基于实时组分追踪的多气源非等温管网仿真方法  

Simulation method of multi-gas source non-isothermal pipeline network based on real-time component tracking

作　　者：陈潜 管奥成 雷鹏正 苗超 黄伟[1] 程遹滔 左丽丽[5] CHEN Qian;GUAN Aocheng;LEI Pengzheng;MIAO Chao;HUANG Wei;CHENG Yutao;ZUO Lili(School of Petroleum Engineering,Yangtze University,Wuhan,Hubei 430100,China;Petroleum Engineering School,Southwest Petroleum University,Chengdu,Sichuan 610500,China;Natural Gas Research Institute,PetroChina Southwest Oil&Gasfield Company,Chengdu,Sichuan 610213,China;Kunming Oil&Gas Transportation Branch,PipeChina Southwest Pipeline Company,Kunming,Yunnan 650217,China;College of Mechanical and Transportation Engineering,China University of Petroleum(Beijing),Beijing102249,China)

机构地区：[1]长江大学石油工程学院 [2]西南石油大学石油与天然气工程学院 [3]中国石油西南油气田公司天然气研究院 [4]国家管网集团西南管道公司昆明输油气分公司 [5]中国石油大学(北京)机械与储运工程学院

出　　处：《天然气工业》2024年第8期133-144,共12页Natural Gas Industry

基　　金：国家自然科学基金面上项目“复杂供气管网大时滞非线性仿真模型构建与智能调控”(编号:52174064)。

摘　　要：在管网公平开放的运行模式下,以天然气气质为基础的能量计量将逐渐成为天然气贸易的主流方式。为实现虚拟预测多气源管网关键节点天然气的实时组分,在结合非等温管流瞬态仿真模型和气体组分运移模型的基础上,建立了多气源天然气管网动态组分追踪模型。首先基于高精度的BWRS方程、稳态流动方程、焓方程、气体混合参数方程和边界条件得到管网运行初始条件,其次采用隐式中心差分法对管流控制方程和对流扩散方程进行离散,最后结合不同时空条件下的物性方程、混合参数关系式及边界条件,通过时间递推法和Newton—Rapshan算法求解得到天然气管网各节点的压力、流量、温度、组分等参数在整个瞬态时间域上的演化规律。研究结果表明:①建立的稳态模型仿真结果与TGNET模拟结果相比,系统压力和温度的最大绝对偏差分别为0.069 MPa和2.81 K,两者的甲烷和氢气摩尔分数结果基本一致;②建立的动态模型仿真结果与TGNET模拟结果相比,关键节点处压力、温度、甲烷和氢气摩尔分数的最大绝对偏差分别为0.078 MPa,2.22 K,1.54%和0.55%,模型预测的准确性高;③提出的多气源混合模型具有较强的通用性,可以适用于复杂拓扑结构的天然气管道系统的运行参数动态仿真和气体组分动态追踪。结论认为,提出的实时组分追踪方法可以为多气源非等温管网气体组分实时预测提供依据,进而为天然气管网能量计量以及管道仿真软件的国产化开发奠定理论基础。Under the fair and open operation mode of pipeline network,energy metering based on natural gas quality will gradually become the mainstream mode of natural gas trade.To realize the virtual prediction of real-time natural gas components at key nodes of the multi-source pipeline network,this paper establishes a dynamic component tracking model of multi-source gas pipeline network based on the transient simulation model of non-isothermal pipe flow and the gas component migration model.Firstly,the initial conditions of the pipeline network are obtained based on high-precision BWRS equation,steady state flow equation,enthalpy equation,gas mixing parameter equation,and boundary condition.Then the pipe flow control equation and the convection-diffusion equation are discretized by the implicit central difference method.Finally,the evolution laws of pressure,flow,temperature,component and other parameters at each node of the pipeline network over the entire transient time domain can be obtained by the time recursion method and the Newton-Rapshan method based on the physical property equations,mixing parameter relations and boundary conditions under different spatiotemporal conditions.The following results are obtained.First,the simulation results of the established steady-state model are basically accordant with the TGNET simulation results in methane and hydrogen mole fractions,and its maximum absolute deviation of system pressure and temperature are 0.069 MPa and 2.81 K respectively.Second,compared with the TGNET simulation results,the established dynamic model has the maximum absolute deviations of 0.078 MPa,2.22 K,1.54%and 0.55%in pressure,temperature,methane molar fraction and hydrogen molar fraction respectively at key nodes,indicating a high accuracy of the model prediction.Third,the proposed multi-source mixture model has a stronger universality,and is applicable to the dynamic simulation of operation parameters and the dynamic tracking of gas components in a topologically complex natural gas pipeline system.In c

关 键 词：天然气管网 能量计量 组分追踪 组分运移模型 中心差分法 瞬态仿真 时空演化规律 水力热力过程 

分 类 号：TE832[石油与天然气工程—油气储运工程]