Numerical simulation and big data analysis for assessing the geothermal utilization potential of deep-buried pipe systems

As a clean and sustainable source, geothermal energy is a key component in the global energy transition. One of the main approaches to geothermal energy utilization involves mid-to-deep layer buried pipe heating technology, emphasizing the efficient and accurate assessment of heat transfer performance. This study presents a three-dimensional, full-scale numerical modeling of heat transfer in deeply buried pipes. Based on the simulation results, a database for evaluating the heat transfer rate of buried pipes is developed. This database serves as the basis for neural network predictions of heat transfer rate. The proposed predictive model is evaluated, and the results indicate that its computational efficiency is over 1000 times higher than that of conventional numerical models. Aside from the initial 1 h of heat transfer, the maximum relative error of the predictions compared to the numerical results is within 0.5 %. This study provides innovative approaches for the theoretical evaluation and technical optimization of deeply buried pipe geothermal systems . The findings contribute to the improvement of energy supply systems, accelerating energy conservation, reducing emissions, and enhancing ecological restoration. • A 3D full-scale numerical model for C-DBHE heat transfer is established. • Orthogonal simulation conditions are designed to study the DBHE heat transfer. • A neural network model is trained based on the orthogonal simulation results. • A DBHE heat transfer prediction platform is constructed using MATLAB.