Optical characterization of historical coloured stained glasses in winter gardens and their modelling in daylight availability simulations. (1st September 2022)

• Record Type: Journal Article
• Authors: Costanzo, V.; Nocera, F.; Evola, G.; Buratti, C.; Lo Faro, A.; Marletta, L.; Domenighini, P.
• Is Part Of: Solar energy. Volume 243(2022)
• Page Start: 22
• Page End: 34
• Publication Date: 2022-09-01
• Subjects: Coloured glazing tiles -- Light transmittance spectra -- Daylight modelling -- Winter garden; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2022.07.043
• Languages: English
• ISSNs: 0038-092X

Abstract

Highlights: Experimental measurement of light transmittance spectra of several coloured glasses. Perceived indoor chromaticity is addressed through CIE colour space. Detailed daylight modelling through suitable simplifications is proposed. Excellent validation of the model is achieved (RMSE = 5.8%; R 2 = 0.89) Abstract: Coloured stained glasses, with their peculiar light transmission properties, affect the perception of indoor spaces especially in winter gardens, where a great amount of coloured glazed surfaces can be easily found. This research aims at advancing the knowledge of coloured stained glasses in architecture and how they can be modelled in daylight simulation software tools. To this aim, the paper first proposes the measurement of light transmittance spectra for a series of coloured glazing tiles installed in a richly decorated Liberty style winter garden in Catania (Southern Italy). The measurements are performed in-place with a portable spectrophotometer, which does not imply dismantling the glazed tiles. A second novelty consists in the identification of the perceived chromaticity for the transmitted daylight, by means of the CIE Colour Space. This analysis reveals that, despite around 75% of the glazed elements are coloured, indoor light is still perceived as colour-neutral because of the wise location of clear panes. Then, the calculated integral transmittance values were used to model the winter garden in the Radiance-based ClimateStudio tool, and the reliability of two different degrees of simplifications in the model was also investigated. The comparison between simulated and measured indoor illuminance values demonstrates that the proposed simplified approach allows very good reliability in daylight modelling, with an average discrepancy below 11%, while also reducing time needed for the preparation of the model. Finally, the validated model is used to assess year-round daylight availability indoors by means of climate based daylight metrics.