 34(1): 8-17. (in Chinese PI Jun, MA Sheng, ZHANG Qiqi, et al. Aero-engine exhaust gas temperature prediction model based on IFOA-GRNN[J]. Journal of Aerospace Power, 2019, 34(1): 8-17. (in Chinese)
[16]	CANDON M, ESPOSITO M, FAYEK H, et al. Advanced multi-input system identification for next generation aircraft loads monitoring using linear regression, neural networks and deep learning[J]. Mechanical Systems and Signal Processing, 2022, 171: 108809. doi: 10.1016/j.ymssp.2022.108809
[17]	WALSH P P. Gas turbine performance[M]. 2nd ed. Oxford, UK: John Wiley & Sons, 2008.
[18]	SELLERS J F, DANIELE C J. DYNGEN: a program for calculating steady-state and transient performance of turbojet and turbofan engines[M]. Cleveland, US: National Aeronautics and Space Administration, 1975.
[19]	HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735-1780. doi: 10.1162/neco.1997.9.8.1735
[20]	伊恩·古德费洛, 约书亚·本吉奥, 亚伦·库维尔. 深度学习[M]. 赵申剑, 黎彧君, 符天凡, 等译. 北京: 人民邮电出版社, 2017.
[21]	曹建国. 航空发动机仿真技术研究现状、挑战和展望[J]. 推进技术, 2018, 39(5): 961-970. CAO Jianguo. Status, challenges and perspectives of aero-engine simulation technology[J]. Journal of Propulsion Technology, 2018, 39(5): 961-970. (in Chinese CAO Jianguo. Status, challenges and perspectives of aero-engine simulation technology[J]. Journal of Propulsion Technology, 2018, 39(5): 961-970. (in Chinese)
[22]	丁有宇. 汽轮机强度计算手册[M]. 北京: 中国电力出版社, 2010. DING Youyu. Handbook of steam turbine strength calculation[M]. Beijing: China Electric Power Press, 2010. (in Chinese DING Youyu. Handbook of steam turbine strength calculation[M]. Beijing: China Electric Power Press, 2010. (in Chinese)
[23]	国林磊. 高温合金、铜合金与模具钢接触换热行为研究[D]. 辽宁大连: 大连理工大学, 2022. GUO Linlei. Study on contact heat transfer behavior of superalloy, copper alloy and die steel[D]. Dalian Liaoning: Dalian University of Technology, 2022. (in Chinese GUO Linlei. Study on contact heat transfer behavior of superalloy, copper alloy and die steel[D]. Dalian Liaoning: Dalian University of Technology, 2022. (in Chinese)
[24]	朱之丽, 陈敏, 唐海龙, 等. 航空燃气涡轮发动机工作原理及性能[M]. 3版. 上海: 上海交通大学出版社, 2023. ZHU Zhili, CHEN Min, TANG Hailong, et al. Working principle and performance of aircraft gas turbine engines[M]. 3rd ed. Shanghai: Shanghai Jiao Tong University Press, 2023. (in Chinese ZHU Zhili, CHEN Min, TANG Hailong, et al. Working principle and performance of aircraft gas turbine engines[M]. 3rd ed. Shanghai: Shanghai Jiao Tong University Press, 2023. (in Chinese)
[25]	KURZKE J, HALLIWELL I. Propulsion and power[M]. Cham: Springer International Publishing, 2018.
[26]	SAPSARD M. Recommended practices for monitoring gas turbine engine life consumption[R]. Quebec, Canada: Canada Communication Group Inc, 2000.
[27]	MIL-HDBK-1783. Engine structural integrity program (ENSIP)[S]. Washington DC: Department of Defense, 1984.