Deep geothermal energy has the potential to provide sustainable, clean and long-lasting energy resources. Therefore, it plays a strategic role in the energy policies for the coming years. Maximizing the efficiency of this technology and ensuring its profitability is a challenge due to the inaccessibility and uncertainty of the geology consisting of highly heterogeneous fractured rocks, and due to the complexity of the coupled physical processes involved of flow, heat transport and mechanical deformation. The expected growth of this technology manifests the need to manage multiple projects optimally and quickly. The GeotermIA project proposes to develop a tool for the real-time management and optimization of deep geothermal resources through the use of AI, which contributes to the development of digital twins of geothermal facilities.
GeotermIA will develop a deep learning algorithm informed by physics. Training in the identification of patterns and trends useful to improve the efficiency of the geothermal system will be based on historical and synthetic data from simplified, high-fidelity and stochastic numerical models. The tool will allow to evaluate the performance of geothermal systems in real time, identify possible problems and propose optimization solutions, thus providing solid assistance in the decision-making process and contributing to the transition towards more sustainable and clean energy sources.
Funding: Programa Momentum CSIC, Plan de Recuperación, Transformación y Resiliencia – Financiado por la Unión Europea – NextGenerationEU – Ref. MMT24-IDAEA-01
