- Email: [email protected]
Evaluation of energy savings potential of variable reflective roofing systems for US buildings
Accepted Manuscript Title: Evaluation of energy savings potential of variable reflective roofing systems for US buildings Author: Jenna Testa Moncef Krarti PII: DOI: Reference:
S2210-6707(16)30278-5 http://dx.doi.org/doi:10.1016/j.scs.2017.01.016 SCS 575
To appear in: Received date: Revised date: Accepted date:
24-8-2016 10-1-2017 11-1-2017
Please cite this article as: Testa, J., and Krarti, M.,Evaluation of energy savings potential of variable reflective roofing systems for US buildings, Sustainable Cities and Society (2017), http://dx.doi.org/10.1016/j.scs.2017.01.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Evaluation of energy savings potential of variable reflective roofing systems for US buildings Jenna Testa and Moncef Krarti*, PhD, PE, LEED-AP Building Systems Program, College of Engineering & Applied Science, University of Colorado, Boulder, CO 80303, USA
ip t
ABSTRACT
M
an
us
cr
This paper explores the potential energy savings of using dynamic cool roofs (DCRs) with seasonally variable reflective surface for various building prototypes when compared to static cool roofs. The analysis conducted demonstrates that the additional energy savings from DCRs depend largely on the climate, insulation level, and reflectance of the roof, as well as but to a lesser extent on the building thermal mass. This study shows that older buildings, with low insulation levels, in colder climates are the best target for retrofit using DCRs. Specifically, the results show that when a variable reflective costing is applied to low insulation buildings, source energy savings can be achieved and range from 4.33 to 19.44 MJ/m2 (i.e., 1.6 to 4.9%) for residential units and from 1.17 to 18.00 MJ/m2 (i.e., 0.3 to 3.9%) for offices. Based on an economic analysis, it is found that the breakeven cost for a variable reflectance coating system with a 22-year life span ranges from 0.80 to 4.84 $/m2 for residential buildings, and from 0.86 to 4.92 $/m2 for commercial buildings.
Keywords:
Ac ce pt e
d
Cool Roofs; Cost Analysis; Energy Use Savings, Variable Reflective Coatings
(*) Correspondent Author: [email protected]
1
Page 1 of 23
1. Introduction
cr
ip t
A significant portion of the energy consumed in the United States (US) attributed to buildings. According to the US Energy Information Administration, commercial and residential buildings consumed 40% of the nation’s total end-use energy in 2015 [1]. In 2010, the residential building sector consumed a majority of its source energy in space heating – 28%, space cooling – 15%, and water heating – 13%. While the commercial building sector consumed a majority of its source energy in lighting – 20%, space heating – 16%, and space cooling – 15% [2]. A large portion of the heating and cooling energy use is associated to building envelope systems such as windows, walls, and roofs.
M
an
us
Heat gain through a roof, specifically for a single-story building, can be a dominant component of a building’s total cooling load since it is highly susceptible to solar radiation [3]. One common solution to reducing the cooling load due to gains through the roof is to increase the solar reflectance of the roof by installing a cool roof. Due to their low solar absorptivity and high thermal emissivity, cool roofs maintain lower surface temperatures and reduce the heat flow into buildings [4]. Indeed, the low solar absorption increases the solar reflectance and the high thermal emittance enhances the roof surface’s ability to radiate any absorbed solar energy [5, 6]. Based on the literature, an average standard roof reflectance ranges from 0.20 to 0.30, an average initial low-sloped cool roof white material ranges from 0.70 to 0.80, and an average aged low-sloped cool roof white material ranges from 0.55 to 0.70 [3, 5, 7, 9]. The typical thermal emittance most roofing materials, regardless of color or reflectance is typically between 0.80 and 0.90 [4, 8, 6].
Ac ce pt e
d
In addition to white cool roofing materials, there has been recent developments in colored cool roofing materials specifically for steep-sloped roofs. A cool colored surface absorbs the visible part of the spectrum in order to appear a specific color, but is highly reflecting in the near-infrared (NIR) part of the spectrum which accounts for nearly 50% of solar radiation [10]. Synnefa et al. [11] have developed prototypes of cool colored coatings using near infrared reflective pigments added to acryl-based coatings which can increase solar reflectance by 0.06 (light blue) and 0.22 (black) [11]. Levinson et al. [12] have developed coating systems for various types of substrates and found that coated steel and glazed clay-tile roofing achieved NIR reflectances as high as 0.50 and 0.75, respectively, using a one-coat system [12]. Pisello et al. [13] have developed a new coating system for colored clay tiles which includes a white englobe above the substrate and a sodium silicate based binder with pigment added for color. The study found that the new coatings are able to reflect the NIR by 75% which is 10% more than traditional tiles with equivalent visible appearance [13]. Several field studies have monitored and documented cooling energy savings and cooling peak energy savings from increasing the reflectance of static cool roofs in US warm climates including California, Florida, and Nevada. The studies found that increasing the reflectance during summer months can reduce both the cooling energy usage and the peak demand. A summary of these studies is provided in Table 1.
Table 1: Summary of benefits for static cool roofs based on measured performance for US climates Daily Reduced Roof Roof Location Building Type System Savings Demand Area m2
R-Value
Duct
Δρ
Wh/m2
%
W/m2
%
2
(K.m /W)
California
2
Page 2 of 23
1.4 3.3 RB 3.3 1.9 3.3 0 1.9 RB 5.3 5.1
Interior Plenum Plenum Ceiling Attic Plenum Interior Attic None None None
0.36 0.35 0.44 0.60 0.59 0.40 0.40 0.40 0.61 0.54 0.61
68 39 4 47 14 10 20 11 72 45 69
Strip mall Residence “ “ “ “ “ “ “ “ “ “ “
1161 130 111 121 121 139 158 167 130 139 167 84 223
1.9 1.9 0 0 1.9 3.3 1.2 4.4 1.9 3.3 3.3 0 5.3
Plenum Attic Attic Attic Attic Attic Attic Attic Attic Attic Attic Attic Attic
0.46 n/a 0.63 0.39 0.52 0.42 0.44 0.51 0.30 0.44 0.42 0.53 0.65
7.5 58 137 116 85 31 73 24 64 23 5.4 68 n/a
Small commercial Small commercial
14.9 14.9
3.2 3.2
None None
0.45 0.45
Retail store
9300
Plenum
0.70
3.34 2.37 1.61 6.78 3.55 n/a n/a n/a 10 5 5
12 8 9 20 25 n/a n/a n/a 50 12 6
25 22 43 26 25 13 20 11 15 10 2 25 17
0.65 1.51 7.75 7.64 5.49 1.61 6.24 n/a 3.44 2.58 1.83 5.92 n/a
29 12 28 29 28 11 23 n/a 16 16 12 30 n/a
31 39
1.2 1.6
n/a n/a
n/a n/a
39
11
3.77
14
us
an
M 2.1
18 13 2 46 63 17 26 39 52 17 4
ip t
2945 2211 3056 89 167 2285 455 557 1600 570 4900
cr
Medical office Medical office Retail store School bungalow Residence Office Museum Hospice Retail store Elementary school Cold storage
d
Davis [18] Gilroy [18] San Jose [18] Sacramento [15] Sacramento [15] Sacramento [16] Sacramento [16] Sacramento [16] Sacramento [21] San Marcus [21] Reedley [21] Florida Cocoa Beach [17] Cape Canaveral [14] Cocoa Beach [14] Cocoa Beach [14] Cocoa Beach [14] Cocoa Beach [14] Merritt Island [14] Merritt Island [14] Miami [14] Palm Bay [14] Palm Bay [14] West Florida [14] Lakeland [14] Nevada Battle Mount. [20] Carlin [20] Texas Austin [19]
Ac ce pt e
In addition to building monitoring studies, several computer energy simulations have been conducted to analyze the impact on energy use for both US residential and commercial buildings. Unlike the case of Table 1 for measured data, the simulation analyses explore the effects of cool roofs in various climate zones and look at both cooling savings and heating penalties. The literature reported simulation based results specific to residential and commercial buildings are summarized in Table 2 for US climates.
Table 2: Summary of simulation results for residential and office buildings
Study [3, 22]
[23]
Building Prototype Location
Wall R
Roof R
(m2K/W)
(m2K/W)
Δρ
Annual Cooling Savings 2 kWh/m (%)
Annual Heating Penalty 2 kWh/m (%)
Residential Old
11 U.S. Cities
1.2
1.9
0.30
1.1-5.2 (7-17)
0-4.98 (2-5)
Residential New
“
1.9
3.3
0.30
0.4-2.8 (6-15)
0-3.81 (0-5)
Office Old
“
1.2
1.9
0.45
1.8-6.0 (5-9)
0-5.57 (0-14)
Office New
“
1.9
3.3
0.45
1.1-3.3 (3-7)
0-3.22 (0-12)
Residential Old
239 U.S. Cities
0.9
1.9
0.30
0.5-8.6 (4-11)
0-6.45 (0-2)
Residential New
“
2.3
5.3
0.30
0.1-3.6 (1-8)
0-2.05 (0-2)
3
Page 3 of 23
[24]
“
1.1
1.9
0.40
4.8-8.2 (4-8)
0-3.81 (0-3)
Office New
“
2.3
5.3
0.40
1.2-2.8 (2-4)
0-1.47 (0-6)
Office Old
236 U.S. Cities
1.1
1.2
0.35
0.5-11.5 (n/a)
0-7.03 (n/a)
Office New
“
2.3
3.3
0.35
0.1-4.1 (n/a)
0-2.93 (n/a)
Residential New
27 World Cities
0.5
1.2
0.40
8-29 (11-35)*
n/a
Residential New
“
0.5
1.2
0.65
9-48 (18-54)*
0-16.99 (n/a)*
ip t
[4]
Office Old
*Numbers for this study are presented as annual load savings and penalty, not energy use savings and penalties
M
an
us
cr
In general, the existing literature on cool roofs demonstrates that location, building construction, building type, and increase in reflectance all have an impact on cool roof energy savings. Akbari and Konopacki [23] results show that the cooling energy savings increases as cooling degree-days (CDDs) increase, and the heating energy penalties increase as the heating degree-days (HDDs) increase [23]. A high reflectance roof can lower roof temperatures which reduce the heat flow from the roof into the buildings, decreasing the need for space cooling during the cooling season. On the other hand, a cool roof may also increase the need for heating energy during the heating period. This can be relatively significant in colder climates [4]. For example, Konopacki et al. [3] found that new offices with an electric heat pump in Chicago and Philadelphia, new residences with an electric heat pump in New York City, and new residences with a gas furnace in Philadelphia all displayed net energy deficits or savings very close to zero. Thus, the literature shows that the best climates for cool roofs are characterized by a long cooling season and a short heating season.
Ac ce pt e
d
Building construction also impacts the cooling savings and heating penalty associated to static cool roofs. Older vintage buildings with low insulation can have higher cooling savings during the cooling season and higher heating penalties during the heating season [3, 4, 22, 23]. Typically, residential buildings tend to have lower absolute cooling savings and higher absolute heating penalties than commercial buildings because they are more shell dominated. Moreover, commercial buildings have higher cooling savings and lower heating penalties since they tend to have higher internal heat gains which increase overall cooling and decrease overall heating. Residential buildings, however, have a higher percentage of cooling savings since a larger portion of the cooling load is attributed to the envelope systems [3, 22, 23]. It is also reported that buildings with longer operation schedules show larger savings per roof area than those with shorter schedules. Konopacki et al. [3] and Akbari et al. [22] found that cooling energy savings and heating energy penalties exhibit a linear relationship with the changes in roof albedo for each building type and location. For example, if the difference in roof reflectance is increased from 0.2 to 0.4, the cooling savings and the heating penalty would roughly double. Thus, to estimate the heating penalties and cooling savings from changes in roof reflectance for the same simulated building, the results can be adjusted by a ratio of the reflectance values [3]. It is clear from the existing literature that while a cool roof has definite benefits, the largest disadvantage is the heating penalty which would be of the greatest concern to a customer evaluating the cos-effectiveness of a cool roof. A building with a cool roof begins to experience an energy penalty as the building transitions to heating mode due to the increase in reflected solar energy that is no longer transferred into heat gain at the roof surface [25]. An ideal alternative would be the ability to have a roofing system which is highly reflective in the times when the building is in cooling mode and highly absorptive when the building transitions to heating mode [25]. The best solution would be a dynamic cool roofs (DRCs) with switchable roof materials or variable reflectance coatings (VRCs) which have the ability to change reflectance throughout the year.
4
Page 4 of 23
an
us
cr
ip t
There are currently only two types of variable reflectance roofing systems that have been developed and explored in literature: thermochromic (TC) coatings and heterogenous directional reflective materials (DRMs). Typically, properties of TC coatings are controlled through temperature variations [26]. A recent experimental analysis has found that for a TC coating applied to a building cool roof the low temperature reflectance is about 0.3 and the high temperature reflectance is about 0.55. The reflectance begins to change at the low temperature set point [25]. The analysis has indicated several challenges with the evaluated TC coating including the fact that the coating requires a 11°C temperature difference to completely shift from the low reflectance to the high reflectance values. In the other hand, a DRM roofing system varies reflectance throughout the year without changing the actual material itself. It is controlled by the angle of the sun at different times of the year; thus it is seasonally controlled. It can be designed to reflect more sun during the cooling months and absorb more sun during the heating months. This would have to be tailored to the location and climate. Hooshangi [27] explores the performance of DRMs with corrugated surfaces including reflective sides facing towards the sky and absorptive sides facing towards the ground. The angle of the corrugation depends on sun location in the summer and winter. The reflective sides reflect sunlight during the summer when the sun is high in the sky and the absorptive sides absorb most of the sunlight during the winter when the sun is low in the sky [27]. Recently, Akbari and Touchaei [28] have developed a model to calculate the hourly reflectance of DRMs as a function of zenith and azimuth angles of the suns position [28]. No analysis has been reported to compare potential energy use savings from DRM roofs compared to static reflectance roofs.
2. Analysis Approach Approach Description
d
M
From the limited reported evaluation [29], it is clear that switchable or variable reflectance roof systems such as DCRs and VRCs could provide additional energy and operational cost savings compared to static cool roofs, there is lack of a detailed analysis to determine the impact of various design and operation conditions on the performance of switchable roof systems for buildings. The main objective of this paper is to assess the additional energy and cost savings potential associated with hypothetical switchable reflectance roof systems including DCRs when compared to static cool roofs for both residential and commercial buildings.
Ac ce pt e
The study outlined in this paper explored the energy savings impact of a switchable roof using a simulation environment developed by Park et al. [29-30]. The simulation environment utilizes RC thermal network and the heat balance technique, and has been validated using EnergyPlus [29]. The tool was selected based on modeling speed, and ease of ability of modeling switchable roof materials. For this study, the roof reflectance has been changed primarily changed on a seasonal basis, but the simulation environment has the ability to change the reflectance on hourly or sub-hourly basis. The analysis, summarized in this paper, includes an estimation of the potential energy savings associated with DCRs as a function of several building parameters: building type, insulation level, thermal mass, location/climate, and roof reflectance. The two building types explored in this study are a small office building and a single-family residential building. The major characteristics of the two buildings have been selected based on the US Department of Energy prototypes [3, 4, 22, 23]. A detailed description of the building geometry, window characteristics, slab-on-grade, and internal loads for both building models is provided in Table 3. Table 3: Model characteristics of residential and commercial buildings Small Office Building
Residential Building
5
Page 5 of 23
Building shape
Total Floor Area (m2) Window-to-Wall Ratio Number of Floors Floor to ceiling height (m)
143 (11.96 m x 11.96 m) 15% for all orientations 1 2.44
ip t
510 (27.68 m x 18.44 m) 21% for all orientations 1 3
U-factor (W / m2 K) SHGC Visible transmittance Emittance
cr
Windows
Slab-On-Grade 2
Ueff (W / m K)
Internal Loads 2
10.76 6.78 26 - 120 W/person
Ac ce pt e
Lighting power density (W/m ) 2 Plug load power density (W/m ) Occupancy HVAC Systems Cooling System Heating System
d
Ventilation
M
an
0.34 Peak: 0.2016 cfm/sf of exterior wall surface area (when fans turn off). Off Peak: 25% of peak infiltration rate (when fans turn on) 0.0025 m3/s x people + 0.0003 m3/s * area = 0.22 m3/s
Infiltration
3.24 0.39 0.40 0.84
us
2.95 0.39 0.40 0.84
Packaged AC unit with COP of 3; electricity Furnace with efficiency of 0.85; natural gas
2
0.70 2
4 cm /m wall surface area
None 7.53 2.69 3 - 100 W/person Packaged AC unit with COP of 3; electricity Furnace with efficiency of 0.85; natural gas
The existing literature indicates that most standards use an aged high reflectance of 0.55 for a cool roof, a low reflectance of 0.25 to 0.30 for a standard roof, and a thermal emittance of 0.90 for both [3, 4, 22, 23]. Additionally, the reported potential for the high reflectance value of a thermochromic coated roof is approximately 0.55 and the low reflectance value is approximately 0.30 with a thermal emittance of approximately 0.90 [25]. Thus, the base high reflectance value used in the analysis for both the cool roof and the switchable roof was 0.55, and the base low reflectance value used for both the standard roof and the switchable roof was 0.30. A thermal emittance of 0.90 was used for all roof types. For the analysis, three different mass levels and insulation levels have been explored: low, medium, and high. The same materials were used for the small office and single-family residential prototypes. The light weight construction consists of wood frame walls and a wood deck roof, the medium weight construction consists of concrete walls with a thin concrete deck, and the heavy weight construction consists of one wythe brick walls with wood frame and a thicker concrete deck roof. In addition to the light, medium, and heavy weight constructions, a fourth extra heavy mass which consisted of a 2 wythe brick wall and very thick concrete roof deck (410 kg/m2 average envelope weight) has been considered to further analyze the effect of thermal mass. In 6
Page 6 of 23
Table 4: Insulation and building mass parameters Low
Medium
High
0.90 1.85
2.12 3.46
3.42 4.85
67 50
150 148
cr
Parameter
215 214
an
Mass (kg/m2) Wall Roof
us
2
R-Value (m K/W) Wall Roof
ip t
addition, three insulation configurations have been evaluated including low, medium, and high insulation levels for each building type. The low and medium insulation levels are similar to the old (pre-1980) and new (post1980) vintages that were used in the literature. A summary of the insulation levels and mass levels analyzed can be found in Table 4.
Ac ce pt e
d
M
Seven different locations across the United States were selected to analyze: New Orleans, LA; Atlanta, GA; Las Vegas, NV; Baltimore, MD; Portland, OR; Chicago, IL; and St. Paul, MN. The locations have been selected to be presentative of ASHRAE US climate zones (2 – 6). The heating degree days (HDD), cooling degree days (CDD), and average daily global horizontal radiation (GHR) for each location are shown in Figure 1. Heating and cooling degree-days shown are at a base temperature of 18oC.
Figure 1: Location and climate characteristics of US cities used in the analysis (HDD and CDD are expressed in o C-days and GHR is expressed in Wh/m2) Finally, the roof reflectance has been also varied in the analysis in order to determine the additional energy savings potential should the range for low-high reflectance values of VRCs expands in the future. In addition to the 0.30/0.55 reflectance switchable roof, a 0.30/0.65, 0.30/0.75, and 0.30/0.85 switchable roof have been also modeled and analyzed.
7
Page 7 of 23
3. Discussion of Selected Results
cr
ip t
As was previously mentioned, the primary purpose of the study presented in this paper is to determine the additional energy savings for dynamic cool roofs with switchable reflectance coatings when compared to traditional static cool roofs, and how specific building design and climatic parameters effect those savings. All annual energy use savings or penalties are normalized per unit of roof area. Therefore, the cooling and heating energy usages are expressed in kWh/m2 while the source energy savings are presented in MJ/m2. Since switchable roofs affect both cooling and cooling thermal loads, source energy savings of dynamic cool roofs are compared to those obtained for static cool roofs are estimated and used as comparative indicators to assess the impact of various parameters on the performance of DCRs. The multipliers used to convert site to source energy uses are 3.14 for electricity and 1.01 for natural gas. It should be noted that the energy use savings associated with static cool roof relative to standard roofs have been estimated from this study and have been found to be generally consistent and within the same range as those listed in Tables 1 and 2 including those reported by Akbari et al. [22].
an
us
For this study, a seasonal control was used such that the roof reflectance was changed twice throughout the year – once at the beginning of the cooling season and once at the end of the cooling season. In order to determine the seasonal switchover period for each climate, a degree-day based method has been utilized. For each month, a high reflectance roof is used when the cooling degree days are higher than the heating degree days. Otherwise, a low reflectance roof is considered. Using detailed simulation analysis, it was found that a reasonable base temperature for the degree day selection method is 15.5°C for the residential building and 13°C for the office building.
Ac ce pt e
d
M
3.1 Seasonal control vs. theoretical optimal control While this study focused on a seasonal switchover in roof reflectance, an optimized control strategy could lead to potential additional energy savings. If a switchable roof had potential to change reflectance from day to day or even hour to hour, then theoretically a switchable roof (i.e., DCR) could have the same heating energy use as a standard roof and the same cooling energy use as a static cool roof. A daily change would be beneficial during swing months when some days may require more cooling and others may require more heating. An hourly switchable roof could take advantage of periods during swing months when heating is necessary during the early morning and late afternoon, but cooling is necessary in the middle of the day. Of course, the energy savings of a switchable roof can only be achieved during hours with daylight. To estimate of DRC optimal controls, energy savings were calculated by combining heating energy use of a low reflectance roof and the cooling energy use of a high reflectance roof throughout the year. The results show that there are generally minimal additional savings using optimal rather than seasonal controls. The additional source savings for US residential buildings range from 0.30 to 2.7 MJ/m2 (0.05 to 0.98%) with cost savings ranging from less than $0.01/m2 to $0.04/m2. The additional source savings for US commercial buildings range from 0.41 to 3.0 MJ/m2 (0.09 to 0.95%) with cost savings ranging from less than $0.01/m2 to $0.03/m2. This analysis shows that a seasonal control is sufficient for switchable reflectance roofs. 3.2 Impact of insulation insulation level The first parameter considered in the parametric analysis of DCR performance was the overall building insulation level. The insulation level has been varied for a light mass, medium mass, and heavy mass building with a switchable roof reflectance of 0.30/0.55 and compared to a cool roof with a reflectance of 0.55. The ranges of source energy savings for each building type and insulation level are summarized in Table 5. Table 5: Ranges of source annual energy savings of different insulation levels for dynamic cool roof when compared to a static cool roof Insulation Level Building Prototype 8
Page 8 of 23
Residential Building Low Medium High
Office Building
MJ/m2
%
MJ/m2
%
4.3 - 19.4 2.3 - 10.2 1.5 - 7.3
1.6 - 4.9 1.3 - 4.1 0.9 - 3.5
1.2 - 18.0 0.4 - 9.4 0.2 - 6.5
0.3 - 3.9 0.1 - 2.6 0.1 - 1.9
Ac ce pt e
d
M
an
us
cr
ip t
The results of Table 5 show that the amount of insulation of the building is an important factor that affects the energy savings resulting from a switchable reflectance roof when compared to a static cool roof. In particular, that annual energy use is lower for DRCs, regardless of the insulation level. However, when the R-value is large (the building is well insulated), the heat transfer between the surface of the roof and the interior of the building is small and the impact on the energy use of using a DRC is not as important. Therefore, implementing a switchable reflectance roof would be more beneficial for building with low insulation levels. Figure 2 illustrates the impact of the insulation R-value for a medium mass office building located in various US sites. In particular, Figure 2 shows that the source energy savings from DRCs decrease with building R-value. A similar trend as that of Figure 2 has been noted for both residential and commercial buildings at all thermal mass levels.
Figure 2: Effect of R-value on source energy savings resulting from a switchable roof when compared to a static cool roof for an office building with a medium mass level
3.3 Impact of building mass mass As noted earlier, the building mass level was varied to include low, medium, and high levels as outlined in Table 4. The impact of the mass level on DRC performance has been investigated when the buildings have DRCs with switchable roof reflectance of 0.30/0.55 and static cool roofs with a reflectance of 0.55. The ranges of source energy savings found for various building types and mass levels are summarized in Table 6. Table 6: Ranges of source energy savings of different mass levels for dynamic cool roofs when compared to static cool roof 9
Page 9 of 23
Mass Level
Building Prototype Residential Building MJ/m
2
Office Building
%
MJ/m2
%
1.5 - 16.9
0.9 - 3.9
0.2 - 15.6
0.1 - 3.2
Medium Heavy
1.7 - 19.4 1.7 - 18.3
1.1 - 4.9 1.1 - 4.8
0.3 - 18.0 0.3 - 17.4
0.1 - 3.9 0.1 - 3.8
ip t
Light
Ac ce pt e
d
M
an
us
cr
Interestingly, Table 6 indicates that the switchable roof savings when compared to a static cool roof do not steadily increase with thermal mass. Rather the savings increase to a maximum level as mass increases, then start to decrease past a certain mass level. To further demonstrate this result, a fourth extra heavy building has been considered in the analysis. Figure 3 presents the effect of thermal mass level for a medium insulated office building located in various US sites. The trend of Figure 3 has been observed for all building types at insulation levels considered in the analysis. Figure 3 shows that once the mass reaches a medium (or heavy level for some climates), the savings begin to decrease.
Figure 3: Effect of building mass on source energy savings for resulting from a switchable roof when compared to a static cool roof for an office building with a medium insulation level The trend noted in Figure 3 is attributed to the fact that as thermal mass increases, a variable reflectance roof initially reduces the building heating thermal load when compared to a static cool roof. Indeed, the lower reflectance roof during winter and swing season experiences more solar gains, a higher mass building initially save more energy due to thermal storage capacity of the building mass that provides free heating during colder times of the day. As can be seen in Figure 3, this initial increase in savings is more prominent when the building is located in heating dominated climates. The savings begin to decrease as the effect of thermal mass diminishes. A static cool roof does not experience as much solar gains as a variable reflectance roof. Therefore, the building with static cool roof does not reach an 10
Page 10 of 23
Ac ce pt e
d
M
an
us
cr
ip t
equilibrium temperature as fast as for the building with variable reflectance roof and thus has more capacity to store heat. This behavior is illustrated in Figure 4 which compares the heating loads for an office building with medium insulation in Atlanta. The slope/rate of decrease in heating load for a static cool roof is larger than that for a variable reflectance roof once the mass gets above a medium threshold level; thus the savings in heating thermal load between the two roof systems decreases as the building mass continues to increase past a medium threshold level.
Figure 4: Comparison of heating load and slope of decrease in thermal heating load for a static cool roof and a variable reflectance roof It should be noted, however, that the impact of thermal insulation on DRC performance is more significant than that of building mass as noted in Figures 5 and 6 showing the annual DRC source energy savings for an office building located in Baltimore, MD. It is clear from Figures 5 and 6 that the insulation level has a much larger impact on the energy savings than mass does. For a building with a low insulation level, DRC can save between 1 and 12 MJ/m2 more source energy than for a building with a high insulation level. For a building with medium mass, DRC can only save between 0.85 and 3 MJ/m2 more source energy than for a building with low or high mass levels. Moreover, the impact on DRC performance diminishes as the insulation level increases, whereas this impact increases to an optimum threshold then decreases as the mass level increases.
11
Page 11 of 23
ip t cr us
Ac ce pt e
d
M
an
Figure 5: Effect of mass on source energy savings for an office building located in Baltimore, MD for various insulation levels
Figure 6: Effect of insulation on source energy savings for an office building located in Baltimore, MD for various mass levels 3.4 Impact of roof reflectance reflectance As part of the parametric analysis, the impact of roof reflectance is evaluated on the DRC performance. As was previously mentioned, the base values used throughout the analysis are 0.55 for high reflectance and 0.30 for low reflectance roof coatings. This parameter is particularly of interest should the technology improves and allows for lower low reflectance or higher high reflectance. Three additional reflectance variations have been modeled including: 0.30/0.65, 0.30/0.75, 0.3/0.85. Figures 7 and 8 summarize the analysis results obtained for 12
Page 12 of 23
d
M
an
us
cr
ip t
an office located in Atlanta and Chicago, respectively. Specifically, Figures 7 and 8 show source energy savings obtained from a static cool roof when compared to a standard roof and those obtained from a switchable reflectance roof when compared to a standard roof. The results provided in Figures 7 and 8 are specific for a light mass residential building with medium insulation, but the trend is the same for all other building configurations and climates. The results show that there is a linear relationship between savings and the difference in roof high-low reflectance values. They also demonstrate that as the high reflectance increases, the DRC energy savings increases at a faster rate than the static cool roof savings. This result is due to the fact that as the reflectance on a static cool roof increases, both the cooling savings and the heating penalties increase. Whereas for a DRC with a switchable reflectance coating, only the cooling savings increase while the heating penalties remain minimal. This behavior is more pronounced for warmer climates when comparing the results obtained Atlanta (i.e., Figure 7) with those obtained for Chicago (i.e., Figure 8).
Ac ce pt e
Figure 7: Effect of high reflectance on cool roof savings and variable roof savings when compared to a standard low reflectance roof in Atlanta, GA
Figure 8: Effect of high reflectance on cool roof savings and variable roof savings when compared to a standard low reflectance roof in Chicago, IL 3.5 Impact of climate climate DRC performance has been evaluated for both residential and commercial buildings located in various US climates. Table 7 summarizes the results of the analysis using the heating degree-days (HDDs) estimated using a 13
Page 13 of 23
base temperature of 18oC. Specifically, Table 7 lists the ranges of annual source energy savings obtained for a switchable roof reflectance of 0.30/0.55 compared to a static cool roof with a reflectance of 0.55 for various US locations considered in the analysis. Since the additional energy savings achieved by a switchable roof come from a decrease in heating energy use, the HDDs were used to represent the climatic characteristics for all US locations. Table 7: Ranges of annual source energy savings for a switchable roof when compared to a static cool roof
ip t
787 1264 1571 2426 2574 3607 4379
cr
New Orleans Las Vegas Atlanta Portland Baltimore Chicago St. Paul
Building Prototype Residential Building Office Building 2 MJ/m % MJ/m2 % 1.5 - 5.5 0.9 - 2.2 0.2 - 2.0 0.1 - 0.6 3.5 - 12.6 1.6 - 3.4 1.2 - 6.8 0.4 - 2.3 3.5 - 11.4 2.0 - 3.8 1.9 - 7.1 0.8 - 2.3 4.5 - 13.8 3.1 - 4.9 2.7 - 9.4 1.7 - 3.9 5.2 - 16.4 2.4 - 4.1 3.9 - 12.8 1.5 - 3.4 6.2 - 17.6 2.1 - 3.3 4.7 - 14.9 1.4 - 3.0 6.9 - 19.4 1.2 - 3.0 6.1 - 18.0 15 - 2.9
us
HDD
an
Location
Ac ce pt e
d
M
It is clear from Table 7 that there is a strong relationship between HDDs and DRC source energy savings. Figures 9 and 10 illustrates the variation of DRC source energy savings as a function of HDD for respectively, a light weight residential and an office building with three insulation levels. A similar trend was found for all building configurations. It should be noted that the results from Portland were not included in the regression analysis because they were significantly lower than the trend. This is most likely because Portland has low solar radiation in the winter months when compared to the other locations. In particular, the results of Table 7 and Figures 9 and 10 show that the DRC energy savings increase as HDD increases. Therefore, implementing a dynamic switchable roof would be more beneficial in colder climates.
Figure 9: Effect of heating degree days on annual DRC source energy savings for a light mass residential building
14
Page 14 of 23
ip t cr us
an
Figure 10: Effect of heating degree days on annual DRC source energy savings for a light mass office building
M
3.5.1 Simplified method for energy savings calculation The relationship between the annual DRC source energy savings and HDD has been obtained for all insulation and mass levels for both residential and commercial buildings as noted by equations (1) through (18) listed in Table 8. The method based on the equations (1) - (18) can be utilized to estimate the potential benefits of dynamic cool roofs compared to static cool roofs.
Ac ce pt e
d
Table 8: Regression analysis results to estimate source energy savings (SES) expressed MJ/m2 as a function of heating degree days (HDD) expressed in oC-days Mass Residential Office Equation R2 Equation R2 Light
SES = 7.120 ln ( HDD ) − 42.5560.97
(1)
SES = 8.416 ln ( HDD ) − 55.411 0.99 (10)
Medium R
SES = 4.120 ln ( HDD ) − 24.8570.98
(2)
SES = 4.719 ln ( HDD ) − 31.449 0.98 (11)
High R
SES = 3.068ln ( HDD ) − 18.8340.99
(3)
SES = 3.427 ln ( HDD ) − 23.027 0.98 (12)
SES = 7.474 ln ( HDD ) − 42.9560.94
(4)
SES = 9.027 ln ( HDD ) − 58.342 0.99 (13)
0.96
(5)
Low R
Medium Low R
Medium R
SES = 4.173ln ( HDD ) − 24.718
SES = 4.926 ln ( HDD ) − 32.553
0.98
(14)
15
Page 15 of 23
SES = 3.094 ln ( HDD ) − 18.6820.98
(6)
SES = 3.583ln ( HDD ) − 23.933 0.98 (15)
Low R
SES = 7.121ln ( HDD ) − 40.6010.93
(7)
SES = 8.631ln ( HDD ) − 55.649 0.99 (16)
Medium R
SES = 4.088ln ( HDD ) − 24.0980.96
(8)
SES = 4.803ln ( HDD ) − 31.696 0.98 (17)
High R
SES = 3.045ln ( HDD ) − 18.2930.97
(9)
SES = 3.538ln ( HDD ) − 23.632 0.98 (18)
High R
us
cr
ip t
Heavy
an
To validate the calculation method summarized by the equations of Table 8, DRC performance is evaluated using the simulation environment for New York City, NY (HDD = 2712 oC-days) for both residential and commercial buildings with various insulation and mass levels. The results obtained for a light mass building at all insulation levels (as defined in Table 4) are summarized in Table 9.
13.92 7.85 5.50
13.71 7.70 5.40
Ac ce pt e
Low R Medium R High R
d
M
Table 9: Comparison between simulated and calculated source energy savings (MJ/m2) for New York, NY Insulation Residential Office level Simulated Calculated Simulated Calculated 11.35 5.98 4.16
11.08 5.84 4.05
Moreover, the DRC source energy savings for all locations and building prototypes obtained from a detailed simulation analysis are compared to the those calculated using equations (1) - (18) listed in Table 8. The results of the comparative analysis are shown in Table 10 and are expressed in terms mean squared error (RMSE) and normalized root mean squared error (NRMSE). The NRMSE values for all building types are generally below 5% indicating that the simplified method outlined by Table 8 is accurate in estimating DRC source energy savings for both residential and commercial buildings. Table 10 RMSE and NRMSE values for calculated savings using equations (1)-(18) for all US locations Mass Residential Office RMSE NRMSE RMSE NRMSE Light 0.55 4% 0.66 5% Low R 0.37 5% 0.35 5% Medium R 0.29 5% 0.21 4% High R Medium 0.74 5% 1.21 9% Low R 0.45 5% 0.46 6% Medium R 0.32 5% 0.28 5% High R Heavy 16
Page 16 of 23
Low R Medium R High R
1.18 0.47 0.31
9% 6% 5%
0.61 0.41 0.31
4% 5% 5%
cr
ip t
3.6 Economic Feasibility In order to assess the cost-effectiveness of DCRs, break-even costs are determined for both residential and commercial buildings in various US sites considered in the analysis. While the life span of variable reflective roofing systems is uncertain from the literature, the average traditional commercial roof life span is approximately 22 years according to Coffelt and Hendirckson [31]. The economic analysis is based on life cycle cost method using estimated DRC annual energy savings combined with electricity and natural gas rates reported by EIA for all US locations [32, 33]. Specifically, equations (19) and (20) are used to determine the initial break even cost if the roofing system life span is 22 years and the inflation rate is 5%.
With,
•
(19)
−N
(20)
rd
USP
M
• W: Uniform Series Present Worth factor
1 − (1 + rd )
an
USPW ( rd , N ) =
us
Break − EvenCost = EC *USPW ( rd , N )
r d:
discount rate and account for both inflation and interest rates
•
Ac ce pt e
life span for the building roof
N:
d
•
∆EC
: annual energy cost savings
Table 11 shows the break-even costs required if switching from a standard roof to a switchable reflectance roof (i.e., DCR) and Table 12 shows the break-even costs required when switching from a static cool roof to a dynamic cool roof for both residential and commercial buildings. Table 11: Initial cost ($/m2) required to break even if switching from a standard roof with a reflectance of 0.30 to a switchable cool roof with a reflectance of 0.3/0.55 Location Residential Office Low R Medium R High R Low R Medium R High R New Orleans, 2A Light Mass Medium Mass Heavy Mass
2.96 3.27 3.08
1.76 1.82 1.77
1.28 1.33 1.30
2.83 3.00 2.78
1.65 1.66 1.60
1.19 1.23 1.21
Las Vegas, 2B Light Mass Medium Mass
4.46 4.84
2.60 2.67
1.88 1.94
4.67 4.92
2.76 2.80
1.99 2.04 17
Page 17 of 23
2.60
1.90
4.57
2.69
1.99
Atlanta, 3A Light Mass Medium Mass Heavy Mass
2.94 3.28 3.10
1.76 1.83 1.77
1.27 1.33 1.30
2.89 3.11 2.91
1.78 1.80 1.73
1.30 1.34 1.32
Portland, 3C Light Mass Medium Mass Heavy Mass
1.47 1.64 1.53
0.94 0.99 0.96
0.72 0.75 0.73
1.82 1.84 1.71
1.19 1.19 1.14
0.89 0.91 0.89
Baltimore, 4A Light Mass Medium Mass Heavy Mass
3.06 3.33 3.13
1.78 1.81 1.76
1.28 1.31 1.28
3.17 3.31 3.07
1.92 1.95 1.88
Chicago, 5A Light Mass Medium Mass Heavy Mass
2.02 2.16 2.04
1.20 1.24 1.20
0.88 0.92 0.90
2.08 2.16 1.99
1.25 1.25 1.20
us
0.91 0.93 0.91
St. Paul, 6A Light Mass Medium Mass Heavy Mass
1.81 1.98 1.87
1.12 1.11 1.08
0.84 0.82 0.80
1.98 2.07 1.93
1.19 1.21 1.16
0.86 0.89 0.87
cr
1.40 1.45 1.43
Ac ce pt e
d
M
ip t
4.59
an
Heavy Mass
Table 12: Initial cost ($/m2) required to break even if switching from a cool roof with a reflectance of 0.55 to a switchable cool roof with a reflectance of 0.3/0.55 Location Residential Office Low R Medium R High R Low R Medium R High R New Orleans, 2A Light Mass Medium Mass Heavy Mass
0.54 0.68 0.67
0.29 0.31 0.32
0.18 0.21 0.21
0.17 0.28 0.28
0.06 0.08 0.07
0.03 0.04 0.04
Las Vegas, 2B Light Mass Medium Mass Heavy Mass
1.36 1.79 1.79
0.76 0.87 0.88
0.51 0.57 0.58
0.68 0.97 0.96
0.30 0.38 0.39
0.17 0.22 0.22
Atlanta, 3A Light Mass Medium Mass Heavy Mass
1.74 2.13 2.08
0.97 1.07 1.06
0.67 0.73 0.73
1.08 1.37 1.33
0.56 0.61 0.59
0.37 0.41 0.40 18
Page 18 of 23
1.68 1.96 1.91
0.92 1.00 1.00
0.64 0.69 0.69
1.15 1.35 1.29
0.58 0.64 0.63
0.39 0.43 0.43
Baltimore, 4A Light Mass Medium Mass Heavy Mass
2.00 2.35 2.28
1.08 1.16 1.15
0.74 0.80 0.80
1.58 1.84 1.76
0.83 0.92 0.90
0.56 0.63 0.63
Chicago, 5A Light Mass Medium Mass Heavy Mass
1.44 1.64 1.59
0.82 0.87 0.86
0.58 0.61 0.61
1.14 1.35 1.31
0.60 0.65 0.64
St. Paul, 6A Light Mass Medium Mass Heavy Mass
1.73 2.02 1.97
1.02 1.06 1.06
0.73 0.75 0.75
us
cr
0.43 0.46 0.45
0.90 0.98 0.96
0.64 0.68 0.67
an
1.66 1.89 1.84
ip t
Portland, 3C Light Mass Medium Mass Heavy Mass
M
It is clear from Tables 11 and 12 that when compared to a standard roof, an office building with a cool roof saves more energy cost per m2 than a residential building. The opposite, however, is true when comparing the cost savings from a static cool roof to a dynamic cool roof. This result is most likely associated to the fact that an office building has higher cooling loads due to high internal loads. Therefore, when compared to a residential building, more savings are seen when the cooling load decreases and less savings are seen when the heating load decreases.
Ac ce pt e
d
Another, more obvious general trend is that the highest break-even costs (and cost savings) are seen in the low insulated building configurations. The results show that a low insulated building can allow for a DRC break-even cost that is more than 1.5 times greater than a high insulated building. Therefore, older buildings would be a better target for retrofit to variable reflectance roofing systems. As noted in the literature review, while some thermochromic (TC) coatings have been developed and tested, their cost is still high [25]. However, the use of DRM roofing system or a mechanically controlled system which only switches reflectance seasonally could be economically feasible based on the benchmark costs listed in Tables 10 and 11.
4. Summary and Conclusions
Based on the analysis results presented in this paper, seasonal switchable reflectance roofs can avoid heating energy use penalties and achieve additional source energy savings and cost savings when compared to static cool roofs for both residential and commercial buildings. In particular, the analysis has indicated that: •
Residential Buildings: For a low insulation building the source energy savings range from 4.33 to 19.44 MJ/m2 (1.6 to 4.9%) and the cost savings range from 0.04 to 0.18 $/m2 (1.9 to 5.0%); for a medium insulation building the source energy savings range from 2.30 to 10.22 MJ/m2 (1.3 to 4.0%) and the cost savings range from 0.02 to 0.09 $/m2 (1.5 to 4.2%); and for a high insulation residential building the source energy savings range from 1.46 to 7.25 MJ/m2 (0.9 to 3.5%) and the cost savings range from 0.01 to 0.06 $/m2 (1.1 to 3.6%).
19
Page 19 of 23
•
Commercial Buildings: For a low insulation office building, the source energy savings range from 1.17 to 18.00 MJ/m2 (0.3 to 3.9%) and the cost savings range from 0.01 to 0.14 $/m2 (0.5 to 4.3%), for a medium insulation office building the source energy savings range from 0.42 to 9.35 MJ/m2 (0.14 to 2.6%) and the cost savings range from 0.0 to 0.07 $/m2 (0.2 to 2.8%, and for a high insulation office building the source energy savings range from 0.21 to 6.49 MJ/m2 (0.1 to 1.9%) and the cost savings range from 0.0 to 0.05 $/m2 (0.1 to 2.1%).
an
us
cr
ip t
Moreover, the analysis results presented in this paper demonstrate that energy savings of dynamic cool roofs depend largely on the climate, insulation level, and reflectance of the roof. The results also show that energy savings generally increase as thermal insulation level decreases for all building types. Making older buildings with low insulation levels a good target to retrofit with dynamic cool roofs rather than well insulated new constructions. In addition, the results of a life cycle cost analysis show that for a residential building the breakeven cost for a 22-year life span ranges depending on the climate from 0.80 to 4.84 $/m2 for switching to a variable reflectance roof from a static standard roof for a residential building, and from 0.18 to 2.35 $/m2 for switching to a variable reflectance roof from a static cool roof. For a commercial building, the break-even cost for a 22-year life span ranges from 0.86 to 4.92 $/m2 for switching to a variable reflectance roof from a static standard roof ranges, and from 0.03 to 1.89 $/m2 for switching to a variable reflectance roof from a static cool roof.
Ac ce pt e
d
M
While more research is needed to improve the performance and lower the cost of switchable reflectance materials like TC coatings, the analysis presented in this paper indicates that seasonally controlled dynamic cool roof systems such as a mechanically controlled roofs or DRM roofing systems have the potential to improve the energy efficiency of both residential and commercial buildings and should be further explored and investigated. Moreover, the potential additional savings for variable thermal emittance coatings could be evaluated using the same analysis presented in this paper. Indeed, a lower thermal emittance provides additional heating savings since it slows the release of long-wave radiation. The lower the thermal emittance, the slower the roof would cool. This technology could be particularly beneficial in providing additional heating savings during the heating period or even at night during swing months.
20
Page 20 of 23
5. References [1] United States Energy Information Administration. Energy consumption estimates by sector.
ip t
[2] D&R International, Ltd. 2011 Buildings Energy Data Book. Washington, D.C.: United States Department of Energy, 2012. [3] Konopacki S, Akbari H, Pomerantz M, Gabersek S, Gartland L. Cooling energy savings potential of light-colored
cr
roofs for residential and commercial buildings in 11 U.S. metropolitan areas. Lawrence Berkeley National Laboratory, Berkeley, CA, 1997. Report LBNL-39433. [4] Levinson R, Akbari H. Potential benefits of cool roofs on commercial buildings: conserving energy, saving money, and reducing emission of greenhouse gases and air pollutants. Energy Efficiency 2010; 3:53-109.
us
[5] Berdahl P, Bretz S. Preliminary survey of the solar reflectance of cool roofing materials. Energy and Buildings 1997; 25: 149-158. [6] ASHRAE, ANSI/ASHRAE. Advanced Energy Design Guide for Small to Medium Office Buildings. 2011, American
an
Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc. Atlanta, GA.
[7] Akbari, H. (1998) ‘Cool roofs save energy’, ASHRAE Transactions, vol 104, no 1B, pp783–788
M
[8] Akbari H, Levinson R, Evolution of Cool-Roof Standards in the US. Advances in Building Energy Research 2008; 2:1-32.
d
[9] Bretz S, Akbari H. Long-term performance of high-albedo roof coatings. Energy and Buildings 1997; 25: 159167.
Ac ce pt e
[10] Synnefa A, Santamouris M. “Mitigating the urban heat with cool materials for building’ fabric.” Urban Climate Mitigation Techniques. Ed. Santamouris M, Kolokotsa D. 2016. [11] Synnefa A, Santamouris M. Cool-colored coatings fight the urban heat-island effect. SPIE – The International Society for Optical Engineering 2007. [12] Levinson R, Berdahl P, Akbari H, Millar W, Joedicke I, Reilly J, Suzuki Y, Vondran M. Methods of creating solar-reflective nonwhite surfaces and their application to residential roofing materials. Solar Energy materials & Solar Cells 2007; 91: 304-314. [13] Pisello A, Cotana F, Brinchi L. On a cool coating for roof clay tiles: development of the prototype and thermal-energy assessment. Energy Procedia 2014; 45: 453-462. [14] Parker D, Huang J. Konopacki S, Gartland L, Sherwin J, Gu L. Measured and simulated performance of reflective roofing systems in residential buildings. ASHRAE Transactions 1998; 104: 963-975. [15] Akbari H, Bretz S, Kurn D, Haniford J. Peak power and cooling energy savings of high-albedo roofs. Energy Building 1997; 25: 117-126. [16] Hildebrandt E, Bos W, Moore R. Assessing the impacts of white roofs on building energy loads. ASHRAE Technical Data Bulletin 1998; 14. [17] Parker D, Sonne J, Sherwin J. Demonstration of cooling savings of light colored roof surfacing in Florida commercial buildings: retail strip mall. Florida Solar Energy Center, Cocoa, FL, 1997. Report FSEC-CR-964-97.
21
Page 21 of 23
[18] Konopacki S, Akbari H, Gartland L, Rainer L. Demonstration of energy savings of cool roofs. Lawrence Berkeley National Laboratory, Berkeley, CA, 1998. Report number LBNL-40673. [19] Konopacki S, Akbari H. Measured energy savings and demand reduction from a reflective roof membrane on a large retail store in Austin. Lawrence Berkeley National Laboratory, Berkeley, CA, 2001. Report number LBNL-47149.
ip t
[20] Akbari H. Measured energy savings from the application of reflective roofs in 2 small non-residential buildings. Energy 2003; 28: 953-967. [21] Akbari H, Levinson R, Rainer L. Monitoring the energy-use effects of cool roofs on California commercial buildings. Energy and Buildings 2005; 37:1007-1101.
cr
[22] Akbari H, Konopacki S, Pomerantz M. Cooling energy savings potential of reflective roofs for residential and commercial buildings in the United States. Energy 1999; 24: 391-407.
us
[23] Akbari H, Konopacki S. Calculating energy-saving potentials of heat-island reduction strategies. Energy Policy 2005; 33: 721-756.
an
[24] Synnefa A, Santamouris M, Akbari H. Estimating the effect of using cool coatings on energy loads and thermal comfort in residential buildings in various climatic conditions. Energy and Buildings 2007; 39: 11671174.
M
[25] Gray C. Application of adaptive albedo roofing coatings in the southeastern united states. A Dissertation – The University of Alabama at Birmingham 2015. [26] Azari S, Bierman J. 2008. System and Method for Energy-Conserving Roofing. US Patent 7,335,419 B2 filed September, 2002 and issued February, 2008.
d
[27] Hooshangi H. Energy Perfomance Modeling of Buildings with Directional Reflective Roofs. A Thesis – Concordia University 2015.
Ac ce pt e
[28] Akbari H, Touchaei A. Modeling and labeling heterogeneous directional reflective roofing materials. Solar Energy Materials & Solar Cells 2014; 124: 192-210. [29] Park B, Krarti, M., Energy performance analysis of variable reflectivity envelope systems for commercial buildings, Energy and Buildings 2015, 124:88-98. [30] Park B, Srubar W, Krarti M. Energy performance analysis of variable thermal resistance envelopes in residential buildings. Energy and Buildings 2015; 103: 317-325. [31] Coffelt D, Hendrickson C. Life-Cycle Costs of Commercial Roof Systems. Journal of Architectural Engineering 2010; March: 29-36. [32] EIA electricity usage
tables
[33] EIA natural gas usage tables https://www.eia.gov/dnav/ng/ng_pri_sum_a_EPG0_PRS_DMcf_a.htm
22
Page 22 of 23
Insulation Type Constant
Los Angeles and Chicago RSI-0.4 (R-2.5) RSI-1.4 (R-8.0) RSI-1.9 (R-11) RSI-3.0 (R-17) RSI-3.9 (R-22) RSI-5.3 (R-30) RSI-1.4/3.0 (R-8.0/17) RSI-1.4/3.9 (R-8.0/22) RSI-1.4/5.3 (R-8.0/30) RSI-1.4/3.0 (R-8.0/17) RSI-1.4/3.9 (R-8.0/22) RSI-0.4/5.3 (R-2.5/30)
an
us
cr
Variable/2-Step
ip t
Table 1: Summary of Insulation Levels and Steps Evaluated
M
Highlights
The impact of using dynamic cool roofs (RCRs) instead of static cool roofs (SCRs) has been evaluated.
•
In all US climates, DCRs outperform SCRs for both residential and commercial buildings.
•
DCRs can achieve annual energy savings up to 5% compared to SCRs.
Ac ce pt e
d
•
23
Page 23 of 23
S2210-6707(16)30278-5 http://dx.doi.org/doi:10.1016/j.scs.2017.01.016 SCS 575
To appear in: Received date: Revised date: Accepted date:
24-8-2016 10-1-2017 11-1-2017
Please cite this article as: Testa, J., and Krarti, M.,Evaluation of energy savings potential of variable reflective roofing systems for US buildings, Sustainable Cities and Society (2017), http://dx.doi.org/10.1016/j.scs.2017.01.016 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Evaluation of energy savings potential of variable reflective roofing systems for US buildings Jenna Testa and Moncef Krarti*, PhD, PE, LEED-AP Building Systems Program, College of Engineering & Applied Science, University of Colorado, Boulder, CO 80303, USA
ip t
ABSTRACT
M
an
us
cr
This paper explores the potential energy savings of using dynamic cool roofs (DCRs) with seasonally variable reflective surface for various building prototypes when compared to static cool roofs. The analysis conducted demonstrates that the additional energy savings from DCRs depend largely on the climate, insulation level, and reflectance of the roof, as well as but to a lesser extent on the building thermal mass. This study shows that older buildings, with low insulation levels, in colder climates are the best target for retrofit using DCRs. Specifically, the results show that when a variable reflective costing is applied to low insulation buildings, source energy savings can be achieved and range from 4.33 to 19.44 MJ/m2 (i.e., 1.6 to 4.9%) for residential units and from 1.17 to 18.00 MJ/m2 (i.e., 0.3 to 3.9%) for offices. Based on an economic analysis, it is found that the breakeven cost for a variable reflectance coating system with a 22-year life span ranges from 0.80 to 4.84 $/m2 for residential buildings, and from 0.86 to 4.92 $/m2 for commercial buildings.
Keywords:
Ac ce pt e
d
Cool Roofs; Cost Analysis; Energy Use Savings, Variable Reflective Coatings
(*) Correspondent Author: [email protected]
1
Page 1 of 23
1. Introduction
cr
ip t
A significant portion of the energy consumed in the United States (US) attributed to buildings. According to the US Energy Information Administration, commercial and residential buildings consumed 40% of the nation’s total end-use energy in 2015 [1]. In 2010, the residential building sector consumed a majority of its source energy in space heating – 28%, space cooling – 15%, and water heating – 13%. While the commercial building sector consumed a majority of its source energy in lighting – 20%, space heating – 16%, and space cooling – 15% [2]. A large portion of the heating and cooling energy use is associated to building envelope systems such as windows, walls, and roofs.
M
an
us
Heat gain through a roof, specifically for a single-story building, can be a dominant component of a building’s total cooling load since it is highly susceptible to solar radiation [3]. One common solution to reducing the cooling load due to gains through the roof is to increase the solar reflectance of the roof by installing a cool roof. Due to their low solar absorptivity and high thermal emissivity, cool roofs maintain lower surface temperatures and reduce the heat flow into buildings [4]. Indeed, the low solar absorption increases the solar reflectance and the high thermal emittance enhances the roof surface’s ability to radiate any absorbed solar energy [5, 6]. Based on the literature, an average standard roof reflectance ranges from 0.20 to 0.30, an average initial low-sloped cool roof white material ranges from 0.70 to 0.80, and an average aged low-sloped cool roof white material ranges from 0.55 to 0.70 [3, 5, 7, 9]. The typical thermal emittance most roofing materials, regardless of color or reflectance is typically between 0.80 and 0.90 [4, 8, 6].
Ac ce pt e
d
In addition to white cool roofing materials, there has been recent developments in colored cool roofing materials specifically for steep-sloped roofs. A cool colored surface absorbs the visible part of the spectrum in order to appear a specific color, but is highly reflecting in the near-infrared (NIR) part of the spectrum which accounts for nearly 50% of solar radiation [10]. Synnefa et al. [11] have developed prototypes of cool colored coatings using near infrared reflective pigments added to acryl-based coatings which can increase solar reflectance by 0.06 (light blue) and 0.22 (black) [11]. Levinson et al. [12] have developed coating systems for various types of substrates and found that coated steel and glazed clay-tile roofing achieved NIR reflectances as high as 0.50 and 0.75, respectively, using a one-coat system [12]. Pisello et al. [13] have developed a new coating system for colored clay tiles which includes a white englobe above the substrate and a sodium silicate based binder with pigment added for color. The study found that the new coatings are able to reflect the NIR by 75% which is 10% more than traditional tiles with equivalent visible appearance [13]. Several field studies have monitored and documented cooling energy savings and cooling peak energy savings from increasing the reflectance of static cool roofs in US warm climates including California, Florida, and Nevada. The studies found that increasing the reflectance during summer months can reduce both the cooling energy usage and the peak demand. A summary of these studies is provided in Table 1.
Table 1: Summary of benefits for static cool roofs based on measured performance for US climates Daily Reduced Roof Roof Location Building Type System Savings Demand Area m2
R-Value
Duct
Δρ
Wh/m2
%
W/m2
%
2
(K.m /W)
California
2
Page 2 of 23
1.4 3.3 RB 3.3 1.9 3.3 0 1.9 RB 5.3 5.1
Interior Plenum Plenum Ceiling Attic Plenum Interior Attic None None None
0.36 0.35 0.44 0.60 0.59 0.40 0.40 0.40 0.61 0.54 0.61
68 39 4 47 14 10 20 11 72 45 69
Strip mall Residence “ “ “ “ “ “ “ “ “ “ “
1161 130 111 121 121 139 158 167 130 139 167 84 223
1.9 1.9 0 0 1.9 3.3 1.2 4.4 1.9 3.3 3.3 0 5.3
Plenum Attic Attic Attic Attic Attic Attic Attic Attic Attic Attic Attic Attic
0.46 n/a 0.63 0.39 0.52 0.42 0.44 0.51 0.30 0.44 0.42 0.53 0.65
7.5 58 137 116 85 31 73 24 64 23 5.4 68 n/a
Small commercial Small commercial
14.9 14.9
3.2 3.2
None None
0.45 0.45
Retail store
9300
Plenum
0.70
3.34 2.37 1.61 6.78 3.55 n/a n/a n/a 10 5 5
12 8 9 20 25 n/a n/a n/a 50 12 6
25 22 43 26 25 13 20 11 15 10 2 25 17
0.65 1.51 7.75 7.64 5.49 1.61 6.24 n/a 3.44 2.58 1.83 5.92 n/a
29 12 28 29 28 11 23 n/a 16 16 12 30 n/a
31 39
1.2 1.6
n/a n/a
n/a n/a
39
11
3.77
14
us
an
M 2.1
18 13 2 46 63 17 26 39 52 17 4
ip t
2945 2211 3056 89 167 2285 455 557 1600 570 4900
cr
Medical office Medical office Retail store School bungalow Residence Office Museum Hospice Retail store Elementary school Cold storage
d
Davis [18] Gilroy [18] San Jose [18] Sacramento [15] Sacramento [15] Sacramento [16] Sacramento [16] Sacramento [16] Sacramento [21] San Marcus [21] Reedley [21] Florida Cocoa Beach [17] Cape Canaveral [14] Cocoa Beach [14] Cocoa Beach [14] Cocoa Beach [14] Cocoa Beach [14] Merritt Island [14] Merritt Island [14] Miami [14] Palm Bay [14] Palm Bay [14] West Florida [14] Lakeland [14] Nevada Battle Mount. [20] Carlin [20] Texas Austin [19]
Ac ce pt e
In addition to building monitoring studies, several computer energy simulations have been conducted to analyze the impact on energy use for both US residential and commercial buildings. Unlike the case of Table 1 for measured data, the simulation analyses explore the effects of cool roofs in various climate zones and look at both cooling savings and heating penalties. The literature reported simulation based results specific to residential and commercial buildings are summarized in Table 2 for US climates.
Table 2: Summary of simulation results for residential and office buildings
Study [3, 22]
[23]
Building Prototype Location
Wall R
Roof R
(m2K/W)
(m2K/W)
Δρ
Annual Cooling Savings 2 kWh/m (%)
Annual Heating Penalty 2 kWh/m (%)
Residential Old
11 U.S. Cities
1.2
1.9
0.30
1.1-5.2 (7-17)
0-4.98 (2-5)
Residential New
“
1.9
3.3
0.30
0.4-2.8 (6-15)
0-3.81 (0-5)
Office Old
“
1.2
1.9
0.45
1.8-6.0 (5-9)
0-5.57 (0-14)
Office New
“
1.9
3.3
0.45
1.1-3.3 (3-7)
0-3.22 (0-12)
Residential Old
239 U.S. Cities
0.9
1.9
0.30
0.5-8.6 (4-11)
0-6.45 (0-2)
Residential New
“
2.3
5.3
0.30
0.1-3.6 (1-8)
0-2.05 (0-2)
3
Page 3 of 23
[24]
“
1.1
1.9
0.40
4.8-8.2 (4-8)
0-3.81 (0-3)
Office New
“
2.3
5.3
0.40
1.2-2.8 (2-4)
0-1.47 (0-6)
Office Old
236 U.S. Cities
1.1
1.2
0.35
0.5-11.5 (n/a)
0-7.03 (n/a)
Office New
“
2.3
3.3
0.35
0.1-4.1 (n/a)
0-2.93 (n/a)
Residential New
27 World Cities
0.5
1.2
0.40
8-29 (11-35)*
n/a
Residential New
“
0.5
1.2
0.65
9-48 (18-54)*
0-16.99 (n/a)*
ip t
[4]
Office Old
*Numbers for this study are presented as annual load savings and penalty, not energy use savings and penalties
M
an
us
cr
In general, the existing literature on cool roofs demonstrates that location, building construction, building type, and increase in reflectance all have an impact on cool roof energy savings. Akbari and Konopacki [23] results show that the cooling energy savings increases as cooling degree-days (CDDs) increase, and the heating energy penalties increase as the heating degree-days (HDDs) increase [23]. A high reflectance roof can lower roof temperatures which reduce the heat flow from the roof into the buildings, decreasing the need for space cooling during the cooling season. On the other hand, a cool roof may also increase the need for heating energy during the heating period. This can be relatively significant in colder climates [4]. For example, Konopacki et al. [3] found that new offices with an electric heat pump in Chicago and Philadelphia, new residences with an electric heat pump in New York City, and new residences with a gas furnace in Philadelphia all displayed net energy deficits or savings very close to zero. Thus, the literature shows that the best climates for cool roofs are characterized by a long cooling season and a short heating season.
Ac ce pt e
d
Building construction also impacts the cooling savings and heating penalty associated to static cool roofs. Older vintage buildings with low insulation can have higher cooling savings during the cooling season and higher heating penalties during the heating season [3, 4, 22, 23]. Typically, residential buildings tend to have lower absolute cooling savings and higher absolute heating penalties than commercial buildings because they are more shell dominated. Moreover, commercial buildings have higher cooling savings and lower heating penalties since they tend to have higher internal heat gains which increase overall cooling and decrease overall heating. Residential buildings, however, have a higher percentage of cooling savings since a larger portion of the cooling load is attributed to the envelope systems [3, 22, 23]. It is also reported that buildings with longer operation schedules show larger savings per roof area than those with shorter schedules. Konopacki et al. [3] and Akbari et al. [22] found that cooling energy savings and heating energy penalties exhibit a linear relationship with the changes in roof albedo for each building type and location. For example, if the difference in roof reflectance is increased from 0.2 to 0.4, the cooling savings and the heating penalty would roughly double. Thus, to estimate the heating penalties and cooling savings from changes in roof reflectance for the same simulated building, the results can be adjusted by a ratio of the reflectance values [3]. It is clear from the existing literature that while a cool roof has definite benefits, the largest disadvantage is the heating penalty which would be of the greatest concern to a customer evaluating the cos-effectiveness of a cool roof. A building with a cool roof begins to experience an energy penalty as the building transitions to heating mode due to the increase in reflected solar energy that is no longer transferred into heat gain at the roof surface [25]. An ideal alternative would be the ability to have a roofing system which is highly reflective in the times when the building is in cooling mode and highly absorptive when the building transitions to heating mode [25]. The best solution would be a dynamic cool roofs (DRCs) with switchable roof materials or variable reflectance coatings (VRCs) which have the ability to change reflectance throughout the year.
4
Page 4 of 23
an
us
cr
ip t
There are currently only two types of variable reflectance roofing systems that have been developed and explored in literature: thermochromic (TC) coatings and heterogenous directional reflective materials (DRMs). Typically, properties of TC coatings are controlled through temperature variations [26]. A recent experimental analysis has found that for a TC coating applied to a building cool roof the low temperature reflectance is about 0.3 and the high temperature reflectance is about 0.55. The reflectance begins to change at the low temperature set point [25]. The analysis has indicated several challenges with the evaluated TC coating including the fact that the coating requires a 11°C temperature difference to completely shift from the low reflectance to the high reflectance values. In the other hand, a DRM roofing system varies reflectance throughout the year without changing the actual material itself. It is controlled by the angle of the sun at different times of the year; thus it is seasonally controlled. It can be designed to reflect more sun during the cooling months and absorb more sun during the heating months. This would have to be tailored to the location and climate. Hooshangi [27] explores the performance of DRMs with corrugated surfaces including reflective sides facing towards the sky and absorptive sides facing towards the ground. The angle of the corrugation depends on sun location in the summer and winter. The reflective sides reflect sunlight during the summer when the sun is high in the sky and the absorptive sides absorb most of the sunlight during the winter when the sun is low in the sky [27]. Recently, Akbari and Touchaei [28] have developed a model to calculate the hourly reflectance of DRMs as a function of zenith and azimuth angles of the suns position [28]. No analysis has been reported to compare potential energy use savings from DRM roofs compared to static reflectance roofs.
2. Analysis Approach Approach Description
d
M
From the limited reported evaluation [29], it is clear that switchable or variable reflectance roof systems such as DCRs and VRCs could provide additional energy and operational cost savings compared to static cool roofs, there is lack of a detailed analysis to determine the impact of various design and operation conditions on the performance of switchable roof systems for buildings. The main objective of this paper is to assess the additional energy and cost savings potential associated with hypothetical switchable reflectance roof systems including DCRs when compared to static cool roofs for both residential and commercial buildings.
Ac ce pt e
The study outlined in this paper explored the energy savings impact of a switchable roof using a simulation environment developed by Park et al. [29-30]. The simulation environment utilizes RC thermal network and the heat balance technique, and has been validated using EnergyPlus [29]. The tool was selected based on modeling speed, and ease of ability of modeling switchable roof materials. For this study, the roof reflectance has been changed primarily changed on a seasonal basis, but the simulation environment has the ability to change the reflectance on hourly or sub-hourly basis. The analysis, summarized in this paper, includes an estimation of the potential energy savings associated with DCRs as a function of several building parameters: building type, insulation level, thermal mass, location/climate, and roof reflectance. The two building types explored in this study are a small office building and a single-family residential building. The major characteristics of the two buildings have been selected based on the US Department of Energy prototypes [3, 4, 22, 23]. A detailed description of the building geometry, window characteristics, slab-on-grade, and internal loads for both building models is provided in Table 3. Table 3: Model characteristics of residential and commercial buildings Small Office Building
Residential Building
5
Page 5 of 23
Building shape
Total Floor Area (m2) Window-to-Wall Ratio Number of Floors Floor to ceiling height (m)
143 (11.96 m x 11.96 m) 15% for all orientations 1 2.44
ip t
510 (27.68 m x 18.44 m) 21% for all orientations 1 3
U-factor (W / m2 K) SHGC Visible transmittance Emittance
cr
Windows
Slab-On-Grade 2
Ueff (W / m K)
Internal Loads 2
10.76 6.78 26 - 120 W/person
Ac ce pt e
Lighting power density (W/m ) 2 Plug load power density (W/m ) Occupancy HVAC Systems Cooling System Heating System
d
Ventilation
M
an
0.34 Peak: 0.2016 cfm/sf of exterior wall surface area (when fans turn off). Off Peak: 25% of peak infiltration rate (when fans turn on) 0.0025 m3/s x people + 0.0003 m3/s * area = 0.22 m3/s
Infiltration
3.24 0.39 0.40 0.84
us
2.95 0.39 0.40 0.84
Packaged AC unit with COP of 3; electricity Furnace with efficiency of 0.85; natural gas
2
0.70 2
4 cm /m wall surface area
None 7.53 2.69 3 - 100 W/person Packaged AC unit with COP of 3; electricity Furnace with efficiency of 0.85; natural gas
The existing literature indicates that most standards use an aged high reflectance of 0.55 for a cool roof, a low reflectance of 0.25 to 0.30 for a standard roof, and a thermal emittance of 0.90 for both [3, 4, 22, 23]. Additionally, the reported potential for the high reflectance value of a thermochromic coated roof is approximately 0.55 and the low reflectance value is approximately 0.30 with a thermal emittance of approximately 0.90 [25]. Thus, the base high reflectance value used in the analysis for both the cool roof and the switchable roof was 0.55, and the base low reflectance value used for both the standard roof and the switchable roof was 0.30. A thermal emittance of 0.90 was used for all roof types. For the analysis, three different mass levels and insulation levels have been explored: low, medium, and high. The same materials were used for the small office and single-family residential prototypes. The light weight construction consists of wood frame walls and a wood deck roof, the medium weight construction consists of concrete walls with a thin concrete deck, and the heavy weight construction consists of one wythe brick walls with wood frame and a thicker concrete deck roof. In addition to the light, medium, and heavy weight constructions, a fourth extra heavy mass which consisted of a 2 wythe brick wall and very thick concrete roof deck (410 kg/m2 average envelope weight) has been considered to further analyze the effect of thermal mass. In 6
Page 6 of 23
Table 4: Insulation and building mass parameters Low
Medium
High
0.90 1.85
2.12 3.46
3.42 4.85
67 50
150 148
cr
Parameter
215 214
an
Mass (kg/m2) Wall Roof
us
2
R-Value (m K/W) Wall Roof
ip t
addition, three insulation configurations have been evaluated including low, medium, and high insulation levels for each building type. The low and medium insulation levels are similar to the old (pre-1980) and new (post1980) vintages that were used in the literature. A summary of the insulation levels and mass levels analyzed can be found in Table 4.
Ac ce pt e
d
M
Seven different locations across the United States were selected to analyze: New Orleans, LA; Atlanta, GA; Las Vegas, NV; Baltimore, MD; Portland, OR; Chicago, IL; and St. Paul, MN. The locations have been selected to be presentative of ASHRAE US climate zones (2 – 6). The heating degree days (HDD), cooling degree days (CDD), and average daily global horizontal radiation (GHR) for each location are shown in Figure 1. Heating and cooling degree-days shown are at a base temperature of 18oC.
Figure 1: Location and climate characteristics of US cities used in the analysis (HDD and CDD are expressed in o C-days and GHR is expressed in Wh/m2) Finally, the roof reflectance has been also varied in the analysis in order to determine the additional energy savings potential should the range for low-high reflectance values of VRCs expands in the future. In addition to the 0.30/0.55 reflectance switchable roof, a 0.30/0.65, 0.30/0.75, and 0.30/0.85 switchable roof have been also modeled and analyzed.
7
Page 7 of 23
3. Discussion of Selected Results
cr
ip t
As was previously mentioned, the primary purpose of the study presented in this paper is to determine the additional energy savings for dynamic cool roofs with switchable reflectance coatings when compared to traditional static cool roofs, and how specific building design and climatic parameters effect those savings. All annual energy use savings or penalties are normalized per unit of roof area. Therefore, the cooling and heating energy usages are expressed in kWh/m2 while the source energy savings are presented in MJ/m2. Since switchable roofs affect both cooling and cooling thermal loads, source energy savings of dynamic cool roofs are compared to those obtained for static cool roofs are estimated and used as comparative indicators to assess the impact of various parameters on the performance of DCRs. The multipliers used to convert site to source energy uses are 3.14 for electricity and 1.01 for natural gas. It should be noted that the energy use savings associated with static cool roof relative to standard roofs have been estimated from this study and have been found to be generally consistent and within the same range as those listed in Tables 1 and 2 including those reported by Akbari et al. [22].
an
us
For this study, a seasonal control was used such that the roof reflectance was changed twice throughout the year – once at the beginning of the cooling season and once at the end of the cooling season. In order to determine the seasonal switchover period for each climate, a degree-day based method has been utilized. For each month, a high reflectance roof is used when the cooling degree days are higher than the heating degree days. Otherwise, a low reflectance roof is considered. Using detailed simulation analysis, it was found that a reasonable base temperature for the degree day selection method is 15.5°C for the residential building and 13°C for the office building.
Ac ce pt e
d
M
3.1 Seasonal control vs. theoretical optimal control While this study focused on a seasonal switchover in roof reflectance, an optimized control strategy could lead to potential additional energy savings. If a switchable roof had potential to change reflectance from day to day or even hour to hour, then theoretically a switchable roof (i.e., DCR) could have the same heating energy use as a standard roof and the same cooling energy use as a static cool roof. A daily change would be beneficial during swing months when some days may require more cooling and others may require more heating. An hourly switchable roof could take advantage of periods during swing months when heating is necessary during the early morning and late afternoon, but cooling is necessary in the middle of the day. Of course, the energy savings of a switchable roof can only be achieved during hours with daylight. To estimate of DRC optimal controls, energy savings were calculated by combining heating energy use of a low reflectance roof and the cooling energy use of a high reflectance roof throughout the year. The results show that there are generally minimal additional savings using optimal rather than seasonal controls. The additional source savings for US residential buildings range from 0.30 to 2.7 MJ/m2 (0.05 to 0.98%) with cost savings ranging from less than $0.01/m2 to $0.04/m2. The additional source savings for US commercial buildings range from 0.41 to 3.0 MJ/m2 (0.09 to 0.95%) with cost savings ranging from less than $0.01/m2 to $0.03/m2. This analysis shows that a seasonal control is sufficient for switchable reflectance roofs. 3.2 Impact of insulation insulation level The first parameter considered in the parametric analysis of DCR performance was the overall building insulation level. The insulation level has been varied for a light mass, medium mass, and heavy mass building with a switchable roof reflectance of 0.30/0.55 and compared to a cool roof with a reflectance of 0.55. The ranges of source energy savings for each building type and insulation level are summarized in Table 5. Table 5: Ranges of source annual energy savings of different insulation levels for dynamic cool roof when compared to a static cool roof Insulation Level Building Prototype 8
Page 8 of 23
Residential Building Low Medium High
Office Building
MJ/m2
%
MJ/m2
%
4.3 - 19.4 2.3 - 10.2 1.5 - 7.3
1.6 - 4.9 1.3 - 4.1 0.9 - 3.5
1.2 - 18.0 0.4 - 9.4 0.2 - 6.5
0.3 - 3.9 0.1 - 2.6 0.1 - 1.9
Ac ce pt e
d
M
an
us
cr
ip t
The results of Table 5 show that the amount of insulation of the building is an important factor that affects the energy savings resulting from a switchable reflectance roof when compared to a static cool roof. In particular, that annual energy use is lower for DRCs, regardless of the insulation level. However, when the R-value is large (the building is well insulated), the heat transfer between the surface of the roof and the interior of the building is small and the impact on the energy use of using a DRC is not as important. Therefore, implementing a switchable reflectance roof would be more beneficial for building with low insulation levels. Figure 2 illustrates the impact of the insulation R-value for a medium mass office building located in various US sites. In particular, Figure 2 shows that the source energy savings from DRCs decrease with building R-value. A similar trend as that of Figure 2 has been noted for both residential and commercial buildings at all thermal mass levels.
Figure 2: Effect of R-value on source energy savings resulting from a switchable roof when compared to a static cool roof for an office building with a medium mass level
3.3 Impact of building mass mass As noted earlier, the building mass level was varied to include low, medium, and high levels as outlined in Table 4. The impact of the mass level on DRC performance has been investigated when the buildings have DRCs with switchable roof reflectance of 0.30/0.55 and static cool roofs with a reflectance of 0.55. The ranges of source energy savings found for various building types and mass levels are summarized in Table 6. Table 6: Ranges of source energy savings of different mass levels for dynamic cool roofs when compared to static cool roof 9
Page 9 of 23
Mass Level
Building Prototype Residential Building MJ/m
2
Office Building
%
MJ/m2
%
1.5 - 16.9
0.9 - 3.9
0.2 - 15.6
0.1 - 3.2
Medium Heavy
1.7 - 19.4 1.7 - 18.3
1.1 - 4.9 1.1 - 4.8
0.3 - 18.0 0.3 - 17.4
0.1 - 3.9 0.1 - 3.8
ip t
Light
Ac ce pt e
d
M
an
us
cr
Interestingly, Table 6 indicates that the switchable roof savings when compared to a static cool roof do not steadily increase with thermal mass. Rather the savings increase to a maximum level as mass increases, then start to decrease past a certain mass level. To further demonstrate this result, a fourth extra heavy building has been considered in the analysis. Figure 3 presents the effect of thermal mass level for a medium insulated office building located in various US sites. The trend of Figure 3 has been observed for all building types at insulation levels considered in the analysis. Figure 3 shows that once the mass reaches a medium (or heavy level for some climates), the savings begin to decrease.
Figure 3: Effect of building mass on source energy savings for resulting from a switchable roof when compared to a static cool roof for an office building with a medium insulation level The trend noted in Figure 3 is attributed to the fact that as thermal mass increases, a variable reflectance roof initially reduces the building heating thermal load when compared to a static cool roof. Indeed, the lower reflectance roof during winter and swing season experiences more solar gains, a higher mass building initially save more energy due to thermal storage capacity of the building mass that provides free heating during colder times of the day. As can be seen in Figure 3, this initial increase in savings is more prominent when the building is located in heating dominated climates. The savings begin to decrease as the effect of thermal mass diminishes. A static cool roof does not experience as much solar gains as a variable reflectance roof. Therefore, the building with static cool roof does not reach an 10
Page 10 of 23
Ac ce pt e
d
M
an
us
cr
ip t
equilibrium temperature as fast as for the building with variable reflectance roof and thus has more capacity to store heat. This behavior is illustrated in Figure 4 which compares the heating loads for an office building with medium insulation in Atlanta. The slope/rate of decrease in heating load for a static cool roof is larger than that for a variable reflectance roof once the mass gets above a medium threshold level; thus the savings in heating thermal load between the two roof systems decreases as the building mass continues to increase past a medium threshold level.
Figure 4: Comparison of heating load and slope of decrease in thermal heating load for a static cool roof and a variable reflectance roof It should be noted, however, that the impact of thermal insulation on DRC performance is more significant than that of building mass as noted in Figures 5 and 6 showing the annual DRC source energy savings for an office building located in Baltimore, MD. It is clear from Figures 5 and 6 that the insulation level has a much larger impact on the energy savings than mass does. For a building with a low insulation level, DRC can save between 1 and 12 MJ/m2 more source energy than for a building with a high insulation level. For a building with medium mass, DRC can only save between 0.85 and 3 MJ/m2 more source energy than for a building with low or high mass levels. Moreover, the impact on DRC performance diminishes as the insulation level increases, whereas this impact increases to an optimum threshold then decreases as the mass level increases.
11
Page 11 of 23
ip t cr us
Ac ce pt e
d
M
an
Figure 5: Effect of mass on source energy savings for an office building located in Baltimore, MD for various insulation levels
Figure 6: Effect of insulation on source energy savings for an office building located in Baltimore, MD for various mass levels 3.4 Impact of roof reflectance reflectance As part of the parametric analysis, the impact of roof reflectance is evaluated on the DRC performance. As was previously mentioned, the base values used throughout the analysis are 0.55 for high reflectance and 0.30 for low reflectance roof coatings. This parameter is particularly of interest should the technology improves and allows for lower low reflectance or higher high reflectance. Three additional reflectance variations have been modeled including: 0.30/0.65, 0.30/0.75, 0.3/0.85. Figures 7 and 8 summarize the analysis results obtained for 12
Page 12 of 23
d
M
an
us
cr
ip t
an office located in Atlanta and Chicago, respectively. Specifically, Figures 7 and 8 show source energy savings obtained from a static cool roof when compared to a standard roof and those obtained from a switchable reflectance roof when compared to a standard roof. The results provided in Figures 7 and 8 are specific for a light mass residential building with medium insulation, but the trend is the same for all other building configurations and climates. The results show that there is a linear relationship between savings and the difference in roof high-low reflectance values. They also demonstrate that as the high reflectance increases, the DRC energy savings increases at a faster rate than the static cool roof savings. This result is due to the fact that as the reflectance on a static cool roof increases, both the cooling savings and the heating penalties increase. Whereas for a DRC with a switchable reflectance coating, only the cooling savings increase while the heating penalties remain minimal. This behavior is more pronounced for warmer climates when comparing the results obtained Atlanta (i.e., Figure 7) with those obtained for Chicago (i.e., Figure 8).
Ac ce pt e
Figure 7: Effect of high reflectance on cool roof savings and variable roof savings when compared to a standard low reflectance roof in Atlanta, GA
Figure 8: Effect of high reflectance on cool roof savings and variable roof savings when compared to a standard low reflectance roof in Chicago, IL 3.5 Impact of climate climate DRC performance has been evaluated for both residential and commercial buildings located in various US climates. Table 7 summarizes the results of the analysis using the heating degree-days (HDDs) estimated using a 13
Page 13 of 23
base temperature of 18oC. Specifically, Table 7 lists the ranges of annual source energy savings obtained for a switchable roof reflectance of 0.30/0.55 compared to a static cool roof with a reflectance of 0.55 for various US locations considered in the analysis. Since the additional energy savings achieved by a switchable roof come from a decrease in heating energy use, the HDDs were used to represent the climatic characteristics for all US locations. Table 7: Ranges of annual source energy savings for a switchable roof when compared to a static cool roof
ip t
787 1264 1571 2426 2574 3607 4379
cr
New Orleans Las Vegas Atlanta Portland Baltimore Chicago St. Paul
Building Prototype Residential Building Office Building 2 MJ/m % MJ/m2 % 1.5 - 5.5 0.9 - 2.2 0.2 - 2.0 0.1 - 0.6 3.5 - 12.6 1.6 - 3.4 1.2 - 6.8 0.4 - 2.3 3.5 - 11.4 2.0 - 3.8 1.9 - 7.1 0.8 - 2.3 4.5 - 13.8 3.1 - 4.9 2.7 - 9.4 1.7 - 3.9 5.2 - 16.4 2.4 - 4.1 3.9 - 12.8 1.5 - 3.4 6.2 - 17.6 2.1 - 3.3 4.7 - 14.9 1.4 - 3.0 6.9 - 19.4 1.2 - 3.0 6.1 - 18.0 15 - 2.9
us
HDD
an
Location
Ac ce pt e
d
M
It is clear from Table 7 that there is a strong relationship between HDDs and DRC source energy savings. Figures 9 and 10 illustrates the variation of DRC source energy savings as a function of HDD for respectively, a light weight residential and an office building with three insulation levels. A similar trend was found for all building configurations. It should be noted that the results from Portland were not included in the regression analysis because they were significantly lower than the trend. This is most likely because Portland has low solar radiation in the winter months when compared to the other locations. In particular, the results of Table 7 and Figures 9 and 10 show that the DRC energy savings increase as HDD increases. Therefore, implementing a dynamic switchable roof would be more beneficial in colder climates.
Figure 9: Effect of heating degree days on annual DRC source energy savings for a light mass residential building
14
Page 14 of 23
ip t cr us
an
Figure 10: Effect of heating degree days on annual DRC source energy savings for a light mass office building
M
3.5.1 Simplified method for energy savings calculation The relationship between the annual DRC source energy savings and HDD has been obtained for all insulation and mass levels for both residential and commercial buildings as noted by equations (1) through (18) listed in Table 8. The method based on the equations (1) - (18) can be utilized to estimate the potential benefits of dynamic cool roofs compared to static cool roofs.
Ac ce pt e
d
Table 8: Regression analysis results to estimate source energy savings (SES) expressed MJ/m2 as a function of heating degree days (HDD) expressed in oC-days Mass Residential Office Equation R2 Equation R2 Light
SES = 7.120 ln ( HDD ) − 42.5560.97
(1)
SES = 8.416 ln ( HDD ) − 55.411 0.99 (10)
Medium R
SES = 4.120 ln ( HDD ) − 24.8570.98
(2)
SES = 4.719 ln ( HDD ) − 31.449 0.98 (11)
High R
SES = 3.068ln ( HDD ) − 18.8340.99
(3)
SES = 3.427 ln ( HDD ) − 23.027 0.98 (12)
SES = 7.474 ln ( HDD ) − 42.9560.94
(4)
SES = 9.027 ln ( HDD ) − 58.342 0.99 (13)
0.96
(5)
Low R
Medium Low R
Medium R
SES = 4.173ln ( HDD ) − 24.718
SES = 4.926 ln ( HDD ) − 32.553
0.98
(14)
15
Page 15 of 23
SES = 3.094 ln ( HDD ) − 18.6820.98
(6)
SES = 3.583ln ( HDD ) − 23.933 0.98 (15)
Low R
SES = 7.121ln ( HDD ) − 40.6010.93
(7)
SES = 8.631ln ( HDD ) − 55.649 0.99 (16)
Medium R
SES = 4.088ln ( HDD ) − 24.0980.96
(8)
SES = 4.803ln ( HDD ) − 31.696 0.98 (17)
High R
SES = 3.045ln ( HDD ) − 18.2930.97
(9)
SES = 3.538ln ( HDD ) − 23.632 0.98 (18)
High R
us
cr
ip t
Heavy
an
To validate the calculation method summarized by the equations of Table 8, DRC performance is evaluated using the simulation environment for New York City, NY (HDD = 2712 oC-days) for both residential and commercial buildings with various insulation and mass levels. The results obtained for a light mass building at all insulation levels (as defined in Table 4) are summarized in Table 9.
13.92 7.85 5.50
13.71 7.70 5.40
Ac ce pt e
Low R Medium R High R
d
M
Table 9: Comparison between simulated and calculated source energy savings (MJ/m2) for New York, NY Insulation Residential Office level Simulated Calculated Simulated Calculated 11.35 5.98 4.16
11.08 5.84 4.05
Moreover, the DRC source energy savings for all locations and building prototypes obtained from a detailed simulation analysis are compared to the those calculated using equations (1) - (18) listed in Table 8. The results of the comparative analysis are shown in Table 10 and are expressed in terms mean squared error (RMSE) and normalized root mean squared error (NRMSE). The NRMSE values for all building types are generally below 5% indicating that the simplified method outlined by Table 8 is accurate in estimating DRC source energy savings for both residential and commercial buildings. Table 10 RMSE and NRMSE values for calculated savings using equations (1)-(18) for all US locations Mass Residential Office RMSE NRMSE RMSE NRMSE Light 0.55 4% 0.66 5% Low R 0.37 5% 0.35 5% Medium R 0.29 5% 0.21 4% High R Medium 0.74 5% 1.21 9% Low R 0.45 5% 0.46 6% Medium R 0.32 5% 0.28 5% High R Heavy 16
Page 16 of 23
Low R Medium R High R
1.18 0.47 0.31
9% 6% 5%
0.61 0.41 0.31
4% 5% 5%
cr
ip t
3.6 Economic Feasibility In order to assess the cost-effectiveness of DCRs, break-even costs are determined for both residential and commercial buildings in various US sites considered in the analysis. While the life span of variable reflective roofing systems is uncertain from the literature, the average traditional commercial roof life span is approximately 22 years according to Coffelt and Hendirckson [31]. The economic analysis is based on life cycle cost method using estimated DRC annual energy savings combined with electricity and natural gas rates reported by EIA for all US locations [32, 33]. Specifically, equations (19) and (20) are used to determine the initial break even cost if the roofing system life span is 22 years and the inflation rate is 5%.
With,
•
(19)
−N
(20)
rd
USP
M
• W: Uniform Series Present Worth factor
1 − (1 + rd )
an
USPW ( rd , N ) =
us
Break − EvenCost = EC *USPW ( rd , N )
r d:
discount rate and account for both inflation and interest rates
•
Ac ce pt e
life span for the building roof
N:
d
•
∆EC
: annual energy cost savings
Table 11 shows the break-even costs required if switching from a standard roof to a switchable reflectance roof (i.e., DCR) and Table 12 shows the break-even costs required when switching from a static cool roof to a dynamic cool roof for both residential and commercial buildings. Table 11: Initial cost ($/m2) required to break even if switching from a standard roof with a reflectance of 0.30 to a switchable cool roof with a reflectance of 0.3/0.55 Location Residential Office Low R Medium R High R Low R Medium R High R New Orleans, 2A Light Mass Medium Mass Heavy Mass
2.96 3.27 3.08
1.76 1.82 1.77
1.28 1.33 1.30
2.83 3.00 2.78
1.65 1.66 1.60
1.19 1.23 1.21
Las Vegas, 2B Light Mass Medium Mass
4.46 4.84
2.60 2.67
1.88 1.94
4.67 4.92
2.76 2.80
1.99 2.04 17
Page 17 of 23
2.60
1.90
4.57
2.69
1.99
Atlanta, 3A Light Mass Medium Mass Heavy Mass
2.94 3.28 3.10
1.76 1.83 1.77
1.27 1.33 1.30
2.89 3.11 2.91
1.78 1.80 1.73
1.30 1.34 1.32
Portland, 3C Light Mass Medium Mass Heavy Mass
1.47 1.64 1.53
0.94 0.99 0.96
0.72 0.75 0.73
1.82 1.84 1.71
1.19 1.19 1.14
0.89 0.91 0.89
Baltimore, 4A Light Mass Medium Mass Heavy Mass
3.06 3.33 3.13
1.78 1.81 1.76
1.28 1.31 1.28
3.17 3.31 3.07
1.92 1.95 1.88
Chicago, 5A Light Mass Medium Mass Heavy Mass
2.02 2.16 2.04
1.20 1.24 1.20
0.88 0.92 0.90
2.08 2.16 1.99
1.25 1.25 1.20
us
0.91 0.93 0.91
St. Paul, 6A Light Mass Medium Mass Heavy Mass
1.81 1.98 1.87
1.12 1.11 1.08
0.84 0.82 0.80
1.98 2.07 1.93
1.19 1.21 1.16
0.86 0.89 0.87
cr
1.40 1.45 1.43
Ac ce pt e
d
M
ip t
4.59
an
Heavy Mass
Table 12: Initial cost ($/m2) required to break even if switching from a cool roof with a reflectance of 0.55 to a switchable cool roof with a reflectance of 0.3/0.55 Location Residential Office Low R Medium R High R Low R Medium R High R New Orleans, 2A Light Mass Medium Mass Heavy Mass
0.54 0.68 0.67
0.29 0.31 0.32
0.18 0.21 0.21
0.17 0.28 0.28
0.06 0.08 0.07
0.03 0.04 0.04
Las Vegas, 2B Light Mass Medium Mass Heavy Mass
1.36 1.79 1.79
0.76 0.87 0.88
0.51 0.57 0.58
0.68 0.97 0.96
0.30 0.38 0.39
0.17 0.22 0.22
Atlanta, 3A Light Mass Medium Mass Heavy Mass
1.74 2.13 2.08
0.97 1.07 1.06
0.67 0.73 0.73
1.08 1.37 1.33
0.56 0.61 0.59
0.37 0.41 0.40 18
Page 18 of 23
1.68 1.96 1.91
0.92 1.00 1.00
0.64 0.69 0.69
1.15 1.35 1.29
0.58 0.64 0.63
0.39 0.43 0.43
Baltimore, 4A Light Mass Medium Mass Heavy Mass
2.00 2.35 2.28
1.08 1.16 1.15
0.74 0.80 0.80
1.58 1.84 1.76
0.83 0.92 0.90
0.56 0.63 0.63
Chicago, 5A Light Mass Medium Mass Heavy Mass
1.44 1.64 1.59
0.82 0.87 0.86
0.58 0.61 0.61
1.14 1.35 1.31
0.60 0.65 0.64
St. Paul, 6A Light Mass Medium Mass Heavy Mass
1.73 2.02 1.97
1.02 1.06 1.06
0.73 0.75 0.75
us
cr
0.43 0.46 0.45
0.90 0.98 0.96
0.64 0.68 0.67
an
1.66 1.89 1.84
ip t
Portland, 3C Light Mass Medium Mass Heavy Mass
M
It is clear from Tables 11 and 12 that when compared to a standard roof, an office building with a cool roof saves more energy cost per m2 than a residential building. The opposite, however, is true when comparing the cost savings from a static cool roof to a dynamic cool roof. This result is most likely associated to the fact that an office building has higher cooling loads due to high internal loads. Therefore, when compared to a residential building, more savings are seen when the cooling load decreases and less savings are seen when the heating load decreases.
Ac ce pt e
d
Another, more obvious general trend is that the highest break-even costs (and cost savings) are seen in the low insulated building configurations. The results show that a low insulated building can allow for a DRC break-even cost that is more than 1.5 times greater than a high insulated building. Therefore, older buildings would be a better target for retrofit to variable reflectance roofing systems. As noted in the literature review, while some thermochromic (TC) coatings have been developed and tested, their cost is still high [25]. However, the use of DRM roofing system or a mechanically controlled system which only switches reflectance seasonally could be economically feasible based on the benchmark costs listed in Tables 10 and 11.
4. Summary and Conclusions
Based on the analysis results presented in this paper, seasonal switchable reflectance roofs can avoid heating energy use penalties and achieve additional source energy savings and cost savings when compared to static cool roofs for both residential and commercial buildings. In particular, the analysis has indicated that: •
Residential Buildings: For a low insulation building the source energy savings range from 4.33 to 19.44 MJ/m2 (1.6 to 4.9%) and the cost savings range from 0.04 to 0.18 $/m2 (1.9 to 5.0%); for a medium insulation building the source energy savings range from 2.30 to 10.22 MJ/m2 (1.3 to 4.0%) and the cost savings range from 0.02 to 0.09 $/m2 (1.5 to 4.2%); and for a high insulation residential building the source energy savings range from 1.46 to 7.25 MJ/m2 (0.9 to 3.5%) and the cost savings range from 0.01 to 0.06 $/m2 (1.1 to 3.6%).
19
Page 19 of 23
•
Commercial Buildings: For a low insulation office building, the source energy savings range from 1.17 to 18.00 MJ/m2 (0.3 to 3.9%) and the cost savings range from 0.01 to 0.14 $/m2 (0.5 to 4.3%), for a medium insulation office building the source energy savings range from 0.42 to 9.35 MJ/m2 (0.14 to 2.6%) and the cost savings range from 0.0 to 0.07 $/m2 (0.2 to 2.8%, and for a high insulation office building the source energy savings range from 0.21 to 6.49 MJ/m2 (0.1 to 1.9%) and the cost savings range from 0.0 to 0.05 $/m2 (0.1 to 2.1%).
an
us
cr
ip t
Moreover, the analysis results presented in this paper demonstrate that energy savings of dynamic cool roofs depend largely on the climate, insulation level, and reflectance of the roof. The results also show that energy savings generally increase as thermal insulation level decreases for all building types. Making older buildings with low insulation levels a good target to retrofit with dynamic cool roofs rather than well insulated new constructions. In addition, the results of a life cycle cost analysis show that for a residential building the breakeven cost for a 22-year life span ranges depending on the climate from 0.80 to 4.84 $/m2 for switching to a variable reflectance roof from a static standard roof for a residential building, and from 0.18 to 2.35 $/m2 for switching to a variable reflectance roof from a static cool roof. For a commercial building, the break-even cost for a 22-year life span ranges from 0.86 to 4.92 $/m2 for switching to a variable reflectance roof from a static standard roof ranges, and from 0.03 to 1.89 $/m2 for switching to a variable reflectance roof from a static cool roof.
Ac ce pt e
d
M
While more research is needed to improve the performance and lower the cost of switchable reflectance materials like TC coatings, the analysis presented in this paper indicates that seasonally controlled dynamic cool roof systems such as a mechanically controlled roofs or DRM roofing systems have the potential to improve the energy efficiency of both residential and commercial buildings and should be further explored and investigated. Moreover, the potential additional savings for variable thermal emittance coatings could be evaluated using the same analysis presented in this paper. Indeed, a lower thermal emittance provides additional heating savings since it slows the release of long-wave radiation. The lower the thermal emittance, the slower the roof would cool. This technology could be particularly beneficial in providing additional heating savings during the heating period or even at night during swing months.
20
Page 20 of 23
5. References [1] United States Energy Information Administration. Energy consumption estimates by sector.
ip t
[2] D&R International, Ltd. 2011 Buildings Energy Data Book. Washington, D.C.: United States Department of Energy, 2012. [3] Konopacki S, Akbari H, Pomerantz M, Gabersek S, Gartland L. Cooling energy savings potential of light-colored
cr
roofs for residential and commercial buildings in 11 U.S. metropolitan areas. Lawrence Berkeley National Laboratory, Berkeley, CA, 1997. Report LBNL-39433. [4] Levinson R, Akbari H. Potential benefits of cool roofs on commercial buildings: conserving energy, saving money, and reducing emission of greenhouse gases and air pollutants. Energy Efficiency 2010; 3:53-109.
us
[5] Berdahl P, Bretz S. Preliminary survey of the solar reflectance of cool roofing materials. Energy and Buildings 1997; 25: 149-158. [6] ASHRAE, ANSI/ASHRAE. Advanced Energy Design Guide for Small to Medium Office Buildings. 2011, American
an
Society of Heating, Refrigerating, and Air-Conditioning Engineers, Inc. Atlanta, GA.
[7] Akbari, H. (1998) ‘Cool roofs save energy’, ASHRAE Transactions, vol 104, no 1B, pp783–788
M
[8] Akbari H, Levinson R, Evolution of Cool-Roof Standards in the US. Advances in Building Energy Research 2008; 2:1-32.
d
[9] Bretz S, Akbari H. Long-term performance of high-albedo roof coatings. Energy and Buildings 1997; 25: 159167.
Ac ce pt e
[10] Synnefa A, Santamouris M. “Mitigating the urban heat with cool materials for building’ fabric.” Urban Climate Mitigation Techniques. Ed. Santamouris M, Kolokotsa D. 2016. [11] Synnefa A, Santamouris M. Cool-colored coatings fight the urban heat-island effect. SPIE – The International Society for Optical Engineering 2007. [12] Levinson R, Berdahl P, Akbari H, Millar W, Joedicke I, Reilly J, Suzuki Y, Vondran M. Methods of creating solar-reflective nonwhite surfaces and their application to residential roofing materials. Solar Energy materials & Solar Cells 2007; 91: 304-314. [13] Pisello A, Cotana F, Brinchi L. On a cool coating for roof clay tiles: development of the prototype and thermal-energy assessment. Energy Procedia 2014; 45: 453-462. [14] Parker D, Huang J. Konopacki S, Gartland L, Sherwin J, Gu L. Measured and simulated performance of reflective roofing systems in residential buildings. ASHRAE Transactions 1998; 104: 963-975. [15] Akbari H, Bretz S, Kurn D, Haniford J. Peak power and cooling energy savings of high-albedo roofs. Energy Building 1997; 25: 117-126. [16] Hildebrandt E, Bos W, Moore R. Assessing the impacts of white roofs on building energy loads. ASHRAE Technical Data Bulletin 1998; 14. [17] Parker D, Sonne J, Sherwin J. Demonstration of cooling savings of light colored roof surfacing in Florida commercial buildings: retail strip mall. Florida Solar Energy Center, Cocoa, FL, 1997. Report FSEC-CR-964-97.
21
Page 21 of 23
[18] Konopacki S, Akbari H, Gartland L, Rainer L. Demonstration of energy savings of cool roofs. Lawrence Berkeley National Laboratory, Berkeley, CA, 1998. Report number LBNL-40673. [19] Konopacki S, Akbari H. Measured energy savings and demand reduction from a reflective roof membrane on a large retail store in Austin. Lawrence Berkeley National Laboratory, Berkeley, CA, 2001. Report number LBNL-47149.
ip t
[20] Akbari H. Measured energy savings from the application of reflective roofs in 2 small non-residential buildings. Energy 2003; 28: 953-967. [21] Akbari H, Levinson R, Rainer L. Monitoring the energy-use effects of cool roofs on California commercial buildings. Energy and Buildings 2005; 37:1007-1101.
cr
[22] Akbari H, Konopacki S, Pomerantz M. Cooling energy savings potential of reflective roofs for residential and commercial buildings in the United States. Energy 1999; 24: 391-407.
us
[23] Akbari H, Konopacki S. Calculating energy-saving potentials of heat-island reduction strategies. Energy Policy 2005; 33: 721-756.
an
[24] Synnefa A, Santamouris M, Akbari H. Estimating the effect of using cool coatings on energy loads and thermal comfort in residential buildings in various climatic conditions. Energy and Buildings 2007; 39: 11671174.
M
[25] Gray C. Application of adaptive albedo roofing coatings in the southeastern united states. A Dissertation – The University of Alabama at Birmingham 2015. [26] Azari S, Bierman J. 2008. System and Method for Energy-Conserving Roofing. US Patent 7,335,419 B2 filed September, 2002 and issued February, 2008.
d
[27] Hooshangi H. Energy Perfomance Modeling of Buildings with Directional Reflective Roofs. A Thesis – Concordia University 2015.
Ac ce pt e
[28] Akbari H, Touchaei A. Modeling and labeling heterogeneous directional reflective roofing materials. Solar Energy Materials & Solar Cells 2014; 124: 192-210. [29] Park B, Krarti, M., Energy performance analysis of variable reflectivity envelope systems for commercial buildings, Energy and Buildings 2015, 124:88-98. [30] Park B, Srubar W, Krarti M. Energy performance analysis of variable thermal resistance envelopes in residential buildings. Energy and Buildings 2015; 103: 317-325. [31] Coffelt D, Hendrickson C. Life-Cycle Costs of Commercial Roof Systems. Journal of Architectural Engineering 2010; March: 29-36. [32] EIA electricity usage
tables
[33] EIA natural gas usage tables https://www.eia.gov/dnav/ng/ng_pri_sum_a_EPG0_PRS_DMcf_a.htm
22
Page 22 of 23
Insulation Type Constant
Los Angeles and Chicago RSI-0.4 (R-2.5) RSI-1.4 (R-8.0) RSI-1.9 (R-11) RSI-3.0 (R-17) RSI-3.9 (R-22) RSI-5.3 (R-30) RSI-1.4/3.0 (R-8.0/17) RSI-1.4/3.9 (R-8.0/22) RSI-1.4/5.3 (R-8.0/30) RSI-1.4/3.0 (R-8.0/17) RSI-1.4/3.9 (R-8.0/22) RSI-0.4/5.3 (R-2.5/30)
an
us
cr
Variable/2-Step
ip t
Table 1: Summary of Insulation Levels and Steps Evaluated
M
Highlights
The impact of using dynamic cool roofs (RCRs) instead of static cool roofs (SCRs) has been evaluated.
•
In all US climates, DCRs outperform SCRs for both residential and commercial buildings.
•
DCRs can achieve annual energy savings up to 5% compared to SCRs.
Ac ce pt e
d
•
23
Page 23 of 23