Determination of Gas–Oil minimum miscibility pressure for impure CO2 through optimized machine learning models

混溶性 材料科学 石油工程 热力学 工程类 复合材料 物理 聚合物
作者
Chenyu Wu,Lu Jin,Jin Zhao,Xincheng Wan,Tao Jiang,Kegang Ling
出处
期刊: [Elsevier BV]
卷期号:242: 213216-213216
标识
DOI:10.1016/j.geoen.2024.213216
摘要

Minimum miscibility pressure (MMP) is one of the most important parameters for designing CO 2 enhanced oil recovery (EOR) and associated storage in depleted oil reservoirs. The injection gas stream often contains a certain concentration of impurities such as N 2 , H 2 S, and CH 4 depending on the source of CO 2 . These impurities have different effects on CO 2 MMP, but there is a lack of widely accepted approaches to account for these effects on MMP calculation. In this study, a series of activities were conducted to develop a machine learning (ML)-based methodology for determining MMP for CO 2 with various impurities. A database containing 234 CO 2 MMP test sets with around 5000 data points was built based on the reported experimental measurements in the public domain. The database was then subgrouped by three specific criteria: CO 2 concentration in the injection gas, type of impurities in the injection gas, and heavier hydrocarbon content in the oil. This subgrouping was essential to capture the impact of different factors on CO 2 MMP. An ensemble ML approach with seven ML models, including random forest, adaptive boosting, light gradient boosting machine, extreme gradient boosting (XGBoost), stacking, artificial neural network, and voting regressor, was employed to calculate MMP based on the subgrouped database. The hyperparameters of these ML models were optimized by the grid search technique to minimize the relative errors between calculated and measured MMP values. The performance of each algorithm was assessed using three regression metrics: average absolute relative error (AARE), R-squared score (R 2 ), and root mean square error (RMSE). All of these metrics exhibited satisfactory values for the optimized ML models. The average values of R 2 , RMSE, and AARE were 0.962, 1.571, and 4.55%, respectively, for the three subgroups, indicating a high accuracy of MMP calculations using the optimized ML models. The XGBoost model emerged as the top performer across the three metrics, with an R 2 of 0.979, an AARE of 2.835%, and an RMSE of 1.183 for a dataset with 190 cases. The overall high level of accuracy confirmed the reliability of these ML models in calculating MMP for CO 2 with different impurities as well as the importance of optimization in the modeling process. • A database with 234 measurements was developed for impure CO 2 MMP investigation. • Seven machine learning models were used to calculate MMP for CO 2 with impurities. • The ML models were optimized by data subgrouping and grid search technique. • The optimized ML models calculated MMP for impure CO 2 accurately. • All three regression metrics confirmed the reliability of the MMP calculation.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
今后的应助被gp采纳,获得10
1秒前
搞材料的电子完成签到 ,获得积分10
2秒前
2秒前
辛勤振家发布了新的文献求助10
2秒前
迷路的手机完成签到,获得积分10
3秒前
研友_Z30GJ8发布了新的文献求助30
3秒前
syvshc的应助被ke采纳,获得10
3秒前
3秒前
脑洞疼的应助被martin采纳,获得30
5秒前
哈呼呼发布了新的文献求助30
6秒前
GIN的应助被zww采纳,获得10
6秒前
粟兮兮完成签到,获得积分10
6秒前
TT发布了新的文献求助10
7秒前
Akim的应助被tongtong采纳,获得10
7秒前
柠檬味电子对儿完成签到,获得积分10
7秒前
情怀的应助被lunhui6453采纳,获得10
7秒前
7秒前
烟花的应助被不散的和弦采纳,获得10
8秒前
ni完成签到,获得积分10
8秒前
莫晓岚完成签到,获得积分10
9秒前
今后的应助被匆匆采纳,获得10
10秒前
研友_VZG7GZ的应助被wuzhi采纳,获得10
10秒前
太渊发布了新的文献求助10
11秒前
哈猪的妈发布了新的文献求助10
12秒前
menyu的应助被淡淡的新筠采纳,获得10
12秒前
脑洞疼的应助被TT采纳,获得10
13秒前
Nole的应助被5Enu采纳,获得10
13秒前
13秒前
二硫碘化钾完成签到,获得积分20
14秒前
李爱国的应助被SuperGai采纳,获得10
15秒前
SJD完成签到,获得积分0
16秒前
17秒前
kali发布了新的文献求助200
18秒前
flash完成签到,获得积分10
19秒前
tongtong发布了新的文献求助10
19秒前
完美世界的应助被贝奇采纳,获得10
19秒前
xiaolizi发布了新的文献求助10
19秒前
大个的应助被weilanhaian采纳,获得10
19秒前
20秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 2000
Agyptische Geschichte der 21.30. Dynastie 510
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
基于结构调控的功能化水诱导发电材料与器件研究 500
基于界面调控的高效湿气发电材料与器件研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8192519
求助须知:如何正确求助?哪些是违规求助? 9579028
关于积分的说明 21863207
捐赠科研通 7792025
什么是DOI,文献DOI怎么找? 3436955
关于科研通互助平台的介绍 2560622
邀请新用户注册赠送积分活动 2483832