Modelling of borehole thermal energy storages: A g-function approach with a novel load aggregation scheme

• Novel load aggregation reduces errors in BTES during sharp load variations. • Cross-validation ensures accurate g-functions for BTES simulations. • Perfect insulation boosts BTES thermal level by up to 3.5 °C. • Model predicts 70% seasonal efficiency for Palermo BTES system. • The Python model enables future Digital Twin for real-time BTES optimization. Borehole Thermal Energy Storage systems can play a pivotal role in enhancing the energy efficiency of building heating, promoting the widespread adoption of heat pumps and solar thermal solutions especially at district-level. In this paper, a novel Python-based model for the analysis of these renewable systems is proposed employing a g-function approach and introducing a new thermal load aggregation scheme to enable accurate and efficient numerical simulations. The proposed model was applied to assess the operation of a recently constructed pilot-scale seasonal thermal storage at the University of Palermo campus, thus demonstrating the feasibility of deploying this technology for building heating in the Mediterranean region. For this case study, multiple g-functions were generated using both the Python pygfunction library and finite element models developed in COMSOL Multiphysics®, allowing investigation of how borehole hydraulic connections, undisturbed thermal profiles, soil thermophysical properties, and storage surface insulation conditions influence the thermal response of the borehole thermal energy storage. The results indicate that the novel load aggregation scheme markedly reduces the errors encountered by conventional algorithms, particularly during the sharp thermal load variations between charging and discharging phases. Moreover, simulations conducted with the new code confirmed a seasonal efficiency of approximately 70 % for the Palermo storage and revealed that perfect surface insulation could raise its thermal level by about 4 °C. The new modelling framework also paves the way for a future Digital Twin that will enable real-time optimization of charge/discharge cycles.