Climate-based daylighting analysis in an open-plan office due to boundary conditions

https://grupo.us.es/grupotep130/es/ TEP 130 ARQUITECTURA, PATRIMONIO Y SOSTENIBILIDAD: ACÚSTICA, ILUMINACIÓN, ÓPTICA Y ENERGÍA ARCHITECTURE, HERITAGE AND SUSTAINABILITY: ACOUSTICS, LIGHTIING, OPTICS AND ENERGY Carmen M. Muñoz, Paula M Esquivias, David Moreno, Ignacio Acosta and Jaime Navarro This is an Accepted Manuscript of an article published by SAGE Journals: Lighting Research and Technology, Volume 46, 3, 2014, Pages 268-280 ISSN 1477-1535 https://doi.org/10.1177/1477153513487005 TEP 130 1 Authors’ names and affiliations: Carmen M. Muñoz a, Paula M. Esquivias a, David Moreno a, Ignacio Acosta a, Jaime Navarro a a TEP 130 Research Group, Universidad de Sevilla Corresponding Author: Carmen Mª Muñoz, TEP 130 Research Group, Universidad de Sevilla, Corresponding Author Email: [email protected] Abstract During the first stages of building design, issues like local climate or surrounding buildings are determinant. Also the configuration of every space depends on its orientation due to solar access, among others. The impact of these boundaries conditions on daylight illuminances in an open-plan office space is studied in terms of Daylight Factor (DF), Daylight Autonomy (DA) and Useful Daylight Illuminances (UDIs). Climate conditions in the location determine the characteristics of the daylight source, such as the ratio between diffuse lighting and global lighting, for a building located in a specific area. However, surrounding buildings and elements which may block daylight in the space hold more weight than the orientation of glazed façades, but are not always taken into consideration in daylighting studies. Keywords : Climate-based Daylight Modelling, Daysim, external conditions, building design, location, orientation, external obstruction 1 Introduction Considering the daylight that enters a building is fundamental, both as a means of reducing the consumption of electric lighting and in terms of its influence on visual comfort conditions, user moods, solar gains and qualitative aspects of the illuminated space [1-2]. This amount of daylight is highly dependent on building design and its boundary conditions. From the mid twentieth century onwards the general calculation method only considered fixed and stable sky conditions, which meant that the study of daylight under an Overcast Sky was only indicative of diffuse illumination conditions, avoiding sunlight, and did not take illumination variability into account. However, the recent appearance of climate-based metrics such as Daylight Autonomy and Useful Daylight Illuminances [3] provides further information on the variable behaviour of daylight throughout the year. This means it is once again possible to take daylight into account when making decisions that affect architectural design. TEP 130 2 In the initial design phases the conditions of the surrounding location are a key factor when drawing up the initial plans for the future building. In this sense, this study analyses the influence of external conditions, including the location of the building, the orientation of façades with openings, and possible frontal obstructions from the surroundings, have on the daylight conditions of an open-plan office space with two opposing glazed façades. This parametric study is the development of the Subtask “Ergonomic analysis of lighting conditions in work or study places” of the Spanish project ‘Tecnocai: efficient and smart technologies oriented to healthy and comfortable interior environments’. This paper focuses mainly on daylighting analysis of boundaries conditions of buildings; spatial aspects (e.g. window-to-wall ratio) and solar protections (e.g. overhangs) will be reported in separate publications in correspondence to the three building design phases established during the project. 2 Antecedents 2.1 Daylight availability The discussion about the limitations of Daylight Factor has been widely reported [3-6]. As Daylight Factor doesn’t include the contribution from sunlight, only diffuse light, it isn’t very useful in climates where clear and sunny skies prevail. Moreover, as it excludes sunlight, the Overcast Sky Model is widely used in correspondence to a possible real sky condition when there is only diffuse daylight. So another limitation of Daylight Factor is the implications of using the Overcast Sky Model. As its luminance is cylindrically symmetrical, the DF does not take the orientation of the glazed surface into account. However, while its simplicity is its major advantage, the sky model does not allow us to consider the variability of annual daylight in the space under study. Climate-based Daylight Metrics, developed recently, allow us to establish the annual daylight amount for a given space using hourly or subhourly calculations of the illuminance of every point on the workplane [6-7]. In order to obtain this information, it is necessary to have the distribution data for the luminance of the sky vault in the same time-step. Perez’s All-weather Sky Model [8] provides information on luminance distribution in the sky vault through irradiance measurements. In the absence of monitored data on luminance distribution, these data can be obtained from climate files [9]. Both Daylight Autonomy (DA) and Useful Daylight Illuminance (UDI) metrics are tools that make possible to process a large amount of illuminance data (up to 4380 hourly values in correspondence with the amount of daytime hours) for each point of the workplane. Both metrics analyse the TEP 130 3 illuminance data by establishing a time range for study and a suitable illumination level for carrying out visual tasks [3-7]. Daylight Autonomy (DA) is defined as the percentage of the year during which there is a minimum threshold of illumination provided only by daylight [3]. The purpose of the Useful Daylight Illuminance is to determine when daylight levels are of use to the user. It is currently measured using three metrics to express the time percentage in which the following illuminance ranges are obtained: less than 100 lx (very dark), more than 2000 lx (excessive light) and between 100 and 2000 lx as a useful range [3, 5]. These ranges may vary as more data on user preference is obtained. The key advantage of dynamic performance metrics for daylight is that they consider the amount and nature of daily and seasonal variations of daylight, as well as irregular meteorological events, for a specific location [3-7, 9-10]. 2.2 Conditions of the location: climate, orientation and obstructions of the surroundings Daylight illumination levels are dynamic and temporal dependent. The climatic local conditions of the geographical location of the building strongly influence on daylight accessibility, since the hours and amount of light available decrease the farther away we move from the equator, among others [10-11]. Climate files incorporate global, diffuse and direct irradiance measurements [12] that are used to obtain the luminance sky distribution through Perez’s All-weather Sky Model, but also global, diffuse and direct illuminance hourly data are incorporated. Criteria for daylight metrics shouldn’t be universal, e.g. the LEED criterion of ‘an average of 2% DF’ throughout the space, as daylight is climate-dependent [4]. Some dynamic daylight studies have been developed for specific locations, but it’s hard to find studies where a wide range of different climate types and latitudes [10, 13] have been taken in consideration. As there is still a discussion about continue using Daylight Factor or move to dynamic daylight metrics [1, 10], it is important to know how much representative is Daylight Factor, based on diffuse illuminance, for the local climate. In this way, Nicol et al. [13] classify the five local climates into the occurrence of three sky conditions, highlighting the difference between northern and southern locations. Today, when an architect designs a building for any location throughout the world, considering the impact of local climate would be very useful to not applying his daylighting solutions unconsciously, e.g. high frequency overcast sky location to high frequency sunny sky location and vice versa. TEP 130 4 If only the geometry of solar trajectories is taken into account, solar incidence in a space depends on the latitude of the building's location and on the orientation of the building's glazed façades. In addition to producing high levels of illumination, direct solar incidence produces a major thermal charge in the summertime. The development of urban planning studies as regards the solar access of buildings is based on the location-orientation binomial [14-15]. Regarding to daylight metrics discussion, meanwhile Daylight Factor is insensible to orientation, a study considering dynamic simulations for every orientation would highlight the difference between those metrics that is very important in places where direct daylight is relevant [5, 10]. In hot and dry climates, it is particularly important to take into consideration the orientation and obstructions in the surroundings, not only to ensure a minimum reception of daylight, but to avoid overheating of their interior spaces [2, 16-17]. In order to reduce the energy demand for HVAC and lighting systems, architectural design must be as compatible with the climate as possible [18-19]. In modern urban cities, many buildings are constructed close to each other, the obstructions of urban surroundings suppress the direct solar component during some or all hours of the day [20]. The shading effect from nearby building can be significant and reflected components can be the main sources of interior lighting [2, 21-22] if they are not shaded [15]. This should be especially taken into account with dense construction as the presence of obstructions has repercussions on the availability of daylight and causes increased energy consumption [23]. D.H.W. Li has a large experience analysing dense urban environments [21, 23], but for the authors knowledge, the effect of the surrounded obstructions are studied in terms of Daylight Factor, so a dynamic daylight analysis would be useful to consider the variability of daylighting performance not only due to building design, but rather the surrounded environment. 3 Methodology The values for DF, DA and UDI are obtained using the Radiance-based software Daysim [9-10]. The international database of climate records for energy calculations (IWEC) provided the climate data used [12]. 3.1 Description of the model The model under study is an open-plan office measuring 20 m x 12 m x 3.5 m. Openings are distributed in two opposing façades. There are 6 windows on each façade measuring 1.35 m x 0.90 m each, with a sill-height of 1 m, and giving a window-to-wall-ratio of 17.50% (10% window-tofloor-ratio). As the authors were looking for the maximum daylight availability of this space, there TEP 130 5 are no interior partitions and glazings have a very high transparency, but not completely transparent as a void, to consider some real and existing cases. Although we’ve placed work stations parallel to the glazed façades, their location is not fixed, so the whole area is treated as the task area. The workplane is found 0.80 m from the floor, with sensor points every 0.20 m (Figure 1). There is a minimum distance of 0.5 m between sensors and walls to avoid boundary effects. Figure 1. Reference open-plan office model: Southern front view, plan and perspective Table 1 shows the optical characteristics of the model: Initial conditions Glazing transparency 90% Floor reflection coefficient 30% Ceiling reflection coefficient 75% Wall reflection coefficient 55% Furniture reflection coefficient 35% Table 1. Optical characteristics of case study. 3.2 Lighting requirements of office spaces In order to calculate the dynamic daylight performance metrics it is necessary to define the time range for which they have to be calculated, as well as the minimum lighting level desired. As it is an office, the time range under study is from 8:00 h to 18:00 h. As regards the minimum lighting level, European Standard EN 12464-1 on Lighting of Workplaces [24] suggests an average maintained illuminance of 500 lx within the working area for office tasks. In terms of DF, an average maintained illuminance of 500 lx can be understood as the minimum DF value to be reached to obtain this level of indoor illumination, considering outdoor illuminance with TEP 130 6 an annual frequency of 50% in the time range specified (Figure 2). Once this minimum DF value is obtained, the percentage of sensors on the workplane exceeding this value is represented in the analysis graphs. This criterion requires the calculation of the value of diffuse horizontal illuminance with a frequency of 50%. To do so, a cumulative curve was calculated for available diffuse horizontal diurnal illuminance, using the data included in the IWEC climate archive for the locations under study and the value for diffuse exterior illuminance for 50% of the year was obtained. Given that we aimed for a minimum illuminance of 500 lx, the minimum DF value for each location was obtained. Figure 2 shows the location of the reference model as an example. Figure 2. Cumulative curve of Diffuse Horizontal Illuminance from 8:00 h to 18:00 h. Values taken from the energyplus weather climate file for Seville (Spain). In terms of DA for the calculation range considered, an average maintained illuminance of 500 lx requires the determination of the time percentage in which each point exceeds this minimum lighting value. Once DA values for each point had been established, the percentage of sensors of the workplane with a DA of 50% or above was represented in a graph. 3.3 Proposals 3.3.1 Location As seen above, the location of the building determines the availability of outdoor daylight based on local climate conditions. In the EnergyPlus weather database [12], arranged by World 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0 10000 20000 30000 40000 50000 60000 % Diffuse Illuminance frequency 8-18 h Diffuse Horizontal Exterior Illuminance SPA SEVILLE WMO 83910 37º 42' E ext = 19800 lx TEP 130 7 Meteorological Organization (WMO) region and country, IWEC climate files can be found for 34 capitals of the countries within WMO region 6 (Figure 3) corresponding to Europe and other nonEuropean countries. These cover latitudes between 31º78’ (Jerusalem, Israel) and 64º13’ (Reykjavik, Iceland). Figure 3. Limits for WMO Regions. In addition to Seville, twelve capitals of countries within WMO region 6 were chosen for this study, following Köppen's climate classification and the value of Eext with 50% frequency from 8:00 h to 18:00 h (the minimum DF value desired), thus obtaining the most representative locations. Köppen's climate classification was considered as there are some capitals with almost the same latitude but different climate and authors suppose that Köppen's classification will highlight the relationship between cloudiness and climate, despite the usual latitude classification. Table 2 shows the different locations, Köppen’s climate classification, latitude, value of diffuse horizontal illuminance with a 50% frequency during daytime and resulting DF to obtain 500 lx on the workplane. Country City WMO Station Köppen latitude Eext DF ISL REYKJAVIK 40300 Cfc 64,13 11300 4.42 NOR OSLO 14880 Dfb 59,90 13100 3.82 RUS MOSCOW 276120 Dfb 55,75 16200 3.09 POL WARSAW 123750 Cfb 52,17 16000 3.13 UKR KIEV 333450 Dfb 50,40 18500 2.70 FRA PARIS 71490 Cfb 48,73 17600 2.84 BIH BANJA LUKA 132420 Cfb 44,78 18300 2.73 TEP 130 8 ROM BUCHAREST 154200 Dfa 44,50 19200 2.60 ITA ROME 162420 Csa 41,80 20400 2.45 PRT LISBOA 85360 Csa 38,73 19000 2.63 ESP SEVILLE 83910 Csa 37,42 19800 2.53 SYR DAMASCUS 400800 Bsh 33,42 17600 2.84 ISR JERUSALEM 401840 Csa 31,78 17800 2.81 Table 2. Region 6 WMO selected locations. Cfc = Maritime Subarctic climates or Subpolar Oceanic climates; Dfb = Moist continental; Cfb = Marine west coastal; Dfa = Humid continental; Csa = Mediterranean climate; Bsh = Hot semi-arid climates. 3.3.2 Orientation If local climate conditions have an influence on the availability of outdoor daylight, building orientation is closely linked to solar geometry. The office spaces studied were designed using bilateral daylight, that is to say, with openings on two opposing façades. In the case under study, light accesses through the north and south façades. Several orientations were suggested for this study, with rotations of 45º, 90º and 135º in relation to the model of reference. 3.3.3 External obstructions In previous sections, models were considered in isolated settings, with no obstructions preventing solar access or blocking off the view of part of the sky vault. Offices tend to be located in urban spaces that are occupied to a greater or lesser density, and therefore, depending on the urban planning conditions, the evolution of the city, the elevation (floor height) of the offices, etc., they have elements in their surroundings that obstruct and limit access of daylight. Although these obstructions can be found in different areas of the outer hemisphere of the façade plane, the greatest obstruction to daylight, involving the greatest reduction of availability, is frontal obstruction (parallel to the façade plane). Just as the solar position is expressed in polar coordinates, the elevation of the obstruction in relation to the windowsill is also represented using polar coordinates. It is possible to examine the different combinations of possible building height and separation by defining different angles. Given that the maximum solar elevation for latitude 37º (Seville, Spain) is 72’8º, four angles between 0º (no obstruction) and 75º have been used to represent the possible panorama. TEP 130 15 southern regions. The lowest difference is up to 13%, and it could be considered to be within the uncertainty for daylight simulation, but this value is a singular case. It is observed that in European regions below latitude 50º the difference is around 50% and this is up to 60% in regions between 35º and 30º. The results highlight that global illuminance should always be taken into account in daylighting studies. In any case, if the decision is made to calculate DF given its simplicity, a specific coefficient should be applied to expand results. As the base case model has a high window-to-wall ratio and bilateral daylighting, the orientation does not play a relevant role, in difference with studies which orientation has been studied in one side daylighting spaces [19, 21]. Although it can be said that one-side-lighting spaces are more common than bilateral spaces, having two opposite glazed façades as lighting source for a common space is not as singular as it is thought. A comparison with equivalent side-lighted spaces should be made to find if there is a direct relationship, and it is possible to add different effects from a simple to a more complex situation. The configuration of this space, its high window-to-wall ratio, determines in a certain way the percentage of the workplane permanently within a UDI range. It can be seen how if this space might be built in any location, there is always around 12.60% of the workplane above 2000 lux; or if the orientation is changed, there is always around 15% of the workplane below 100 lux. These values are showing intrinsic characteristics of the daylight performance for this type of space. Once these characteristics are known, some architectural design decisions can be made to reduce the percentage of the workplane above 2000 lux and below 100 lux. Also it has to be considered that as illuminances above 2000 lux represent a risk of glare, and the portion of the workplane within glare risk seems to be almost constant, there are some factors involved in discomfort glare that have not been considered in this study. Compared to the proposed reference model, and due to its geometrical characteristics, the built-up surroundings are the factor with most influence on daylight conditions. The presence of obstacles reduces till 80% the percentage of the workplane achieving a Daylight Autonomy of 500 lux, and up to 40% the ratio above DF criterion. An angle of obstruction of 45º combines the maximum value of UDI with a value of 50% of the workplane achieving DA500, and that could represent a threshold affecting to urban planning. Curiously this angle is linked with rule of thumbs about skylines used for ages to sure a solar access [22]. TEP 130 16 Looking the results it is appreciated that the obstacles can be as relevant in daylight conditions as window size. However, UDI100-2000 seems to be insensitive to these changes, probably due to the wide range of illuminances that it covers. Although DAYSIM incorporates the commonly UDI range, the limits are based on human factor considerations, so these limits could change as further studies on human daylight satisfaction are carried out. However, some intermediate limits are suggested by Nabil and Mardaljevic 5 that could give more accuracy and sensitive performance to the presence of obstacles, and to any variation of any parameter involved in daylighting in general. Funding Acknowledgments This paper presents part of the results obtained during the development of the Subtask “Ergonomic analysis of lighting conditions in work or study places” for the TECNOCAI research project (Efficient and smart technologies oriented to healthy and comfortable interior environments), contracted for Acciona Infraestructuras, S.A. [C.P.1168, CGT 0010] within the Industrial R&D Initiative CENIT, funded by the Ministry of Science and Innovation. This Subtask was carried out in the Institute of Architecture and Building Science (IUACC) of the University of Seville by members of the TEP-130 research group. The authors are grateful for the technical and financial support provided and wish to thank all those involved for their valuable collaboration. References 1. Andersen M, Mardaljevic J and Lockley SW. A framework for predicting the non-visual effects of daylight - Part I: photobiology-based model. Light. Res. Technol. 2012; 44: 37-53. 2. Li DHW. A review of daylight illuminance determinations and energy implications. Appl. Energy 2010; 87: 2109-2118. 3. Reinhart CF, Mardaljevic J and Rogers, Z. Dynamic daylight performance metrics for sustainable building design. Leukos 2006; 3: 1-25. 4. Mardaljevic J, Hesching L and Lee E. Daylighting metrics and energy savings. Light. Res. Technol. 2009; 41: 261-283. 5. Nabil A and Mardaljevic J. Useful daylight illuminances: A replacement for daylight factors. Energy Build. 2006; 38: 905-913.6. Mardaljevic J. Simulation of annual daylighting profiles for internal illuminance. Light. Res. Technol. 2000; 32: 111-118. 7. Reinhart CF and Wienold J. The daylighting dashboard-A simulation - based design analysis for daylit spaces. Build. Environ. 2011; 46: 386-396. TEP 130 17 8. CIE S 011:2003. Spatial distribution of Daylight - CIE Standard General Sky 9. Reinhart CF. Tutorial on the Use of Daysim Simulations for Sustainable Design, http://daysim.com/pub/Daysim3.0.Tutorial.pdf. (2010, accessed 10 September 2010) 10. Kleindienst S, Bodart M and Andersen M. Graphical representation of climate-based daylight performance to support architectural design. Leukos 2008; 5: 39-61. 11. Li DHW, Lam TNT and Chu VWC. Relationship between the total solar radiation on tilted surfaces and the sunshine hours in Hong Kong. Sol. Energy 2008; 82: 1220-1228. 12. EEREU.S. Department of Energy. EnergyPlus Energy Simulation Software: Weather Data, http://apps1.eere.energy.gov/buildings/energyplus/cfm/weather_data2.cfm (accesed 30 september 2010) 13. Nicol F, Wilson M and Chiancarella C. Using field measurements of desktop illuminance in European offices to investigate its dependence on outdoor conditions and its effect on occupant satisfaction, and the use of lights and blinds. Energy Build. 2006; 38: 802-813. 14. Van Esch MME, Looman RHJ and Bruin-Hordijk GJ. The effects of urban and building design parameters on solar access to the urban canyon and the potential for direct passive solar heating strategies. Energy Build. 2012; 47: 189-200. 15. Mesa NA, Corica L and Pattini A. Evaluation of the potential of natural light to illuminate buildings in dense urban environment. a study in Mendoza, Argentina. Renewable Energy 2011; 36: 2414-2423. 16. Manioglu G and Yilmaz Z. Energy efficient design strategies in the hor dry area of Turkey. Build. Environ. 2008; 43: 1301-1309. 17. Belakehal A, Tabet Aoul K and Bennadji A. Sunlighting and daylighting strategies in the traditional urban spaces and buildings of the hot arid regions. Renewable Energy 2004; 29: 687702. 18. Faizi F, Noorani M, Ghaedi A and Mahdavinejad M. Design an optimum pattern of orientation in residential complexes by analyzing the level of energy consumption. (Case study: Maskan Mehr Complexes, Tehran, Iran). Procedia Engineering 2011; 21: 1179-1187. 19. Morrissey J, Moore T and Horne RE. Affordable passive solar design in a temperate climate: An experiment in residential building orientation. Renewable Energy 2011; 36: 568-577. TEP 130 18 20. Iversen A, Nielsen TR and Svendsen SH. Illuminance Level in the Urban Fabric and in the Room. Indoor Built Environ. 2011; 20: 456-463. 21. Li DHW, Wong SL, Tsang CL and Cheung GHW. A study of the daylighting performance and energy use in heavily obstructed residential buildings via computer simulation techniques. Energy Build. 2006; 38: 1343-1348. 22. Lukman N, Hayman S and Hyde R. Rule of thumb for daylighting of rooms with external obstructions. Architectural Science Review . 2009; 52: 151-160. 23. Li DHW, Cheung GHW, Cheung KL and Lam JC. Simple method for determining daylight illuminance in a heavy obstructed environment. Build. Environ. 2009; 44: 1074-1080. 24. EN 12464-1: Lighting of Workplaces: Indoor workplaces. 2001.