Sustainability, climate resiliency, and mitigation capacity of geothermal heat pump systems in cold regions

•Air temperature can be used to estimate ground temperatures using ANN.•The performance of shallow exchangers drastically drops during the operation season.•Thermal depletion is observed during the early years of operation.•Seasonal balancing alleviates depletion and improves the performance of the...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Geothermics 2021-03, Vol.91, p.101979, Article 101979
Hauptverfasser: Gheysari, Ali Fatolahzadeh, Holländer, Hartmut M., Maghoul, Pooneh, Shalaby, Ahmed
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Air temperature can be used to estimate ground temperatures using ANN.•The performance of shallow exchangers drastically drops during the operation season.•Thermal depletion is observed during the early years of operation.•Seasonal balancing alleviates depletion and improves the performance of the system.•GHP technology reduces the carbon footprint of residential and commercial buildings. Geothermal heat pump (GHP) systems have recently gained popularity in providing heat to buildings. In this study, the short-term and long-term performance of closed-loop horizontal GHP systems in cold regions and the effectiveness of seasonal balancing were investigated. A period of 50 years was simulated to cover the life of a conventional system. A repeating block of heat exchangers at three depths was modeled in a series of 3D multi-physics finite element simulations. Two operation modes were defined, representing the non-balanced mode and the seasonal balancing through the injection of lake water in summer. The climate conditions at the ground surface were explicitly modeled by defining a temperature boundary condition. A novel AI framework was developed to generate the projections of ground surface temperatures from air temperature and other atmospheric variables. An artificial neural network was trained using the air and ground temperature measurements at the nearby weather station. Several training approaches were compared to minimize prediction errors. The model was then used to convert the air temperatures from downscaled outputs of the CanESM2 climate model into surface temperatures. Finite element models, representing the two operation modes under three climate change scenarios and various heat exchanger spacing, were simulated and verified by comparing the resulting ground temperatures with the local measurements. The outputs were processed into extraction power, thermal output, and carbon emissions. Results suggest that in order to attain a stable heat extraction throughout a year, heat exchangers should be placed at depths that are not affected by seasonal variations of surface conditions. Simulations revealed thermal depletion and its inverse correlation with heat exchanger spacing. The seasonal balancing operation approach showed enhancement in total thermal output by 55%. The system was found to be resilient when subjected to the major climate pathways. In a rough estimate based on the average carbon footprint of electricity production in Canada, a syste
ISSN:0375-6505
1879-3576
DOI:10.1016/j.geothermics.2020.101979