Energy 233 (2021) 121165

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Energy performance of integrated adaptive envelope systems for

residential buildings
Ammar H.A. Dehwah, Moncef Krarti*
Building Systems Program, Civil Environmental and Architectural Engineering Department, University of Colorado Boulder, CO, USA

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, the energy performance of an integrated adaptive envelope system (AES) is evaluated when
Received 27 January 2021 applied to detached houses in four US climates. Three main technologies are part of the AES including
Received in revised form cool roofs, movable PV-integrated shading devices (MPVISDs), and switchable insulation systems (SISs).
10 April 2021
For this study, the AES is operated to minimize annual heating and cooling energy uses. The analysis
Accepted 2 June 2021
results clearly indicate that the integrated AES have high potential for cooling energy savings for resi-
Available online 7 June 2021
dential buildings. Specifically, MPVISDs offer the highest contribution followed by attic and walls inte-
grated SISs. Overall, the integrated AES allows on-site electricity generation and offers savings between
Keywords:
Adaptive envelope systems
234 kWh/yr and 949 kWh/yr in cooling energy depending on the US climate. The deployment of AES
Residential buildings alone allows US homes to almost reach net-zero energy designs especially in mild and hot climates.
Switchable insulation systems © 2021 Elsevier Ltd. All rights reserved.
Attics
Dynamic cool roofs

1. Introduction the use of tunable thermal conductivity materials and thermally

anisotropic systems [18]. In particular, switchable insulation sys-
Significant fractions of the total US electricity use (39%) and tems (SISs) have been proposed as effective technologies in
natural gas demand (16%) are consumed by residential buildings reducing energy consumption when applied to walls and attics of
[1,2]. In particular, space heating and cooling make up more than residential and commercial buildings. SISs showed high potential in
half of the annual energy use of US residential sector. Heat losses energy savings when operated using simple schedule-based [19], 2-
and gains from the envelope contribute substantially to the heating step temperature-based [20,21] and optimal control strategies [22].
and cooling energy needs of residential buildings [3,4]. In the last In a recent study, Dehwah and Krarti [23] performed a techno-
decade, a special attention has been given to the development of economic analysis for the potential of attic-integrated SISs
high-performance envelope elements to lower thermal loads of applied to prototypical homes in 44 US cities. Their study indicated
buildings [5e9]. In particular, adaptive envelope systems (AESs) that SISs, operated using 2-step temperature-based controls, offer
have been demonstrated to be effective in reducing building ther- savings between 14 kWh/m2 and 51 kWh/m2 varying widely by
mal loads and improve thermal comfort levels [10,11]. Typically, climate. Moreover, Jin et al. [24] evaluated the performance of
AESs allow the thermal and optical properties of the building en- adaptive insulation when applied to a south-oriented wall of an
velope elements to be controlled based on predefined control office room when operated using temperature based and optimal
strategies. Adaptive envelope technologies that have been pro- controls. It is concluded that an adaptive wall insulation can pro-
posed and evaluated in the last decade include dynamic insulations vide 10e50% savings in energy use and thermal comfort improve-
[12], breathing walls [13], variable reflective coatings [14], dynamic ments when compared to static R-value insulation.
shading elements [15,16] and electrochromic windows [17]. Cool roof systems, or reflective roofs, are effective technologies
Moreover, the US Department of Energy (DOE) is leading efforts to in reducing cooling thermal loads particularly in hot climates [25].
develop dynamic envelope technologies that can provide tangible Cool roofs, typically with high solar reflectance and thermal emit-
grid services through load shifting and shedding strategies through tance, reflect most of the received as well as the absorbed solar
radiation fluxes back to the sky resulting in less heat dissipation
through the roof. It has been demonstrated that roof construction
and insulation levels can influence the performance of reflective
* Corresponding author.
E-mail address: Krarti@colorado.edu (M. Krarti).
roof systems [14,26]. For instance, Akbari and Konopacki [26]

https://doi.org/10.1016/j.energy.2021.121165
0360-5442/© 2021 Elsevier Ltd. All rights reserved.
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Nomenclature EER Energy Efficiency Ratio for Cooling Mode Operation

of a Heat Pump [Btu/W]
AC Air Conditioner HVAC Heating, Ventilation, and Air Conditioning
ACH Air Change per Hour LPD Lighting Power Density [W/m2]
AES Adaptive Envelope System MPVISD Movable Photovoltaic Integrated Shading Device
AFUE Annual Fuel Utilization Efficiency R-value Thermal Resistance in IP Unit [hr.oF.ft2/Btu]
ASHRAE American Society for Heating, Refrigerating, and Air RSI Thermal resistance in SI Unit [oC.m2/W]
Conditioning Engineers PV Photovoltaic
BIPV Building Integrated Photovoltaic SPVISD Static Photovoltaic Integrated Shading Device
COP Coefficient of Performance SHGC Solar Heat Gain Coefficient
DOE Department of Energy SIS Switchable Insulation System
DX Direct Expansion StaticCR Static Cool Roof
DynamicCR Dynamic Cool Roof VT Visible Transmittance
EMS Energy Management System WWR Window-to-Wall Ratio
EPD Equipment Power Density [W/m2]

evaluated the performance of static cool roofs when applied to US compared to an exposed roof. The reduction in heating load is due
residential and commercial buildings considering two construction to heat gains from the PV panel combined with lower radiative
vintages, i.e. pre 1980 and post 1980. Their evaluation showed that losses from the roof during nighttime. The study concluded that as
there is a direct correlation between the cooling energy savings and the roof reflectance increases, the absolute annual heating load
cooling degree-days as well as the heating energy penalty and savings increases especially for poorly insulated ceilings. These
heating degree-days. In addition, savings up to 11% and 8% can be savings reduced significantly when the ceiling insulation level is
achieved when static cool roofs are applied to pre-1980 and post increased from RSI-0.9 to RSI-5.3. Similarly, Kapsalis and Karamanis
1980 US homes, respectively, with a 2% heating energy penalty. [33] evaluated the impact of PV panels when installed on top of a
Dynamic cool roof systems are ideal alternatives to static cool roofs, roof with reflectance of 0.45 and RSI-1.25 thermal insulation using
primarily, for the advantage of eliminating heating penalties. A a TRNSYS model validated against experimental data. Temperature
thermochromic coating, that can control optical properties based measurements indicated that during winter, the nighttime roof
on temperature variation, is the main technology used for variable temperature is warmer when a PV panel is installed while it is
reflective materials including cool roofs [14,27]. significantly lower during daytime when compared to an exposed
Recently, the application of building integrated photovoltaics roof (i.e., without any PV panel). Based on annual simulations, the
(BIPVs) as a roofing system has gained more attention specifically analysis showed a 7% (168 kWh/m2) increase in heating loads
for their dual functions: acting as an additional layer to the building compared to the baseline case with an exposed roof.
element and generating on-site electricity [28,29]. A BIPV roofing Movable shading devices have been shown to be efficient
system is another alternative to cool roof systems, for summer technologies in enhancing the energy performance and indoor vi-
applications, due to their indirect shading impact and ability to sual comfort [16,35]. PV integrated shading devices (PVISDs) offer
produce electricity especially with decreasing PV costs. For the additional benefit of converting solar radiation, otherwise
instance, Wang et al. [30] reported that PV ventilated roofs have the wasted, to electricity that can be used on-site and hence help
same impact on cooling load reduction compared to that achieved mitigating CO2 emissions. Moreover, optimized controls for
by a cool roof with a reflectance of 0.65. However, the impact of movable PV-integrated shading devices (MPVISDs) have been
installing PV on top of a cool roof system on heating energy has not proposed to minimize thermal loads and maximize electricity
been duly investigated in the literature. Deploying PV panels on top generation. For example, Jayathissa et al. [36] investigated the
of a static cool roof system has different impacts on heating and optimal orientation of a MPVISD that maximizes PV output and
cooling thermal loads compared to deploying PV on top of a stan- minimizes lighting and thermal loads. Their analysis estimated that
dard non-reflective roof. Based on experimental and simulation 20e80% net energy savings can be achieved when MPVISD is uti-
studies, it has been reported that installing a PV panel on a standard lized compared to an equivalent static shade. This study utilizes a
roof (dark roof) generally reduces the cooling energy use due to the MPVISD for a different purpose, mainly to avoid the heating penalty
shading effects. However, its impact on heating energy use is found that would occur as the result of the static cool roof system during
to depend highly on the building integrated Photovoltaic (BIPV) the winter season (refer to section 2.1.1). Most of the aforemen-
configuration as well as the climate [31,32]. Based on the review of tioned reviewed studies have quantified the performance of AESs
the existing literature, a more detailed investigation is needed to individually. No reported study has considered integrated AESs that
assess the specific energy impact of installing PV panels on static encompass various dynamic building envelope technologies. In this
cool roofs [33,34]. William et al. [34] evaluated the impact of paper, a novel integrated adaptive envelope system is evaluated
rooftop PVs on heating and cooling thermal loads for five US cli- when applied to US homes. Three technologies make up the pro-
mates. Specifically, they showed that the yearly savings in heating posed AES including cool roofs, MPVISDs and SISs. The integrated
and cooling loads for various roof types depend on the level of roof AES can be dynamically operated using seasonal and/or hourly
insulation (R-5 to R-30) as well as the roof reflectance (0.05e0.87). control strategies. First, the design features and control mecha-
The numerical model, used for the analysis, estimates the heat flow nisms of the AES are described. Then, detailed description of the
through only the roof and does not consider the thermal in- building model used for the analysis is presented. The performance
teractions for the whole building. The model was validated against of each technology of the AES is assessed individually. Then, the
summer and winter experimental results. The analysis results benefits of the integrated AES are assessed in terms of energy ef-
showed that installing an offset BIPV on top of a cool roof in cold ficiency and peak demand reduction under various design and
climates helps in reducing the roof-associated heating loads climatic conditions.

2
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

2. Materials and methods at which roofs can radiate the absorbed solar radiation and even-
tually indicate how rapidly it can cool. Static cool roofs (StaticCRs)
2.1. Description of the adaptive envelope systems are generally effective in cooling-dominant climates but they
typically result in a heating penalty [39]. The heating penalty and
In this study, the energy performance of an integrated adaptive the cooling savings associated with StaticCRs can vary with loca-
envelope system (AES) is evaluated when applied to US homes. tion, insulation level, and building type [26,40]. The heating penalty
Three main technologies are involved in this AES including cool can be avoided when considering dynamic cool roof (DynamicCR)
roofs, movable PV-integrated shading devices (MPVISDs) and systems [14]. Ideally, dynamic cool roofs lead to the same heating
switchable insulation systems (SISs) as illustrated in Fig. 1. The energy use as standard roofs and the same cooling energy use
proposed AESs are suitable for homes with pitched roofs (e.g., gable savings as StaticCRs. The proposed MPVISD is designed to make the
and hipped) as indicated in Fig. 1 but can be also adopted for StaticCR to act as a dynamicCR since the overall optical properties of
buildings with flat roofs with a slight modification of the MPVISD the roof changes with the MPVISD position. To confirm this
system. In this study, the roof has a static reflective coating (i.e., cool assumption, a dynamic cool roof, presented in Ref. [14], is modeled
roof). However, the AES can consider instead a dynamic reflective and evaluated as part of the current analysis. The dynamic cool roof
coating. Moreover, a movable PV-integrated shading device is system switches between a high reflectance of 0.55 during the
installed and allowed to slide along the roof slope (with an angle of cooling season and a low reflectance of 0.3 during the heating
18
) to take two positions depending on a predefined control season. This switching range is considered in the analysis to be
schedule (i.e., positions a and b) as indicated in Fig. 1 to manage conservative and consistent with the reported results on the per-
solar heat gains through the windows. In addition, attic and wall formance of dynamic cool roofs [14]. Indeed, higher reflectance
integrated SISs are deployed to switch the thermal resistance of the values up to 0.96 have been reported for StaticCRs [41].
attic and/or wall assemblies between high and low values accord- Movable shading devices, when well designed, have been
ing to specific temperature-based rulesets. Specifically, the roof of demonstrated to be effective systems to reduce the total energy
the home is coated with a cool material to reflect the solar radiation consumption of residential buildings [16]. When integrated with PV
and reduce the cooling loads during summer seasons. As shown in panels, shading devices provide the additional benefits of gener-
Fig. 1, a movable PV-integrated shading device (MPVISD) is ating electricity that can be used by the building with any excess
deployed and allowed to switch between two positions including supplied to the grid. The basic operation scheme of the proposed
position “a” (i.e., partially off the roof) and position “b” (i.e., on top MPVISD is illustrated in Fig. 2. In particular, during the summer
of the roof). During summer, the MPVISD, set at position “a”, acts a season, the MPVISD is set to shade the windows facing east, south,
shading device and hence has no impact on the roof reflectance. and west while the cool roof is exposed as shown in Fig. 2 (a). In this
However, during winter, the MPVISD moves to position “b” and MPVISD position, the cool roof system reflects most of the solar
hence changes the optical properties of the roof since the PV panels radiation striking the roof surface while the MPVISD shades win-
now cover the roof. Specific details pertinent to each technology dows to minimize the cooling thermal needs. In contrast, during
involved in the AES are described in the following subsections. the winter season, the MPVISD moves upward covering the roof
surface and maintaining the windows unshaded to maximize free
2.1.1. Cool roofs and movable PV-integrated shading devices solar heat gains as shown in Fig. 2 (b).
The application of cool roofs can reduce cooling thermal loads DynamicCRs can be controlled to switch the reflectivity of the
by utilizing highly reflective materials or applying white coatings to roof's surface on seasonal or hourly bases. Optimized control stra-
the exterior roof surface [37,38]. Typical cool roofs have light- tegies can determine the best roof reflectivity level during each
colored materials or coatings with low solar absorption and high hour, however, they offer minimal additional savings compared to
thermal emittance. The thermal emittance is used to define the rate seasonal controls [14]. Hence, the analysis presented in this study
considers only seasonal control options for DynamicCRs. Specif-
ically, the MPVISDs are set to shade the windows and expose the
roof to be reflective during the summer season and to be on top of
the roof to reduce its overall reflectivity as well as allow full solar
heat gains through the windows during the winter season. Fig. 3
shows 3D renderings of the house model illustrating the shadow
effects when the MPVISDs are deployed at different times of the day
during July 21st.
As discussed in Section 1, the thermal impact of installing PV on
top of cool roofs on heating energy use is rather uncertain and
depends greatly on the BIPV configuration as well as the climate
conditions [31,32]. To account for the uncertainty in determining
the impact of rooftop PV deployment on heating thermal loads, the
analysis in this study considers the two extreme scenarios, with the
assumption that the actual performance would be between the
results obtained for these extreme cases. Specifically, it assumed
that the PV panels, when deployed on a static cool roof, can either
reduce or increase the heating thermal loads. These two options are
accounted for by defining equivalent reflectance values to mimic
both possible scenarios. In particular, during the heating season the
equivalent roof solar reflectance is set either to be 0.9 or 0.1.
The PV electricity output is estimated using EnergyPlus ac-
counting for the MPVISD position, i.e. as BIPV when it is on the
rooftop during winter and as a decoupled surface when it acts as a
Fig. 1. Basic overview of the Adaptive envelope system (AES). shade device during summer. All required model inputs are
3
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 2. 3D illustrations of the cool roof and the MPVIDS technologies during (a) summer and (b) winter seasons.

Fig. 3. 3D renderings displaying shadows from MPVISD on wall and window surfaces during July 21st at (a) 9 a.m., (b) 12 p.m. and (c) 3 p.m.

retrieved from manufacturer's datasheet. The PV panels are limited settings for both walls and attic dependent on US climate zone
to the roof areas of the south, east and west oriented shading de- based on ASHRAE 90.2 standard according to the location of the
vices which are 23.6 m2, 16.4 m2 and 16.4 m2, respectively. Spe- house [44]. SISs are applied to the attic and exterior walls of the
cifically, the PV modules are of mono-crystalline silicon type with house model, described in more details in Section 2.3, and are
17.5% efficiency, 290 Wp output and a temperature coefficient modulated independently with the main objective to reduce
of 0.427%/
C. heating and cooling thermal loads. In particular, for this study, the
low R-value is set to 0.5 m2 K/W (RSI-0.5) and 0.6 m2 K/W (RSI-0.6)
for attic-integrated and wall-integrated SISs respectively. Moreover,
2.1.2. Switchable insulation systems the analysis in this study considers that the SIS for each wall (a total
Recently, SISs have been shown to be effective in enhancing the of 8 walls; 4 in each floor) and for the attic is controlled
energy efficiency of residential buildings [21,42]. The SIS technol- independently.
ogy, considered in this study for the AES, uses a rotating mechanism The time and the level of the thermal resistance switchover are
that controls the position of insulation panels to make full contact determined using rule sets based on the indoor and outdoor con-
or to create separation as exemplified in Fig. 4. This allows the SIS to ditions. From the results of a comparative analysis of a wide range
switch between high thermal resistance value during the fully of control strategies [42], temperature-based control strategies are
closed configuration [Fig. 4(a)] and low thermal resistance value employed in this study for their simplicity and effectiveness to
during the fully open configuration [Fig. 4(b)]. For the latter control both the attic-integrated and wall-integrated switchable
configuration, the rotating insulation layers create openings insulation system as depicted in Fig. 5. Specifically, during the
allowing the air to flow through the wall and attic cavities to heating mode, the controller switches the thermal resistance of the
decrease their R-value as demonstrated through experimental overall attic/wall assembly from the default high R-value setting to
testing [43]. The assembly constructions for both walls and attic low R-value if the mean indoor air temperature (IAT) is less than the
dictate the specific low thermal resistance values. The high R-value

Fig. 4. Basic operation of the attic/wall integrated SIS at the (a) fully closed position and (b) fully opened position.

4
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 5. Simplified temperature-based ruleset used to operate attic/wall integrated SISs.

attic/wall outside surface temperature (Tso). In particular, the resistance allows the trapped heat to be rejected to the ambient
controller switches the R-value of the attic/wall assemblies to their environment and thus benefiting from free cooling. In other words,
low settings of RSI-0.5/RSI-0.6 offering free heating since the heat the SIS operated by the temperature-based control strategy aims to
flows into the house. During the cooling season, the thermal re- maximize heat losses while minimizing heat gains during the
sistances of the attic/wall assemblies are set to their low value cooling season and the opposite occurs during the heating season.
when the mean indoor air temperature (IAT) is higher than the
attic/wall outside surface temperature (Tso). Lowering the thermal

5
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

2.2. Building model description 3. Results and discussion

For this study, an energy model for a single-family detached This section starts by reporting the energy end-uses for the
house, developed by DOE [45], encompassing a 220 m2 floor area is baseline residential building when located in Boulder, CO to assess
considered. This model represents a code-compliant typical new the contribution of heating and cooling loads in the total energy
single-family home [46]. The house model is based on US con- consumption. Next, it discusses the performance of the various AES
struction practices and is widely used for the development of en- technologies when deployed individually and when integrated for
ergy efficiency codes [45]. The main characteristics of the baseline the prototypical detached home when located in Boulder, CO. Then,
house energy model, made-up of brick walls and a light-weight the performance of the integrated AESs is analyzed when applied to
wooden roof structure, is summarized in Table 1. The house has a a prototypical house in representative US climates considering
hipped roof, with the top layer made of shingles having a solar various metrics including energy use, peak demand, and thermal
reflectance of 0.25, and an unconditioned sealed attic space. The comfort.
ceiling has a thermal resistance of RSI-6.8 with the insulation
installed on the attic floor. The HVAC system operates on heating
mode throughout the heating season with a setpoint temperature 3.1. Annual energy end use of baseline residential building
of 22.2
C and switches over to cooling mode during the cooling
season with a setpoint of 23.8
C. Thermal properties of the floor, As described in section 2.2, the baseline house, located in
exterior wall and roof construction materials are listed in Table 2. In Boulder, CO, has a static roof insulation and a solar reflectance of
particular, the insulation levels are chosen to meet ASHRAE 0.25. The HVAC system operates based on seasonal heating and
90.2e2018 standard requirements when the baseline house is cooling schedules. The heating and cooling seasons are determined
located in Boulder, CO [44]. Specifically, the thermal transmittance based on the monthly degree days for base temperatures of 18.3
C
(U-value) of the roof, ceiling and exterior walls are 3.1 W/m2.K, and 10
C respectively [46]. Fig. 6 shows that, for Boulder, CO, the
0.148 W/m2.K and 0.34 W/m2.K, respectively. The baseline internal heating season spans from October through April while the cooling
loads and associated schedules for occupancy, lighting and equip- season lasts from May through September. The total site and source
ment are illustrated in Table 3. For this study, Energyplus, a state- energy consumptions of the prototypical home is found to be
of-the-art whole building energy simulation tool, is used to eval- 32,097 kWh/yr and 56,228 kWh/yr indicating a site and source
uate the energy performance of the AESs [47]. Energyplus features energy use intensities (EUI) of 145.3 kWh/m2/yr and 254.6 kWh/
the capability of modeling advanced technologies such as switch- m2/yr respectively. Fig. 7 depicts the percent distribution of the
able insulation systems and dynamic cool roofs through the energy monthly source energy end-uses for the baseline house located in
management system (EMS) [48]. Furthermore, the annual energy Boulder, CO. Interior equipment accounts for 43% of the annual
simulation is performed considering a 3-min timestep. baseline source energy consumption, followed by space heating
(27%), interior lighting (7%), space cooling (7%) and finally fans (5%).

Table 1
Main characteristics of the baseline house model.

Characteristics

3D Rendering

Floor area 220 m2

(12.1 m  9.1 m)
Floor-ceiling height 2.6 m
WWR 15%
Windows Double-glazed
U-value: 1.7 W/m2.K, SHGC: 0.3, VT ¼ 0.8
Heating operation Setpoint: 22.2
C, No setback
Gas Furnace, AFUE: 0.8
Cooling operation Setpoint: 23.8
C, No setup
Central DX AC, COP: 3.2
Infiltration 0.24 ACH
Ventilation 0.028 m3/s outdoor air

6
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Table 2
Construction and material properties for the envelope elements of the baseline house model.

Construction material Thickness (cm) Conductivity (W/m.K) Density (kg/m3) Specific Heat (J/kg.K)

Roof
Roof Shingles 0.63 0.082 1121 1255
Oriented Strand Board 1.27 0.12 545 1213
Ceiling
Insulation 21.00 0.032 28 960
Drywall 1.27 0.16 800 1088
Exterior Walls
Stucco 0.30 0.087 400 879
Sheathing 1.27 0.094 685 1172
Oriented Strand Board 1.11 0.116 545 1213
Insulation 13.97 0.057 120 1036
Drywall 1.60 0.16 800 1090
Floor
Plywood 1.91 0.115 545 674
Carpet 2.54 0.060 32 837

3.2. Cool roofs earlier, for the integrated AES analysis, the two extreme scenarios
are considered to account for the uncertainty in determining the
In this section, the impacts of applying reflective coatings to the impact of rooftop PV deployment on top of the cool roof on heating
roof surface on HVAC energy use, electrical peak demand, and thermal loads since most likely the actual performance would fall
thermal comfort levels are investigated when the house is located between the results obtained for these two cases.
in Boulder, CO. First, two static cool roof (StaticCR) cases are eval- To assess the indoor thermal comfort levels, unmet hours during
uated to demonstrate the impact of implementing this technology. the heating and cooling seasons are reported with a tolerance of
These cases represent an off-white coating with a reflectance of ±0.2
C. Specifically, the unmet hours represent the total number of
0.55 and a solar reflectance index (SRI) of 63 as well as a highly hours throughout one year in which heating and cooling set points
reflective white EPDM coating with a reflectance of 0.70 and RSI of are not met by the HVAC system when the zone is occupied. In
84 [39]. Fig. 8 shows daily profiles of the outside roof surface particular, the unmet hours during the cooling mode are counted
temperatures as well as cooling energy uses for the baseline and based on the number of hours when indoor air temperatures (IATs)
the static cool roof cases for a specific cooling day (i.e. May 6th). It are higher than the cooling set point (i.e., 23.8
C in this study).
can be seen that the peak roof outside surface temperature is Similarly, the unmet hours during the heating mode include the
reduced, as the result of reflecting solar radiation compared to a hours when the IATs are lower than the heating set point (i.e.,
standard roof, by 17
C and 27
C when static cool roofs with a 22.2
C in this study). The total number of unmet hours during both
reflectance of 0.55 and 0.70 are deployed, respectively. The cooling heating and cooling seasons are listed in Table 4 for the standard
energy use for a StaticCR with a reflectivity of 0.55 and 0.70 is roof, StaticCR and DynamicCR cases. Table 4 indicates that the set
reduced, respectively, by 12% and 19% due to the temperature drop point is met in all hours during the heating season while the cooling
during May 6th. Table 4 summarizes the annual energy perfor- set point is unmet for 17 h during the cooling season for the
mance for the analyzed roof types. Considering the annual energy baseline, and ranges between 17.5 h and 18.0 h for the StaticCR and
performance, a StaticCR with a reflectivity of 0.55 results in a 1% DynamicCR cases.
increase in heating energy use and 4% decrease in cooling energy
use relative to the baseline with a standard dark roof. Furthermore, 3.3. PV integrated shading devices
a StaticCR with a reflectance of 0.55 can reduce the peak demand by
2%. When increasing the reflectance of the cool roof from 0.55 to The impact of the static PV-integrated shading devices
0.70, 2% additional savings in annual cooling energy use and (SPVISDs) on annual heating and cooling energy use is depicted in
additional 0.8% increase in heating energy use occur with a 1% Table 5 when deployed on only one facade as well as on all three
additional reduction in electrical peak demand. facades. As expected, south orientation shows the highest impact
As described in Section 2.1, a DynamicCR system can be effec- on heating and cooling energy uses since south windows have the
tively deployed as part of the proposed AES using movable PV in- highest solar heat gains throughout the day. Deploying static
tegrated shading devices (MPVISDs) that change position to either shading on the south façade results in a 4.6% reduction in cooling
cover the roof during the winter or to act as a shading device during energy use while a 1% increase in heating energy use. The shading
the summer. Deploying this system results in a 4% savings in device covers the south windows and walls partially reducing the
cooling energy use, that is, the same savings achieved by a StaticCR amount of solar radiation received on the south window/wall (and
compared to the standard roof. In the best scenario, where covering eventually surface temperatures) and thus helps in lowering the
the roof by the PV panels (i.e., MPVISDs) has the same effect as a cooling energy demand while increasing the heating energy needs
dark roof surface with an equivalent reflectance of 0.1, 0.4% savings since shading reduces the opportunity of free solar heat intake.
in heating energy use can result compared to a standard roof. Deploying static shades on east and west orientations results in a
However, the same action may result in a 3% heating penalty in the small heating penalty of 1% while providing 4.6% and 3.9% savings
worst scenario when covering the roof with PV panels is equivalent in cooling energy use, respectively. Furthermore, the maximum
to a roof reflectance of 0.9. A theoretical DynamicCR [14] that electrical peak demand saving of 3.6% is achieved when shades are
operates between a reflectance of 0.55 during the summer and 0.3 deployed on the west façade as expected since the peak electricity
during the winter is evaluated. The results show that this theo- demand occurs during the late afternoon hours. When all three
retical DynamicCR system results in a 1.8% heating penalty and a shades are deployed statically throughout the year, the heating
4.2% savings in cooling energy savings as shown in Table 4. As noted energy use increases by 5.4% (penalty) while the cooling energy use
7
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Table 3
Internal load details of the baseline house model.

Internal Loads Description Schedule

Occupancy 3 people

Lighting LPD ¼ 1.6 W/m2

Plug loads EPD ¼ 2.84 W/m2

is decreased by 13% relative to the baseline case where no shades resources. In particular, the south orientation has the highest
are considered. As part of the AES, movable PV-integrated shading contribution to the total electricity generated by the PV system with
devices (MPVISDs) are set to manage solar gains through the 5619 kWh/yr. The east-facing MPVISD has a slightly higher PV
windows during the summer season and to cover the roof during output compared to that of the west-facing with 3287 kWh/yr and
the winter season as described in section 2.1. In summary, MPVISD 3048 kWh/yr, respectively. The MPVISD facing south, east and west
results in 16.3% savings in cooling energy use while the heating represent, respectively, 57%, 34% and 31% of the total building
penalty drops to 1%. electricity needs when deployed individually. To assess if the
The additional benefit that MPVISD offers is the in-situ elec- MPVISD system is sufficient for the house to reach a net zero status
tricity generation. Figs. 9 and 10 depict the daily and monthly based on source energy, the site electricity and natural gas uses are
electricity generated by PV when the MPVISD is facing east, west converted using conversion factors of 2.8 and 1.05, respectively.
and south. The PV capacity is 3.5 kWp, 2.3 kWp and 2.3 kWp for When all MPVISDs are deployed, the total electricity generated by
south, east and west MPVISDs, respectively. As expected, the PV PV is 11,955 kWh/yr resulting in annual savings of site and source
output is higher during the summer season compared to that energy consumptions of 40% and 68%, respectively.
during the winter season due to availability of higher solar

8
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 6. Monthly (a) HVAC load and (b) degree days for the baseline home located in Boulder, CO.

Fig. 7. Monthly source energy consumption when the home is located in Boulder, CO.

3.4. Attic-integrated switchable insulation energy use while it offers only 1% savings in heating energy use. The
attic SIS is active for 9% and 3% of the summer and winter season
Daily profiles of temperatures, energy use and thermal resis- hours, respectively. Furthermore, attic SIS seems to have minimal
tance settings are presented to illustrate the performance of attic- impact on the electrical peak demand since the peak load occurs in
integrated switchable insulation when applied to a typical house hot summer days where it is preferable to maintain the insulation
located in Boulder, CO. The attic-integrated SIS is allowed to switch at high R-value during the peak hours. It is worth noting that up to
between a low value of RSI-0.5 and a high R-value setting of RSI-6.8. 29% in cooling energy savings and 3% in heating energy savings
Fig. 11 depicts the hourly profiles of (a) indoor air temperature could be achieved with flexible control strategies but at the detri-
(IAT), ceiling outside surface temperature (Tso), outdoor air tem- mental cost of lower levels of thermal comfort within the occupied
perature (OAT) and thermal resistance setting and (b) the cooling zone.
energy use for the attic-integrated SIS (RSI-0.5/RSI-6.8) compared
to the baseline with static R-value (RSI-6.8) during May 6th. Fig. 11
3.5. Wall-integrated switchable insulation
(a) shows that the ceiling thermal resistance is set to its low R-value
only when there is a need for cooling and IAT is higher than Tso
Fig. 12 illustrates the energy performance of wall-integrated SIS
which occurs at hour 18 (i.e., 6 p.m.) and lasts for about an hour.
operated using a 2-step temperature-based control strategy when
Switching the attic-integrated SIS to its low R-value during this
the house is located in Boulder, CO, by presenting the daily tem-
hour allows the heat trapped indoors to escape the house resulting
perature, thermal resistance settings, and energy use profiles dur-
in a reduction of 12% in cooling energy use during May 6th as
ing May 6th. The wall-integrated SIS is allowed to switch between a
shown in Fig. 11(b). Based on annual performance, the attic-
low value of RSI-0.6 and a high R-value setting of RSI-2.9 when
integrated SIS is more effective to reduce energy use during the
applied to the house model located in Boulder, CO. Specifically,
cooling season. Specifically, operating the attic-integrated SIS using
Fig. 12 (a) and (b) show the wall outdoor temperatures and the
the ruleset defined in Section 2.1.2 offers a 10% saving in cooling
switching activity of the wall-integrated SIS for all first-floor wall
9
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 8. Daily profiles of (a) outside roof temperature and (b) cooling energy use for standard and static cool roofs during May 6th when the house is located in Boulder, CO.

Table 4
Annual heating and cooling energy savings, peak demand savings and unmet hours achieved by static and dynamic cool roofs compared to standard roof when the house is
located in Boulder, CO.

Reflectance Difference in heating energy Difference in cooling energy Peak demand Unmet hours during Unmet hours during
usea,b (%) usea,b (%) savingsa (%) heating (hrs) cooling (hrs)

Baselinec 0.25 12,886 1186 3.140 0 17.0

StaticCR 0.55 1.2% 4.2% 1.9% 0 17.5
StaticCR 0.70 1.9% 6.4% 2.9% 0 18
DynamicCR 0.10/0.55 0.4% 4.2% 1.9% 0 17.5
DynamicCR 0.30/0.55 1.8 4.2% 1.9% 0 17.5
DynamicCR 0.90/0.55 2.7% 4.2% 1.9% 0 17.5
a
Difference is relative to the baseline house with standard roof.
b
Negative values indicate an increase in energy consumption while positive values indicate savings.
c
Energy use (kWh/yr) specific to the baseline case (i.e., dark roof with reflectance of 0.25).

orientations during May 6th. It is clear that the wall SIS starts that take longer to cool due to the sun path. Switching the wall
switching to low R-value for the north and east walls at 4:30 p.m., insulation to low thermal resistance settings results in rejecting
which is, respectively, 0.5 h and 2.5 h earlier than when south and heat within the thermal zone to the outdoors and hence reducing
west walls take actions. Indeed, the outside surface temperatures of the cooling energy use by 9% during May 6th as shown in Fig. 12 (c).
north and east walls cool faster compared to south and west walls Deploying SIS for only the second-floor walls show similar results

10
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Table 5
Annual heating and cooling energy savings and peak demand savings achieved by static and movable PV-integrated shading devices compared to the baseline home located in
Boulder, CO.

Difference in heating energy usea,b Difference in cooling energy usea,b Peak demand savingsa,b Unmet hours during heating Unmet hours during cooling
(%) (%) (%) (hrs) (hrs)

Baselinec 12,886 1186 3.140 0 17

Static East 1.0% 4.6% 0.7% 0 17
Static 1.0% 3.9% 3.6% 0 21
West
Static 3.2% 4.7% 2.9% 0 17
South
Static All 5.4% 13.0% 5.7% 0 21
MPVISD 1.0% 16.3% 6.4% 0 10
All
a
Difference is relative to the baseline house with standard roof.
b
Negative values indicate an increase in energy consumption while positive values indicate savings.
c
Energy use (kWh/yr) specific to the baseline case (i.e., no shading device).

Fig. 9. PV electricity generation from MPVISD oriented east, west and south as well as electricity demand during (a) August 8th and (b) December 21st when the house is located in
Boulder, CO.

Fig. 10. Monthly PV electricity generation from MPVISD of east, west and south orientations.

to those obtained when only the first-floor walls have SIS. When modulate independently. During the heating season, SIS is mostly
deploying SIS to all walls (i.e. first and second floors), the per- active when applied to the southern exterior wall, which receives
centage saving increases to 14%. Indeed, including all walls in- high solar radiation during daytime, resulting in a 2% savings in
creases the surface area and hence increases the potential of the SIS heating energy use. During the cooling season, SIS applied to walls
to reject heat to the outdoor environment. facing north and south are more effective compared to the other
Table 6 depicts the annual performance of SIS when applied to wall orientations as it can save about 4% in cooling energy use. The
four wall orientations (i.e. first and second floors) and allowed to annual cooling and heating energy uses are reduced by 2% and 10%,
11
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 11. Hourly profiles of (a) temperatures and overall thermal resistance of the attic and (b) cooling energy use specific to static and SIS during May 6th when the house is located
in Boulder, CO.

respectively, when the 2-step controlled SIS is applied to exterior insulation technology, it is clear that the attic and wall SISs are
walls of the prototype house. It is worth mentioning that the po- mostly active during the cooling season. For the attic-integrated SIS
tential for walls to reject/accept heat to/from the surroundings during the heating season, the maximum switching activity (5% of
depends on the surface area as well as the temperature differential the hours) occurs primarily on the warmer months of March and
across the wall. April. When the SIS is applied to walls, the activity level of the north
and west walls is minimal during the heating season mainly
3.6. Integrated AES performance because as indicated by the ruleset either there is no heating load or
the wall outside surface temperatures are low. During the cooling
This section discusses the potential energy benefits of inte- season, July and August have the highest level of actions when SIS is
grating all AES technologies to the same prototypical house located integrated with the attic and the walls. Based on the ruleset, SIS
in Boulder, CO. The MPVISD, acting as both a shading and a operates only when there is a cooling need. Indeed, July and August
DynamicCR, is allowed to switch positions on seasonal bases while experience high outdoor temperatures and higher cooling loads
the attic and wall SISs are controlled on hourly basis using the providing more opportunity for switching specially during night-
rulesets outlined in Fig. 5. Table 7 shows on a monthly basis the time hours.
state of each AES element including MPVISD position and associ- Table 8 shows the monthly heating and cooling energy savings,
ated roof solar reflectance, as well as the number of hours of low relative to the baseline case with a standard roof reflectance of 0.25,
thermal resistance selected by the attic and wall SIS controllers. It when integrating AES technologies individually and simulta-
can be seen that during the winter season, the position of the neously. Generally, these AES technologies work best during the
MPVISD is on top of the roof and hence the equivalent roof cooling season. In particular, the results show that the highest
reflectance is either 0.1 or 0.9 to span all the possible cases for the percent savings in cooling energy use occur in May with 51% when
actual performance. During the summer season, the MPVISD moves all AES technologies are integrated. Using the absolute energy
down to act as a shading device and hence the equivalent roof savings as the performance metrics, July and August have the
reflectance is 0.55 acting as a cool roof. For the switchable highest savings with 102 kWh/yr and 104 kWh/yr, respectively. In
12
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 12. Hourly profiles of (a) temperatures, (b) overall thermal resistance of the different wall orientations and (c) cooling energy use specific to static and SIS during May 6th when
the house is located in Boulder, CO.

13
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Table 6
Annual heating and cooling energy savings and percent of hours of low R-value setting corresponding to the wall-integrated SIS when the house is located in Boulder, CO.

Savings in heating energy use relative to Savings in cooling energy use relative to Percent hours of low R-value in Percent hours of low R-value in
baseline (%) baseline (%) winter (%) summer (%)

Baselinea 12,886 1186 0% 0%

All North 0.0% 3.9% 0% 17%
All East 0.5% 3.0% 8% 16%
All South 1.9% 4.3% 10% 16%
All West 0.0% 2.9% 1% 16%
All Walls 2.2% 9.6%
a
Energy use (kWh/yr) specific to the baseline case (i.e., static high R-value wall insulation).

Table 7
Monthly settings of the different adaptive envelope systems deployed for the house model located in Boulder, CO.

Month MPVISD position Roof reflectance Hours of low R-value setting from SIS (%)

Attic North Wall East Wall South Wall West Wall

Jan On Roof 0.1/0.9 3% 0% 5% 13% 1%

Feb On Roof 0.1/0.9 4% 0% 8% 9% 1%
Mar On Roof 0.1/0.9 5% 0% 10% 7% 0%
Apr On Roof 0.1/0.9 5% 0% 10% 4% 0%
May Shading device 0.55 4% 4% 7% 6% 3%
Jun Shading device 0.55 9% 8% 12% 13% 10%
Jul Shading device 0.55 16% 12% 13% 18% 17%
Aug Shading device 0.55 15% 16% 15% 17% 18%
Sep Shading device 0.55 8% 10% 11% 9% 8%
Oct On Roof 0.1/0.9 1% 0% 5% 3% 0%
Nov On Roof 0.1/0.9 3% 0% 7% 11% 1%
Dec On Roof 0.1/0.9 1% 0% 4% 13% 1%
Heating season On Roof 0.1/0.9 3% 0% 7% 9% 1%
Cooling Season Shading device 0.55 10% 10% 12% 13% 11%

Table 8
Heating and cooling energy savings/penalties of the individual and integrated adaptive envelope systems when the house is located in Boulder, CO.

Baselinea DynamicCR All Shades AtticSIS WallSIS Integrated AESs

Jan 2750 0% 1% 0% 3% 2%

Feb 2323 0% 1% 1% 2% 2%
Mar 1706 1% 1% 2% 3% 4%
Apr 874 1% 2% 3% 2% 4%
May 84 9% 32% 9% 11% 51%
Jun 232 5% 17% 9% 9% 33%
Jul 369 3% 12% 9% 8% 28%
Aug 325 3% 14% 11% 11% 32%
Sep 175 4% 23% 12% 11% 39%
Oct 706 1% 5% 0% 0% 3%
Nov 1621 0% 1% 0% 2% 2%
Dec 2905 0% 1% 0% 2% 2%
Heating season 12,886 0% 1% 1% 2% 2%
Cooling season 1186 4% 16% 10% 10% 33%
a
Energy use (kWh/yr) specific to the baseline case (i.e., static high R-value insulation, no shading device and dark roof with reflectance of 0.25).

Table 9
Annual heating and cooling energy savings and unmet hours for the AES integrated in a prototypical house located in four US climates.

City, State Season Baseline Integrated AES

Energy use (kWh) Unmet hours (hrs) Savings in energy use (%)a Unmet hours (hrs)

Phoenix, AZ Heating 2482 0 3% 0

Cooling 6054 20 16% 20
San Francisco, CA Heating 3709 0 2% 0
Cooling 359 0 65% 0
Boulder, CO Heating 12,886 0 2% 0
Cooling 1186 17 33% 2
Burlington, VT Heating 18,219 0 0.5% 0
Cooling 1027 8 34% 9
a
Negative values indicate an increase in energy consumption while positive values indicate savings relative to baseline.

14
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

Fig. 13. Monthly electricity consumption specific to the baseline and integrated adaptive envelope systems as well as MPVISD generation when the house is located in Boulder, CO.

addition, the integrated system offers 2%e4% savings in heating (climate zone 2B), San Francisco, CA (climate zone 3C), Boulder, CO
energy use except for November where there is a 3% increase. This (climate zone 5B), and Burlington, VT (climate zone 6A). For the
higher energy need is mainly caused by the MPVISD where the attic-integrated SIS, the low R-value setting is kept at RSI-0.5 while
penalty cannot be offset by the other systems during the month of the high R-value settings are set at RSI-5.7 for Phoenix, AZ, and RSI-
November. Overall, the integrated AESs can reduce the annual 6.8 for Boulder, CO, San Francisco, CA, and Burlington, VT. For wall-
cooling energy use by 33% (394 kWh/yr) and the heating energy use integrated SIS, the low R-value is kept at RSI-0.6 while the high R-
by 2% in the best-case scenario. In the worst case (i.e., when the value settings are set at RSI-2 for Phoenix, AZ, RSI-2.9 for San
BIPV has a negative impact on heating thermal loads), the inte- Francisco, CA, and Boulder, CO, and RSI-3.7 for Burlington, VT.
grated AESs results in a 1% penalty in heating energy use. In addi- Tables 9 and 10 summarize the main results of annual energy use
tion, it can reduce the peak electric demand by 7%. Furthermore, and peak demand savings as well as PV output when applying the
the heating set point is met for all occupied hours during the integrated AES in four US locations for the best-case scenario (i.e.,
heating season (i.e., zero unmet hours) while the unmet hours roof reflectance of 0.1). It is evident from Table 9 that the integrated
during the cooling mode dropped from 17 h for the baseline case to AES offers high potential savings in cooling energy use while it can
2 h only for the integrated AES as shown in Table 9. Fig. 13 compares result in low savings and, sometimes, penalties in heating energy
the monthly electricity consumption for the baseline and inte- use specially for cold climates. For instance, the highest percent
grated AESs as well as the electricity generated by the PVISDs. The energy saving in cooling energy use is achieved in San Francisco, CA
total electricity generated from the east, west and south PVISDs is (65%) characterized by its mild climate, followed by Burlington, VT,
11,955 kWh/yr (i.e., 54 kWh/m2/yr). This on-site generated elec- and Boulder CO with 34% and 33%, respectively. These two locations
tricity allows the AES to save 40% and 68% of the total site and are characterized by their cold winters and warm summers.
source energy consumptions, respectively, compared to the base- Phoenix, AZ, characterized by its hot summers and mild winters,
line case. show modest savings of 16% in cooling energy use when compared
to the baseline home. However, based on the absolute annual
3.7. Impact of climate on AES performance savings, the highest cooling energy savings occurs when the
dwelling is located in Phoenix, AZ (949 kWh/yr), followed by
In this section, the impact of climate on the performance of the Boulder, CO (394 kWh/yr), Burlington, VT (349 kWh/yr), and finally
integrated AESs applied to the single-family house described in San Francisco, CA (234 kWh/yr). The potential savings are corre-
section 2.3 is investigated. While the shape and size of the house lated to the cooling energy use; the higher the cooling needs, the
model described in section 2.3 are kept the same, the energy effi- higher the potential absolute energy use savings from the inte-
ciency features (i.e., R-values of walls, roofs, and windows as well as grated AES. Many factors influence the AES performance since there
HVAC systems’ efficiencies and schedules) are modified to corre- are three different technologies that act interactively together.
spond to each climate zone. Four US locations to represent different Generally, the integrated AES has low potential for heating energy
climates are considered in the analysis including Phoenix, AZ savings. Specifically, when the integrated AES is deployed, heating

Table 10
Energy efficiency and electrical peak demand performance for the AES integrated in a prototypical house located in four US climates.

City, State Baseline Savings from Integrated AES (%)a PV generation (kWh)

Site energy (kWh) Electric peak demand (kW) Site energy Electrical peak demand

Phoenix, AZ 25,283 4.88 63% 8.2% 14,729

San Francisco, CA 21,230 2.51 61% 14.2% 12,546
Boulder, CO 32,097 3.14 40% 10.8% 11,955
Burlington, VT 37,754 3.37 27% 7.5% 9689
a
Savings are relative to the baseline house.

15
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

energy savings of 3%, 2%, 2% and 0.5% can be achieved when the
single-family house is located in Phoenix, AZ, Boulder, CO, San ICSIS ¼ DECA *USPWðrd ; NÞ (1)
Francisco, CA and Burlington, VT, respectively. As indicated in
Table 9, the integrated AES can maintain similar if not better 1  ð1 þ rd ÞN
thermal comfort levels compared to the baseline for most US USPWðrd ; NÞ ¼ (2)
rd
climates. Where.
Furthermore, savings of between 7 and 15% in peak electrical
demand can be achieved by the AES (including the on-site power  ICSIS : Initial cost of AES technology to breakeven ($/m2)
generated by the PV system) as depicted in Table 10. Most of these  DECA : Annual energy cost savings ($/m2)
savings are attributed to the MPVISDs as well as cool roof system  USPW: Uniform series present worth factor
that significantly reduce the cooling energy use as discussed in  rd : discount rate
sections 3.1 and 3.2. When accounting for the worst-case scenario  N: service life
(i.e., the roof is exposed and has an effective reflectance of 0.9), the
penalties in heating energy use for Phoenix, AZ, San Francisco, CA The application of the AES technology results in a total site
and Burlington, VT are, respectively, 2%, 4.2% and 3%. In addition to annual energy cost savings of $1,727, $2,138, $1,256, $1540 for
solar shading management, the utilization of the MPVISD offers the Phoenix, AZ, San Francisco, CA, Boulder, CO, and Burlington, VT,
benefit of generating electricity that can be used on-site or supplied respectively, relative to the baseline building. These high savings
to the grid through net-metering or other mechanisms. Indeed, the are obtained mainly due to the high electricity production from the
PV production depends on the availability of solar radiation and the PV integrated shading devices. The breakeven costs, considering 20
outdoor conditions. The PV system, of 8.1 kWp, can generate up to years lifetime and 5% discount rate, range from $720 to $1611 for
14,729 kWh/yr of electricity when located in Phoenix, AZ due to its the four US climates as indicted in Table 11. As indicated earlier,
high solar resources. Considering the integrated AES, the added these breakeven costs represent the extra implementation costs for
benefit of the PV system results in a total site energy savings of 63% the AES technology (SIS and MPVISD) to be economically viable
in Phoenix, AZ. Furthermore, the PV system generates 12,546 kWh/ alternatives to static systems when applied to homes located in
yr, 11,955 kWh/yr and 9689 kWh/yr resulting when combined with various US locations considered in this study. Most of the additional
the energy efficiency benefits of AES in annual savings of 61%, 40% costs for both MPVISD and SIS are attributed to motors and con-
and 27% of the site energy when the house is located in San Fran- trollers. The breakeven costs are higher for longer lifetime periods
cisco, CA, Boulder, CO and Burlington, VT, respectively. Conse- and lower discount rates as depicted in Table 11.
quently, AES alone allow the prototypical homes to reach nearly
net-zero energy designs since it reduces the overall source en-
ergy of these homes by 95%, 82%, 68%, and 50% for respectively, San 4. Summary and conclusions
Francisco, CA, Phoenix, AZ, Boulder, CO, and Burlington, VT.
In this paper, the performance of an integrated adaptive enve-
lope system (AES) is evaluated when applied to a prototypical
house located in four US locations. The AES includes a cool roof, a
movable PV-integrated shading device (MPVISD), and switchable
3.8. Breakeven cost analysis insulation systems (SISs) deployed in the exterior walls and the
attic. The proposed MPVISD, that can be automatically set to cover
To evaluate the cost-effectiveness of the adaptive envelope the roof during the winter and to act as a shading device during the
system, a breakeven cost analysis is performed considering the life summer, is designed to render the roof to perform as a dynamic
cycle cost (LCC) approach [14]. The breakeven costs are the highest cool roof (DynamicCR) system. In addition, SISs are applied to the
additional capital costs relative to the static systems (i.e., insulation attic and exterior walls of the prototypical house and are controlled
and rooftop PV) for the AES technology to be cost-effective. Annual independently according to specific temperature-based rule-set
site energy costs are estimated using the prevalent electricity and controls. The analysis results indicated that, generally, the AESs
natural gas utility rates for the four US locations considered in this have higher potential for saving cooling energy rather than heating
study. Specifically, the electricity prices considered are 10.9 US energy. Based on the analysis of the individual technologies part of
cents/kWh, 16.6 US cents/kWh, 10 US cents/kWh and 15.1 US cents/ the AES when applied to a prototypical house located in Boulder,
kWh [49] while the natural gas prices are 1.3 US $/therm, 1.28 US CO, MPVISD showed the highest contribution to the cooling energy
$/therm, 0.74 US $/therm, 1.27 US $/therm for Phoenix, AZ, San savings (16.5%), followed by attic SIS (10%), and wall SIS (9.6%). In
Francisco, CA, Boulder, CO, and Burlington, VT, respectively [50]. terms of heating energy use, MPVISD showed a penalty of 1.1%
The breakeven cost for AES in each US location is calculated based while attic and wall SISs showed respective savings of 1% and 2%.
on Eq. (1) and Eq. (2), considering a 20e40 years life span for the Integrated AES results in 33% savings in annual cooling energy
roof system and discount rate ranging from 2% to 8% [14]. savings as well as 11% reduction in the peak electric demand. In

Table 11
Annual site energy cost savings and breakeven costs for the AES technology relative to static systems.

Climate Annual site energy cost savings ($) Break-even cost ($)

Service life, n (when r ¼ 5%) Discount rate, r (when

n ¼ 20yrs)

20 yrs 30 yrs 40 yrs 2% 8%

Phoenix, AZ 1727 1611 1987 2218 2113 1269

San Francisco, CA 2138 720 888 991 944 567
Boulder, CO 1256 721 890 993 946 568
Burlington, VT 1540 927 1144 1277 1217 731

16
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

addition, it results in either 2% savings (in the best case) or a 1% [11] Berge A, Hagentoft CE, Wahlgren P, Adl-Zarrabi B. Effect from a variable U-
value in adaptive building components with controlled internal air pressure.
penalty (worst case) in heating energy use. Moreover, it is found
Energy Procedia 2015;78:376e81. https://doi.org/10.1016/
that the performance of the AES varies significantly with the j.egypro.2015.11.677.
climate. The highest percent energy saving in cooling energy use is [12] Pflug T, Nestle N, Kuhn TE, Siroux M, Maurer C. Modeling of facade elements
achieved in San Francisco, CA (65%) characterized by its mild with switchable U-value. Energy Build 2018;164:1e13. https://doi.org/
10.1016/j.enbuild.2017.12.044.
climate, followed by Burlington, VT, Boulder CO, and Phoenix, AZ [13] Alongi A, Angelotti A, Mazzarella L. Experimental validation of a steady pe-
with 34%, 33% and 16% respectively. Based on the absolute annual riodic analytical model for Breathing Walls. Build Environ 2020;168:106509.
savings, the highest cooling energy savings occurs when the https://doi.org/10.1016/j.buildenv.2019.106509.
[14] Testa J, Krarti M. Evaluation of energy savings potential of variable reflective
dwelling is located in Phoenix, AZ (949 kWh/yr), followed by roofing systems for US buildings. Sustain. Cities Soc. 2017;31:62e73. https://
Boulder, CO (394 kWh/yr), Burlington, VT (349 kWh/yr), and finally doi.org/10.1016/j.scs.2017.01.016.
San Francisco, CA (234 kWh/yr). In addition, AES can achieve sav- [15] Pesenti M, Masera G, Fiorito F. Exploration of adaptive origami shading con-
cepts through integrated dynamic simulations. J Architect Eng 2018;24:1e14.
ings of up to 14% in peak electrical demand mostly attributed to the https://doi.org/10.1061/(ASCE)AE.1943-5568.0000323.
MPVISD as well as the cool roof system. The MPVISD can generate [16] Krarti M. Evaluation of energy performance of dynamic overhang systems for
on-site up to 14,729 kWh of electricity annually allowing the house US residential buildings. Energy Build 2021;234:110699. https://doi.org/
10.1016/j.enbuild.2020.110699.
to save 63% of the site energy consumption, compared to the [17] T€allberg R, Jelle BP, Loonen R, Gao T, Hamdy M. Comparison of the energy
baseline case, when the prototypical house is located in Phoenix, saving potential of adaptive and controllable smart windows: a state-of-the-
AZ. When considering their overall energy performance, AESs can art review and simulation studies of thermochromic, photochromic and
electrochromic technologies. Sol Energy Mater Sol Cells 2019;200:109828.
achieve net-zero energy operation level for most US homes espe-
https://doi.org/10.1016/j.solmat.2019.02.041.
cially those located in mild and hot US climates including San [18] C. Harris, S. Mumme, M. LaFrance, M. Neukomm, K. Sawyer, Grid-Interactive
Francisco, CA, and Phoenix, AZ. Efficient Buildings Technical Report Series: Windows and Opaque Envelope,
Further research work is recommended to determine the actual (n.d.). https://doi.org/10.2172/1577965.
[19] Dehwah AHA, Krarti M. Control strategies for switchable roof insulation sys-
thermal performance of cool roofs covered with PV panels through tems applied to US residential homes. Energy Build; 2020.
experimental testing and field implementation. In addition, future [20] Dehwah AHA, Krarti M. Impact of switchable roof insulation on energy per-
work should consider optimization techniques combined with formance of US residential buildings. Build Environ 2020;177:106882. https://
doi.org/10.1016/j.buildenv.2020.106882.
model predictive controls to optimize the performance of the [21] Menyhart K, Krarti M. Potential energy savings from deployment of Dynamic
adaptive envelope system. Insulation Materials for US residential buildings. Build Environ 2017;114:
203e18. https://doi.org/10.1016/j.buildenv.2016.12.009.
[22] Favoino F, Jin Q, Overend M. Design and control optimisation of adaptive
Author contribution statement insulation systems for office buildings. Part 1: adaptive technologies and
simulation framework. Energy 2017;127:301e9. https://doi.org/10.1016/
Ammar Dehwah: Investigation, Data curation, Software, Meth- j.energy.2017.03.083.
[23] Dehwah AHA, Krarti M. Cost-benefit analysis of retrofitting attic-integrated
odology, Validation, Writing-Original draft. Moncef Krarti: switchable insulation systems of existing US residential buildings. Energy
Conceptualization, Reviewing, and Editing. 2021;221. https://doi.org/10.1016/j.energy.2021.119840. 119840.
[24] Jin Q, Favoino F, Overend M. Design and control optimisation of adaptive
insulation systems for office buildings. Part 2: a parametric study for a
Declaration of competing interest temperate climate. Energy 2017;127:634e49. https://doi.org/10.1016/
j.energy.2017.03.096.
[25] Synnefa A, Santamouris M, Akbari H. Estimating the effect of using cool
The authors declare that they have no known competing
coatings on energy loads and thermal comfort in residential buildings in
financial interests or personal relationships that could have various climatic conditions. Energy Build 2007;39:1167e74. https://doi.org/
appeared to influence the work reported in this paper. 10.1016/j.enbuild.2007.01.004.
[26] Akbari H, Konopacki S. Calculating energy-saving potentials of heat-island
reduction strategies. Energy Pol 2005;33:721e56. https://doi.org/10.1016/
References j.enpol.2003.10.001.
[27] Hu J, Yu XB. Adaptive thermochromic roof system: assessment of performance
[1] Energy Information Administration (EIA). Electricity explained. 2020. https:// under different climates. Energy Build 2019;192:1e14. https://doi.org/
www.eia.gov/energyexplained/electricity/. 10.1016/j.enbuild.2019.02.040.
[2] Energy Information Administration (EIA). Natural gas explained. 2020. https:// [28] Zhang X, Lau SK, Lau SSY, Zhao Y. Photovoltaic integrated shading devices
www.eia.gov/energyexplained/natural-gas/. (PVSDs): a review. Sol Energy 2018;170:947e68. https://doi.org/10.1016/
[3] Saroglou T, Theodosiou T, Givoni B, Meir IA. A study of different envelope j.solener.2018.05.067.
scenarios towards low carbon high-rise buildings in the Mediterranean [29] Shukla AK, Sudhakar K, Baredar P. Recent advancement in BIPV product
climate - can DSF be part of the solution? Renew Sustain Energy Rev technologies: a review. Energy Build 2017;140:188e95. https://doi.org/
2019;113:109237. https://doi.org/10.1016/j.rser.2019.06.044. 10.1016/j.enbuild.2017.02.015.
[4] Hart R, Selkowitz S, Curcija C. Thermal performance and potential annual [30] Wang Y, Tian W, Zhu L, Ren J, Liu Y, Zhang J, Yuan B. Interactions between
energy impact of retrofit thin-glass triple-pane glazing in US residential building integrated photovoltaics and microclimate in urban environments.
buildings. 2019. p. 79e86. J. Sol. Energy Eng. Trans. ASME. 2006;128:168e72. https://doi.org/10.1115/
[5] Alam M, Singh H, Suresh S, Redpath DAG. Energy and economic analysis of 1.2188533.
Vacuum Insulation Panels (VIPs) used in non-domestic buildings. Appl Energy [31] Wang D, Qi T, Liu Y, Wang Y, Fan J, Wang Y, Du H. A method for evaluating
2017;188:1e8. https://doi.org/10.1016/j.apenergy.2016.11.115. both shading and power generation effects of rooftop solar PV panels for
[6] Batard A, Duforestel T, Flandin L, Yrieix B. Prediction method of the long-term different climate zones of China, vol. 205; 2020. p. 432e45. https://doi.org/
thermal performance of Vacuum Insulation Panels installed in building ther- 10.1016/j.solener.2020.05.009.
mal insulation applications. Energy Build 2018;178:1e10. https://doi.org/ [32] Dominguez A, Kleissl J, Luvall JC. Effects of solar photovoltaic panels on roof
10.1016/j.enbuild.2018.08.006. heat transfer. Sol Energy 2011;85:2244e55. https://doi.org/10.1016/
[7] Lee KO, Medina MA, Sun X, Jin X. Thermal performance of phase change j.solener.2011.06.010.
materials (PCM)-enhanced cellulose insulation in passive solar residential [33] Kapsalis V, Karamanis D. On the effect of roof added photovoltaics on build-
building walls. Sol Energy 2018;163:113e21. https://doi.org/10.1016/ ing's energy demand. Energy Build 2015;108:195e204. https://doi.org/
j.solener.2018.01.086. 10.1016/j.enbuild.2015.09.016.
[8] Berardi U, Sprengard C. An overview of and introduction to current researches [34] Miller WA, Brown E, Livezey RJ. Building-integrated photovoltaics for low-
on super insulating materials for high-performance buildings. Energy Build slope commercial roofs. 2005. https://doi.org/10.1115/1.1877514.
2020;214:109890. https://doi.org/10.1016/j.enbuild.2020.109890. [35] Zhang W, Lu L, Peng J. Evaluation of potential benefits of solar photovoltaic
[9] K.A. Cort, J.A. McIntosh, G.P. Sullivan, T.A. Ashley, C.E. Metzger, N. Fernandez, shadings in Hong Kong. Energy 2017;137:1152e8. https://doi.org/10.1016/
Testing the Performance and Dynamic Control of Energy-Efficient Cellular j.energy.2017.04.166.
Shades in the PNNL Lab Homes, (n.d.). https://doi.org/10.2172/1477792. [36] Jayathissa P, Luzzatto M, Schmidli J, Hofer J, Nagy Z, Schlueter A. Optimising
[10] Luo Y, Zhang L, Bozlar M, Liu Z, Guo H, Meggers F. Active building envelope building net energy demand with dynamic BIPV shading. Appl Energy
systems toward renewable and sustainable energy. Renew Sustain Energy Rev 2017;202:726e35. https://doi.org/10.1016/j.apenergy.2017.05.083.
2019;104:470e91. https://doi.org/10.1016/j.rser.2019.01.005. [37] Pisello AL. State of the art on the development of cool coatings for buildings

17
A.H.A. Dehwah and M. Krarti Energy 233 (2021) 121165

and cities. Sol Energy 2017;144:660e80. https://doi.org/10.1016/ doi.org/10.1016/j.enbuild.2020.110649.

j.solener.2017.01.068. [43] Dabbagh M, Krarti M. Evaluation of the performance for a dynamic insulation
[38] Kolokotroni M, Shittu E, Santos T, Ramowski L, Mollard A, Rowe K, Wilson E, system suitable for switchable building envelope. Energy Build 2020;222:
Filho JP de B, Novieto D. Cool roofs: high tech low cost solution for energy 110025. https://doi.org/10.1016/j.enbuild.2020.110025.
efficiency and thermal comfort in low rise low income houses in high solar [44] ASHRAE, ANSI/ASHRAE/IES. Standard 90.2-2018, energy efficient design of
radiation countries. Energy Build 2018;176:58e70. https://doi.org/10.1016/ low-rise residential buildings. Atlanta, GA: American Society of Heating,
j.enbuild.2018.07.005. Refrigeration and Air-Conditioning Engineers; 2018.
[39] Testa J, Krarti M. A review of benefits and limitations of static and switchable [45] US Department of Energy (DOE). Residential prototype building models. Build.
cool roof systems. Renew Sustain Energy Rev 2017;77:451e60. https:// Energy Codes Progr; 2020 (accessed March 28, 2021), https://www.
doi.org/10.1016/j.rser.2017.04.030. energycodes.gov/development/commercial/prototype_models.
[40] Levinson R, Akbari H. Potential benefits of cool roofs on commercial buildings: [46] V Mendon V, Lucas RG, Goel S. Cost-effectiveness analysis of the 2009 and
conserving energy, saving money, and reducing emission of greenhouse gases 2012 IECC residential provisions. Richland, WA: DOE; 2012.
and air pollutants. Energy Effic 2010;3:53e109. https://doi.org/10.1007/ [47] US Department of Energy (DOE). EnergyPlusTM version 9.1.0 documentation:
s12053-008-9038-2. engineering reference. 2019.
[41] Anand J, Sailor DJ, Baniassadi A. The relative role of solar reflectance and [48] US Department of Energy (DOE). EnergyPlus EMS application guide. 2019.
thermal emittance for passive daytime radiative cooling technologies applied [49] Energy Information Administration (EIA), State Electricity Profiles, (n.d.).
to rooftops. Sustain. Cities Soc. 2021;65:102612. https://doi.org/10.1016/ https://www.eia.gov/electricity/state/(accessed July 10, 2020).
j.scs.2020.102612. [50] Energy Information Administration (EIA), Natural Gas Prices, (n.d.).https://
[42] Dehwah AHA, Krarti M. Control strategies for switchable roof insulation sys- www.eia.gov/dnav/ng/ng_pri_sum_a_EPG0_PRS_DMcf_a.htm (accessed July
tems applied to US residential homes. Energy Build 2021;231:110649. https:// 10, 2020).

18