| Citation: |
MA Xianlin, LIU Zhenzhi, ZHAN Jie, et al. Prediction method of CO2 plume distribution based on physics-informed neural networks [J]. Petroleum Drilling Techniques, 2024, 52(5)：69−75. DOI: 10.11911/syztjs.2024090
|
To enhance the effectivity and safety of CO2 geological storage, accurate prediction of CO2 plume distribution and migration in formations has become essential. Therefore, the partial differential equation (PDE) constraints of multiphase flow were embedded into the loss function of the model by using automatic differential technique, and deep neural network models were developed to predict CO2 plume distributions, with constraints imposed by multiphase flow mechanics, ensuring that the model’s prediction results not only conform to the distribution law of training data samples but also strictly abide by the physical law of fluid seepage described by the PDE. To validate the model’s effectiveness, two PINN models were constructed using a multi-layer perceptron (MLP) and a long short-term memory (LSTM) network. These were applied in a practical case study on CO2 storage within a depleted oil reservoir. The results show that compared with pure data-driven models, the PINNs-based models demonstrate superior prediction accuracy. The findings of this research provide technical support for the design and implementation of CO2 geological storage projects while offering a theoretical foundation for the practical application of this technology.
| [1] |
周守为,朱军龙. 助力“碳达峰、碳中和”战略的路径探索[J]. 天然气工业,2021,41(12):1–8. doi: 10.3787/j.issn.1000-0976.2021.12.001
ZHOU Shouwei, ZHU Junlong. Exploration of ways to helping “Carbon Peak and Neutrality” strategy[J]. Natural Gas Industry, 2021, 41(12): 1–8. doi: 10.3787/j.issn.1000-0976.2021.12.001
|
| [2] |
霍宏博,刘东东,陶林,等. 基于CO2提高采收率的海上CCUS完整性挑战与对策[J]. 石油钻探技术,2023,51(2):74–80.
HUO Hongbo, LIU Dongdong, TAO Lin, et al. Integrity challenges and countermeasures of the offshore CCUS based on CO2-EOR[J]. Petroleum Drilling Techniques, 2023, 51(2): 74–80.
|
| [3] |
杨术刚,李兴春,蔡明玉,等. 国外CO2地质封存管理制度、标准体系分析及其启示[J]. 天然气工业,2023,43(12):130–137.
YANG Shugang, LI Xingchun, CAI Mingyu, et al. Overseas management systems and standards for CO2 geological storage and their implications for China[J]. Natural Gas Industry, 2023, 43(12): 130–137.
|
| [4] |
柏明星,张志超,白华明,等. 二氧化碳地质封存系统泄漏风险研究进展[J]. 特种油气藏,2022,29(4):1–11.
BAI Mingxing, ZHANG Zhichao, BAI Huaming, et al. Progress in leakage risk study of CO2 geosequestration system[J]. Special Oil & Gas Reservoirs, 2022, 29(4): 1–11.
|
| [5] |
李凤霞,王海波,周彤,等. 页岩油储层裂缝对CO2吞吐效果的影响及孔隙动用特征[J]. 石油钻探技术,2022,50(2):38–44.
LI Fengxia, WANG Haibo, ZHOU Tong, et al. The influence of fractures in shale oil reservoirs on CO2 huff and puff and its pore production characteristics[J]. Petroleum Drilling Techniques, 2022, 50(2): 38–44.
|
| [6] |
李阳,王敏生,薛兆杰,等. 绿色低碳油气开发工程技术的发展思考[J]. 石油钻探技术,2023,51(4):11–19.
LI Yang, WANG Minsheng, XUE Zhaojie, et al. Thoughts on green and low-carbon oil and gas development engineering technologies [J]. Petroleum Drilling Techniques, 2023, 51(4): 11–19.
|
| [7] |
张涛,杨若凡,常文杰,等. CO2伴生气混合过程的数值模拟研究[J]. 西南石油大学学报(自然科学版),2023,45(3):143–153.
ZHANG Tao, YANG Ruofan, CHANG Wenjie, et al. Numerical simulation of CO2 associated gas mixing process[J]. Journal of Southwest Petroleum University(Science & Technology Edition), 2023, 45(3): 143–153.
|
| [8] |
赵鹏,朱海燕,张丰收. CO2增强页岩气开采及地质埋存的三维数值模拟[J]. 天然气工业,2024,44(4):104–114.
ZHAO Peng, ZHU Haiyan, ZHANG Fengshou. Three-dimensional numerical simulation of CO2 injection to enhance shale gas recovery and geological storage[J]. Natural Gas Industry, 2024, 44(4): 104–114.
|
| [9] |
LI Dong, PENG Suping, GUO Yinling, et al. CO2 storage monitoring based on time-lapse seismic data via deep learning[J]. International Journal of Greenhouse Gas Control, 2021, 108: 103336. doi: 10.1016/j.ijggc.2021.103336
|
| [10] |
SINHA S, DE LIMA R P, LIN Youzuo, et al. Normal or abnormal? Machine learning for the leakage detection in carbon sequestration projects using pressure field data[J]. International Journal of Greenhouse Gas Control, 2020, 103: 103189. doi: 10.1016/j.ijggc.2020.103189
|
| [11] |
ZHONG Zhi, SUN A Y, YANG Qian, et al. A deep learning approach to anomaly detection in geological carbon sequestration sites using pressure measurements[J]. Journal of Hydrology, 2019, 573: 885–894. doi: 10.1016/j.jhydrol.2019.04.015
|
| [12] |
KARNIADAKIS G E, KEVREKIDIS I G, LU Lu, et al. Physics-informed machine learning[J]. Nature Reviews Physics, 2021, 3(6): 422–440. doi: 10.1038/s42254-021-00314-5
|
| [13] |
RAISSI M, PERDIKARIS P, KARNIADAKIS G E. Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations[J]. Journal of Computational Physics, 2019, 378: 686–707. doi: 10.1016/j.jcp.2018.10.045
|
| [14] |
赵暾,周宇,程艳青,等. 基于内嵌物理机理神经网络的热传导方程的正问题及逆问题求解[J]. 空气动力学学报,2021,39(5):19–26.
ZHAO Tun, ZHOU Yu, CHENG Yanqing, et al. Solving forward and inverse problems of the heat conduction equation using physics-informed neural networks[J]. Acta Aerodynamica Sinica, 2021, 39(5): 19–26.
|
| [15] |
李野,陈松灿. 基于物理信息的神经网络:最新进展与展望[J]. 计算机科学,2022,49(4):254–262.
LI Ye, CHEN Songcan. Physics-informed neural networks: recent advances and prospects[J]. Computer Science, 2022, 49(4): 254–262.
|
| [16] |
JAGTAP A D, KHARAZMI E, KARNIADAKIS G E. Conservative physics-informed neural networks on discrete domains for conservation laws: applications to forward and inverse problems[J]. Computer Methods in Applied Mechanics and Engineering, 2020, 365: 113028. doi: 10.1016/j.cma.2020.113028
|
| [17] |
KHARAZMI E, ZHANG Zhongqiang, KARNIADAKIS G E M. hp-VPINNs: variational physics-informed neural networks with domain decomposition[J]. Computer Methods in Applied Mechanics and Engineering, 2021, 374: 113547. doi: 10.1016/j.cma.2020.113547
|
| [18] |
YANG Liu, MENG Xuhui, KARNIADAKIS G E. B-PINNs: Bayesian physics-informed neural networks for forward and inverse PDE problems with noisy data[J]. Journal of Computational Physics, 2021, 425: 109913. doi: 10.1016/j.jcp.2020.109913
|
| [19] |
LU Lu, MENG Xuhui, MAO Zhiping, et al. DeepXDE: a deep learning library for solving differential equations[J]. SIAM Review, 2021, 63(1): 208–228. doi: 10.1137/19M1274067
|
| [20] |
RACKAUCKAS C, NIE Qing. DifferentialEquations. jl: a performant and feature-rich ecosystem for solving differential equations in Julia[J]. Journal of Open Research Software, 2017, 5: 15. doi: 10.5334/jors.151
|
| [21] |
薛亮,戴城,韩江峡,等. 油藏渗流物理和数据联合驱动的深度神经网络模型[J]. 油气地质与采收率,2022,29(1):145–151.
XUE Liang, DAI Cheng, HAN Jiangxia, et al. Deep neural network model driven jointly by reservoir seepage physics and data[J]. Petroleum Geology and Recovery Efficiency, 2022, 29(1): 145–151.
|
| [22] |
SHOKOUHI P, KUMAR V, PRATHIPATI S, et al. Physics-informed deep learning for prediction of CO2 storage site response[J]. Journal of Contaminant Hydrology, 2021, 241: 103835. doi: 10.1016/j.jconhyd.2021.103835
|
| [23] |
EBIGBO A, CLASS H, HELMIG R. CO2 leakage through an abandoned well: problem-oriented benchmarks[J]. Computational Geosciences, 2007, 11(2): 103–115. doi: 10.1007/s10596-006-9033-7
|
| [24] |
LALLAHEM S, MANIA J, HANI A, et al. On the use of neural networks to evaluate groundwater levels in fractured media[J]. Journal of Hydrology, 2005, 307: 92–111.
|
| [25] |
HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Computation, 1997, 9(8): 1735–1780. doi: 10.1162/neco.1997.9.8.1735
|
| 1. |
潘冠昌，杨斌，张浩，常坤，冯云辉. 超深层碳酸盐岩裂缝面形态与摩擦因数研究. 断块油气田. 2022(06): 794-799 .
|
Supervisor：SINOPEC GROUP
Sponsor：Sinopec Research Institute of Petroleum Engineering
Serial Number：CN 11-1763/TE, ISSN 1001-0890
Chief Editor： WANG Minsheng
Supported by: Beijing Renhe Information Technology Co., Ltd. 
Copyright © 2009《Petroleum Drilling Techniques 》 All Rights Reserved.
DownLoad: