中国石油大学学报（自然科学版）2025，Vol.49Issue(3)：158-168,11.DOI:10.3969/j.issn.1673-5005.2025.03.017
基于连接元的二氧化碳驱模拟及窜流通道预测方法
Simulation of CO2 flooding and prediction method for conformance channels based on connection elements
摘要
Abstract
To address the issue of gas channeling affecting the efficiency of the oil displacement in the CO2 flooding,a con-nectivity element-based numerical model for multi-component CO2 flooding was proposed.By discretizing the reservoir into connection elements,the flow characteristics between nodes were characterized by the connection conductivity and the con-nection volume.A fully implicit method is employed to solve the pressure,saturation,and component molar fractions,and a path-tracking algorithm was used to characterize inter-node connectivity.At the same time,a simultaneous stochastic pertur-bation approximation algorithm was used for historical fitting to realize the automatic inversion of model parameters.Results indicate that in the conceptual case,the historical fitting accuracy of the model is 95.17%,the single simulation time is only 23.33 s,and the computational efficiency is significantly higher than that of the traditional grid method with 140.28 s.In the actual reservoir application,the fitting accuracy of cumulative oil production and cumulative gas production reaches 96.1%and 94.2%,respectively,and the average fitting accuracy of single well daily oil production and daily gas production is 91.21%.The model can effectively identify the gas channeling channels in the RA0050 well group,which is consistent with the tracer monitoring results.关键词
CO2驱/全组分模型/连接元法/流动路径追踪/气窜识别Key words
CO2 flooding/multi-component model/connectivity elements method/flow path tracing/gas channeling identifi-cation分类
能源科技引用本文复制引用
徐云峰,刘伟,赵辉,陈奇龙..基于连接元的二氧化碳驱模拟及窜流通道预测方法[J].中国石油大学学报（自然科学版）,2025,49(3):158-168,11.基金项目
国家自然科学基金面上项目(52274030) （52274030）
新疆维吾尔自治区自然科学基金项目(2022D01A330) （2022D01A330）
新疆维吾尔自治区"天池英才"引进计划项目(2022) （2022）
国家自然科学基金优秀青年科学基金项目(51922007) （51922007）
