Performance prediction of gas turbine engines driven by test and overall performance simulation data
-
摘要:
以某燃气轮机为研究对象，利用基于部件法的燃气轮机总体性能仿真模型，通过敏感性分析选取合适的部件特性修正系数，然后耦合试车数据设计共同工作方程组，并利用多点优化对部件特性进行修正，最终获得高精度的燃气轮机总体性能仿真模型。基于修正后总体性能仿真模型获得的数据，使用多层感知机（MLP）建立燃气轮机性能预测模型。研究结果表明：与试车数据相比，基于修正后燃气轮机总体性能仿真数据建立的预测模型，对各稳态工况点性能评估的误差均在1%以内，满足工程精度需求，并且比修正后的总体性能仿真模型快74.7%，计算耗时远小于总体性能仿真模型，能够为燃气轮机总体性能实时预测提供技术支撑。
Abstract:Taking a certain gas turbine as the research object, a component-based overall performance computational model was developed. Sensitivity analysis was conducted to select appropriate correction coefficients for component characteristics. Test data was then coupled to design a coordinated operating equation set, and multi-point optimization was applied to refine the component characteristics. This methodology ultimately yielded a high-precision overall performance simulation model for the gas turbine. Based on the data obtained from the modified overall performance simulation model, a gas turbine performance prediction model was established using a multilayer perceptron (MLP). The results showed that, compared with the test data, the prediction model based on the modified gas turbine overall performance simulation data had an error of less than 1% in the performance evaluation of each steady-state operating point, which met the engineering accuracy requirements. It was also 74.7% faster than the corrected model, significantly reducing the computing time, and providing technical support for real-time prediction of the overall performance of gas turbine.
-
图 1 分轴式燃气轮机示意图
0 大气环境； 1 进气道进口截面； 2 压气机进口截面；3 压气机出口截面； 31 燃烧室进口截面；4 燃烧室出口截面和高压涡轮进口截面；44 高压涡轮出口截面； 45 动力涡轮进口截面；5 动力涡轮出口截面； 8 排气道出口截面。
Figure 1. Schematic diagram of split shaft gas turbine
图 3 单试车数据工况点修正流程图
Figure 3. Flowchart of correction for single test data operating condition point
图 5 修正系数预测模型网络结构图
Figure 5. Correction coefficient predictive model network structure diagram
图 8 修正前后燃气轮机部件特性图比较
Figure 8. Comparison of component characteristic maps of compressor before and after correction
图 9 修正前后动力涡轮进口压力分析
Figure 9. Analysis of the inlet pressure of the power turbine before and after correction
图 10 基于性能模型修正前后仿真数据的预测
Figure 10. Prediction based on simulation data before and after performance model correction
表 1 特性修正系数编号
Table 1. Correction coefficients of characteristic map
编号 物理意义 符号 1 高压压气机流量修正系数 α1 2 高压压气机压比修正系数 α2 3 高压压气机效率修正系数 α3 4 高压涡轮流量修正系数 α4 5 高压涡轮压比修正系数 α5 6 高压涡轮效率修正系数 α6 7 动力涡轮流量修正系数 α7 8 动力涡轮压比修正系数 α8 9 动力涡轮效率修正系数 α9 
下载: 导出CSV
表 2 MLP的超参数设置
Table 2. Hyperparameter settings for MLP
参数 数值/说明 参数 数值/说明 输入层 5 隐含层层数 4 第1隐含层节点数 512 第2隐含层节点数 256 第3隐含层节点数 128 第4隐含层节点数 64 输出层 5 激活函数 ReLu 优化器 Adam 学习率 10−4 Batch size 32 Epochs 800
下载: 导出CSV
表 3 稳态工况点提取
Table 3. Extraction of steady-state operating points
工况点编号 ncor,hc/% ncor,pt/% 1 95.57 171.02 2 99.63 188.36 3 99.18 186.98 4 99.10 186.95 5 99.26 187.05
下载: 导出CSV
-
[1] 刘大响, 金捷, 刘邓欢. 数值仿真技术在航空动力研制中的地位和作用[J]. 航空动力学报, 2022, 37(10): 2017-2024. Liu Daxiang, Jin Jie, Liu Denghuan. Position and function of numerical simulation technology in aero-engine development[J]. Journal of Aerospace Power, 2022, 37(10): 2017-2024. (in Chinese doi: 10.13224/j.cnki.jasp.20220103Liu Daxiang, Jin Jie, Liu Denghuan. Position and function of numerical simulation technology in aero-engine development[J]. Journal of Aerospace Power, 2022, 37(10): 2017-2024. (in Chinese) doi: 10.13224/j.cnki.jasp.20220103 [2] 王富宁, 杨晨, 张小玉, 等. 耦合17级压气机通流模型的燃气涡轮发动机混合维度仿真研究[J]. 工程热物理学报, 2024, 45(9): 2612-2621. Wang Funing, Yang Chen, Zhang Xiaoyu, et al. Hybrid-dimensional simulation of gas turbine coupled with 17-stage compressor through-flow model[J]. Journal of Engineering Thermophysics, 2024, 45(9): 2612-2621. (in ChineseWang Funing, Yang Chen, Zhang Xiaoyu, et al. Hybrid-dimensional simulation of gas turbine coupled with 17-stage compressor through-flow model[J]. Journal of Engineering Thermophysics, 2024, 45(9): 2612-2621. (in Chinese) [3] 王涛. 变几何三轴式燃气轮机总体性能及控制规律研究[D]. 北京: 中国科学院大学, 2019. Wang Tao. Characteristic and control law investigations on variable geometry triaxial gas turbine[D]. Beijing: Chinese Academy of Sciences, 2019. (in ChineseWang Tao. Characteristic and control law investigations on variable geometry triaxial gas turbine[D]. Beijing: Chinese Academy of Sciences, 2019. (in Chinese) [4] 董威, 尹家录, 郑培英, 等. 航空发动机及燃气轮机整机性能仿真综述[J]. 航空发动机, 2023, 49(5): 8-21. Dong Wei, Yin Jialu, Zheng Peiying, et al. Review: engine-level performance simulation of aeroengine and gas turbines[J]. Aeroengine, 2023, 49(5): 8-21. (in ChineseDong Wei, Yin Jialu, Zheng Peiying, et al. Review: engine-level performance simulation of aeroengine and gas turbines[J]. Aeroengine, 2023, 49(5): 8-21. (in Chinese) [5] Kayadelen H K, Ust Y. Thermodynamic, environmental and economic performance optimization of simple, regenerative, STIG and RSTIG gas turbine cycles[J]. Energy, 2017, 121(C): 751-771. [6] 焦洋. 民用大涵道比涡扇发动机建模与控制研究[D]. 南京: 南京航空航天大学, 2016. Jiao Yang. Research on modeling and control of civil turbofan engine with large bypass ratio[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2016. (in ChineseJiao Yang. Research on modeling and control of civil turbofan engine with large bypass ratio[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2016. (in Chinese) [7] 林新智. 双轴燃气轮机性能仿真及故障特征研究[D]. 北京: 北京化工大学, 2020. Lin Xinzhi. Research on performance simulation and fault characteristics of dual shaft gas turbine[D]. Beijing: Beijing University of Chemical Technology, 2020. (in ChineseLin Xinzhi. Research on performance simulation and fault characteristics of dual shaft gas turbine[D]. Beijing: Beijing University of Chemical Technology, 2020. (in Chinese) [8] Zhang Peng, Feng Kun, Liu Baoxia, et al. Operational data-based adaptive improvement method of gas turbine component characteristics for performance simulation[J]. Journal of Mechanical Science and Technology, 2023, 37(12): 6691-6709. doi: 10.1007/s12206-023-1040-2 [9] Deng Weimin, Xu Yibing, Ni Ming, et al. Multi-fidelity simulation of gas turbine overall performance by directly coupling high-fidelity models of multiple rotating components[J]. Journal of Thermal Science, 2024, 33(4): 1357-1378. doi: 10.1007/s11630-024-1975-3 [10] Kurzke J. How to get component maps for aircraft gas turbine performance calculations[R]. ASME Paper 96-GT-164, 1996. [11] Tamiru A L, Hashim F M, Rangkuti C. Generating gas turbine component maps relying on partially known overall system characteristics[J]. Journal of Applied Sciences, 2011, 11(11): 1885-1894. doi: 10.3923/jas.2011.1885.1894 [12] Spina P R. Gas turbine performance prediction by using generalized performance curves of compressor and turbine stages[R]. ASME Paper GT2002-30275, 2002. [13] Kong C, Ki J. Components map generation of gas turbine engine using genetic algorithms and engine performance deck data[J]. Journal of Engineering for Gas Turbines and Power, 2007, 129(2): 312-317. doi: 10.1115/1.2436561 [14] Tsoutsanis E, Li Y G, Pilidis P, et al. Part-load performance of gas turbines: Part Ⅰ a novel compressor map generation approach suitable for adaptive simulation[R]. ASME Paper GTIndia2012-9580, 2013. [15] 路绪坤. 燃气轮机压气机特性曲线拟合研究[D]. 北京: 中国科学院大学, 2021. Lu Xukun. Study on fitting characteristic curve of gas turbine compressor[D]. Beijing: University of Chinese Academy of Sciences, 2021. (in ChineseLu Xukun. Study on fitting characteristic curve of gas turbine compressor[D]. Beijing: University of Chinese Academy of Sciences, 2021. (in Chinese) [16] 林京, 张博瑶, 张大义, 等. 航空燃气涡轮发动机故障诊断研究现状与展望[J]. 航空学报, 2022, 43(8): 626565. Lin Jing, Zhang Boyao, Zhang Dayi, et al. Research status and prospect of fault diagnosis for gas turbine aeroengine[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(8): 626565. (in ChineseLin Jing, Zhang Boyao, Zhang Dayi, et al. Research status and prospect of fault diagnosis for gas turbine aeroengine[J]. Acta Aeronautica et Astronautica Sinica, 2022, 43(8): 626565. (in Chinese) [17] 俞明帅. 航空发动机模型组态与修正技术研究[D]. 南京: 南京航空航天大学, 2012. Yu Mingshuai. Research on configuration and modification technology of aero-engine model[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2012. (in ChineseYu Mingshuai. Research on configuration and modification technology of aero-engine model[D]. Nanjing: Nanjing University of Aeronautics and Astronautics, 2012. (in Chinese) [18] 王佳雯, 黄向华, 嵇润民. 基于特性数据的燃气涡轮发动机修正方法[J]. 航空发动机, 2024, 50(3): 114-121. Wang Jiawen, Huang Xianghua, Ji Runmin. Correction method of component characteristics for gas turbine based on characteristic data[J]. Aeroengine, 2024, 50(3): 114-121. (in Chinese doi: 10.13477/j.cnki.aeroengine.2024.03.017Wang Jiawen, Huang Xianghua, Ji Runmin. Correction method of component characteristics for gas turbine based on characteristic data[J]. Aeroengine, 2024, 50(3): 114-121. (in Chinese) doi: 10.13477/j.cnki.aeroengine.2024.03.017 [19] 钟文城, 汪勇, 宋劼, 等. 一种面向航空发动机数学模型的新型修正方法[J]. 航空动力学报, 2023, 38(11): 2776-2784. Zhong Wencheng, Wang Yong, Song Jie, et al. A new correction method for aero-engine mathematical model[J]. Journal of Aerospace Power, 2023, 38(11): 2776-2784. (in Chinese doi: 10.13224/j.cnki.jasp.20220114Zhong Wencheng, Wang Yong, Song Jie, et al. A new correction method for aero-engine mathematical model[J]. Journal of Aerospace Power, 2023, 38(11): 2776-2784. (in Chinese) doi: 10.13224/j.cnki.jasp.20220114 [20] 樊巍, 施洋, 曹铭栋, 等. 一种新的航空发动机总体性能稳态模型修正方法[J]. 燃气涡轮试验与研究, 2020, 33(6): 28-31. Fan Wei, Shi Yang, Cao Mingdong, et al. A new aero-engine steady-state performance model modification method[J]. Gas Turbine Experiment and Research, 2020, 33(6): 28-31. (in ChineseFan Wei, Shi Yang, Cao Mingdong, et al. A new aero-engine steady-state performance model modification method[J]. Gas Turbine Experiment and Research, 2020, 33(6): 28-31. (in Chinese) [21] 郑斐华. 基于系统辨识的航空发动机建模研究[D]. 北京: 中国科学院大学, 2018. Zheng Feihua. Research on aero-engine modeling based on system identification[D]. Beijing: University of Chinese Academy of Sciences, 2018. (in ChineseZheng Feihua. Research on aero-engine modeling based on system identification[D]. Beijing: University of Chinese Academy of Sciences, 2018. (in Chinese) [22] 陈煜, 黄金泉, 罗启君, 等. 基于遗传算法的涡喷发动机身份证模型建立[J]. 航空发动机, 2015, 41(4): 24-28. Chen Yu, Huang Jinquan, Luo Qijun, et al. Establishment of turbojet engine identification model based on genetic algorithms[J]. Aeroengine, 2015, 41(4): 24-28. (in ChineseChen Yu, Huang Jinquan, Luo Qijun, et al. Establishment of turbojet engine identification model based on genetic algorithms[J]. Aeroengine, 2015, 41(4): 24-28. (in Chinese) [23] 刘策. 基于智能优化机制的先进民用航空发动机性能模型偏差修正方法研究[D]. 天津: 中国民航大学, 2023. Liu Ce. Research on deviation correction method of advanced civil aeroengine performance model based on intelligent optimization mechanism[D]. Tianjin: Civil Aviation University of China, 2023. (in ChineseLiu Ce. Research on deviation correction method of advanced civil aeroengine performance model based on intelligent optimization mechanism[D]. Tianjin: Civil Aviation University of China, 2023. (in Chinese) [24] Liu Zuming, Karimi I A. Gas turbine performance prediction via machine learning[J]. Energy, 2020, 192(C):116627 doi: 10.1115/1.0003040v [25] Goyal V, Xu M Y, Kapat J, et al. Prediction of gas turbine performance using machine learning methods[R]. ASME Papaer GT2020-15232, 2020 [26] Guan Jin, Wang Xusheng, Lv Xiaojing, et al. Multidimensional analysis and performance prediction of heavy-duty gas turbine based on actual operational data[J]. Applied Thermal Engineering, 2023, 234: 121280. doi: 10.1016/j.applthermaleng.2023.121280 [27] Xiao Dasheng, Xiao Hong, Wang Zhanxue. IndRNN-based data-driven modeling integrated with physical knowledge for engine performance monitoring[J]. Journal of Engineering for Gas Turbines and Power, 2025, 147(2): 021012. doi: 10.1115/1.4066292 [28] 邹泽龙, 黄金泉, 周鑫, 等. VCE身份证动态模型部件特性快速自动修正方法[J]. 航空动力学报, 2024, 39(9): 20220680. Zou Zelong, Huang Jinquan, Zhou Xin, et al. Fast automatic correction method for component characteristics of the identification dynamic model of VCE[J]. Journal of Aerospace Power, 2024, 39(9): 20220680. (in Chinese doi: 10.13224/j.cnki.jasp.20220680Zou Zelong, Huang Jinquan, Zhou Xin, et al. Fast automatic correction method for component characteristics of the identification dynamic model of VCE[J]. Journal of Aerospace Power, 2024, 39(9): 20220680. (in Chinese) doi: 10.13224/j.cnki.jasp.20220680 -
点击查看大图
计量
- 文章访问数: 336
- HTML浏览量: 181
- PDF量: 28
- 被引次数: 0