Abstract
The increasing use of ground-source heat pumps (GSHPs) for heating and cooling of buildings raises questions regarding the technical potential of GSHPs and their impact on the temperature in the shallow subsurface. In this paper, we develop a method using Machine Learning to estimate the technical potential of shallow GSHPs, which enables such an estimation for Switzerland with limited data and computational resources. A training dataset is constructed based on meteorological and geological data across Switzerland. We analyse correlations and the importance of each of the input data for estimating the GSHP potential and compare different input feature sets and Machine Learning models. The Random Forest algorithm, trained on the full dataset, provides the best performance to estimate the GSHP potential. The resulting model yields an R2 score of 0.95 for the annual energy potential, 0.86 for the heat extraction rate, and 0.82 for the potential number of boreholes per GSHP system.
Highlights
The extraction of shallow geothermal energy using ground-source heat pumps (GSHPs) of depths up to 400 m is growing worldwide
We develop a method using Machine Learning to estimate the technical potential of shallow GSHPs, which enables such an estimation for Switzerland with limited data and computational resources
This study presents a Machine Learning method to estimate the technical potential of GSHP systems, their heat extraction rate, and the number of boreholes across a wide range of borehole field geometries, meteorological conditions, and geological properties
Summary
The extraction of shallow geothermal energy using ground-source heat pumps (GSHPs) of depths up to 400 m is growing worldwide. A dense installation of BHEs may, lead to an excessive long-term cooling of the subsurface due to thermal interactions between boreholes. Thermal interaction must be considered when quantifying the technical potential of BHEs, that is, the long-term maximum energy which can be extracted annually using GSHPs [2,3]. While analytical methods exist to quantify this potential, they are computationally intensive and require the availability of high-resolution data of the thermal properties of the shallow ground. These limitations may hinder the use of analytical models to quantify the technical potential of shallow GSHPs at national scale. Machine Learning (ML) can handle missing data in analytical models and, once trained, is computationally highly efficient [4]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
