The impact of a combined dynamic shading system on the thermal performance of building perimeter zones.
With the development of energy-conscious building design techniques and solar architecture in particular, the study and use of dynamic facade elements is gaining momentum. They not only act as daylight harvesting systems, but also as solar heat gain regulators. Apart from maximizing daylight utilization, they also dynamically control solar heat gains, allowing solar radiation into the room during the winter months and keeping part of the solar heat gain out in the summer. External shading is more effective than internal, since solar gains are efficiently rejected (Raeissi and Taheri, 1997). Moreover, overhangs reduce overheating and cooling load during summer months at a time of the day when electricity is most expensive. During winter, they allow some solar heat which can be desirable in cold climates (Dubios, 1998, 1999). A study to assess these performance parameters for an overhang shading a private office with one exterior wall (window) was carried out to help designers select and optimize such devices (Rao and Tzempelikos, 2010). It was concluded that the overhang is an effective external shading device, but needs to be used in combination with other shading device(s) for best results as far as visual comfort of occupants is concerned. External overhangs provide good solar protection as they are external shading devices and are effective in keeping out unwanted solar radiation in the summer, while they allow some solar heat in the winter. Light shelves enhance illuminance distribution in the room and facilitate deeper daylight penetration. However, when compared to no shading, they reduce the overall illuminance in rooms and may require a custom design depending on room configuration, ceiling characteristics, ceiling geometry, etc. (Claros and Soler, 2001, Aghemo et al., 2008). Nonetheless, research shows that active control of light shelves results in higher potential for energy savings even for simple room configurations. Automated roller shades either cover the entire window or leave the window completely open. Partial opening of shades has to be researched further as it needs to be carefully designed to avoid glare.
For this reason, a new combined dynamic shading system consisting of three different shading devices is proposed. The system consists of (i) a fixed exterior reflective overhang, (ii) a tiltable fabric internal light shelf on the top window part (auxiliary aperture) and (iii) automated bottom-up roller shades on the bottom window part (view/main aperture). In order to justify its use as a shading system, detailed transient thermal analysis models with dynamic view factors implementations and non-linear heat transfer coefficients were implement and solved explicitly using the finite difference method to compare its performance with that of a standard advanced shading system-automated internal roller shades.
APPROACH AND METHODOLOGY
The objective of this study is to quantify and assess typical performance indicators (annual cooling and heating energy, peak demand, etc) that help to determine the effectiveness of a shading device and compare these values for a variety of shading types, shade fabrics and glazing properties. Two specific shading types were modeled and compared; (i) automated internal roller shades (Base Case Shading) and (ii) a combined shading system consisting of an exterior reflective overhang, a tiltable fabric internal light shelf and automated bottom-up roller shades (Advanced Shading System). For both systems, the shades were controlled at each time step to block direct light. In this way, bright sun spots were eliminated reducing the possibility of glare. An effort was made to create simulation models that are flexible, efficient and reliable, both during program implementation and at run time. The analysis simulates a typical south-facing classroom located in Chicago, IL that measures 10m x 10m (32ft x 32 ft) with a 3m (10 ft) ceiling. It must be noted that when the window is shaded by the automated roller shades only, depending on the weather conditions, either the entire window is covered by them or the window remains completely non-shaded. However, in case of the advanced system, the large window is divided into two 'sections'; the View Aperture (bottom window) and the Auxiliary Aperture (top window). The position of the light shelf and the overhang determine their exact areas. The curtain wall, part of the building's south oriented facade, was made up of a lower spandrel section (insulation sandwiched between glass panes) and large glazing occupying about 72% of the curtain wall area. The glazing types used are presented in Table 1.
Table 1. Glazing properties used in the simulation Property Glazing 1 Glazing 2 Glazing 3 Inner Pane Clear Low-el Low-e2 Coated (IP) Uncoated Coated Glass Glass Glass U-value 3.55 1.92 1.74 W/[m.sup.2] W/[m.sup.2] W/[m.sup.2]K K K (0.625 Btu / (0.338 Btu / (0.306 Btu / [ft..sup.2] [ft..sup.2] [ft..sup.2] hr [degrees] hr [degrees] hr [degrees] F) F) F) Solar 61% 61% 47% Transmittance (Assembly) Absorptance 5% 17% 14% (IP) Visual 81% 78% 77% Transmittance
For the base case system, the roller shades operates in two modes, on and off (retracted). Between sunrise and sunset, whenever the facade receives direct sunlight the shade closes completely, thus protecting the occupants from sunlight. In order to maximize daylight utilization, on overcast days or at time steps when there is no direct light incident on the window, the roller shades withdraw completely allowing all the diffuse sunlight to enter the classroom. An obvious drawback with this control strategy is that occupants' view to the outside is greatly affected. Practically, a manual override may be provided to tackle this issue. However, this would have a negative impact on overall energy efficiency and may also increase occupant thermal and visual discomfort. In case of the advanced studied system, the exterior reflective overhang is fixed and remains in the horizontal position at all times. Since the room being modeled is surrounded by identical spaces, using profile angle calculations, it was concluded that the overhang extends far enough on either sides of the room, and therefore, the effect of the solar azimuth on its shading pattern can be neglected. The remaining portion of the view aperture (sill upwards) is shaded by internal automated bottom-up roller shades (Figure 1); this reduces the possibility of glare. So, at any given time during the day, the overhang, together with the roller shades provides complete shading of the view aperture allowing only diffuse natural light to enter the room. Between sunset and sunrise, the roller shades cover the entire bottom window.
[FIGURE 1 OMITTED]
The tilt angle of the internal light shelf during the day depends on the altitude angle of the sun. The light shelf extends all along the window and therefore the effect of a small patch of sunlit area obtained in the top portions of the west and east vertical walls was neglected. These were only observed close to sunrise and sunset, i.e. for higher solar azimuth angles. Figure 2 and Table 2 shows the variation of light shelf tilt angle, k, as a function of the solar altitude angle).
[FIGURE 2 OMITTED]
Table 2. Light Shelf Tilt Angle as a function of solar altitude Altitude Angle Light Shelf Tilt Angle a > 45 [degrees] [zeta] = 0 [degrees] 45 [degrees] >= a > 30 [degrees] [zeta] = 30 [degrees] 30 [degrees] >= a > 15 [degrees] [zeta] = 60 [degrees] 15 [degrees] > a [zeta] = 90 [degrees]
Solar heat gains are greatly affected by facade parameters such as shading type (external or internal), glazing properties, etc. In case of automated roller shades, the entire facade outer surface was exposed to all the direct solar radiation (if applicable) along with the diffuse radiation from the sky and radiation reflected off the ground. Considering the sky to be an anisotropic source of diffuse sunlight, the all-weather sky model (Perez et al., 1993) was used to calculate diffuse solar radiation incident on the facade, including sky and ground diffuse components. Since double pane windows were modeled, the radiation incident on each of the glass panes was determined using the angle-dependent properties from WINDOW6[R]. A scenario for no shading was also analyzed.
[[alpha].sub.max] = [tan.sup.-1] (1 - sin [kappa]/cos [kappa]) (1)
For the advanced shading system, the irradiation on each outer window portion resulting from the (Figure 3) was computed as follows:
[FIGURE 3 OMITTED]
[I.sub.ovg] = [DNI * sin [alpha] + DHI] * [[rho].sub.ovg] * [F.sub.twind, ovg] (2)
[I.sub.aux] = [I.sub.beam] + [I.sub.sky, diff] + [I.sub.gnd] * ([[zeta].sub.1]/90 [degrees]) + [I.sub.ovg] (3)
[I.sub.shaded] = [I.sub.gnd] + [I.sub.sky, diff] * ([[zeta].sub.2]/90 [degrees]) (4)
[I.sub.unshaded] = [I.sub.beam] + [I.sub.shaded] (5)
where, DNI is the direct normal irradiance, DHI is the diffuse horizontal irradiance, [I.sub.beam] is the beam solar radiation incident on a facade, [I.sub.sky,diff] is the incident sky diffuse solar radiation, .[I.sub.gnd] is the irradiation reflected off the ground, [I.sub.ovg] is the radiation incident on the external overhang, [[rho].sub.ovg] is the reflectance of the overhang and [F.sub.twind,ovg] is the view factor from the top window to the overhang. The angles [[zeta].sub.1] and [[zeta].sub.2] are compensation factors that account for the fact that the top window is unable to see the entire ground surface and similarly the bottom window does not receive all the diffuse radiation from the sky.
Internal heat gains from sources within the room have to be accounted for. Scheduled heat gains (9am-5pm) were modeled and classified into three broad categories. The intensity of these internal heat gains were set based on ASHRAE (2009) recommended standard values: occupants (100W or 340 Btu/hr); lighting (10W/[m.sup.2] or 3.4 Btu/hr [ft.sup.2]); and equipment (5W/[m.sup.2] 1.7 Btu/hr [ft.sup.2]).
A thermal network was developed for each analysis (base case shading and the advanced case). Thermally massive elements (ceiling and floor) formed three separate nodes correspond to the two outer surfaces and the thermal mass. Assuming identical properties on either side of the roller shades and no heat transfer between the two surfaces, they were modeled as a single node. In the advanced case, since the light shelf reflects a portion of direct light on to the ceiling, the exact position is traced using one bounce ray tracing and the incident source on the ceiling is accordingly adjusted. Assuming well mixed room air, the air mass is considered to be one node, positioned in the center of the room. At each time step, surface temperatures were computed for a Typical Meteorological Year and the instantaneous sensible heating/cooling demands were generated. The heating/cooling demand at each time step was then used to determine peak and annual heating/cooling demand for the space. With the enclosure surface temperatures known, for a given infiltration rate and internal heat gain, the sensible heating/cooling demand (Q) can be found using the following formula.
Q = [n.summation over(i=1)] ([T.sub.i] - [T.sub.air]) * [h.sub.i] + ([T.sub.o] - [T.sub.air]) * [U.sub.oa] + [Q.sub.ihg] (6)
where, at each time step, [T.sub.i] is the temperature of the [i.sup.th] room surface, [h.sub.ia] is the heat transfer coefficient between the room air and the [i.sup.th] internal surface, [T.sub.o] is the outside air temperature, [U.sub.oa] is the infiltration conductance and [Q.sub.ihg] is the internal heat gain. Solar heat gains and thermal mass effects are taken into account when calculating surface temperatures.
The model was verified for base case shading against predicted values from two software packages, Parasol (Hellstrom et al., 2007) and IESVE (www.iesve.com). An analysis was run for four different cases, i.e. (i) non-shaded windows with Glazing I, (ii) Glazing I with automated internal roller shades, (iii) Glazing II with automated internal roller shades and (iv) Glazing III with automated internal roller shades. Figure 4 compares peak heating and cooling loads; the numbers show the absolute difference in predicted results with reference to the thermal model for all four cases. A variation of about 8% was observed in predicting peak cooling load, while the annual cooling demand differed by approximately 4%. When clear glass was replaced by Glazing II, variation in annual cooling demand for the classroom was a mere 0.5%, but the difference in peak heating load was 11%. The discrepancies in predicted annual and peak heating/cooling demands can be attributed to a multitude of factors such as variations in weather data, difference in modeling techniques used to assess glazing components, dissimilarities in the modeling of thermal mass in the space, etc.
[FIGURE 4 OMITTED]
A parametric analysis was carried out to study the impact of shading type, overhang length and shade material properties on the annual and peak energy demand of the classroom. Numerous fabrics were evaluated, with five different material transmittances; 1%, 4%, 7%, 10% and 13%, each with three possible reflectance values; 40%, 60% and 80%. Automated internal roller shades along with three different configurations for the advanced case for overhang lengths 0.5m, 0.7m and 1m were studied. For each of the cases, the annual and peak heating and cooling demands were computed.
For the base case, the results showed that higher shade absorptance (and lower reflectance) results in lesser rejection of solar radiation incident on the shades, consequently increasing shade temperatures. Therefore, higher annual and peak cooling demands were observed. When compared to a standard material with 36% absorptance, a 59% absorbing shade results in up to 33% higher annual cooling energy demand, with peak cooling load increasing from 5.36 kW (19.2 kBtu/hr) to 6.88 kW (23.5 kBtu/hr), a 28% increase.
For a typical fabric material with 4% transmittance and 60% reflectance, the advanced system can reduce annual cooling demand by about 43% when compared to the automated roller shades, with a 51% reduction in peak cooling demand, decreasing from 5.36 kW (19.2 kBtu/hr) to 2.64 kW (9 kBtu/hr). The benefits of the advanced shading system are more prominent when a material with higher absorptance is selected. Moreover, it was observed that for a large 1m overhang, the effect of shade properties diminished and the shelf fabric properties are dominant.
The effect of overhang length on space heating and cooling demand was also investigated. For a material with 60% reflectance and 13% transmittance, 1m overhangs cuts down annual cooling demand by 5% when compared to a 0.5m overhang. The decrease in peak cooling load is about 12% for a negligible (0.2%) increase in peak heating demand. The annual heating demand, however, goes up by about 2% from 1810 W (6176 Btu/hr) to 1848 W (6305 Btu/hr). The effect of overhang length becomes less significant for low absorptance shade fabrics. In order to further study the impact of shade and shelf material on annual and peak thermal loads, three specific fabrics were studied (Table 3).
Table 3. Fabric types and properties studied Transmittance Reflectance Absorptance Fabric 1 1% 40% 59% Fabric 2 4% 60% 36% Fabric 3 13% 80% 7%
It may be inferred that for the base case, as shade absorptance decreases, annual cooling energy goes down, for instance, it decreases from 7224 kWh for fabric 1 (59% absorptance) to 3712 kWh for fabric 3 (7% absorptance). The variation in instantaneous cooling demand for a classroom equipped with automated internal roller shades for three shade fabrics is plotted in Figure 5. It is clear that cooling demand is directly related to shade temperature. Higher cooling demand is shown for September 24th because of higher shade temperatures (high intensity of solar radiation).
[FIGURE 5 OMITTED]
Figure 6 presents results for the base case and the advanced system with the three selected fabrics. It was also found that larger overhangs reduce the area of the roller shades making the shelf reflectance an important factor in determining cooling demands. This is due to higher ceiling temperatures resulting from the increased reflection of solar radiation off the shelf, since both the shelf and the shades are made of the same material. This effect is clearly shown by the fact that overhang length does not greatly affect annual cooling energy use for a low absorptance shade material.
[FIGURE 6 OMITTED]
In case of the advanced system, as discussed in the previous section, shelf reflectance also plays an important role. Therefore, for a small overhang, peak cooling demand decreases from 3.419 kW (11.67 kBtu/hr) for fabric 1 to 2.852 kW (9.73 kBtu/hr) for fabric 3 as a consequence of decreasing material absorptance. Although the effect of fabric absorptance is clear for a small overhang, in case of a 1m overhang the trend isn't necessarily the same. Minimum peak cooling demand is obtained for fabric 2 which is 2.646 kW (9.03 kBtu/hr). Hence, it was concluded that in case of high reflectance shades, smaller overhangs result in larger shade areas, consequently rejecting a larger portion of solar radiation, and reducing peak cooling demand. However, in case of fabric 1, larger overhang reduces shade area, resulting in lower shade temperatures and consequently decreasing peak cooling demand.
Figure 7 shows the variation of bottom-up shade temperature of the view aperture for three representative days in the summer, September 22nd, 23rd and 24th for a 0.5m (1.6 ft) overhang. Temperature of the roller shades is important not only because it greatly affects thermal loads in the space, but also because it has an impact on thermal comfort of occupants. Therefore, excessively high shade temperatures need to be avoided. Shade temperatures are primarily affected by the window area they cover. At night (before 8 am), shade temperatures for all the three fabrics are almost the same. However, after 8 am, when they cover a large portion of the window and begin receiving direct sunlight, their temperature increases. Maximum shade temperatures are observed around 2 pm when, the shades occupy about 54% of the window area while receiving a large amount of solar radiation from the afternoon sun. For a high reflectance shade with low absorptance (fabric 3), the maximum temperature on September 23rd was about 32 [degrees] C (89.6 [degrees] F). On the same day, the temperature of fabric 2 was 38 [degrees] C (100.4 [degrees] F), while that of fabric 3 was 33% higher at over 42 [degrees] C (107.6 [degrees] F).
[FIGURE 7 OMITTED]
This study presents a comprehensive simulation study about the impact of a combined fixed exterior-movable interior overhang/light-shelf system on the thermal performance of buildings, in conjunction with controlled roller shades in the view aperture (main window), as well as results for the performance of standard automated roller shades. The results showed that selection of shade fabric for the base case (automated roller shades) is a key design decision that needs to be made keeping in mind important design parameters such as thermal loads and the thermal comfort of occupants. Annual and peak cooling demand can be greatly affected by the shade material properties. High reflectance shade fabrics do not overheat since they reject a large portion of the incident solar radiation, reducing cooling loads in the summer for a nominal increase of heating demand in the winter months.
In general, the combined shading system is more effective in reducing cooling demand when compared to the base case. Its effectiveness decreases as a high reflectance shade and shelf fabric is chosen, but the effect needs to be assessed on a case by case basis to know if the reflectance of the shade is more dominant than the effects of the light shelf made from the same fabric material. However, for standard shade fabrics the energy savings are considerable and capital costs can be decreased since smaller capacity HVAC equipment could be installed. Finally, various factors such as structural limitations should be considered when, longer overhangs (up to 1 m or 3 ft) are selected for better solar protection.
Aghemo, C., Pellegrino, A. and LoVerso, V. 2007. The approach to daylighting by scale models and sun and sky simulators: A case for different shading systems. Building and Environment, 43(5): 917-927.
American Society of Heating Refrigeration and Air-Conditioning. 2009. ASHRAE Handbook-Fundamentals. Atlanta, USA.
Claros, T., Soler, A. 2001. Indoor Daylight Climate-Comparison between Light Shelves and Overhang Performances in Madrid for Hours with Unit Sunshine Fraction and Realistic Values of Model Reflectance. Solar Energy, 71(4): 233-239. 2001.
Dubios, M.C. 1998. Awnings and Solar-Protective Glazing for Efficient Energy Use in Cold Climates. Proceedings of Renewable Energy technologies in Cold Countries. 1998.
Dubios, M.C. 1999. The design of seasonal awnings for low cooling and heating loads in offices. Proceedings of 5th Symposium on Building Physics in the Nordic Countries. 1999.
Raeissi, S., Taheri, M. 1998. Optimum Overhang Dimensions for Energy Savings. Building and Environment, 33(5): 293-302.
Perez, R., Ineichen, P., Seals, R., Michalsky, J., Stewart, R. 1990. Modeling daylight availability and irradiance components from direct and global irradiance. Solar Energy 44(5): 271-289.
Rao, S., Tzempelikos, A. 2010. Impact of exterior overhangs on the daylighting performance of an office space. Proceedings of 1st High Performance Building Conference at Purdue, West Lafayette, IN, USA, 2010.
ASHRAE student Member
Athanasios Tzempelikos, Ph.D.
ASHRAE Associate Member
Sagar Rao is a graduate student and Athanasios Tzempelikos is an Assistant Professor in the School of Civil Engineering, Purdue University, West Lafayette, IN.
|Printer friendly Cite/link Email Feedback|
|Author:||Rao, Sagar; Tzempelikos, Athanasios|
|Date:||Jan 1, 2012|
|Previous Article:||Neural networks approach for energy consumption in air-conditioned administrative building.|
|Next Article:||Design and application considerations for smoke management fans.|