Probabilistic modelling of building response to induced seismicity from geothermal operations

Geothermal energy provides great potential for a carbon-neutral, weather-independent energy supply. However, induced seismicity poses a challenging issue, particularly because of the proximity of geothermal power plants to populated areas, ensuring the provision of energy near the consumer. In consequence, even minor events might be perceivable by the residents and raise public concern. This work proposes a computational framework for assessing the risk of the serviceability of existing buildings, considering the parameter variation and the dynamic model uncertainty. The structural response prediction is generated using a parametric finite element (FE) model combined with a multi-modal response spectrum analysis. In order to capture the stochastic nature of the dynamic system, Monte Carlo simulations are employed. The seismic input is modelled using newly developed Ground Motion Prediction Equations (GMPEs), and the statistical distributions of the building parameters are derived from the national building database according to the area of interest. The resulting risk assessment is expressed by fragility curves, integrating probabilistic estimates, parameter distributions, and model uncertainties. This framework provides a comprehensive basis for assessing and managing the risk for the serviceability of existing buildings under induced seismicity.