A concept to evaluate dynamic daylight glare

Sammanfattning: A method towards a dynamic daylight glare evaluation in the early stage of design was developed in this thesis, using existing glare metrics under multiple observer positions and gazes. The results pointed out that it is possible for a zonal characterization regarding glare sensation to be achieved with the combination of multiple existing metrics and a weighting system, named the adaptation possibility. Initially, a literature review was conducted in order to specify the limitations of the existing glare metrics and the different practices used for glare evaluation. Then the thesis continued in the definition of the proposed methodology and the case study on which the method would be tested. The proposed methodology for a dynamic and zonal glare evaluation would account for different gaze directions and observer positions, as well as, the adaptation possibility in terms of head rotation and glare duration. Multiple glare metrics were used in the method as the literature review indicated that, there is a need either for an enhanced glare metric that would meet the current limitations or for a combination of the existing ones. Thus, the existing glare metrics of DGP, DGPs and DGI were implemented in the method. For the extraction of one glare value per position, two indicators were created named the Total Point Glare (TPG) and the Space-Time Glare (STG) based on a point-in-time analysis and a period of time respectively. It is important to note that, the method included arbitrary assumptions which needed to be validated ‘in situ’. The case study was that of a hypothetical library room in Stockholm with one fully glazed façade facing South. Additionally, the performance of different shading systems, such as venetian blinds and roller shades, were tested. After a direct sun analysis at eye level, a specific grid was formed indicating the different possible positions of the observer in the space. The observer would be able to rotate his head in each position and, thus, the study would account for multiple gazes with a fixed visual centre directed towards a computer screen. The time step of the analysis was that of 1 hour. The results of the study indicate that provided glare free area under roller shades reached 100% during a small period of time. However, the glare free area coverage fluctuated a lot between extremes (from 100% to only 20% of the total space). Such rapid variations could potentially have a negative impact on visual comfort counteracting any temporary glare protection. On the contrary, the performance of the venetian blinds seemed to be more stable, even though the glare free area coverage did not reach 100% of the space. It was found that, even though the method allows the placement of the furniture/partition at an early design stage, it is still quite expensive in terms of computational time. This issue could be solved in the future when supercomputers will be available to the general public and lighting simulation may rely on more efficient engines. Since the method derived from researched rules of thumb and reasoneable hypothesis implemented in a virtual model, should be further investigated and validated with empirical studies with human subjects. Regarding the case study, it was found that, different metrics could evaluate the space differently according to the parameter that they are based on (e.g. Contrast, beam illuminance, dim space conditions), indicating the importance to implement specific metrics under the corresponding circumstances. Different shading systems could provide different choices for the glare assessment. The examined statistic shading systems indicated that the venetian blinds could partially shade the glare source and regulate the contrast, while the roller shades limited the possibilities of adaptation at each case. It was interesting to observe that the DGP indicated low glare risk in 100% of the space under roller shades, while almost 100% problematic area under the venetian blinds.