A key part of planning CO2 storage sites is to devise a monitoring strategy. The aim of this strategy is to fulfill the requirements of legislations and lower cost of the operation by avoiding operational problems. If CCS is going to be a widespread technology to deliver energy without CO2 emissions, cost-efficient monitoring programs will be a key to reduce the storage costs. A simulation framework, previously used to estimate flow parameters at Sleipner Layer 9 [1], is here extended and employed to identify how the number of measurements can be reduced without significantly reducing the obtained information. The main part of the methodology is based on well-proven, stable and robust, simulation technology together with adjoint-based sensitivities and data mining techniques using singular value decomposition (SVD). In particular we combine the simulation framework with time-dependent (seismic) measurements of the migrating plume. We also study how uplift data and gravitational data give complementary information.
We apply this methodology to the Sleipner project, which provides the most extensive data for CO2 storage to date. For this study we utilize a vertical-equilibrium (VE) flow model for computational efficiency as implemented in the open-source software MRST-co2lab.
However, our methodology for deriving efficient monitoring schemes is not restricted to VE-type flow models, and at the end, we discuss how the methodology can be used in the context of full 3D simulations.
The Roles of Low-Noise Stations, Arrays and Ocean-Bottom Seismometers in Monitoring UK Offshore Seismicity associated with Subsurface Storage of Carbon Dioxide
School of Geosciences, University of Edinburgh(爱丁堡大学地球科学学院)
Bayesian Full-waveform Monitoring of CO2 Storage with Fluid-flow Priors via Generative Modeling
Stanford University(斯坦福大学);Institute of GeoEnergy Engineering, School of Energy, Geoscience, Infrastructure and Society, Heriot-Watt University(赫瑞-瓦特大学能源、地球科学、基础设施与社会学院地质能源工程研究所)
Connect the Dots: In Situ 4D Seismic Monitoring of CO2 Storage with Spatio-temporal CNNs
Los Alamos National Laboratory(洛斯阿拉莫斯国家实验室);Michigan State University(密歇根州立大学)
4D震波成像已被广泛用于CO $_2$封存项目,以监测未通过井采样的体积地下区域中的流体流动。理想情况下,实时监测和近期预测将为现场运营商提供深刻的见解,以了解地下储层的动态并评估任何潜在风险。然而,由于诸如高部署成本、采集设备的可用性、表面结构周围的禁区等障碍,在监测期间只能获得非常稀疏的震波成像数据。随着时间的推移,这导致了不可避免的和不断扩大的知识差距。运营商需要了解整个项目生命周期内的流体流动,而地震资料只能在有限的时间内获得。这对于理解储层动态是不够的。为了克服这些挑战,我们开发了基于时空神经网络的模型,可以有效和高效地产生高保真的内插或外推图像。具体来说,我们的模型建立在自动编码器上,并将长短期记忆(LSTM)结构与光流正则化的新损失函数相结合。我们验证我们的模型使用真实的4D叠后震波成像数据获得的Sleipner CO $_2$封存领域的性能。我们采用两种不同的策略来评估我们的模型。在数值上,我们使用经典的基于像素的指标将我们的模型与不同的基线方法进行比较。我们还进行了一项盲测,并从领域专家那里收集了总共20份回复,以评估我们的模型生成的数据质量。通过数值和专家评估,我们得出结论,我们的模型可以产生高质量的二维/三维震波成像数据在一个合理的成本,提供了实时监测,甚至不久的将来预测的CO $_2$存储水库的可能性。
Deep Learning-Accelerated 3D Carbon Storage Reservoir Pressure Forecasting Based on Data Assimilation Using Surface Displacement from InSAR
Lawrence Livermore National Laboratory(劳伦斯利弗莫尔国家实验室);TotalEnergies E&P Research and Technology(道达尔能源勘探与生产研究技术中心);University of North Dakota, Energy & Environmental Research Center(北达科他大学能源与环境研究中心)
通过同化监测数据快速预测地质碳封存储层压力分布是一个具有挑战性的问题。由于高钻井成本,GCS项目通常具有来自井的空间稀疏测量,导致储层压力预测的高不确定性。为了应对这一挑战,我们建议使用低成本的干涉合成孔径雷达(InSAR)数据作为监测数据来推断储层压力的建立。我们开发了一个深度学习加速的工作流程,以同化从InSAR解释的地表位移图,并预测动态储层压力。该工作流程采用Encourage Smoother Multiple Data Assimilation(ES-MDA)框架,更新三维(3D)地质属性,并预测具有量化不确定性的储层压力。我们使用一个合成的商业规模的GCS模型双峰分布的渗透率和孔隙度,以证明工作流程的有效性。一个两步CNN-PCA方法被用来参数化的双峰字段。工作流程的计算效率通过两个基于剩余U-Net的代理模型分别用于地表位移和储层压力预测来提高。该工作流程可在个人计算机上在半小时内完成数据同化和储层压力预测。