Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process

SUN Shoutai; XUE Yali; WANG Mingchun; SUN Li

doi:10.19693/j.issn.1673-3185.02914

Volume 18 Issue 3

Jun. 2023

Turn off MathJax

Article Contents

Abstract

References

Chinese Journal of Ship Research > 2023 > 18(3): 222-230. > DOI: 10.19693/j.issn.1673-3185.02914 CSTR: 32390.14.j.issn.1673-3185.02914

SUN S T, XUE Y L, WANG M C, et al. Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process[J]. Chinese Journal of Ship Research, 2023, 18(3): 222–230. DOI: 10.19693/j.issn.1673-3185.02914

Citation:

PDF (2133 KB)

Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process

1.
Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, School of Energy and Environment, Southeast University, Nanjing 210018, China
2.
State Key Laboratory of Electric Power Systems, Department of Energy and Power Engineering, Tsinghua University, Beijing 100084, China

More Information

Received Date: May 17, 2022
Revised Date: July 06, 2022
Available Online: July 18, 2022

This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Graphical Abstract

Abstract

Abstract

Objective In order to overcome the influence of the nonlinear time-varying characteristics of gas turbines on dynamic control and performance monitoring, this paper combines the time series memory and nonlinear relation expression ability of a long short-term memory neural network (LSTM) with the interval probability estimation ability of Gaussian process regression (GPR) to propose an online parameter identification algorithm for the key dynamic parameters of gas turbines based on an LSTM and GPR-based hybrid deep learning model (LSTM-GPR).
Methods First, the dynamic mechanism model of a gas turbine is established, and a large amount of training data is generated by taking fuel calorific value, compressor efficiency and load power moment as the parameters to be identified. Next, the parameter identification network model of LSTM-GPR is constructed, and the training data is input for network training and weight coefficient learning. Finally, the trained LSTM-GPR hybrid deep learning model is used to identify the dynamic operating parameters of the gas turbine model online, and the identification results are analyzed to verify the effectiveness of the proposed algorithm.
Results The simulation results show that the online identification results of the proposed LSTM-GPR hybrid model algorithm are accurate, with a recognition error of less than 1% and good real-time performance. Compared with the LSTM single model, the proposed algorithm can obtain a better mean estimation effect and provide a reliable confidence interval range.
Conclusions The LSTM-GPR hybrid algorithm can be effectively applied to the online parameter identification of a gas turbine model, laying a foundation for its further application to the dynamic operation parameter identification of practical units.
- gas turbine model,
- online parameter identification,
- long short-term memory,
- Gaussian process regression

FullText(HTML)

References (35)

References

[1]	KURZKE J, HALLIWELL I. Propulsion and power: an exploration of gas turbine performance modeling[M]. Cham, Switzerland: Springer International Publishing, 2018.
[2]	PIEREZAN J, MAIDL G, YAMAO E M, et al. Cultural coyote optimization algorithm applied to a heavy duty gas turbine operation[J]. Energy Conver-sion and Management, 2019, 199: 111932. doi: 10.1016/j.enconman.2019.111932
[3]	RIGO-MARIANI R, ZHANG C, ROMAGNOLI A, et al. A combined cycle gas turbine model for heat and power dispatch subject to grid constraints[J]. IEEE Transactions on Sustainable Energy, 2019, 11(1): 448–456.
[4]	TSOUTSANIS E, MESKIN N. Dynamic performance simulation and control of gas turbines used for hybrid gas/wind energy applications[J]. Applied Thermal Engineering, 2019, 147: 122–142. doi: 10.1016/j.applthermaleng.2018.09.031
[5]	KULIKOV G G, THOMPSON H A. Dynamic modelling of gas turbines: identification, simulation, condition monitoring and optimal control[M]. [S.1.] : Springer Science & Business Media, 2004.
[6]	WANG C, LI D, LI Z, et al. Optimization of controllers for gas turbine based on probabilistic robustness[J]. Journal of engineering for gas turbines and power, 2009, 131(5): 054502. doi: 10.1115/1.2981174
[7]	MOHAMED O, KHALIL A. Progress in modeling and con-trol of gas turbine power generation systems: a survey[J]. Energies, 2020, 13(9): 2358. doi: 10.3390/en13092358
[8]	YOUSEFI I, YARI M, SHOOREHDELI M A. Modeling, identification and control of a heavy duty industrial gas turbine[C]//2013 IEEE International Conference on Mechatronics and Automation. Takamatsu, Japan: IEEE, 2013: 611-615.
[9]	ASGARI H, CHEN X Q, SAINUDIIN R. Modelling and simulation of gas turbines[J]. International Journal of Modelling, Identification and Control, 2013, 20(3): 253–270. doi: 10.1504/IJMIC.2013.057137
[10]	张显库, 祝慧颖. 基于正弦函数处理新息的船舶模型参数辨识新算法[J]. 中国舰船研究, 2021, 94(5): 158–162. doi: 10.19693/j.issn.1673-3185.02122 ZHANG X K, ZHU H Y. A new algorithm for para-meter identification of ship model based on sinusoidal func-tion processing innovation[J]. Chinese Journal of Ship Research, 2021, 94(5): 158–162 (in Chinese). doi: 10.19693/j.issn.1673-3185.02122
[11]	郑涵, 俞孟蕻, 袁伟. 基于反馈粒子滤波的船舶模型参数辨识[J]. 中国舰船研究, 2019, 80(3): 158–162, 178. doi: 10.19693/j.issn.1673-3185.01323 ZHENG H, YU M H, YUAN W. Parameter identification of ship model based on feedback particle filter[J]. Chinese Journal of Ship Research, 2019, 80(3): 158–162, 178 (in Chinese). doi: 10.19693/j.issn.1673-3185.01323
[12]	SHALAN H E, HASSAN M A M, BAHGAT A B G. Parameter estimation and dynamic simulation of gas turbine model in combined cycle power plants based on actual operational data[J]. Journal of American Science, 2011, 7(5): 303–310.
[13]	贾宏杰, 戚冯宇, 徐宪东, 等. 微型燃气轮机型综合能源系统的建模与辨识[J]. 天津大学学报(自然科学与工程技术版), 2017, 50(2): 215–223. JIA H G, QI F G, XU X D, et al. Modeling and iden-tification of integrated energy system of micro gas turbine[J]. Journal of Tianjin University (Natural Sci-ence and Engineering Technology Edition), 2017, 50(2): 215–223 (in Chinese).
[14]	MAZAHERI A, MANSOURI M, SHOOREDELI M A. Parameter estimation of Hammerstein-Wiener ARMAX systems using unscented Kalman filter[C]//2014 Second RSI/ISM International Conference on Robotics and Mechatronics (ICRoM). Tehran, Iran: IEEE, 2014: 298-303.
[15]	高林, 夏俊荣, 戴义平, 等. 基于遗传算法的联合循环机组建模和参数辨识[J]. 电力系统自动化, 2010, 34(4): 34–37, 83. GAO L, XIA J R, DAI Y P, et al. Modeling and para-meter identification of combined cycle unit based on genetic algorithm[J]. Power System Automation, 2010, 34(4): 34–37, 83 (in Chinese).
[16]	HOU G, GONG L, HUANG C, et al. Fuzzy modeling and fast model predictive control of gas turbine system[J]. Energy, 2020, 200: 117465. doi: 10.1016/j.energy.2020.117465
[17]	LIU Z, KARIMI I A. Gas turbine performance prediction via machine learning[J]. Energy, 2020, 192: 116627. doi: 10.1016/j.energy.2019.116627
[18]	ZHANG S, LIU S, WEI F, et al. Parameter identifi-cation of hypersonic vehicle based on LSTM algori-thm[C]//The 40th Chinese Control Conference (CCC). Shanghai, China: IEEE, 2021: 1230–1235.
[19]	陈龙. 基于长短记忆模式递归神经网络(LSTM-RNN)的异步电机参数辨识研究[D]. 安徽大学, 2018. CHEN L. Research on parameter identification of in-duction motor based on long-short memory mode recurrent neural network (LSTM-RNN). [D]Anhui: Anhui University, 2018 (in Chinese) .
[20]	SHENG H, XIAO J, WANG P. Lithium iron phosphate battery electric vehicle state-of-charge estimation based on evolutionary Gaussian mixture regression[J]. IEEE Transactions on Industrial Electronics, 2016, 64(1): 544–551.
[21]	CHEN H, ZENG Y, HUANG J, et al. Deep neural network Gaussian process regression method for end-to-end driving behavior learning[C]//The 3rd International Conference on Mechatronics, Robotics and Automation (ICMRA). Hangzhou, China: IEEE, 2020: 64–73.
[22]	WANG Y, FENG B, HUA Q S, et al. Short-term solar power forecasting: a combined long short-term memory and Gaussian process regression method[J]. Sustainability, 2021, 13(7): 3665.
[23]	DAILY J W. Statistical thermodynamics: an engineering approach[M]. [S.1.]: Cambridge University Press, 2018.
[24]	YEE S K, MILANOVIC J V, HUGHES F M. Overview and comparative analysis of gas turbine models for system stability studies[J]. IEEE Transactions on power systems, 2008, 23(1): 108–118. doi: 10.1109/TPWRS.2007.907384
[25]	RAZAK A M Y. Industrial gas turbines: performance and operability [M]. [S.1.] Elsevier Publishing House, 2007.
[26]	COHEN H, ROGERS G F C, SARAVANAMUTTOO H I H. Gas turbine theory [J]. England: Pearson Education Ltd. , 2009.
[27]	SUN G, JIANG C, WANG X, et al. Short-term build-ing load forecast based on a data-mining feature selec-tion and LSTM-RNN method[J]. IEEJ Transactions on Electrical and Electronic Engineering, 2020, 15(7): 1002–1010. doi: 10.1002/tee.23144
[28]	FISCHER T, KRAUSS C. Deep learning with long short-term memory networks for financial market pre-dictions[J]. European Journal of Operational Research, 2018, 270(2): 654–669. doi: 10.1016/j.ejor.2017.11.054
[29]	GONZALEZ J, YU W. Non-linear system modeling using LSTM neural networks[J]. IFAC-PapersOnLine, 2018, 51(13): 485–489. doi: 10.1016/j.ifacol.2018.07.326
[30]	SCHULZ E, SPEEKENBRINK M, KRAUSE A. A tutorial on Gaussian process regression: modelling, exploring, and exploiting functions[J]. Journal of Ma-thematical Psychology, 2018, 85: 1–16. doi: 10.1016/j.jmp.2018.03.001
[31]	LIU K, HU X, WEI Z, et al. Modified Gaussian process regression models for cyclic capacity prediction of lithium-ion batteries[J]. IEEE Transactions on Trans-portation Electrification, 2019, 5(4): 1225–1236. doi: 10.1109/TTE.2019.2944802
[32]	WANG Y. A new concept using LSTM neural networks for dynamic system identification[C]//2017 American control conference (ACC). Seattle, WA: IEEE, 2017: 5324–5329.
[33]	CHEN F, RUAN C, YU T, et al. Effects of fuel variation and inlet air temperature on combustion stability in a gas turbine model combustor[J]. Aerospace Science and Technology, 2019, 92: 126–138. doi: 10.1016/j.ast.2019.05.052
[34]	MARIN G E, MENDELEEV D I, AKHMETSHIN A R. Analysis of changes in the thermophysical parameters of the gas turbine unit working fluid depending on the fuel gas composition[C]//2019 International Multi-Conference on Industrial Engineering and Modern Technologies (FarEastCon). Vladivostok: IEEE, 2019: 1–4.
[35]	LIU Z, KARIMI I A. New operating strategy for a combined cycle gas turbine power plant[J]. Energy Conversion and Management, 2018, 171: 1675–1684. doi: 10.1016/j.enconman.2018.06.110

[1]	LI Shijie, HE Jiawei, LIU Jialun, LIU Taixu, XU Chengqi. Ship heading and trajectory control method based on L1-GPR[J]. Chinese Journal of Ship Research, 2025, 20(1): 278-288. DOI: 10.19693/j.issn.1673-3185.03888
[2]	FAN Yuwei, GUO Tengbo, LI Zhe, HONG Liangyou, LIU Chao, JIANG Dongxiang. Structural dynamic load prediction method based on long short-term memory network[J]. Chinese Journal of Ship Research, 2024, 19(6): 228-236. DOI: 10.19693/j.issn.1673-3185.03463
[3]	WEI Ye, ZHANG Zhenguo. Modeling and characteristic analysis of unsteady broadband excitation spectrum of thruster based on OU-Gaussian process[J]. Chinese Journal of Ship Research. DOI: 10.19693/j.issn.1673-3185.04109
[4]	HOU Muyu, GONG Shuhong, XIAO Donghai, ZUO Yanchun, LIU Yu. Bagging-GPR method for ship RCS extrapolation in frequency domain[J]. Chinese Journal of Ship Research, 2023, 18(4): 103-110. DOI: 10.19693/j.issn.1673-3185.03147
[5]	YANG Liu, XU Donghao. Aircraft carrier landing process simulation based on extremely short-term prediction of ship motion[J]. Chinese Journal of Ship Research, 2018, 13(4): 99-103. DOI: 10.19693/j.issn.1673-3185.01193
[6]	HAO Jinyu, ZHANG Xiaobin, YANG Yuanlong. 船用蒸汽蓄热器放汽过程动态特性数值模拟[J]. Chinese Journal of Ship Research, 2015, 10(3): 98-101,107. DOI: 10.3969/j.issn.1673-3185.2015.03.016
[7]	Yan Fuyu, Zhu Xiaojun, Peng Fei. Gaussian Sampling Path Planning of Ship Assembly/Disassembly Based on RRTConCon Algorithm[J]. Chinese Journal of Ship Research, 2011, 6(5): 108-112. DOI: 10.3969/j.issn.16733185.2011.05.022
[8]	Shen Xiaohong, Wu Qirui, Xiao Di, Deng Aimin. 双断级滑行艇的阻力回归公式[J]. Chinese Journal of Ship Research, 2010, 5(2): 59-63. DOI: 10.3969/j.issn.1673-3185.2010.02.012
[9]	Zong Zhi, Bi Junying, Wu Qirui. 国外航空母舰主尺度与航速关系的回归分析[J]. Chinese Journal of Ship Research, 2007, 2(4): 14-19,25. DOI: 10.3969/j.issn.1673-3185.2007.04.004
[10]	Li Dongqin, Wang Lizheng, Wang Chengfang. Method of Support Vector Regression in Modeling Ship Principal Particulars[J]. Chinese Journal of Ship Research, 2007, 2(3): 18-21,39. DOI: 10.3969/j.issn.1673-3185.2007.03.004

Cited By

Cited by

Periodical cited type(3)

1.	张浩晢，杨智博，焦绪国，吕成兴，雷鹏. 基于增强Bi-LSTM的船舶运动模型辨识. 中国舰船研究. 2025(01): 76-84 . 本站查看
2.	孙守泰，汤冰，薛亚丽，孙立. 基于混合深度学习的压气机喘振快速诊断及自抗扰控制方法. 中国舰船研究. 2024(02): 187-196 . 本站查看
3.	魏澈，苏开元，邱银锋，谢小荣，李国香，车久玮. 海上油田群新型电力系统的构建路径与关键技术. 电网技术. 2024(08): 3287-3298 .

Other cited types(0)

Citation

XML

PDF in Chinese

Article views (586) PDF downloads (78) Cited by(3)

Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process

Abstract

References

Related Articles

Cited by

Periodical cited type(3)

Other cited types(0)

Catalog

Related

Hybrid deep learning-based online identification method for key parameters of gas turbine dynamic process

Abstract

References

Related Articles

Cited by

Periodical cited type(3)

Other cited types(0)

Catalog

Related

Export File

Citation

Format

Content