Abstract
Ground source heat pump (GSHP) systems are using vertical ground heat exchangers, known as Borehole Heat Exchangers (BHEs), as a heat source or sink. The performance of the GSHP system strongly relies on the ground temperature surrounding the BHEs. This temperature depends on many parameters and varies during the operating time. Therefore, the determination of the ground temperature is crucial to define the design and the proper size of the BHEs so that the performance of the GSHP system can be kept at the desired level. The current study aims to formulate a complex structure of artificial neural network (ANN) model in a mathematical equation that expresses the change in the ground temperature around BHEs due to heat injection in the long run. To fulfill this aim, a numerical model of BHEs was created using the ANSYS (Analysis System) software to generate data. The generated data was then used to train the ANN model, which was built for this study. The simulation results show that the ANN model estimates the ground temperature (T-g) in the target GSHP system with higher accuracy.
Holger Wallbaum
Full Professor, Vice-Head of Department and Vice-Dean for Research
Holger is a Full Professor in sustainable building at the Division of Building Technology, research group Sustainable Building, and in the Area of advance Building Futures. Holger works within sustainable building on concepts, tools and strategies to enhance the sustainability performance of construction materials, building products, buildings as well as entire cities.