Experimental Study and Model Optimization of Carbon Dioxide Dissolved Gas Crude Oil

doi:10.12388/j.issn.1673-2677.2026.01.015

Abstract

Abstract: With the technical progress in carbon dioxide （CO₂） flooding for enhanced oil recovery （CO₂-EOR），it is of great significance to study the influence mechanism of CO₂ injection on the viscosity of crude oil. This study systematically investigated the synergistic effects of temperature （20-40 ℃），pressure （0.5-3.5 MPa），and water content （0%-60%） on the viscosity of saturated CO₂ dissolved gas crude oil through laboratory experiments. The results showed that under constant pressure，for every 5℃ increase in temperature，the viscosity decreases by approximately 60%-80%. At a constant temperature，for every 0.5 MPa increase in pressure，the viscosity decreases by approximately 1%-5%. In terms of water content，a 10% increase in water content leads to a rise in viscosity by 15%-50% before the inverted point （water content < 40%）. After the inverted point （water content>40%），for every 10% increase in water content，the viscosity decreases by 50%-80%. Based on the experimental data，the prediction performance of three classic models，namely Standing，Glaso and Vazquez & Beggs （V&B），was evaluated. It was found that the V&B model had the best robustness. The XGBoost algorithm combined with Bayesian optimization was used to dynamically correct the V&B model，and a multi-parameter mapping relationship of temperature，pressure and water content was established. The average relative error of the model was reduced from 20.1% to 6.9%，and the average absolute error was decreased by approximately 66%. This study provides a theoretical basis and intelligent prediction tool for the process design and dynamic optimization of oil and gas gathering and transportation systems under CO₂ flooding conditions.

Key words: CO₂ , dissolved gas crude oil；viscosity model；XGBoost；Bayesian optimization；viscosity reduction mechanism

摘要： 随着CO₂驱油提高采收率（CO₂-EOR）技术的发展，研究CO₂注入对原油黏度的影响机制具有重要意义。通过室内实验系统考察了温度（20~40 ℃）、压力（0.5~3.5 MPa）和含水率（0%~60%）对饱和CO₂溶气原油黏度的协同影响。结果表明：在恒定压力下，温度每升高5 ℃，黏度下降约60%~80%；在恒定温度下，压力每增加0.5 MPa，黏度下降约1%~5%；反相点前（含水率<40%），含水率每增加10%，黏度上升15%~50%；反相点后（含水率>40%），含水率每增加10%，黏度下降50%~80%。基于实验数据，评估了Standing、Glaso和Vazquez & Beggs（V&B）三种经典模型的预测性能，发现V&B模型具有最优的鲁棒性。采用XGBoost算法结合贝叶斯优化对V&B模型进行动态修正，建立了温度-压力-含水率的多参数映射关系，将模型平均相对误差从20.1%降低至6.9%，平均绝对误差降低约66%。研究成果可为CO₂驱油条件下油气集输系统的工艺设计与动态优化提供理论依据和智能预测工具。

关键词: CO₂溶气原油, 黏度模型, XGBoost, 贝叶斯优化, 降黏机理

CLC Number:

TE357.45

HE Yanan, XIONG Xiaoqin, HUANG Delin, WANG Peiyao, MA Qizhao, LIAO Tao.

Experimental Study and Model Optimization of Carbon Dioxide Dissolved Gas Crude Oil [J]. Xinjiang Oil & Gas, 2026, 22(1): 133-142.

贺雅楠, 熊小琴, 黄德林, 王培尧, 马启钊, 廖涛.

二氧化碳溶气原油黏度实验研究及模型优化 [J]. 新疆石油天然气, 2026, 22(1): 133-142.

References

［1］伦小秀，王强，吴博.二氧化碳驱油封存技术全球进展［J］.安全与环境学报，2025，25（9）：3362-3370.LUN Xiaoxiu，WANG Qiang，WU Bo. Global innovations in carbon dioxide capture，utilization，and storage technology［J］. Journal of Safety and Environment，2025，25（9）：3362-3370.

［2］国丽萍，王海洋，石辰子，等.原油及其乳化体系性质受CO₂影响的研究进展［J/OL］.化工进展，1-11［2026-01-09］.https：//doi.org/10.16085/j.issn.1000-6613.2025-0703.GUO Liping，WANG Haiyang，SHI Chenzi，et al. Research progress on the influence of CO₂ on the properties of crude oil and its emulsification system［J/OL］. Chemical Industry and Engineering Progress，1-11［2026-01-09］.https：//doi.org/10.16085/j.issn.1000-6613.2025-0703.

［3］杨红霞.二氧化碳作用下稠油降粘机制研究［D］.中国石油大学（华东），2021.YANG Hongxia. Study on the viscosity reduction mechanism of heavy oil under the action of carbon dioxide［D］. China University of Petroleum （East China），2021.

［4］姜会心，杨红霞，崔梦雨，等. 二氧化碳对稠油中沥青质分子聚集行为的影响［J］. 油田化学，2023，40（1）：87-91.JIANG Huixin，YANG Hongxia，CUI Mengyu，et al. Effects of CO₂ on the aggregation behavior of asphaltene molecules in heavy oil［J］. Oilfield Chemistry，2023，40（1）：87-91.

［5］王爽. 二氧化碳驱采出液多相体系相间传质机理及界面特性研究［D］. 大庆：东北石油大学，2025.WANG Shuang. Study on interphase mass transfer mechanism and interface characteristics of multiphase system of carbon dioxide flooding produced liquid［D］. Daqing：Northeast Petroleum University，2025.

［6］程显闻. 溶气高含水含蜡原油的低温集输技术研究［D］. 北京：中国石油大学（北京），2022.CHENG Xianwen. Study on the low temperature gathering and transportation of gas-saturated waxy crude oil during high water-cut period［D］. Beijing：China University of Petroleum （Beijing），2022.

［7］钟海全，郑传根，任呈祥，等. 塔河油田稠油黏度预测模型研究与修正［J］. 中国科技论文，2023，18（1）：45-49、56.ZHONG Haiquan，ZHENG Chuangen，REN Chengxiang，et al. Research and correction of heavy oil viscosity prediction model in Tahe Oilfield［J］. China Sciencepaper，2023，18（1）：45-49，56.

［8］王海燕，宋时凤，张永年，等.含蜡原油流变性描述及降黏机制研究进展［J］.当代化工，2024，53（10）：2460-2467.WANG Haiyan，SONG Shifeng，ZHANG Yongnian，et al. Research progress in rheological description and viscosity reduction mechanism of waxy crude oil［J］. Contemporary Chemical Industry，2024，53（10）：2460-2467.

［9］周奥.溶气油水两相流流动特性及水力计算模型研究［D］.青岛：中国石油大学（华东），2022.ZHOU Ao. Study on the flow characteristics of dissolved gas oil-water two-phase flow and hydraulic calculation model［D］. Qingdao：China University of Petroleum （East China），2022.

［10］王春兰，王毅杰，刘文臣，等. 溶气原油CO₂溶解度预测模型［J］. 石油与天然气化工，2024，53（4）：36-43.WANG Chunlan，WANG Yijie，LIU Wenchen，et al. Prediction model of carbon dioxide solubility in crude oil［J］. Chemical Engineering of Oil and Gas，2024，53（4）：36-43.

［11］戚亚明，罗新占.玛湖油田溶气原油流动特性研究［J］.山东化工，2022，51（7）：189-194.QI Yaming，LUO Xinzhan. Study on the flow characteristics of crude oil with dissolved gas in the Mahu Oil Field［J］. Shandong Chemical Industry，2022，51（7）：189-194.

［12］罗海军.关联原油物性的油水混合体系黏度与转相预测模型研究及应用［D］.大庆：东北石油大学，2024.LUO Haijun. Research and application of viscosity and phase inversion prediction model for crude oil-water mixed system with correlation of crude oil physical properties［D］. Daqing：Northeast Petroleum University，2024.

［13］孔维敏. 高含水溶气原油不加热集输流动特性研究［D］. 北京：中国石油大学（北京），2021.KONG Weimin. Research on flow characteristics of high water-cut dissolved gas crude oil gathering and transportation without heating［D］. Beijing：China University of Petroleum （Beijing），2021.

［14］田慧洋，刘瑞恒，李冬爽，等. DeepSeek大模型在石油工程中的应用前景与展望［J］. 新疆石油天然气，2025，21（2）：55-63.TIAN Huiyang，LIU Ruiheng，LI Dongshuang，et al. Applications and prospect of Deepseek large language model in petroleum engineering［J］. Xinjiang Oil & Gas，2025，21（2）：55-63.

［15］李欣泽，王德中，张海帆，等. 新疆油田超临界二氧化碳管道投产置换过程温压及相变特性［J］. 新疆石油天然气，2024，20（4）：77-86.LI Xinze，WANG Dezhong，ZHANG Haifan，et al. Temperature，pressure and phase transition characteristics of supercritical CO₂ pipeline during commissioning replacement process in Xinjiang Oilfield［J］. Xinjiang Oil & Gas，2024，20（4）：77-86.

［16］王思哲. 基于机器学习的混合原油凝点、黏度预测方法研究［D］. 北京：中国石油大学（北京），2022.WANG Sizhe. Prediction of gel point and viscosity of mixed crude oil based on machine learning［D］. Beijing：China University of Petroleum （Beijing），2022.

［17］赵艳伟，胡正祥，乔登攀，等. 一种基于贝叶斯优化和XGBoost的膏体流变参数预测模型［J］. 有色金属（矿山部分），2024，76（5）：118-128.ZHAO Yanwei，HU Zhengxiang，QIAO Dengpan，et al. A prediction model for paste rheological parameters based on Bayesian optimization and XGBoost［J］. Nonferrous Metals （Mining Section），2024，76（5）：118-128.

［18］王金咏. 特征优化结合深度集成学习的造纸废水出水指标预测研究［D］. 南京：南京林业大学，2024.WANG Jinyong. Feature optimization combined with deep ensemble learning for modeling effluent indicators of papermaking wastewater［D］. Nanjing：Nanjing Forestry University，2024.

［19］冼天朗，李小虎，张赟，等. 基于XGBoost和贝叶斯优化的静力触探地层分类预测方法［J］. 岩土工程技术，2025，39（5）：656-666.XIAN Tianlang，LI Xiaohu，ZHANG Yun，et al. Stratigraphic classification prediction method for static cone penetrate test based on XGBoost and Bayesian optimization［J］. Geotechnical Engineering Technique，2025，39（5）：656-666.

［20］刘涛. 基于XGBoost算法的碳酸盐岩储层横波速度预测研究［D］. 成都：成都理工大学，2023.LIU Tao. Prediction of shear wave velocity in carbonate reservoir based on XGBoost algorithm［D］. Chengdu：Chengdu University of Technology，2023.

［21］蒲万芬，靳星，唐晓东，等. 基于BP神经网络的低渗透底水油藏油井见水模式预测模型［J］. 新疆石油天然气，2024，21（2）：37-47.PU Wanfen，JIN Xing，TANG Xiaodong，et al. Prediction model of water breakthrough patterns of low-permeability bottom water reservoirs based on BP neural network［J］. Xinjiang Oil & Gas，2024，21（2）：37-47.

［22］彭家骏. 基于GP的贝叶斯优化算法超参数优化研究［J］. 中国新技术新产品，2025，（11）：38-40.PENG Jiajun. Research on hyperparameter optimization of Bayesian optimization algorithm based on GP［J］. New Technology and New Products of China，2025，（11）：38-40.

［23］彭白雪. 基于XGBoost和SHAP的环境试验箱制冷系统故障诊断方法［D］. 淮南：安徽理工大学，2025.PENG Baixue. Fault diagnosis method for refrigeration system of environmental test chambers based on XGBoost and SHAP［D］. Huainan：Anhui University of Science and Technology，2025.

［24］熊玥雯，向海昀，杨宏楠.随机森林算法在石油领域的应用［C］//西安石油大学，陕西省石油学会.2024油气田勘探与开发国际会议论文集Ⅰ.西南石油大学计算机与软件学院；西南石油大学石油与天然气工程学院；2024：616-626.XIONG Yuewen，XIANG Haiyun，YANG Hongnan. Application of random forest algorithm in petroleum［C］// Xi'an Shiyou University，Shaanxi Petroleum Society. Proceedings of the 2024 International Conference on Oil and Gas Field Exploration and Development Volume I. School of Computer and Software，Southwest Petroleum University；School of Petroleum and Natural Gas Engineering，Southwest Petroleum University；2024：616-626.

［25］闫锋，陈思杭，李其抚，等. 原油分子组成及物性模型研究［J］. 石油工程建设，2025，51（1）：54-58、63.YAN Feng，CHEN Sihang，LI Qifu，et al. Study on the molecular composition and physical property model of crude oil［J］. Petroleum Engineering Construction，2025，51（1）：54-58，63.

[1]	WANG Sui , SONG Shaohua, , CHEN Sen , ZHANG Liyi , CHEN Zhijun , LI Yanpeng, . CO2 Corrosion Resistance Evaluation of Packer Rubber in CCUS Wells [J]. Xinjiang Oil & Gas, 2024, 20(4): 50-59.
[2]	WU Zheng, LING Haochuan, WANG Jijun, ZHAO Zhuo, LI Shan. EOR Scheme Optimization of CO2 Miscible Flooding in Bohai BZ Reservoir [J]. Xinjiang Oil & Gas, 2024, 20(4): 70-76.
[3]	XU Jiangwen, WANG Mingxing, WANG Junchao, SUN Haoran, WANG Liang. Practice and Understanding of Volume Fracturing Technology for Horizontal Wells in Jimsar Shale Oil Reservoir [J]. Xinjiang Oil & Gas, 2024, 20(3): 23-30.
[4]	MOU Jianye , YAN Xiaolun , ZHANG Shicheng , SHI Lei , LI Dong , MA Xinfang. Research on CO2 Matrix Penetration Distance in Tight Reservoirs [J]. Xinjiang Oil & Gas, 2024, 20(3): 64-71.
[5]	XIAO Wenlian, CHEN Shengen, YI Yonggang, CHEN Haoyu, REN Jitian. NMR-Based Experiments of Fracturing Fluid Assisted CO₂ Huff-n-Puff for Enhancing Shale Oil Recovery [J]. Xinjiang Oil & Gas, 2024, 20(3): 83-90.
[6]	CHEN Shengen, XU Jinshan, ZHANG Xin, WANG Rui, ZHANG Liwei, LI Haifu. CO2 Flooding Connectivity Characteristics of the Strong Water-Sensitive Conglomerate Reservoir in Xinjiang Oilfield [J]. Xinjiang Oil & Gas, 2024, 20(1): 44-51.
[7]	YANG Tianmeng, YANG Fan, REZEY Rehemtuli, GOU Guolei, LI Xiuhui, HOU Junwei. Research Progress on Capture of CO2 by Alcohol Amine Absorption Method [J]. Xinjiang Oil & Gas, 2024, 20(1): 52-60.
[8]	LI Qiang, , WANG Fuling, , ZHOU Chang , Chen Jiashuo , LI Qingchao, , Wang Yanling. Effect of Siloxane-Based Thickener on Properties of Carbon Dioxide Fracturing Fluid [J]. Xinjiang Oil & Gas, 2023, 19(1): 73-80.
[9]	CHEN Shengen, WANG Rui, YI Yonggang, ZHANG Xin, XU Jinshan, SHANG Jin. Experimental Study on Film Permeability and Stability of Nanofiber Enhanced Foam [J]. Xinjiang Oil & Gas, 2023, 19(1): 81-89.
[10]	CAO Liyuan , QIAN Weiming , GONG Ping , GUAN Shijie , DUAN Yutao , HE Yuan. Application Practices and Evaluation of Gas Injection Technology for CO2Flooding in Subei Oilfield#br# [J]. Xinjiang Oil & Gas, 2022, 18(2): 46-50.