面向综合能源系统运行的气网聚合建模及辨识方法  

Gas Network Aggregate Modeling and Identification Method for Integrated Energy System Operation

作　　者：孙勇 高子航 侯泽胤 李德鑫 王尧 张海锋 陆帅[2] SUN Yong;GAO Zihang;HOU Zeyin;LI Dexin;WANG Yao;ZHANG Haifeng;LU Shuai(State Grid Jilin Electric Power Company,Changchun 130021,Jilin Province,China;School of Electrical Engineering,Southeast University,Nanjing 211189,Jiangsu Province,China)

机构地区：[1]国网吉林省电力有限公司,吉林省长春市130021 [2]东南大学电气工程学院,江苏省南京市211189

出　　处：《发电技术》2025年第2期274-283,共10页Power Generation Technology

基　　金：国家重点研发计划项目(2022YFB2404000)。

摘　　要：【目的】天然气网络是综合能源系统的重要组成部分,可为综合能源系统的运行提供可观的灵活性。建立精确的气网模型是综合能源系统运行控制的基础,然而现有的气网机理模型复杂度较高,部分参数在实际工程中不可辨识。为此,提出一种面向综合能源系统运行的气网聚合建模及辨识方法。【方法】以气网为研究对象,提出了气网聚合模型的概念并推导了其表达式,设计了对应的离线参数估计方法。由于气网聚合模型的推导依赖于特定的工况,当工况发生变化时,模型参数亦会随之改变,离线估计模型会产生不可忽略的误差。针对该问题,提出了基于遗忘因子递推最小二乘法的模型滚动校正方法。【结果】在气体流速变化较小的情况下,离线参数估计方法具有较好准确性;但是在流速快速变化情况下,离线估计无法精准描述气网运行状态,滚动校正可以实现对模型的精确拟合。【结论】所提气网聚合模型离线估计与在线滚动校正方法可为综合能源系统的运行分析与控制提供实用、高效的流体网络建模技术。[Objectives]Natural gas networks are important components of integrated energy systems,which can provide considerable flexibility for the operation of the integrated energy system.Developing an accurate gas network model is the foundation for the operation and control of integrated energy systems.However,existing mechanism models for gas networks show high complexity,with certain parameters being unidentifiable in practical engineering applications.Therefore,this study proposes a gas network aggregate modeling and identification method specifically for integrated energy system operation.[Methods]Taking the gas network as the research object,the concept of the gas network aggregate model is proposed,its formula is derived,and a corresponding offline parameter estimation method is designed.Since the derivation of the gas network aggregate model depends on specific operating conditions,the model parameters vary with condition changes,leading to nonnegligible errors in the offline estimation model.To address this issue,a rolling calibration method based on forgetting factor recursive least squares is proposed.[Results]Under small gas flow variations,the offline parameter estimation method demonstrates satisfactory accuracy.However,under rapid flow variations,the offline estimation fails to accurately describe the operation status of the gas network,and rolling calibration can achieve accurate fitting of the model.[Conclusions]The proposed gas network aggregate model integrating offline estimation and online rolling calibration can provide a practical and efficient fluid network modeling solution for the operation analysis and control of integrated energy systems.

关 键 词：天然气 综合能源系统 新能源消纳 气网 聚合模型 参数估计 在线校正 

分 类 号：TK01[动力工程及工程热物理] TM711[电气工程—电力系统及自动化]