Uncertainty Quantification of Borehole Thermal Energy Storage Facilities

Borehole thermal energy storages (BTES) have become a common implement for extracting and/or storing heat energy from and into the soil. Building these facilities is expensive, especially the drilling of boreholes, into which borehole heat exchangers are inserted. To cut costs, drilling methods, which can produce inaccuracies of varying degree, are utilized. This brings into question how much these inaccuracies could potentially affect the energy storage/extraction performance of a planned facility. To this end, we performed an uncertainty quantification for seasonally operated BTES facilities, where we studied the influence of geometries deviating from the planned layout and other sources of uncertainty, such as varying soil and material parameters.
In our research, we make use of a 3D simulation model for BTES facilities in a patch of soil with optional groundwater flow, designed as a system of partial differential equations (PDEs). The system is solved with a simulation toolkit, which was programmed as an extension for the finite element method solver KARDOS. The toolkit builds on previous work for the simulation tool BASIMO and was validated with benchmarks calculated with the commercial software FEFLOW, which specializes in heat transfer in porous media among other things. For the uncertainty quantification, we utilize an adaptive, anisotropic stochastic collocation method, which uses solutions of the PDE system as samples. We present the method and apply it to an illustrative as well as a practical example. Lastly, we discuss the results and assess the impact of deviating borehole paths on the performance of BTES facilities.