Physical and hybrid modelling techniques for earth-air heat exchangers in reducing building energy consumption: Performance, applications, progress, and challenges

• Published in: Solar energy 2021-03, Vol.216, p.274-294
• Main Authors: Ahmed, S.F., Liu, G., Mofijur, M., Azad, A.K., Hazrat, M.A., Chu, Yu-Ming
• Format: Article
• Language: English
• Subjects: Building energy consumption; Buildings; Earth-air heat exchanger; Energy consumption; Energy efficiency; Heat exchangers; Heat transfer; Modelling technique; Solar energy; Solar radiation; Training
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/j.solener.2021.01.022

•Physical and hybrid EAHE modelling techniques used in buildings have been reviewed.•The prospects, benefits, progress, and challenges of these modelling techniques are analyzed.•The hybrid modelling technique is found more effective based on overall performance.•Physical models offer improved performance in terms of generalization capability. Noteworthy advancements are seen in developing the earth-air heat exchanger (EAHE) models in the past several decades to reduce building energy consumption. However, it is still an ongoing challenge in selecting and implementing the most suitable and appropriate EAHE modelling technique in buildings based on the climates, performance, and limitations of the techniques. Therefore, this paper aims to review the published research related to the physical, and hybrid EAHE modelling techniques used in buildings, and highlight the prospects, benefits, progress, and challenges of these techniques. This is the first study that comprehensively evidences the prospects and technical challenges caused by unmeasured disturbances, assumptions, or the uncertainties generated in experimental and numerical works of all EAHE modelling techniques. Nevertheless, this study found that hybrid modelling is more effective than physical models for accurate prediction. On the contrary, the hybrid models suffer from high complexity if EAHE operating conditions and all key parameters are considered during the model development. Regarding the generalization capability, the physical models offer improved performance followed by the hybrid models. A minimum number of training data is needed for developing physical models, whereas medium training data is required for the hybrid models. The outcome of this study also provides valuable information regarding the physical and hybrid EAHE modelling techniques to the scientists, researchers, and so on in adopting the most appropriate EAHE modelling technique for their climates.