Clustering of visible and infrared solar irradiance for solar architecture design and analysis

RENEWABLE ENERGY

卷:165

Article

Duan, Qiuhua¹  Feng, Yanxiao¹  Wang, Julian¹ 

[1] Penn State Univ, Dept Architectural Engn, 108 Engn Unit A, University Pk, PA 16802 USA

关键词: Solar radiation;    Solar architecture design;    Classification trees;    Prediction model;    Solar energy;    Building energy;   

DOI  :  10.1016/j.renene.2020.11.080

来源: Elsevier

【 摘 要 】

Incoming solar radiation is a key factor influencing solar architecture design. It determines the thermal and optical regime of the building envelope and affects the solar heat and light transfer between the indoors and outdoors. Computational analysis is an essential tool in solar architecture design. Usually, an entire year's weather data in a conventional weather file can be imported into such computational analyses. Solar irradiance data used in a conventional solar architecture design analytics are broadband (the total of UV, VIS, and NIR); however, these three components play different roles in building energy efficiency. So, analyzing individual solar components separately can be desirable. This research is to develop estimation models of the VIS and NIR components that can be captured efficiently from readily available datasets collected from the ground weather stations; such a model can then be conveniently implemented into current solar architecture design and research. We explored and tested classificationbased modeling methods for decomposing hourly broadband global horizontal solar irradiance data in conventional weather files into hourly global horizontal solar VIS and NIR components. Furthermore, a workflow of how to implement these models in solar architecture design and analysis has been developed and discussed herein. (C) 2020 Elsevier Ltd. All rights reserved.

【 授权许可】

Free