Novel method of thermal behavior prediction of evacuated tube solar collector

• Published in: Solar energy 2020-07, Vol.204, p.761-768
• Main Authors: Nokhosteen, Arman, Sobhansarbandi, Sarvenaz
• Format: Article
• Language: English
• Subjects: Domain names; Evacuated tube solar collector; Heat; Heat pipes; Heat transfer; Mathematical models; Numerical analysis; Proper Orthogonal Decomposition; Resistance network; Robustness (mathematics); Solar collectors; Solar energy; Solar heating; Temperature distribution; Thermal behavior; Thermodynamic properties; Water heaters; Water heating
• ISSN: 0038-092X; 1471-1257
• DOI: 10.1016/j.solener.2020.05.008

•Several shortcomings of RN models for thermal behavior prediction are mitigated.•This model is able to accurately calculate the collector’s temperature distribution.•Any desired temporal resolution of temperature distribution can be achieved.•The modified RN model can extrapolate the thermal behavior of the col-lector.•The proposed model is more computationally efficient than DNS methods. Heat pipe evacuated tube solar collectors (HPETCs) are a type of solar collectors with appealing characteristics for the application in solar water heating (SWH) technologies. In order to better understand the heat transfer phenomena in HPETCs and improve their efficiency, there is a need for a fast and robust numerical tool. Due to the complexity of the heat transfer processes involved in modeling a collector’s performance, direct numerical analysis solutions (DNS) are computationally cumbersome. Recent studies have shown that resistance network (RN) models are suitable tools for studying the performance and thermal behavior of HPETCs. In this work, a novel method of resistance network based proper orthogonal decomposition (RNPOD) is presented which can not only consider the geographical and meteorological characteristics of the ambient surroundings, but also take into account the peripheral temperature distribution of a single HPETCs. Once the temperatures at each instance in time have been calculated, a POD method is used to predict the thermal behavior of the collector with desired temporal accuracy. The obtained results of this study are cross-validated with the previous experimental work of the authors, illustrating that the model is able to predict the peripheral temperature distribution with a maximum error of 10%.