Unobstructed view metric (Graphic by author). 

Context in source publication

Context 1

... necessity to utilise natural light in the interior of buildings is strongly supported by numerous health benefits, energy saving and environmental factors [1]. Traditional design principals guide the vertical and horizontal orientation of shading devices for buildings with east, west and equatorial façades [2]. However there is little research that contributes to the knowledge of shading systems suited to contemporary architecture, characterised by double curved surfaces. A research through design process and state of the art software is utilised to model and simulate the performance of a daylight responsive skin suited to hot climate double curved façades. The system being optimised is an operable envelope, which regulates the amount of light transmitted from the exterior to the interior of the building over the course of the day. Findings from the research conclusively support the performance of the two resultant designs and also give an insight to the characteristics associated with shading devices suited to double curved façades. The methodology developed, establishes unique metrics for performance measurements. The process consists of an evidence-based approach to design evolution, where each iterative design decision responds to the findings from the former measured results. The skin system is parametrically modelled to simulate the automated response the independently moving panels have to the changing daylight conditions over the course of a day. An initial design is tested through computational simulations, to evaluate its performance and to gain an understanding of where improvement can be made. Interior and Workplace Lighting standards state, commercial lighting levels should be 160 Lux for background environments and 320 Lux for task lighting [3]. Therefore the aim of the skin system is to regulate the internal mean natural light levels at 240 Lux over the course of the day, which is the midpoint between these two lighting levels. Further to this, the system being designed also needs to be suitable for a surface that has a double curvature so that it is a feasible resolution to the issues associated with shading organic building façades. An operable panel system has been devised as the most appropriate design response due to its ability to shade whilst maintaining an external view. Five iterations of the design are generated sequentially through performance analysis, via the computational simulation. Both summer and winter solstices are selected as simulation days to test the full range in sun altitudes. Analyses are performed at seven hourly increments between 9am and 3pm to evaluate the system’s performance over the course of the day. In total, five design iterations are tested, resulting in 210 calculations. This number is compounded by the kinetic function of the individually responsive panels. The complexities in modelling this system are most clearly represented through quantifying the amount of panel angle variations. For example the third, fourth and fifth skin iterations consist of modules containing four individually moving panels. As these modules are arrayed 300 times over the testing surface, the complete skin consists of 1,200 individually responsive panels. To simulate a design at the 7 hourly increments over the two solstices, the system requires the control of over 16,000 individual panel angle variations. Due to this degree of complexity, an approach that automates the movement of the skin’s response to the sun’s position is required. Parametric software is utilised to produce an algorithm that controls the response each individual panel has to the sun’s changing position in the sky, over the course of the day. As well as controlling this movement, the algorithm also simultaneously measures the performance of the design. The package used to create these parameters is Grasshopper, which is a graphical algorithm editor, that is tightly integrated with McNeel’s Rhinoceros 3D. The algorithm gives the designer a highly refined degree of control over the system. The 1,200 individual panel’s sensitivity to sunlight can be fine-tuned with the adjustment of a numerical value, to achieve the desired internal lighting levels. This tuning process occurs through utilising Radiance illuminance simulations as a direct feedback loop, in order to achieve the desired interior mean Lux of 240. The algorithm developed for this research utilises a node that moves along a curve in the modelling environment, which has an associative relationship to the rotation of the skin’s panels. As the curve in space is defined as a sun path, with an equatorial orientation, the node traveling this path simulates the movement of the sun. Additionally Geco components for Grasshopper are utilised to create a live link between McNeel’s Rhinoceros 3D and Autodesk’s Ecotect to define the sun path. These components are supplementary plugins for Grasshopper, developed by Ursula Frick and Thomas Grabner; the directors of UTO at the University of Innsbruck [4]. To measure the performance of the daylight responsive system, three metrics have been defined. Existing methods sourced from the literature have been adapted specifically to suit the performance measurement of this daylight responsive system. A pavilion is modelled as a testing space to simulate how the five module options perform, arrayed over the double curved façade. Specific material surface reflectivity is assigned to the various elements of the pavilion (Table 1). Studies by Hartig et al. show that having access to an external view has been an associative factor of good health and well-being [5]. Therefore the ability for the skin to achieve a high degree of external view is a vitally important aspect in understanding the design’s performance. To derive the most appropriate method for determining the amount of external views, an occupant would receive, research on window to wall ratios by Xing Su et al. is investigated [6]. Due to the double curvature of the pavilion’s façade, a flattening process is required to measure the area of external unobstructed view. Planar area calculations are performed on the elevated projections of the skin, which are represented as a percentage that correlates to the amount of unobstructed view received by the occupant (Figure 1). It is commonly known that in hot climates it is desirable to avoid direct sunlight entering an internal space as it increases the building’s heat gain, meaning it is more reliant on cooling systems [7]. Therefore the ability for the skin to shade the space is a vitally important aspect in understanding the design’s performance. In order to measure this performance, area calculations are again conducted utilising an additional projection method. To simulate the shadows generated from the skin, an outline of its geometry is projected onto the internal floor at a vector direction based on the sun’s rays at the given testing time. Following this, area calculations of the floor and projected geometry are performed to generate a percentage of the amount of floor area illuminated by direct sunlight (Figure 2). Glare is a subjective condition that is experienced by a high contrast in light, which can cause irritation, fatigue and headaches [8]. Therefore the skin system’s ability to reduce the glare perceived by building occupants is also of vital importance. Jan Weinold and Jens Christoffersen at the Fraunhofer Institute for Solar Energy Systems devised a new method for analysing glare, known as Daylight Glare Probability (DGP) [9]. DGP is able to gather data on the complexities of glare as the method computes the directional properties associated with light. To analyse the DGP a virtual camera with a 180 degree fish eye lens is positioned inside the pavilion 1700mm from the ground, to replicate an occupants perspective (Figure 3). From this camera location a simulation takes place, using Evalglare, a programme that uses the DGP algorithm to compute the probability of glare. Evalglare also produces a High Dynamic Range image, which serves as an instrument to visually determine the location of the glare source (Figure 4 ). The results from DGP are represented as a percentage, which can be categorised using the index depicted in the table below (Table 2). The following section of the paper displays elevations of the five iterative skin types. Graphs illustrate the performance results of each of the iterations according to the three performance metrics. The reasons for each of the sequential design iterations in the optimisations process are also discussed. The following five skin systems are successful in maintaining a mean interior Lux of 240 over the course of the day, for both summer and winter ...