With the energy demand arising globally, geothermal recovery by Enhanced Geothermal Systems (EGS) becomes a promising option to bring a sustainable energy supply and mitigate CO2 emission. However, reservoir management of EGS primarily relies on reservoir simulation, which is quite expensive due to the reservoir heterogeneity, the interaction of matrix and fractures, and the intrinsic multi-physics coupled nature. Therefore, an efficient optimization framework is critical for the management of EGS. We develop a general reservoir management framework with multiple optimization options. A robust forward surrogate model fl is developed based on a convolutional neural network, and it successfully learns the nonlinear relationship between input reservoir model parameters (e.g., fracture permeability field) and interested state variables (e.g., temperature field and produced fluid temperature). fl is trained using simulation data from EGS coupled thermal-hydro simulation model by sampling reservoir model parameters. As fl is accurate, efficient and fully differentiable, EGS thermal efficiency can be optimized following two schemes: (1) training a control network fc to map reservoir geological parameters to reservoir decision parameters by coupling it withfl ; (2) directly optimizing the reservoir decision parameters based on coupling the existing optimizers such as Adam withfl. The forward model fl performs accurate and stable predictions of evolving temperature fields (relative error1.27±0.89%) in EGS and the time series of produced fluid temperature (relative error0.26±0.46%), and its speedup to the counterpart high-fidelity simulator is 4564 times. When optimizing withfc, we achieve thermal recovery with a reasonable accuracy but significantly low CPU time during inference, 0.11 seconds/optimization. When optimizing with Adam optimizer, we achieve the objective perfectly with relatively high CPU time, 4.58 seconds/optimization. This is because the former optimization scheme requires a training stage of fc but its inference is non-iterative, while the latter scheme requires an iterative inference but no training stage. We also investigate the option to use fc inference as an initial guess for Adam optimization, which decreases Adam's CPU time, but with excellent achievement in the objective function. This is the highest recommended option among the three evaluated. Efficiency, scalability and accuracy observed in our reservoir management framework makes it highly applicable to near real-time reservoir management in EGS as well as other similar system management processes.

Energy extraction from the Enhanced Geothermal System (EGS) is highly dependent on the transmissivity of fractures. However, due to the heterogeneity and complex multi-physics nature, high-fidelity physics-based forward simulation can be computationally intensive, creating a barrier to efficient reservoir management. A robust and fast optimization framework for maximizing the thermal recovery from EGS is needed. We developed a general reservoir management framework which is combining a low-fidelity forward surrogate model (fl) with gradient-based optimizers to speed up reservoir management process. thermo-hydro-mechanical (THM) EGS simulation model is developed based on the finite element method. We parameterized the fracture aperture and well controls and ran the THM model to generate 2500 datasets. Further, we used two different deep neural networks (DNNs) with the datasets to predict the dynamics of pressure and temperature, and this ultimately becomes the fl for calculating the energy production. Instead of performing optimization workflow with large amount of simulations from fh, we directly optimize the well control parameters based on geological input to the fl. As the fl can reach the high accuracy with fast prediction, also it is differentiable, gradient-based optimization was utilized to find the maximized total energy production optimum with temperature constraint at producer. Based on the simulation datasets, we evaluated the fracture aperture and temperature evolution and demonstrated that the spatial fracture aperture distribution dominates the thermal front movements. The fracture aperture expansion is highly correlated with temperature change inside of the fracture, mainly from thermal stress changes. Compared to the full-fledged physics simulator, our forward surrogate model based on DNN not only provides a computational speedup of around 1500 times, but also brings a high R2 value about 99% in predicting subsurface responses. With the aid of the efficient forward model fl, gradient-based optimizers show efficient optimization with 10-68 times faster than the derivative-free global optimization method. The proposed reservoir management framework shows both efficiency and scalability, which enables each optimization process to be executed within half a minute.

No abstract

No abstract

In this study we developed mathematical model for thermo-hydro-mechanical process occurs within the geothermal reservoir with variable rock/fracture/fluid parameters. The influence of fracture network on the cold plume movement, pore pressure, changes in the rock/fracture effective stress under the same operating conditions. The injected fluid transport to extraction well from injection well within the interconnected fractures. In the same direction variation of the effective stress, pore pressure both in rock matrix and fractures was observed. Due to the variation of effective stress in the fracture, it will undergo shearing and alter the fracture aperture. This variation of fracture aperture will create a micro-seismic moment in the fractured geothermal reservoir. The magnitude of micro-seismic moment and hyper center were changing with time and highly sensitive to the fracture connectivity of each fracture set. The developed mathematical model was observed these variations efficiently. Thus, the developed model can be utilized to address the variations occurred throughout the heat extraction in the fractured geothermal reservoir in conjunction with the activation of fracture and location of hyper center of each seismic moment.