Efficient heat exploitation strategies from geothermal systems demand for accurate and efficient simulation of coupled flow-heat equations on large-scale heterogeneous fractured formations. While the accuracy depends on honouring highresolution discrete fractures and rock heterogeneities, specially avoiding excessive upscaled quantities, the efficiency can be maintained if scalable model-reduction computational frameworks are developed. Addressing both aspects, this work presents a multiscale formulation for geothermal reservoirs. To this end, the nonlinear time-dependent (transient) multiscale coarse-scale system is obtained, for both pressure and temperature unknowns, based on elliptic locally solved basis functions. These basis functions account for fine-scale heterogeneity and discrete fractures, leading to accurate and efficient simulation strategies. The flow-heat coupling is treated in a sequential implicit loop, where in each stage, the multiscale stage is complemented by an ILU(0) smoother stage to guarantee convergence to any desired accuracy. Numerical results are presented in 2D to systematically analyze the multiscale approximate solutions compared with the fine scale ones for many challenging cases, including the outcrop-based geological fractured field. These results show that the developed multiscale formulation casts a promising framework for the real-field enhanced geothermal formations.

No abstract

Thermochemical Energy Storage (TCES), specifically the calcium oxide (CaO)/calcium hydroxide (Ca(OH)2) system is a promising energy storage technology with relatively high energy density and low cost. However, the existing models available to predict the system’s internal states are computationally expensive. An accurate and real-time capable model is therefore still required to improve its operational control. In this work, we implement a Physics-Informed Neural Network (PINN) to predict the dynamics of the TCES internal state. Our proposed framework addresses three physical aspects to build the PINN: (1) we choose a Nonlinear Autoregressive Network with Exogeneous Inputs (NARX) with deeper recurrence to address the nonlinear latency; (2) we train the network in closed-loop to capture the long-term dynamics; and (3) we incorporate physical regularisation during its training, calculated based on discretized mole and energy balance equations. To train the network, we perform numerical simulations on an ensemble of system parameters to obtain synthetic data. Even though the suggested approach provides results with the error of 3.96×10−4 which is in the same range as the result without physical regularisation, it is superior compared to conventional Artificial Neural Network (ANN) strategies because it ensures physical plausibility of the predictions, even in a highly dynamic and nonlinear problem. Consequently, the suggested PINN can be further developed for more complicated analysis of the TCES system.