Energy 239 (2022) 122266

Contents lists available at ScienceDirect

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

Parametric study of venetian blinds for energy performance

evaluation and classification in residential buildings
Yutong Tan a, b, Jinqing Peng a, b, *, Yimo Luo a, b, Jing Gao a, b, Zhengyi Luo a, b,
Meng Wang c, Dragan C. Curcija d
a
College of Civil Engineering, Hunan University, Changsha, 410082, Hunan, China
b
Key Laboratory of Building Safety and Energy Efficiency of the Ministry of Education, Changsha, 410082, Hunan, China
c
School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, 410114, Hunan, China
d
Energy Storage and Distributed Resources Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA, 94720, USA

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

Article history: Venetian blinds are widely used to improve the performance of window systems, while their energy-
Received 5 April 2021 saving potentials vary with different materials. This paper aims to undertake a standardized evalua-
Received in revised form tion of the annual energy-saving potential of venetian blinds in residential buildings, and identify the
26 August 2021
most influential parameters for design optimization. To achieve this goal, nine parameters were chosen
Accepted 1 October 2021
to establish venetian blinds models in WINDOW. Then, 9072 shading models were coupled with building
Available online 6 October 2021
models for annual energy consumption simulation in EnergyPlus, where the shading schedule was
derived from questionnaires. Further, an experiment was conducted to verify the reliability of the joint
Keywords:
Venetian blinds
simulation between WINDOW and EnergyPlus. Finally, annual energy performance (AEP) was adopted
Thermal and optical characteristics for a standardized energy efficiency evaluation of various venetian blinds products, and the sensitivity
Sensitivity analysis analysis of AEP to various parameters was conducted to identify the key parameters. The results indi-
Annual energy performance cated that by adjusting the key parameters of the venetian blinds, cooling energy saving can reach 44%
Standardized evaluation and the heating energy saving can be up to 65%. The significant energy saving potential of venetian blinds
indicates the prospect and importance of their use in residential buildings.
© 2021 Elsevier Ltd. All rights reserved.

1. Introduction requirements, which limit their large-scale application.

Another fundamental passive design strategy to reduce building
Reports have indicated that 51% of energy consumption in res- energy consumption is the usage of shade devices [6]. Well-
idential buildings was for air conditioning and space heating [1], designed shading devices can regulate solar heat gain [7],
and this proportion will increase with a larger window-to-wall improve the indoor daylighting environment [8], and ensure ther-
ratio, typically adopted in modern buildings. Air conditioning en- mal comfort for the occupants [9]. Among the various shading
ergy consumption is closely related to thermal insulation perfor- devices available, venetian blinds are one of the most commonly
mance of the building envelope, in particular the performance of used shading devices in the U.S [10]. Because they can protect
the windows. Inappropriate window arrangement results in privacy [11], redirect natural light [12], and regulate transmitted
excessive heat gain in summer and severe heat loss in winter. To solar radiation. In addition, they can be used in combination with
address this issue, numerous technologies have emerged for other passive technologies like Trombe walls, which are a mature
regulating solar radiation, such as electrochromic windows [2], and popular solar heating technology [13]. Researchers have found
thermochromic windows [3], multi-pane windows [4], phase that indoor temperature regulation though the integration of
change material windows [5], etc. However, these new products are venetian blinds with Trombe walls can reduce the cooling energy
usually accompanied by high costs and complex installation consumption by 4.6% [14,15]. Venetian blinds can not only be used
with traditional buildings, but also new technologies. Therefore,
the research into the energy efficiency performance of venetian
* Corresponding author. College of Civil Engineering, Hunan University, Chang- blinds is meaningful in the building energy conservation era.
sha, 410082, Hunan, China. Thermal performance of venetian blinds is usually evaluated
E-mail address: [email protected] (J. Peng).

https://doi.org/10.1016/j.energy.2021.122266
0360-5442/© 2021 Elsevier Ltd. All rights reserved.
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

together with glazing systems through total heat transfer coeffi- venetian blinds were selected and each parameter was defined
cient (U-value) and solar heat gain coefficient (SHGC). A double with three levels of values; by permuting and combining these
glazing with an interstitial venetian blind has been studied through values of parameters, a total of 9072 cases were modelled in
experimental and numerical methods, and the results have indi- WINDOW. Secondly, building models in both cooling-dominated
cated that the U-value of window systems can be improved by and heating-dominated climates were established based on the
12%e28% with venetian blinds at different slat angles when residential prototype building models of the Department of Energy
compared with a double-glazing unit [16]. Another study examined (DOE) [28]. The shade models were then coupled with building
not only the U-value but also the radiative heat transfer through models for annual air conditioning energy consumption simulation
windows with venetian blinds, and the results revealed that the in EnergyPlus. Though the above software has been widely used
radiative heat transfer can be reduced by 42% and 37% when and validated, an experiment was conducted in this study to vali-
compared with a single glazed window and a double-pane window, date the reliability of the joint simulation between WINDOW and
respectively. Besides, the U-value can be improved by 12% when the EnergyPlus. Finally, the energy-saving potential of different vene-
blind emissivity is reduced from 0.8 to 0.4 [17]. Moreover, the tian blinds was analyzed by a dimensionless indicator, annual en-
variation of blind reflectance has a larger impact than emissivity on ergy performance (AEP), and the most influential parameters were
the U-value. When the long-wavelength reflectivity decreases from discussed.
0.93 to 0.1, the U-values of a single window and a double-glazed
window with venetian blinds increase by 28% and 20%, respec- 2. Research methodology
tively [18]. Additionally, a mathematical model was developed to
study the influence of optical properties, slat spacing, and slat angle A flowchart of the research methodology is presented in Fig. 1,
on SHGC, and the high reflectance venetian blind with a small slat which includes simulations, experiments, and data analysis. As for
distance and small radius of curvature was found to reduce SHGC the simulation work, WINDOW, which is developed by Lawrence
[19]. Berkeley National Laboratory, was used to set up window models
From the literature review, it is evident that the U-value and with the integration of different venetian blinds. These venetian
SHGC of window systems can be effectively improved by various blinds models were obtained by permuting and combining their
measures. However, they cannot reflect the dynamic energy saving optical, thermal, and geometrical parameters. Then, these window
potential, because they are two static indicators calculated under models were coupled with the building model, which was estab-
certain boundary conditions. In this regard, Lee et al. [20] figured lished in Sketchup, for annual energy consumption simulation in
out that dynamic venetian blinds at 45
blind angle can reduce the EnergyPlus. Since most of the shading in residential buildings is
average daily cooling energy by 7e15%, while 17e32% can be saved manually controlled, human behavior has great impacts on shading
at a 0
blind angle. An analysis of heat transfer through a double- schedule. Therefore, our colleagues designed questionnaires for
pane window with venetian blinds [21] showed that the reduc- occupants to obtain the shading schedule to be used in the resi-
tion of heating load and cooling load were 5% and 30%, respectively. dential energy simulation. Though the main purpose of this
Cooling energy-saving in the above studies is significant, and research is the energy performance of venetian blinds, daylighting
Kunwar et al. [22] conducted a full-scale experiment to quantify the environment simulation was also conducted in Radiance to ensure
cooling energy saving potential of venetian blinds in a cool-humid that the shading schedule derived from questionnaires can improve
climate under sunny, cloudy, and overcast sky conditions from the daylighting environment. The daylighting environment was
March 28 to September 14, 2017. The results showed that an evaluated from the perspective of daylight glare probability (DGP),
average of 25.4% cooling energy savings was achieved compared which is the most sophisticated and appropriate metric in assessing
with the baseline room without venetian blinds and lighting glare issues [29] through a full evaluation of vertical illuminance as
control. perceived by the observers. The categories of DGP are listed in
Previous studies have investigated the energy-saving potential Table 1.
of venetian blinds, and Wang et al. [23] have emphasized the In addition, an experiment was conducted to verify the accuracy
importance of optical properties to solar gains through window of the joint simulation of WINDOW and EnergyPlus, which was
systems. Moreover, inappropriate values of design parameters lead described specifically in section 5. The inner surface temperature of
to extra energy consumption [24], especially in highly glazed window systems was measured and collected for verification.
buildings. Different heating and cooling demands in diverse climate Further, the annual energy consumption simulation was performed
zones require different characteristics of venetian blinds for in both heatinge and coolingedominated climates in EnergyPlus,
energy-saving purposes. Therefore, a parametric study on venetian and the weather data in a typical meteorological year format was
blinds is crucial and meaningful. However, few shading devices, adopted for this analysis. Finally, the sensitivity analysis, perfor-
including venetian blinds, have been designed for energy savings mance evaluation, and classification of venetian blinds were con-
purpose [25]. Besides, a standardized evaluation mechanism to ducted based on the simulation results.
evaluate the energy-saving potential of different venetian blinds is
currently lacking in the existing literature. 3. Window model
Therefore, this paper aims to evaluate and clarify the annual
energy-saving potential of different venetian blinds in residential 3.1. Model of venetian blinds
buildings, and identify the most influential parameters for further
design optimization in different climate zones. To achieve this goal, Venetian blinds have a high level of flexibility as the slats can be
a series of software packages, namely WINDOW, Radiance, and adjusted according to sun position to avoid glare and control solar
EnergyPlus, were adopted. WINDOW, a validated program, was heat gain. In this study, venetian blinds were installed inside the
used to establish the window model with venetian blinds [25]. window. Nine parameters, namely outer emissivity, inner emis-
Radiance was used for the daylighting simulation, which has been sivity, solar transmittance, solar reflectance, thermal infrared
validated for a range of buildings and shading devices [26]. Ener- transmittance, thermal openness, conductivity, left-right gap, and
gyPlus is a whole building energy simulation program, whose ac- top-bottom gap, were chosen to establish the venetian blinds
curacy has been validated by multiple studies [27]. The study was models. Thermal openness is also called permeability factor, it is
carried out from the following steps: firstly, nine parameters of the airflow permeability for the material in the shading system and
2
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

Fig. 1. Flowchart of the research methodology.

Table 1 AEP distribution.

Categories of DGP [30]. As shown in Fig. 2, for each case of the venetian blinds, four
Level DGP typical slat angles, namely 0
, 45
, 45
, and 90
were considered.
Imperceptible glare <0.35
A slat angle of 0
means that the slat is horizontal while 90
means
Perceptible glare 0.35e0.4 it is vertical. Furthermore, the slat width of blinds is 25.4 mm and
Disturbing glare 0.4e0.45 the spacing between the adjacent blinds is 19 mm. Full shade
Intolerable glare 0.45 means that the height of venetian blinds is the same as the win-
dow's height, as illustrated in Fig. 3. The detailed shading status of
venetian blinds, considering both slat angles and heights, were
is used to quantify convective heat transfer through the shading listed in Table 3. Apart from the slat angle, the energy performance
layer. To include as many products as possible, each parameter has of venetian blinds was also largely affected by human behaviors
three representative values. The values of each of the nine pa- relating to privacy protection, comfortable indoor daylight envi-
rameters are listed in Table 2. The low value of each parameter was ronment, and the outside view, especially in residential buildings.
determined based on the theoretical lower limit, while the high Therefore, a questionnaire investigating human behavior and
value corresponds to the theoretical upper limit. The medium value shading status, the latter of which is comprised of three statuses:
was determined based on the characteristics of the mainstream full shade, half shade, and no shade, was conducted to obtain the
products in the market. By permuting and combining these values, shading schedule, which was used in the following simulation.
a total of 19,683 cases were obtained initially. However, the optical
properties of venetian blinds are interrelated. Therefore, some 3.2. Model of window systems
constraints were implemented to avoid unreasonable cases:
Two window models were established in WINDOW, namely a
(1) The sum of solar transmittance, solar reflectance, and emis- baseline window and an “adiabatic” window. The baseline window
sivity does not exceed 1; (WB ) consists of two clear panes and a 12 mm air gap, and venetian
(2) The sum of thermal infrared transmittance and emissivity blinds were installed inside the baseline window (WS ). The adia-
does not exceed 1; batic window (WA ) is an ultra-insulating window, for which the
SHGC and infiltration were set to be 0. The overall heat transfer
By eliminating these unreasonable cases from the original set, coefficient (U-value) and SHGC of each WS were calculated under
9072 cases were finally acquired. Even if very few cases are not standard environmental conditions [31] and the results were
ruled out, they will not have obvious impacts on our research exported in idf format for the simulation in EnergyPlus.
because the key parameters were determined based on the overall

4. Building model

Table 2
The residential building models were established in accordance
Level of values for each parameter.
with the International Energy Conservation Code (IECC) [32] in
Parameters Abbr. Low level Medium level High level both heatinge and cooling-dominated climates. Houston was
Emissivity_1 Emis_1 0.05 0.2 0.9 chosen to represent the hot (2A) climatic zone, while Minneapolis
Emissivity_2 Emis_2 0.05 0.2 0.9 was selected as the representative city for the cold (6A) climatic
Thermal IR transmittance Tir 0.05 0.2 0.9
zone [33]. As shown in Fig. 4, the top and bottom parts of each
Solar transmittance Tsol 0.01 0.1 0.8
Solar reflectance Rsol 0.05 0.1 0.5 window were separated because EnergyPlus cannot simulate the
Thermal openness Cond 0 0.02 0.1 half-shade status directly. For the full shade status, both top and
Conductivity (W/m$K) Ther_o 0.05 1 160 bottom parts were WS . For the no shade status, both parts were WB ;
Left and right gap (mm) TB_gap 0 3 12 for the half shade mode, the top part of the window was covered
Top and bottom gap (mm) LR_gap 0 3 12
with WS while the bottom part was WB . More detailed information
3
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

Fig. 2. Different angles of venetian blinds (a: 0
VB (horizontal), b: 45
VB, c: 45
VB, d: 90
VB (vertical)).

energy. The conversion factor of electricity to primary energy is

3.167, and the conversion factor of natural gas to primary energy is
1.084 [34].

5. Experimental validation

As mentioned in the introduction, the software adopted in this

study are well established, therefore the key point that requires
validation is the reliability of the joint simulation between WIN-
DOW and the EnergyPlus software. The zone load calculation in
EnergyPlus is based on the temperature difference between the
inner surface of the building envelopes and the indoor tempera-
ture. Therefore, accurate inner surface temperature is a prerequisite
for reliable joint energy consumption simulation. Since this study
Fig. 3. Shading status of venetian blinds (a: no shade status, b: half shade status, c: full contains over 9000 cases and each case has different statuses, it is
shade status). (For interpretation of the references to color in this figure legend, the almost impossible to conduct comprehensive experiment. Among
reader is referred to the Web version of this article.) the different shade status, full shade status is the simplest status to
maintain because the slats with full shade status were connected
vertically, as shown in Fig. 2 (d). Therefore, a shade device was
Table 3
Detailed status of venetian blinds.
installed with full shade status inside the window in the test room.
The exterior and interior layout of the test room are shown in Fig. 5
ID Status Abbr. Upper part Lower part
(2) and (3). To measure the inner surface temperature, temperature
1 Open-deployed Open_1 No shade No shade sensors were evenly arranged on the window shade during the test
2 Open_2 0
0
period from January 12th to January 13th, 2021. The meteorological
3 Half-deployed Half_3 45
45

4 Half_4 45
45

data, including horizontal global solar irradiance, horizontal diffuse
5 Half_5 90
No shade solar irradiance, outdoor dry bulb temperature, and wind speed
6 Half_6 45
No shade were recorded by a weather station, the results of which are pre-
7 Half_7 45
No shade sented in Fig. 6. More detailed information about the instruments
8 Full-deployed Full_8 90
90

used is listed in Table 5.
For verification of joint simulation between the WINDOW and
the EnergyPlus, the corresponding models of the window system
and the test room were established. The window model was set up
based on the spectral optical properties of each sheet of the win-
dow as well as the shade device, as measured by a UV-VIS-NIR
spectrophotometer, LAMBDA PE 1050. The spectral optical charac-
teristics of shade sample are depicted in Fig. 8 (c). It can be seen that
the average solar transmittance of the shade sample used for
validation is around 0.6 and its average solar reflectance is
approximately 0.3. The building model of the test room was con-
structed in accordance with the Design Standard for Energy Effi-
ciency of Public Buildings [36] and the Design Standard for Energy
Efficiency of Public Buildings in Hunan Province [37]. The U-values
of the external wall, roof, and floor are 0.71 W/(m2$K), 0.42 W/
Fig. 4. External view of building model [35]. [][][]. (m2$K), and 1.24 W/(m2$K), respectively. A schematic of the
building model was shown in Fig. 7 and the simulation was con-
ducted in EnergyPlus using the real-time onsite weather data. All
of the building model is listed in Table 4. Different air conditioning experiment apparatus used in this experiment were shown in
systems in Houston and Minneapolis results in different types of Fig. 8.
energy consumption. For a standardized comparison and analysis, The measured and simulated inner surface temperatures of the
both electricity and natural gas were converted into primary window system are depicted in Fig. 9. The relative error (RE) and
4
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

Table 4
Parameter setting of the building model [35].

parameters Setting values

Dimension of building 10.56 m  10.56 m  2.6 m (Each floor)

Dimension of window 1.2 m  1.5 m (Each window)
Baseline window U ¼ 2.68 W/m2$K
SHGC ¼ 0.60
Window-wall ratio WWR ¼ 13%
U-value W/(m2$K) Houston External wall 0.467
Internal wall 0.238
Floor 0.259
Roof 0.347
Interior ceiling 0.300
Minneapolis External wall 0.347
Internal wall 0.238
Floor 0.259
Roof 0.347
Interior ceiling 0.134
Air conditioning system Houston:heat pump
Minneapolis: gas furnace and air conditioning
Thermostat settings Heating:21.1
C
Cooling:23.9
C
Internal loads According to the residential building model of Department
of Energy (RPBM, 2012)
Weather data Typical meteorological year for Houston and Minneapolis

simulation temperature, ei represents the experimental tempera-

ture, and N is the number of test hours. The averaged relative error
over the whole experiment period is 7.5%, while the rRMSE is 8.3%.
The error was caused by: (1) The thermal properties of building
envelopes. Though both experimental platform and building
models are established in accordance with the energy-saving
standard, the standard only stipulates the thermal properties of
building envelopes a range. There may be deviations in the values
of experimental platform and model within the range. (2) The ac-
curacy of the experimental instrument itself. (3) The accuracy of the
software itself. ASHARE Guideline 14e2014 stated that the simu-
lation model is regarded to be acceptable in terms of accuracy is the
rRMSE is lower than 30% [1]. Therefore, the methodology is at the
acceptable accuracy level, and can be applied to other building
models.

jsi  ei j
RE ¼  100% (1)
ei
Fig. 5. (1) Experimental platform, (2) exterior layout of the test room, (3) interior
layout of the test room.

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
PN
1 2
relative root mean square error (rRMSE), as described in Eqs. (1) N i¼1 ðsi  ei Þ
rRMSE ¼  100% (2)
and (2), were adopted to estimate the agreement between the m
two data sets quantitatively. In the Eqs. (1) and (2), si represents the

Fig. 6. Temperature and radiation during test period in Jan.12 to Jan. 13, 2021.

5
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

Table 5
Instrument specification.

Instruments Type Range Accuracy

Temperature sensor Pt-100 50-200
C ±0.15
C

Weather station Solar tracker STR-21G-S2 0e360 ＜0.01

Solar pyronometer EKO MS-80A 0e4000 W/m2 ±0.2%
Temperature sensor R.M.Young 41,382 50-50
C ±0.3
C
Spectrophotometer LAMBDA PE 1050 175e3300 nm 0.05 nm
Datalogger Graphtec midi logger GL840 e e

Fig. 7. Schematic of the building model.

Fig. 8. Experiment apparatus (a) weather station, (b) LAMBDA PE 1050, (c) spectral optical properties of shade sample. (For interpretation of the references to color in this figure
legend, the reader is referred to the Web version of this article.)

Fig. 9. Inner surface temperature of experiment and simulation.

6. Daylighting environment evaluation chosen as typical days to represent heating season, transition sea-
son, and cooling season [38], and the simulation was run for the
The daylighting simulation was conducted in Radiance with a daytime period from 8:00 to 18:00.
shading schedule derived from questionnaires to ensure a The results of the daylighting simulation with the baseline
comfortable daylighting environment. The case that has all nine window indicated that the strongest glare occurred at 13:00 on
parameters set to their medium value, as listed in Table 2, was January 1st in Houston, and at 15:00 on January 1st in Minneapolis.
selected as the daylighting research object. The south-facing The corresponding glare source pictures and false color pictures are
baseline window was chosen as a comparison object, for which presented in Fig. 10. It is seen that the glare of the baseline window
the DGP was calculated. January 1st, April 15th, and July 1st were is intolerable, therefore, arrangement of venetian blinds is

6
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

necessary to ensure a comfortable indoor daylighting environment.

Further, the hourly DGPs with different shade statuses were
calculated for Houston and Minneapolis, the results of which are
shown in Fig. 11. The results indicate that the baseline window
(Open_1) in both Houston and Minneapolis led to a poor daylight
environment on January 1st since the DGP exceeded the critical
value (0.35). Moreover, the status of Half_3 and Full_8 have the
strongest reduction ability for DGP, for example, the DGP at 15:00
on January 1st was reduced by 50% compared with Open_1 for the
above two shade statuses. For strong solar radiation (January 1st) in
Houston, although some shade statuses can reduce DGP, it still lies
in the unacceptable range such as Open_2, Half_6 and Half_7. On
April 15th and July 1st, the DGP was almost below 0.35, therefore,
the occupants can choose the shade status according to their
preference. The situation in Minneapolis is similar to that of
Houston, although the DGP value is larger on January 1st. Therefore,
it can be concluded that the venetian blinds are beneficial for
reducing DGP, and the improvement varies significantly across
different shade statuses. Each shade status in this study performs
better than the baseline window, therefore, the final status was
found to be strongly related to human behavior. The shading status
derived from the questionnaires in the following simulation is
effective at improving the indoor daylighting environment in terms
of DGP.

7. Results and discussions

Fig. 11. Hourly DGP with different shade statuses in (1) Houston and (2) Minneapolis.
(For interpretation of the references to color in this figure legend, the reader is referred
7.1. Annual energy performance index
to the Web version of this article.)

The annual energy performance (AEP), which has been pro-

posed in our previous study, was adopted to evaluate the energy
performance of the venetian blinds. It is a dimensionless indicator,
calculated by Eqs. (3)e(5) [35]:

Fig. 10. Glare source images and false color images of the baseline window (1) in Houston at 13:00 on January 1st and (2) in Minneapolis at 15:00 on January 1st. (For interpretation
of the references to color in this figure legend, the reader is referred to the Web version of this article.)

7
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

EBS ¼ EB  ES (3)

EBA ¼ EB  EA (4)

EBS
AEP ¼ (5)
EBA

where EB is the energy of the building with the baseline window

(GJ), ES is the energy of the building with the window with inte-
grated venetian blinds (GJ), and EA is the energy of the building
with the “adiabatic” windows (GJ). EB-S is the annual cooling or
heating energy savings achieved by the venetian blinds (GJ), and
EB-A is the annual cooling or heating energy consumption caused
by the baseline window (GJ). It can be seen that AEP is a stable
indicator only dependent on the venetian blinds, independent of
window type, building dimensions, climatic conditions, etc.
Fig. 12 depicts the AEP of each case in the cooling-dominated
and heating-dominated climatic zones. The case with all parame-
ters set to the medium value is treated as the performance of the
current products available on the market because the medium
value is determined according to such mainstream products. The
Fig. 13. Variations of (a) AEPc and (b) AEPh compared to the benchmark.
AEP of this case in cooling-dominated climates is 0.15 while that in
heating-dominated climates is 0.24. The different cooling and
heating demands in different climate zones lead to different re- notable increment of 126% will occur when it is increased from the
quirements for the venetian blinds being installed. Consequently, medium level to the high level. However, AEPc drops by 13% when
the annual energy performance for cooling (AEPc) and for heating Rsol is set to the low level. The high level of solar transmittance
(AEPh) are discussed separately. The AEP of approximately 60% of leads to a 29% increase in AEPh, while the low level leads to a
cases cannot reach the current level in both cooling and heating- reduction of 4%. In contrast, when Rsol is set to the low level, AEPh
dominated climate zones. Therefore, venetian blinds result in a increases by 4%, and a reduction of 42% occurs with the high level.
significant improvement, and sensitivity analysis on the factors Furthermore, the influences of the outer and inner emissivities are
affecting AEP is crucial to find the dominant parameters for further more obvious on AEPh than on AEPc, which is especially the case
design optimization. for the inner emissivity. Previous studies have shown that reducing
the surface emissivity can save air conditioning energy consump-
7.2. Influence of parameters on AEP tion, and the energy saving is better in winter than in summer [27].
Therefore, inner emissivity has a greater impact on AEP than the
The impact of each individual parameter's impact on AEPc and outer emissivity (Emis_1), and the influence is more obvious on
AEPh were depicted in Fig. 13 (a) and (b). It is seen that conductivity AEPh than AEPc.
(Cond) has almost no impact on AEPc and AEPh because the
venetian blinds are not in physical contact with the wall or the 7.3. Average annual energy performance
window, thus there is little heat conduction. As for the influence of
thermal openness (Ther_o), top-bottom gap (TB_gap), and left- The physical characteristics of venetian blinds, especially optical
right gap (LR_gap), they are inversely proportional to AEP. It is characteristics including transmittance, reflectance, and absorp-
worth noting that solar transmittance (Tsol) and solar reflectance tion, are usually interrelated. Therefore, the average annual energy
(Rsol) have opposing impacts on AEPc and AEPh. Compared to the performance for cooling and heating, considering the interrelations
benchmark, when Tsol is set to a low level, AEPc increased by 7%, among these parameters, was proposed as a metric to analyze the
while the high level of Tsol leads to a 33% reduction. As for Rsol, a influence of each parameter on AEP. It is calculated as the average

Fig. 12. Annual energy performance of each case in (a) cooling-dominated climate and (b) heating-dominated climate.

8
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

better for venetian blinds in cooling-dominated climates.

Fig. 15 describes the sensitivity of AEPh to various parameters in
a heating-dominated climate, represented by Minneapolis. The
AEPh is greatly influence by Tsol and Rsol, and is inversely pro-
portional to Rsol while it is directly proportional to Tsol. AEPh rises
from 0.17 to 0.27, which is an increase of 59%, when Tsol changes
from the medium value to the high value. As for Rsol, when the
level changes from medium to high, AEPh drops from 0.22 to 0.01.
In addition, Ther_o, Emis_1, and Emis_2 have obvious effects on
AEPh. AEPh rises from 0.14 to 0.26 when Tir changes from the low
to the medium level. Once set to the high level, AEPh reaches 0.43.
As for emissivity, Emis_2 has a greater impact on AEPh than Emis_1.
Moreover, Cond, Ther_o, as well as gaps have almost the same effect
on AEPh in the heating-dominated climate as in the cooling-
dominated climate.
Fig. 14. Sensitivity of AEPc to various parameters in the cooling-dominated climate.
7.4. Classification of venetian blinds

Based on the simulation results, the annual energy performance

of venetian blinds was divided into five grades. As mentioned in
section 7.1, the case with all parameters set to the medium level
represents the mainstream products available on the market.
Therefore, AEPc and AEPh for this case d 0.15 and 0.24 d are used
as benchmarks in the classification. The highest AEPs in the cool-
inge and heatingedominated climates were considered to be the
excellent grade. Next, the AEP value between the excellent grade
and the benchmark grade was evenly divided into three segments
which are corresponding to different grades of normal, good and
best. Details of the classification method are provided in Table 6.
Taking the grade of AEPc as example, the current level of AEPc is
0.15, and the highest AEPc in the database (9072 cases) is 0.44.
Therefore, the AEPc is classified according to the range of 0.15e0.44.
The product is considered to be a good product if the AEP increases
by one-third compared to the benchmark value. The best product
Fig. 15. Sensitivity of AEPh to various parameters in the heating-dominated climate.
can improve the AEP by two-thirds. When AEPc is less than 0.15, it
is marked as a poor product. When AEPc larger than 0.44, it is
all the cases when a certain parameter is set to a certain value. marked as excellent product. The classification of venetian blinds is
Fig. 14 depicts the sensitivity of AEPc to various parameters in expected to help the fair and orderly development of the venetian
the cooling-dominated climate d Houston. It is worth noting that blinds market, meanwhile, it can help consumers intuitively un-
Tsol and Rsol result in the greatest changes of AEPc. Specifically, derstand the annual energy performance of the product.
AEPc is directly proportional to Rsol, while it is inversely propor-
tional to Tsol. When Rsol and Tsol rise from the low to medium 8. Conclusions
value, AEPc has a relatively small variation. However, once Tsol or
Rsol rises from medium value to high value, the change in AEPc is Venetian blinds are a commonly used and effective measure to
considerable. AEPc drops from 0.19 to 0.09, which is a 53% reduc- improve the performance of window systems. However, their
tion, when Tsol changes from the medium value to the high value. energy-saving potential varies with the different materials to pro-
AEPc increases by 167% when Rsol is set to the high value instead of duce them. In this paper, a parametric study was conducted to
the medium value. The least influential parameters on AEPc are Tir perform a standardized evaluation of the energy-saving potential of
and Cond, hence, manufacturers can focus less on these parameters venetian blinds using a dimensionless indicator, annual energy
when improving the energy performance of venetian blinds. As for performance (AEP). To achieve this goal, nine parameters were
the gaps of the shades and window opening, both the TB_ gap and chosen to establish 9072 different venetian blind cases, with
the LR_ gap are inversely proportional to AEPc. Additionally, Emis_1 different statuses in WINDOW, and then coupled these cases to
and Emis_2 are directly proportional to AEPc, but their impact on building models in EnergyPlus for annual energy consumption
AEPc is smaller than that of the gaps. A lower value of Ther_o is simulations. The accuracy of these models was experimentally
validated. The AEP was used to evaluate the year-round energy

Table 6
Classification method for venetian blinds.

Grades Values of AEPc Values of AEPh Meaning of grade

Excellent 0.44 0.65 Better than the best product in the database
Best 0.35 0.52 A two-third improvement on current level of AEP
Good 0.25 0.38 A one-third improvement on current level of AEP
Normal 0.15 0.24 Current level of AEP
Poor <0.15 <0.24 Worse than the current product

9
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

performance of the venetian blinds, and the most influential pa- https://doi.org/10.1016/j.enbuild.2021.110778.
[4] Fang Y, Memon S, Peng J, Tyrer M, Ming T. Solar thermal performance of two
rameters on AEP were identified in different climates. The main
innovative configurations of air-vacuum layered triple glazed windows.
conclusions are as follows: Renew Energy 2020;50:167e75. https://doi.org/10.1016/
j.renene.2019.12.115.
(1) The whole window model including the shadings was [5] Zhong K, Li S, Sun G, Li S, Zhang X. Simulation study on dynamic heat transfer
performance of PCM-filled glass window with different thermophysical pa-
experimentally validated. The relative error and relative root rameters of phase change material. Energy Build 2015;106:87e95. https://
mean square error between the experimental measured and doi.org/10.1016/j.enbuild.2015.05.014.
the simulated inner surface window temperature are 7.5% [6] Do CT, Chan YC. Evaluation of the effectiveness of a multi-sectional façade
with Venetian blinds and roller shades with automated shading control
and 8.3%, respectively. strategies. Sol Energy 2020;212:241e57. https://doi.org/10.1016/
(2) For coolingedominated climate, solar transmittance and j.solener.2020.11.003.
solar reflectance are the two most influential parameters on [7] Klems JH. A new method for predicting the solar heat gain of complex
fenestration systems - I. Overview and derivation of the matrix layer calcu-
AEPc. AEPc is directly proportional to solar reflectance with a lation. Build Eng 1994;100:1065e72.
slope of 0.49 while it is inversely proportional to solar [8] Gomes MG, Santos AJ, Rodrigues AM. Solar and visible optical properties of
transmittance with a slope of 0.13. glazing systems with Venetian blinds: numerical, experimental, and blind
control study. Build Environ 2014;71:47e59. https://doi.org/10.1016/
(3) For heatingedominated climate, thermal infrared trans- j.buildenv.2013.09.003.
mittance and inner emissivity are the two most influential [9] Yao J. An investigation into the impact of movable solar shades on energy,
parameters. AEPh is directly proportional to thermal infrared indoor thermal and visual comfort improvements. Build Environ 2014;71:
24e32. https://doi.org/10.1016/j.buildenv.2013.09.011.
transmittance with a slope of 0.34, and it is inversely pro-
[10] Tzempelikos A, Athienitis AK, Nazos A. Integrated design of perimeter zones
portional to inner emissivity with a slope of 0.29. with glass facades. Build Eng 2010;116:461e77.
(4) Conductivity has little impact on AEP in both the heatinge [11] Chanvit C, Fisher DE. Comparative analysis of one-dimensional slat-type blind
and cooling-dominated climates, because there is a gap be- models. In: SimBuild 2004, IBPSA-USA national conference boulder, CO; 2004.
[12] Chan YC, Tzempelikos A. Efficient Venetian blind control strategies consid-
tween the shades and windows that weakens the heat con- ering daylight utilization and glare protection. Sol Energy 2013;98:241e54.
duction process. https://doi.org/10.1016/j.solener.2013.10.005.
(5) The energy-saving potential of venetian blinds can be [13] Saadatian O, Sopian K, Lim CH, Asim N, Sulaiman MY. Trombe wall: a review
of opportunities and challenges in research and development. Renew Sustain
improved by adjusting the material parameters. AEPc can be Energy Rev 2021;16:6340e51. https://doi.org/10.1016/j.rser.2012.06.032.
improved by 44% by increasing the values of outer emissivity [14] Islam N, Irshad K, Zahir MH, Islam S. Numerical and experimental study on the
and solar reflectance, while decreasing other parameters. As performance of a Photovoltaic Trombe wall system with Venetian blinds.
Energy 2021;218:119542. https://doi.org/10.1016/j.energy.2020.119542.
for AEPh, it can be increased by 65% by increasing solar [15] Hong X, Leung MKH, He W. Effective use of Venetian blind in Trombe wall for
transmittance and thermal infrared transmittance, while solar space conditioning control. Appl Energy 2019;250:452e60. https://
decreasing the other parameters. doi.org/10.1016/j.apenergy.2019.04.128.
[16] Sun Y, Wu Y, Wilson R, Lu S. Experimental measurement and numerical
simulation of the thermal performance of a double glazing system with an
interstitial Venetian blind. Build Environ 2016;103:111e22. https://doi.org/
Credit author statement 10.1016/j.buildenv.2016.03.028.
[17] Shahid H, Naylor D. Energy performance assessment of a window with a
horizontal Venetian blind. Energy Build 2005;37:836e43. https://doi.org/
Yutong Tan: Conceptualization, Methodology, Software, Vali- 10.1016/j.enbuild.2004.11.008.
dation, Data curation, Writing e original draft preparation. Jinqing [18] Fang X. A study of the U-factor of the window with a high-reflectivity
Venetian blind. Sol Energy 2000;68:207e14. https://doi.org/10.1016/S0038-
Peng: Conceptualization, Methodology, Supervision. Yimo Luo:
092X(99)00065-1.
Writing-Reviewing and Editing. Jing Gao: Software. Zhengyi Luo: [19] Chaiyapinunt S, Khamporn N. Heat transmission through a glass window with
Writing-Reviewing and Editing. Meng Wang: Writing-Reviewing a curved Venetian blind installed. Sol Energy 2014;100:71e82. https://
and Editing. Dragan C. Curcija: Resources. doi.org/10.1016/j.solener.2014.08.041.
[20] Lee ES, Dibartolomeo DL, Selkowitz SE. Thermal and daylighting performance
of an automated Venetian blind and lighting system in a full-scale private
Declaration of competing interest office. Energy Build 1998;29:47e63. https://doi.org/10.1016/S0378-7788(98)
00035-8.
[21] Cho SH, Shin KS, Zaheer-Uddin M. The effect of slat angle of windows with
The authors declare that they have no known competing Venetian blinds on heating and cooling loads of buildings in South Korea.
financial interests or personal relationships that could have Energy 1995;20:1225e36. https://doi.org/10.1016/0360-5442(95)00074-Q.
appeared to influence the work reported in this paper. [22] Kunwar N, Cetin KS, Passe U, Zhou X, Li Y. Energy savings and daylighting
evaluation of dynamic Venetian blinds and lighting through full-scale
experimental testing. Energy 2020;197:117190. https://doi.org/10.1016/
Acknowledgements j.energy.2020.117190.
[23] Wang Y, Chen Y, Li C. Energy performance and applicability of naturally
ventilated double skin façade with Venetian blinds in Yangtze River Area.
This work was supported by the National Natural Science Sustainable Cities and Society 2020;61:102348. https://doi.org/10.1016/
Foundation of China (51978252), the Training Program for Excellent j.scs.2020.102348.
Young Innovators of Changsha (No. kq2009041), the Science and [24] Singh R, Lazarus IJ. Energy and daylighting performances of highly glazed
buildings. Building simulation conference proceedings. 2015. p. 2363e70.
Technology Innovation Program of Hunan Province (2017XK2015), [25] Hart R. Numerical and experimental validation for the thermal transmittance
and Science and Technology Project Plan of Ministry of Housing and of windows with cellular shades. Energy Build 2018;166:258e371. https://
Urban-Rural Development of China (2020-K-165). doi.org/10.1016/j.enbuild.2018.02.017.
[26] Ward G, Shakespeare R. In: Rendering with RADIANCE, the art and science of
lighting visualization. Morgan Kaufmann Publishers; 1998. p. 9.
References [27] Delgarm N, Sajadi B, Kowsary F, Delgarm S. Multi-objective optimization of
the building energy performance: a simulation-based approach by means of
[1] U.S. Energy Information Administration. Energy use in homes, use of energy particle swarm optimization (PSO). Appl Energy 2016;170:293e303. https://
explained. https://www.eia.gov/energyexplained/use-of-energy/homes.php. doi.org/10.1016/j.jobe.2020.101505.
[Accessed 8 March 2021]. [28] Office of energy efficiency and renewable energy, building energy Code pro-
[2] Mallikarjuna K, Shinde MA, Kim H. Electrochromic smart windows using 2D- gram. https://www.energycodes.gov/prototype-building-models#Residential.
MOS2 nanostructures protected silver nanowire based flexible transparent [Accessed 8 March 2021].
electrodes. Mater Sci Semicond Process 2020;17:105176. https://doi.org/ [29] Wienold J, Christoffersen J. Evaluation methods and development of a new
10.1016/j.mssp.2020.105176. glare prediction model for daylight environments with the use of CCD cam-
[3] Liang R, Kent M, Wilson R, Wu Y. The effect of thermochromic windows on eras. Energy Build 2006;38:743e57. https://doi.org/10.1016/
visual performance and sustained attention. Energy Build 2021;236:110778. j.enbuild.2006.03.017.

10
Y. Tan, J. Peng, Y. Luo et al. Energy 239 (2022) 122266

[30] Suk JY, Schiler M, Kensek K. Development of new daylight glare analysis impact of window attachments on air conditioning energy consumption. Sol
methodology using absolute glare factor and relative glare factor. Energy Energy 2020;202:136e43. https://doi.org/10.1016/j.solener.2020.03.096.
Build 2013;64:113e22. https://doi.org/10.1016/j.enbuild.2013.04.020. [36] Ministry of Housing and Urban-Rural Development of the Peoples’ Republic of
[31] National Fenestration Rating Council. Procedure for determining fenestration China. Design standard for energy efficiency of public buildings (GB 20189-
product U-factors, NFRC-100-2010E0A1. 2010. 2015). Beijing: China Architecture Publishing & Media Co., Ltd.; 2015
[32] American National Standards Institute. International energy conservation [Chinese)].
Code. IECC); 2012. [37] Department of Housing and Urban-Rural Development of Hunan Province.
[33] American National Standards Institute. Climatic data for building design Design standard for energy efficiency of public buildings in hunan Province
standards. ANSI/ASHRAE Standard 2013;169e2013. (DBJ43/003-2017). Beijing: China Architecture Publishing & Media Co., Ltd.;
[34] US Department of Energy. Output details and examples. EnergyPlusTM 2017 [Chinese)].
version 8.7 documentation. [38] Yao J. Effect of stochastic adjustment of solar shading on indoor thermal and
[35] Tan Y, Peng J, Curcija C, Hart R, Josson J, Selkowitz S. Parametric study of the visual conditions. Tianjin: Tianjin University; 2013.

11