An empirical validation of daylighting tools: Assessing radiance parameters and simulation settings in Ladybug and Honeybee against field measurements. (1st September 2020)

• Record Type: Journal Article
• Authors: Kharvari, Farzam
• Is Part Of: Solar energy. Volume 207(2020)
• Page Start: 1021
• Page End: 1036
• Publication Date: 2020-09-01
• Subjects: Validation -- Daylighting software -- Radiance -- Ladybug -- Honeybee -- Field measurements; Solar energy -- Periodicals; Solar engines -- Periodicals; 621.47
• DOI: 10.1016/j.solener.2020.07.054
• Languages: English
• ISSNs: 0038-092X

Abstract

Graphical abstract: Highlights: Parameters other than the suggested ones do not improve the simulation results. Increasing the ambient bounces to more than 8 does not affect the accuracy of results. The effect of disabling ambient accuracy (-aa = 0) in Radiance is nominal. ±0.1 change in reflectance factors significantly affects the accuracy. Decreasing ambient bounces to less than 6 creates biased results. Abstract: Most of the new studies in Radiance rely on industry-accepted values for setting Radiance parameters. Although Radiance is validated against the analytical test cases of CIE 171:2006, there is no scientific basis for these settings. Therefore, the aim of this study is to assess the effect of Radiance parameters and model-related components on the accuracy of simulation results in Radiance by comparing field measurements and simulation results. This study uses a point by point illuminance comparison of simulation results and field measurements under an overcast sky with measured illuminance in HoneybeePlus version 0.0.04 which uses Radiance version 5.1. The accuracy is rechecked with three other measured illuminance levels at different times under the same sky. Furthermore, this study assesses the effect of simulation parameters and model settings on the accuracy of simulation results. Results show that the recommended Radiance settings for having accurate results are accurate enough with a bias below 15%. Setting ambient bounces to more than 8 and disabling interpolation by setting ambient accuracy to 0 does not significantly affect the simulation results. By disabling interpolation, Monte Carlo ray tracing provides accurate results for simple geometries. Moreover, ±0.1 difference in reflectance factors has a significant effect on the accuracy of results. This study makes a scientific basis for setting Radiance parameters and measuring reflectance factors in future studies. The outcome of this paper creates a validated model for future studies on shaders, light shelves, and related designable components that rely on daylight in general.