王能,冯旭波
WANG Neng,FENG Xubo

• 1:广西投资集团有限公司创新研究院

摘要(Abstract)：

锅炉-汽轮机系统是超超临界发电机组中最主要的工作系统之一，具有高度非线性、强耦合性等特点，并受到各种不确定性和干扰的影响。因此，提出一种基于CWHLO（复合加权人类学习网络）的多变量建模方法，以智能分区、局部建模和实时融合的形式建立动态线性模型，可基于负荷需求实时更改局部模型权重，来描述锅炉-汽轮机组的非线性运行过程。仿真结果表明，在强耦合和非线性的情况下CWHLO建模效果仍较显著，可为其他控制方法提供理论依据。
As core subsystems in ultra-supercritical power generation units, boiler-turbine systems exhibit high nonlinearity, strong coupling, and vulnerability to various uncertainties and disturbances. To address these challenges, this paper proposes a multivariate modeling approach utilizing the composite weighted human learning optimization network(CWHLON). The method establishes dynamic linear models through intelligent zoning, localized modeling, and real-time fusion, enabling adaptive adjustment of local model weights according to load demand to accurately represent the nonlinear operational processes of boiler-turbine systems. Simulation results demonstrate that the CWHLON-based modeling maintains excellent performance even under strong coupling and nonlinear conditions, providing a theoretical foundation for other control strategies.

关键词(KeyWords)： 复合加权人类学习网络;多变量建模;动态线性模型;锅炉-汽轮机;超超临界机组
CWHLON;multivariate modeling;dynamic linear model;boiler-turbine system;ultra-supercritical unit

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金（U24A20592）;; 广西广投北海发电有限公司智慧电厂开发项目（GTS2024-257）

作者(Author): 王能,冯旭波
WANG Neng,FENG Xubo

DOI: 10.19585/j.zjdl.202511003

参考文献(References)：

• [1]王建录，张晓东，侯明军.超超临界二次再热机组汽轮机应用现状与展望[J].热力发电，2017,46(8):11-15.WANG Jianlu,ZHANG Xiaodong,HOU Mingjun.Development and expectation of application of ultra-supercritical double-reheat steam turbines[J].Thermal Power Generation,2017,46(8):11-15.
• [2]袁世通，韩璞，孙明.基于大数据的多变量系统建模方法研究[J].系统仿真学报，2014,26(7):1454-1459.YUAN Shitong,HAN Pu,SUN Ming.Modeling research of multivariable system based on big data[J]. Journal of System Simulation,2014,26(7):1454-1459.
• [3]金宏伟，王策，张新胜，等.超超临界机组湿态给水控制策略的设计与应用[J].浙江电力，2023,42(10):107-113.JIN Hongwei,WANG Ce,ZHANG Xinsheng,et al.Design and application of the autonomous object oriented control strategies for wet water supply system of thermal power units[J].Zhejiang Electric Power,2023,42(10):107-113.
• [4]李先锋，胡晨刚，卜莉敏，等.储能电池实时荷电状态联合估计方法[J].浙江电力，2024,43(5):73-82.LI Xianfeng,HU Chengang,BU Limin,et al.A joint realtime SOC estimation method for energy storage batteries[J].Zhejiang Electric Power,2024,43(5):73-82.
• [5]徐心怡，贺兴，艾芊，等.基于随机矩阵理论的配电网运行状态相关性分析方法[J].电网技术，2016,40(3):781-790.XU Xinyi,HE Xing,AI Qian,et al.A correlation analysis method for operation status of distribution network based on random matrix theory[J].Power System Technology,2016,40(3):781-790.
• [6]翟国富，王淑娟，姜守旭，等.电器可靠性失效分析中极大似然估计方法的研究[J].中国电机工程学报，2001,21(3):90-92.ZHAI Guofu,WANG Shujuan,JIANG Shouxu,et al.The research on method of maximum likelihood estimation for apparatus reliability failure analysis[J].Proceedings of the CSEE,2001,21(3):90-92.
• [7]张华军，赵金，王瑞，等.基于非线性共轭梯度的同时扰动随机逼近方法[J].华中科技大学学报（自然科学版），2009,37(1):85-87.ZHANG Huajun,ZHAO Jin,WANG Rui,et al.Nonlinear conjugate gradient method based SPSA[J]. Journal of Huazhong University of Science and Technology(Nature Science Edition),2009,37(1):85-87.
• [8]孙慧君，闫志彬，余蕾，等.微电网多目标优化运行综述[J].山东电力技术，2020,47(2):37-42.SUN Huijun,YAN Zhibin,YU Lei,et al.Review of multiobjective optimization operation of micro-grid[J]. Shandong Electric Power,2020,47(2):37-42.
• [9] WANG L,NI H Q,YANG R X,et al.A simple human learning optimization algorithm[C]//Computational Intelligence,Networked Systems and Their Applications. Berlin,Heidelberg:Springer Berlin Heidelberg,2014:56-65.
• [10] WANG L,AN L,PI J X,et al.A diverse human learning optimization algorithm[J].Journal of Global Optimization,2017,67(1):283-323.
• [11] WANG L,YANG R X,NI H Q,et al.A human learning optimization algorithm and its application to multidimensional knapsack problems[J].Applied Soft Computing,2015,34:736-743.
• [12] CAO J,YAN Z,XU X Y,et al.Optimal power flow calculation in AC/DC hybrid power system based on adaptive simplified human learning optimization algorithm[J].Journal of Modern Power Systems and Clean Energy,2016,4(4):690-701.
• [13] CHENG C L,PENG C,ZHANG T F. Fuzzy K-means cluster based generalized predictive control of ultra supercritical power plant[J]. IEEE Transactions on Industrial Informatics,2021,17(7):4575-4583.
• [14]程传良，彭晨，曾德良，等.基于复合加权人类学习网络的超超临界机组建模与仿真[J].系统仿真学报，2022,34(7):1430-1438.CHENG Chuanliang,PENG Chen,ZENG Deliang,et al.Modeling and simulation of ultra supercritical unit using A composite weighted human learning network[J]. Journal of System Simulation,2022,34(7):1430-1438.
• [15] LIU X J,LIU J Z,GUAN P.Neuro-fuzzy generalized predictive control of boiler steam temperature[J].Journal of Control Theory and Applications,2007,5(1):83-88.
• [16] LIU X J,KONG X B,HOU G L,et al.Modeling of a1 000 MW power plant ultra super-critical boiler system using fuzzy-neural network methods[J]. Energy Conversion and Management,2013,65:518-527.
• [17]李贵海，孙有坤，高坚，等.超超临界机组城市中水综合实践应用及零排放试验研究[J].山东电力技术，2024,51(10):84-90.LI Guihai,SUN Youkun,GAO Jian,et al.Comprehensive practice application and zero discharge test research on urban reclaimed water of ultra supercritical units[J]. Shandong Electric Power,2024,51(10):84-90.
• [18]张迎喜.660 MW超超临界汽轮机启动过程胀差偏大原因分析[J].山东电力技术，2021,48(6):68-71.ZHANG Yingxi. Cause aanalysis of large differential expansion during start-up of 660 MW ultra supercritical steam turbine[J].Shandong Electric Power,2021,48(6):68-71.
• [19] CSERHáTI T,KóSA A,BALOGH S. Comparison of partial least-square method and canonical correlation analysis in a quantitative structure-retention relationship study[J]. Journal of Biochemical and Biophysical Methods,1998,36(2/3):131-141.
• [20] YUAN N M,FU Z T,ZHANG H,et al. Detrended partial-cross-correlation analysis:a new method for analyzing correlations in complex system[J].Scientific Reports,2015,5:8143.
• [21] BURGES C,SHAKED T,RENSHAW E,et al.Learning to rank using gradient descent[C]//Proceedings of the22nd International Conference on Machine LearningICML'05. August 7-11,2005. Bonn,Germany. ACM,2005:89-96.
• [22]王福胜，甄娜，李晓桐.R-线性收敛的重要样本抽样随机梯度下降算法[J].工程数学学报，2023,40(5):833-842.WANG Fusheng,ZHEN Na,LI Xiaotong. R-linear convergence of importance sampling stochastic gradient decent algorithm[J]. Chinese Journal of Engineering Mathematics,2023,40(5):833-842.
• [23]吕伟锋.RC积分微分电路实验的误差分析方法[J].实验室研究与探索，2009,28(11):34-35.LüWeifeng.Error analysis in RC integral and differential circuit experiment[J].Research and Exploration in Laboratory,2009,28(11):34-35.
• [24]林涵，王珊珊.基于绝对指数积分的永磁同步电机控制[J].安徽理工大学学报（自然科学版），2023,43(5):17-23.LIN Han,WANG Shanshan. Control of permanent magnet synchronous motor based on absolute exponential integration[J]. Journal of Anhui University of Science and Technology(Natural Science),2023,43(5):17-23.
• [25] CHENG C L,PENG C,ZENG D L,et al.An improved neuro-fuzzy generalized predictive control of ultrasupercritical power plant[J]. Cognitive Computation,2021,13(6):1556-1563.