电工技术学报  2022, Vol. 37 Issue (21): 5451-5461    DOI: 10.19595/j.cnki.1000-6753.tces.211208
电力系统与综合能源 |
基于神经网络的数据驱动互联电力系统负荷频率控制
陈宗遥1, 卜旭辉1,2, 郭金丽1
1.河南理工大学电气工程与自动化学院 焦作 454003;
2.河南省煤矿装备智能检测与控制重点实验室(河南理工大学) 焦作 454003
Neural Network Based Data-Driven Load Frequency Control for Interconnected Power Systems
Chen Zongyao1, Bu Xuhui1,2, Guo Jinli1
1. School of Electrical Engineering and Automation Henan Polytechnic University Jiaozuo 454003 China;
2. Henan Key Laboratory of Intelligent Detection and Control of Coal Mine Equipment Henan Polytechnic University Jiaozuo 454003 China
全文: PDF (2875 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 针对高度复杂的电力系统存在的建模误差和不确定性等问题,该文基于无模型自适应控制算法提出一种不依赖电力系统模型信息的负荷频率控制策略。首先将电力系统的动力学模型抽象为一般的非线性函数,在其I/O数据之间引入时变的伪偏导数,将非线性电力系统等效为动态线性数据模型;然后构建一个径向基神经网络在线估计系统的伪偏导数,并使用优化理论设计数据驱动的负荷频率控制方案,在理论上严格分析了闭环电力系统的稳定性和径向基神经网络估计方法的收敛性;最后在互联电力系统上验证该负荷频率控制方法在不利用模型信息的前提下,能够取得良好的跟踪性能。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
陈宗遥
卜旭辉
郭金丽
关键词 互联电力系统负荷频率控制无模型自适应控制径向基函数神经网络数据驱动控制    
Abstract:To the problems of modeling errors and uncertainties in highly complex power systems, a load frequency control (LFC) strategy was proposed in this paper without using any model information of power system based on model-free adaptive control (MFAC) algorithm. First, the dynamic model of the power system was abstracted as a general nonlinear function. By introducing a time-varying pseudo partial derivative (PPD) between historical I/O data, the nonlinear power system was equivalent to a dynamic linear data model. Secondly, an RBF neural network was constructed to estimate the PPD of the system online, and the optimization theory was used to design the data-driven LFC scheme. In theory, the stability of the closed-loop power system and the convergence of the RBF neural network estimation method were strictly analyzed. Finally, it is verified on the interconnected power system that the LFC method in this paper can achieve good tracking performance without using model information.
Key wordsInterconnected power system    load frequency control    model-free adaptive control    radial basis function (RBF) neural network    data driven control   
收稿日期: 2021-08-04     
PACS: TM732  
  TP273  
基金资助:自然科学基金(61573130, U1804147)、河南省高校科技创新团队(20IRTSTHN019)、河南理工大学创新型科技团队项目(T2019-2, T2017-1)和河南省创新型科技团队项目(CXTD2016054)资助
通讯作者: 卜旭辉 男,1981年生,教授,博士生导师,研究方向为数据驱动控制,电力系统运行控制,网络化控制。E-mail:buxuhui@gmail.com   
作者简介: 陈宗遥 男,1998年生,硕士研究生,研究方向为电力系统的数据驱动控制。E-mail:1556771859@qq.com
引用本文:   
陈宗遥, 卜旭辉, 郭金丽. 基于神经网络的数据驱动互联电力系统负荷频率控制[J]. 电工技术学报, 2022, 37(21): 5451-5461. Chen Zongyao, Bu Xuhui, Guo Jinli. Neural Network Based Data-Driven Load Frequency Control for Interconnected Power Systems. Transactions of China Electrotechnical Society, 2022, 37(21): 5451-5461.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.211208          https://dgjsxb.ces-transaction.com/CN/Y2022/V37/I21/5451