Search

To understand how daylight gives shape and life to architectural spaces, whether existing or imagined, requires quantifying its dynamism and energy. Maintaining these details presents a challenge to simulation and analysis methods that flatten data into discrete images or virtual sensor points from a point and time. To address this challenge, this thesis presents a new method for sampling and evaluating simulated daylight. It is intended as a bridge between image-based and sensor-based methods; one that can produce image-level high accuracy directional distribution data with much shorter simulation times that are closer to sensor-based methods. Instead of producing a fixed grid of points, pixels, and sun directions, an iteratively guided sampling approach structured by the discrete wavelet transformation captures the distribution of light incident on a viewpoint with a variable density. By storing the direction vector and effective solid-angle of each sampled ray, the data can be directly evaluated for any luminance-based quantity and view direction. Coupled with daylight coefficients, where the contributions from regions of the sky-dome are recorded rather than associating a value to every possible sky distribution, these methods reduce simulation time at three stages by reducing the number of samples. With fewer samples, it takes less time to render, less time to combine with sky values, and less time to evaluate potential glare sources. Through a series of simulated reference validations, this thesis shows that the proposed method can reproduce the value predicted by a high quality reference simulation in a small fraction of the time. The method achieves higher accuracy results compared to those of existing simulation methods that rely on simplifications. The method is robust across a wide range of daylight conditions and can be tuned to the desired scope of the output. The method can be applied to render a single image in high detail or used to calculate glare metrics in order to evaluate a whole building zone. This method enables a more complete characterization of daylight and visual comfort across an occupied zone; one that is less biased by proxy measurements, representative point selection, or assumptions about glare-causing pathways. Using these methods, we can more reliably understand how daylight will respond to proposed interventions, which should be useful for guiding design, regulatory standards, and improving performance.

Avoiding discomfort glare is one of the critical aspects of maintaining visual comfort, which significantly influences occupants' overall satisfaction with their indoor environment. Discomfort glare prediction models have been developed for various lighting conditions, and their equations commonly use either contrast and/or saturation terms to account for glare caused by excessive luminance contrast or excessive overall brightness, respectively. Daylight Glare Probability (DGP), which includes both terms (hybrid model) and thus accounts for both effects, is one of the more robust ones. However, the predictive performance of discomfort glare models is limited to the luminous conditions and light source types found in the dataset from which they were developed. As daylight glare models like DGP are typically derived from glare evaluations in brightly lit environments, their performance in dimmer conditions, such as those found in deep open-plan workspaces away from the window, could be limited. This thesis, therefore, aims to extend the prediction range of existing discomfort glare models to reliably cover low-light ranges as well. To determine which conditions are most critical to focus on, the luminous ranges covered in previous laboratory studies were compared to those that can be expected in open-plan offices where low-light conditions may occur. The identified range of missing lighting conditions was then used to design and conduct two new user studies in semi-controlled dim daylit conditions to supplement prior discomfort glare datasets. In the first study, participants evaluated four scenes in which the size and luminance of the glare source were varied, and in the second, participants evaluated four scenes in which it was the size and position of the glare source were varied. In parallel, the models that currently perform best in dimmer conditions were identified using an existing dataset of glare evaluations in daylit conditions: contrast-driven models were shown to outperform saturation-driven models in dim conditions, while the more versatile hybrid metrics tended to perform well overall. Hence, the hybrid format was found to be better suited for extending glare models. To create a comprehensive training dataset for this model extension, the data obtained from the two user studies were combined with experimental data from other recent studies containing high-contrast scenarios, namely with the sun disc visible through fabric shadings and low-transmittance glazing. A new best-fit discomfort glare prediction model was then proposed for the broader targeted range of luminous conditions. Based on preliminary performance checks, the new hybrid model appears to fit the participants' glare responses better in low-light conditions than the reference model, DGP, while maintaining a fit equivalent to DGP in brightly lit conditions. The new user studies also revealed an unexpected finding: the summation of multiple glare sources in glare model equations can sometimes result in an over-prediction of discomfort glare. More research will be needed to validate the newly proposed model using a larger test dataset that does not include data points used for any model development. This thesis' findings may help to advance daylight glare prediction in indoor spaces by broadening the range of validity of prediction models to dimmer conditions, thus improving the overall reliability of visual comfort appraisal in the built environment.

Recent advances in biology and neuroscience have elucidated pathways in which visible light initiates a signaling process, via the eye, responsible for activating direct and indirect biological responses. While the full impact of unnatural light patterns on human biology remains poorly understood, controlled laboratory studies, and a handful of field studies, have strongly indicated that exposure to light (and darkness) outside of the natural light-dark cycle is a factor in increased disease prevalence and general malaise related to anomalous wake and sleep behavior enabled by modern lighting. Consequently, there exists a mismatch between our evolved needs – based on the natural day-night cycle – and the needs driven by modern lifestyle enabled by technology – which typically include an over exposure to artificial light at night and restricted access to natural light during the day. In the same way that our nutritional diet has evolved with modern convenience, so has our spectral diet. From a neurobiological perspective, we can think of discrete instances of spectral composition and intensity of light received over the day as micronutrients, that when concatenated over a 24h period, compose a diet. Relating spectral diet information to neurobiological outcomes (i.e., "neurophotic" effects), is a central aim of multidisciplinary research teams. Unfortunately, collecting spectral information in the field is challenging due to constraints associated with existing encoders. The central aim of this thesis is to circumvent these constraints by approaching spectral sensing from the angle of compressed sensing and information theory by developing SpecRA: an adaptive reconstruction algorithm. In order to demonstrate the efficacy of SpecRA in practice, we develop a novel compact spectrometer \textit{Spectrace}, in tandem and test its performance against existing state-of-the-art devices. The approach taken herein allows us to maximize the amount of information recovered from cheaper, low-fidelity, filter-array sensors by pushing beyond the limits of conventional signal reconstruction though the exploitation of regularities in the natural world. As a first application of this research, we investigate the diversity of so-called spectral diets in a limited subset of commuters with the aim of elucidating structures and properties of modern spectral diets in populations with shared behaviors, in order to better anticipate their prevalence and potential impact on health and well-being. Indeed, by applying this methodology to other domains, we could also develop more efficient sensing methods of other types of chaotic signals thereby mitigating constraints and enabling greater opportunity for data collection across a diversity of physical phenomena.