Daylight luminous efficacy: An overview. (1st November 2021)

• Record Type: Journal Article
• Authors: Aghimien, Emmanuel I.; Li, Danny H.W.; Chen, Wenqiang; Tsang, Ernest K.W.
• Is Part Of: Solar energy. Volume 228(2021)
• Page Start: 706
• Page End: 724
• Publication Date: 2021-11-01
• Subjects: Daylighting -- Luminous efficacy -- Energy savings -- CIE. Standard skies -- Machine learning; RMSE Root mean square error -- MBE Mean bias error -- R Correlation coefficient -- R2 Coefficient of determination -- RE Relative error -- T-stat T statistics -- MAD Mean absolute deviation -- MPD Mean percentage deviation -- RMSRE Root Mean Squared Relative Errors; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2021.05.018
• Languages: English
• ISSNs: 0038-092X

Abstract

Abstract: The building sector accounts for 40% of global energy consumption, and artificial lighting is a contributory factor. An approach to resolving lighting issues is daylighting. For daylight to deliver energy-efficient buildings, illuminance data is required. However, few measuring stations collect this data. A preferred data generation approach is luminous efficacy. This study systematically reviewed existing studies on luminous efficacy. The Perez brightness and clearness, sky clearness index and sky ratio were discovered to be crucial in sky characterization. Although the 15 CIE skies globally represent the whole skies, it has low adoption. Solar altitude was found to be a key input parameter in luminous efficacy. Also, constant value luminous efficacies showed good predictive abilities despite being a secondary alternative. Furthermore, most studies adopted empirical models for predictions. However, the machine learning approach can be used due to its accuracy. Information on model evaluation metrics, comparative models and modeling techniques were also identified. The study further presented research gaps for future research.