Printer Friendly

A simulation and experimental study of the impact of passive and active facade systems on the energy performance of building perimeter zones.


Facade design is a key point for achieving high performance buildings. It has a significant impact on heating, cooling, and lighting loads as well as on thermal and visual comfort in perimeter zones of commercial buildings. Solar energy utilization can reduce heating load in winter and at night, reduce lighting energy consumption (Tzempelikos and Athienitis, 2007), and improve lighting quality (Selkowitz and Lee, 1998) but also has some drawbacks such as increasing peak cooling load, cooling energy consumption and thermal discomfort due to high radiant temperatures (Atmaca et al., 2007).

Design of passive facade systems includes the selection of types, sizes, materials and configurations for glazing, shading, insulation, and thermal mass systems in the perimeter zone. Active systems refer to the design of control algorithms and the control operations that directly impact the function, position, performance and the physical properties of facades that affect the whole building energy consumption. Such active (dynamic) systems include motorized shading, glazing products with controllable properties, variable set points, controlled lighting etc that have the potential of saving energy or improving comfort for the occupants (van Moeseke et al., 2007; Jonsson and Roos, 2010). Dynamic facades with appropriate adjustment to changing outside conditions allow flexibility in the integration of design considerations. This concept may lead to a balance between positive and negative impacts of solar energy utilization in perimeter zones and improve energy performance and comfort conditions for the occupants. The optimal solution for each building would be specific to its unique characteristics such as type, size, properties, climate and orientation. A flexible and accurate model that can utilize integrated system options and different control algorithms is needed to predict the energy performance and occupant comfort (Franzetti et al., 2004). This indicates the importance of integration of lighting and thermal models in order to get a comprehensive solution (Lee et al., 1998). However, static and dynamic facade systems are often engaged in the conflict between lighting and air-conditioning savings. A typical example is automated shading: allowing more light would reduce lighting consumption while at the same time, increasing peak cooling demand and energy consumption. An alternative solution to the above problem is the development of multi-sectional and multifunctional facades with embedded intelligent components (Tzempelikos et al., 2007, 2010). These facades consist of different systems along their height (e.g. a lower "vision" part protected from the sun and a top part primarily used for daylighting) that can be controlled independently in order to optimize indoor conditions.

Experimental measurements are always important in order to validate and fine-tune specific models and algorithms (Clarke 2001). Although experiments are time-consuming and expensive, complex cases can be performed and analyzed easier than with analytical validation (Manz et al., 2006). The objectives of this paper are 1) to develop a dynamic transient thermal model in conjunction with a lighting model to evaluate the energy performance of different passive, active and multifunctional facade systems and compare the results with experimental measurements and 2) to set the basis for building a tool that could be used for optimized facade design using innovative components and controls at the early stage.


The big challenge of integrating a daylighting and thermal models is to determine dynamic linking parameters which influence both the daylighting model and the thermal model (Tzempelikos, et al., 2005, 2010) and the feedback loop between two models. This means that the output of the daylighting model will be input to the thermal model and vice versa in a sequential or coupled simulation approach. In any case, the two models are inter-dependent (Figure. 1).


Baseline Building Description

The baseline building (private office) is located in West Lafayette, Indiana and consists of a single room with large windows facing south. The total floor area is 26 [m.sup.2] (5.2m by 5m) and the ceiling height is 4m. Walls, floor, and ceiling are constructed with high resistance materials (R-19 and R-20), and with thermal mass (concrete slabs) in both floor and ceiling. The top part of the facade (0.62 m high) is double-glazed, low-e glass (U-value= 1.668 W/[m.sup.2]-C) and the bottom part (2.15 m high) is double glazed windows with a selective low-e coating (U-value = 1.649 W/[m.sup.2]-C). The infiltration rate is 0.25 ach/hr which varies based on the room construction details (e.g., openings) and outside conditions (e.g., wind speed and wind direction). The room is equipped with a controllable lighting system and automated interior roller shades.

Solar Radiation Model

Typical Meteorological Year (TMY3) data is interpolated to represent a typical weather pattern of this city as the input value for both thermal and daylighting models. For the direct illuminance and solar radiation, they can be calculated directly from beam normal solar radiation, surface tilt angle, orientation, and solar angle which vary with time. For the diffuse part, the Perez et al. model (1990) is used, which provides a method to calculate solarradiation on tilted surfaces.

Absorbed Solar Gains

Solar radiation that strikes the exterior wall is absorbed and reflected; the part that is incident on the windows is absorbed, reflected, and transmitted into the room. The angle-dependent glazing properties (transmittance, reflectance, and absorptance) are obtained from WINDOW6 and matched with experimental measurements. The transmittance of roller shades depends on the fabric material properties; for blinds and light shelves, it depends on the profile angle, slat (or shelf) tilt angle, and slat configuration and properties. When the roller shades close, the inter-reflections between the glazing and the fabric are taken into count when calculating the transmittance of the entire fenestration system:

[[tau].sub.dir-system] = [[tau].sub.dir-glass] * [[tau].sub.shades] + [[tau].sub.dir-glass] * ([[tau].sub.shades] * [[rho].sub.shades] * [[rho].sub.dif-glass]) / (1 - [[rho].sub.shades] * [[rho].sub.dif-glass])(1)

where: [[tau].sub.dir-system] is the direct transmittance of the facade system, [[tau].sub.dir-glass] is the direct transmittance of the glass, [[tau].sub.shades] is the transmittance of the roller shades, [[rho].sub.shades] is the shade reflectance and [[rho].sub.dif-glass] is the diffuse glass reflectance. In the absence of shading, transmitted solar radiation can be tracked on the different room interior surfaces using basic ray-tracing techniques. However, this is a computationally expensive procedure, especially when a yearly analysis is involved. A simple assumption was used to account for the direct solar radiation absorbed in room surfaces: 30% of direct solar radiation comes into the room is assumed to fall on the three walls and 70% is assumed to fall on the floor. This assumption, although creating errors, should change depending on the room geometry. After the first absorption, solar radiation will be reflected to other surfaces. The amount of diffuse solar radiation and reflected solar radiation absorbed by each surface is estimated by area and reflectance of each surface. The equation for the absorbed radiation in the floor can be written as:


where: [alpha] is the absorptance, [rho] is reflectance, A is area, l is solar radiation, and n is the total opaque surface number. When the shades are closed, the absorbed diffuse radiation is calculated as combined area- and absortptance-averaged considering all interior surfaces.

Thermal Model Description

A thermal network approach is used to predict indoor thermal environmental conditions and annual energy consumption on perimeter zones equipped with combinations of passive and active facade systems such as selective glazings, translucent panels, motorized shades and blinds, in conjunction with daylight-linked lighting and shading controls. Each component of the building system is presented as a node (ex: interior surface, thermal mass, exterior surface, shading layers, air). For each node, a heat balance governing equation can be written based on its connections with surrounding nodes by conduction, convection, and radiation and the heat generation and heat storage in control volume. The convections for both outside and inside surfaces are obtained by using the following ASHRAE recommended relation (ASHRAE, 2009). The radiosity method with non-linear coefficients is used for radiation exchange. The convection heat transfer between the shades and gap air is computed using an advanced method (EnergyPlus, 2007). The equation sets are finally solved by the implicit finite difference method. The governing equation for the temperature of the [i.sup.tb] node at a time step p+1 can be written as:


Where: [DELTA]t is the time step, [C.sub.i] is the thermal capacity of the control volume, [U.sub.ij] is the U-values between node i and surrounding nodesy j, [T.sub.j.sup.p+1] is the temperature of the surrounding nodes j, [q.sub.i.sup.p+1] is the heat input (e.g., absorbed solar radiation) on node i at the same time step, and [T.sub.i.sup.p] is the temperature at previous time step. To maintain accuracy, numerical stability, and computing efficiency, a suitable time step is determined from building details, efficiency and sensitivity analysis and a modified Gauss-Siedel iteration algorithm is employed to solve the iteration of equation sets. It requires larger computational memory but increases the convergence rate and speeds up the convergence time. For given infiltration rate and internal heat gains, the load at each time step is:


where: n is the total number of interior surface nodes, [T.sub.i] is the interior surface temperature for all surfaces (i), [T.sub.air] is the indoor air temperature, [T.sub.0] is the outside air temperature, [h.sub.i-air] is the convection coefficient of surface i, [U.sub.infiltration] is the air conductance due to infiltration and [Q.sub.ihg] is internal convective heat gain which comes from lighting, people, and other equipment. The load can be positive (cooling) or negative (heating).

In the multi-sectional facade cases, the top and bottom sections are treated as different surfaces. For each of them, separate nodes are used for outside and inside surface temperatures and as well as separate shading device properties, surface temperatures and controls.

Lighting Model Description

A previously developed daylighting model (Shen and Tzempelikos, 2012) is used to predict work plane illuminance levels and lighting demand. For diffuse illuminance, the window is assumed to be a perfect diffuser. For direct illuminance, we track the exact locations and areas that daylight falls on and use those as separated surfaces. The view factors between interior surfaces are calculated first, and the radiosity matrix is solved to get the illuminance on each surface. Configuration factors between each surface and selected points on the work plane are then calculated and the final work plane illuminance distribution is determined by the configuration factor and illuminance on each surface. The lighting electricity demand then can be determined from the illuminance values for each time step. If the daylight supplies enough amount of illuminance for the workers, the model will automatically dim or turn off the electric lights. Under direct sunlight, the model will activate the shades. These control actions affect the status of the thermal nodes and the internal heat gains, therefore this information is passed to the thermal model at the same time step.


Figure 2 shows a representative annual load profile for the baseline model in West Lafayette for the case with a constant set point (23 [degrees]C), and open shades. The room is exposed to high solar gains throughout the year, so the annual cooling load is high even in the winter. Peak cooling demand reaches 4 KW in September while peak heating demand reaches 2.3 KW in February.


Effects of Set Point and Shading Operation

The impact of set point control and shading operation on energy performance was investigated using the dynamic thermal-lighting model. Figure 3 presents results for representative cases with variable and constant set points as well as with closed, open and controlled roller shades. The baseline buildinghas five exterior surfaces and a large glass facade therefore the heating and cooling energy is much higher than the lighting energy demand. Open shades allow utilization of solar energy and reduce the heating load, resulting in overheating in almost all seasons. Note that this is not a realistic case (showed only for comparison purpose) since the shades need to close to protect from glare whenever there is sunlight incident on the facade. Closed shades significantly reduce cooling energy use (reflective fabrics were considered). The controlled shading results presented in Figure. 3 refer to automatic closing of shades when there is incident direct radiation on the facade (higher than a small threshold, 20W/[m.sup.2]). This is a strict criterion used as a base case -- there are no standards in shading control operation at the moment. If the threshold is increased, the energy performance is clearly improved (also due to lighting energy savings). A detailed analysis of the impact of shade properties and control can be found in Tzempelikos and Athienitis (2007) and in Shen and Tzempelikos (2012).


Considering the set point control, it is obvious that a constant set point (during day and night) is not efficient. Figure. 3 presents results for a few different cases, including variable set points during non-office hours (22-24 [degrees]C and 18-26 [degrees]C) as well as system shut down during night time when the temperatures are not very high or very low. The latter might be problematic since there will be high peaks and thermal discomfort when the system is activated early in the morning. Ramp-shaped set point increase and model-predictive controls provide better solutions.

Model Validation

Experimental measurements in full-scale outdoor test office spaces at Purdue University were conducted in order to compare with simulation results. Figure 4 shows a schematic section view of the experimental setup and a picture of the full-scale office spaces with multi-section facade. Measured quantities include incident and transmitted solar radiation and illuminance, surface temperatures and indoor and outdoor air temperatures.


Representative comparative indoor air temperature results (free-floating) are shown in Figure 5 with both open and closed shades in December together with the respective outdoor air temperature and incident solar radiation values. For closed shades, a consistent trend is observed betweeen simulated and experimental results, with a maximum difference of 0.4 [degrees]C. For open shades on sunny days, simulated and experimental results also have good agreement. However, in party cloudy days, the maximum difference exceed 4[degrees]C. This is due to small uncertainties in the construction and also due to assumptions made in the modeled absorbed solar gains and fabric properties. The distinciton in two cases indicates that the model is very sensitive to solar radiation. One of the challenges dealing with solar radiation is how ot split diffuse and direction portion

In both cases, the lowest temperatures occurs at 7:30 am (before sunrise) and the highest temperature occurs between 12 pm and 1pm, when solar radiation and outdoor temperatures reach their highest levels. The concreate slabs have an effect on the peak shift (2 hrs), as expected with medium thermal mass.


Figure 6 presents another comparison of measured versus experimental results for different temperature ranges using 15-min data collected during three days of measurements. Again, for close shades, the results are in good agreement. But the results are diverse with open shades. More experiments for different cases are planned, including variations of multi-sectional facades with independently controlled sections.



This paper presents a method for assessing the integrated energy performance of passive and active multisection facade systems combined with lighting and thermal controls of perimeter building zones using an open source language. A thermal network approach is used to predict indoor thermal environmental conditions and annual energy consumption of perimeter zones equipped with combinations of passive and active facade systems. The model uses anisotropic sky models for accurate prediction of solar gains, variable angular glazing properties and non-linear interior and exterior convection and radiation heat transfer coefficients together with transient internal gains obtained from transient lighting simulation. A private perimeter office with a glass facade and equipped with automated shading and lighting systems was used as a base case. The results show that changes of control strategies and components properties may result in large amounts of energy savings. There is always a trade-off between lighting, heating, and cooling demand. Examination of the overall energy performance based on building geometry and weather conditions is useful for both designers and engineers to achieve a comfortable environment and at the same time saving money. Further considerations such as cost and feasibility should be taken into account in the real decision making process. For example, some advanced control strategies may be implemented by installing expensive detecting and monitoring systems.

Multifunctional and multi-sectional dynamic facade systems can provide solutions to problems related to the balance between air-conditioning and lighting energy demand. Each section could be independently controlled so as to optimize energy use and maintain comfort for the occupants. Detailed and full-scale experimental studies can help in validating and fine-tuning models and in investigating ways of utilizing solar energy in perimeter spaces of commercial buildings. More work is needed for development of systematic design and optimization methods for integrated facade concepts.


ASHRAE. 2009. ASHRAE Handbook--Fundamentals Atlanta: American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc.

Atmaca, I., Kaynakli O., Yigit A. 2007. Effects of radiant temperature on thermal comfort. Building and Environment 42(9): 3210-3220.

Clarke, J. A. 2001. Energy simulation in building design. Oxford, Butterworth-Heinemann: 362 p.

Franzetti, C, Fraisse G., Achard G. 2004. Influence of the coupling between daylight and artificial lighting on thermal loads in office buildings. Energy and Buildings 36(2): 117-126.

Jonsson, A., Roos A. 2010. Evaluation of control strategies for different smart window combinations using computer simulations. Solar Energy 84(1): 1-9.

Lee, E. S., DiBartolomeo D. L., Selkowitz S. E. 1998. Thermal and daylighting performance of an automated venetian blind and lighting system in a full-scale private office. Energy and Buildings 29(1): 47-63.

Manz, H, Loutzenhiser P., Frank T., Strachan P. A., Bundi R., Maxwell G. 2006. Series of experiments for empirical validation of solar gain modeling in building energy simulation codes--Experimental setup, test cell characterization, specifications and uncertainty analysis. Building and Environment 41(12): 1784-1797.

Selkowitz, S. Lee E. S. 1998. Advanced fenestration systems for improved daylight performance. Daylighting '98 Conference Proceedings, Ottawa, Ontario, Canada.

Tzempelikos, A., Athienitis, A.K., 2005. Integrated daylighting and thermal analysis of office buildings. ASHRAE Transactions 111(1), 227-238.

Perez, R., Ineichen, P., Seals, R., 1990. Modeling daylight availability and irradiance components from direct and global irradiance. Solar Energy 44(5), 271-289.

Shen, H., Tzempelikos, A. 2012. Daylighting and energy analysis of private offices with automated interior roller shades. Solar Energy, in press.

Tzempelikos, A. Athienitis A. K.. 2007. The impact of shading design and control on building cooling and lighting demand. Solar Energy 81(3): 369-382.

Tzempelikos, A., Athienitis A. K., Nazos A. 2010. Integrated Design of Perimeter Zones with Glass Facades. ASHRAE Transactions 116(1): 461-477.

van Moeseke, G., Bruyere I., De Herde A. 2007. Impact of control rules on the efficiency of shading devices and free cooling for office buildings. Building and Environment 42(2): 784-793.

WINDOW 6.3/THERM NFRC Simulation Manual. 2011. Lawrence Berkeley National Laboratory.

EnergyPlus Engineering Document: the reference to EnergyPlus calculations. US Department of Energy, 2007.

Ying-Chieh Chan

Student Member ASHRAE

Athanasios Tzempelikos, PhD

Associate Member ASHRAE

Ying-Chieh Chan is a Ph.D student and Athanasios Tzempelikos is an Assistant Professor of Architectural Engineering in the School of Civil Engineering, Purdue University, West Lafayette, Indiana, USA.
COPYRIGHT 2012 American Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Chan, Ying-Chieh; Tzempelikos, Athanasios
Publication:ASHRAE Transactions
Article Type:Report
Geographic Code:1USA
Date:Jul 1, 2012
Previous Article:Beyond the shoebox: thermal zoning approaches for complex building shapes.
Next Article:Uncertainties in achieving energy savings from HVAC maintenance measures in the field.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters