Optimizing Window Glass Design for Energy Efficiency in South Korean Office Buildings: A Hierarchical Analysis Using Energy Simulation

Abstract

The world is emphasizing the need for building design that considers energy performance to deal with climate problems. South Korea has constantly been tightening the design standards for saving building energy but with a focus on thermal performance and equipment systems. Accordingly, this study conducted an energy simulation experiment on office buildings with different window-to-wall ratios (WWRs) to propose a smart glazing plan to improve energy performance. An energy simulation experiment was performed on office buildings with varying WWRs to hierarchically analyze the influence of building window performance elements, including the heat transmission coefficient (U-value), visible light transmittance (VLT), and solar heat gain coefficient (SHGC), on building energy performance. The analysis showed that SHGC had the most significant impact on the heating and cooling load, by 22.13%, with the influences of the variables being 12.4% for the U-value, 4.78% for VLT, and 82.83% for SHGC. The results showed that the solar heat gain coefficient (SHGC) had the greatest impact on energy performance among window performance elements, and the effect increased significantly in certain WWRs. Moreover, to improve the energy performance of buildings with higher WWRs, it is essential to reflect the optimum composition of the U-value and SHGC on the window plan. This study’s findings propose measures to supplement existing window plans focusing on thermal performance. Furthermore, these results hold academic value in providing concrete grounds for that.

1. Introduction

The increase in greenhouse gases due to the excessive use of fossil fuels worldwide is causing serious environmental problems related to global warming and abnormal climate [1,2]. According to the Global Alliance for Buildings and Construction (GlobalABC), carbon emissions in construction account for approximately 38% of the world’s carbon emissions, and energy consumption represents about 35% of the world [3]. The European Union (EU) has recently released a draft of the so-called ‘Net-Zero Industry Act’ on media to reinforce the net-zero implementation technology and accelerate the transition to climate neutrality [4]. Countries worldwide announced goals to reach net-zero emissions as a response to climate change, and energy performance has become a critical element to consider in building design [5]. The South Korean government has also strengthened and implemented the roadmap for turning public buildings in 2020 and private buildings in 2025 into zero-energy buildings [6,7,8]. Moreover, the performance standards for thermal insulation and windows have been constantly tightened to improve the building envelope performance for minimized building energy requirements [8]. However, these standards were focused on supplying high insulation and airtight construction technology to physically strengthen building energy performance [9,10]. Although thermal performance is a critical element of energy performance, it is a disadvantage in window plans considering lighting and views [11,12,13]. Building regulations and performance evaluations are focused on the heat transmission coefficients of windows [11].

Building energy efficiency can be increased by glazing windows to reduce solar heat gains in addition to heat transmission coefficients [14,15,16]. Building energy performance varies depending on optical performance, such as the solar heat gain coefficient (i.e., SHGC) and visible light transmittance (i.e., VLT), and optical performance, in particular, significantly affects the cooling load in summer [9,11].

SHGC represents an index of the sum of solar heat directly transmitted through the window and heat absorbed by solar radiation and radiated inside. A lower SHGC indicates reduced solar heat transmission and increased energy efficiency [17]. VLT is an index representing how much visible light is transmitted. A lower VLT indicates less light transmission and glare, which increases visual comfort [18]. A U-value is used as a heat transfer coefficient, and a lower U-value indicates less heat loss and improved insulation performance [19]. Therefore, the following glazing factors must be applied to satisfy the efficiency of both the heating and cooling load, and an optimized window plan must be established by comprehensively comparing and analyzing each performance element.

Windows not only provide lighting and a view but also affect heating and cooling loads through heat transfer [20,21]. Heat is transferred inside due to solar heat gain, which raises the building temperature. Particularly, this results in an increased cooling load for the building in summer. Sbar et al. [22] evaluated electrochromic (EC) window performance with different glazing characteristics, including the U-value, VLT, and SHGC, for energy savings in office buildings. The application of EC windows has been found to reduce carbon emissions by up to 35% in new buildings and to 50% in remodeled buildings. Mesloub et al. [23] explained that switchable suspended particle device (SPD) glazing (VLT, U-value, SHGC) is a good alternative to standard glazing in hot desert climates for reducing energy usage of buildings and providing visual comfort. Ko et al. [24] also performed verification simulations on SPD window models and confirmed that the cooling power consumption decreased by 29.1%, resulting in a total annual power consumption reduction of 4.1%. Baetens et al. [25] conducted research on window prototypes and smart windows to maximize daylight and energy efficiency through the optical performance of the windows. While the thermal transmittance (U-value) of glass affects building energy performance, optical performance is also highlighted as a crucial factor [26,27]. Many previous studies propose alternatives using glazing characteristics such as windows applying the U-value and SHGC or smart glazing windows that can adjust them [24,28,29]. Glazing significantly impacts building energy, applying to both design and energy performance, such as the penetration and solar radiation control functions of glass [21,22]. Previous studies mention that the glazing performance of windows can enhance the energy performance of buildings, with a particular focus on the significant role of optical performance, specifically the SHGC. However, many of these studies have conducted experiments only on fixed window sets rather than analyzing the performance of each individual element, including the U-value, VLT, and SHGC.

Hee et al. [30] conducted an experiment to improve window performance through the optical and thermal properties of glass to determine the effect of window glass on the energy and daylight performance of buildings. Windows’ thickness, coating, and coloring are key mediating variables that determine heat and daylight; thus, it is important to optimize the glazing attributes (U-value, SHGC, VLT). Kim et al. [31] also acknowledged the need for research on the impact of each glazing element on energy demand. As confirmed in prior research, it is necessary to understand the impact of each glazing element on energy performance and their interrelationships to save building energy. Yoon et al. [32] analyzed the effect of glazing performance on building energy. The analysis showed that buildings with a higher WWR have the lowest energy demand when they have high thermal transmittance and low SHGC. Generally, reducing thermal transmittance to improve insulation performance decreases energy demand. However, research by Yoon et al. presented different findings. This suggests that a comprehensive consideration of window design variables is necessary to meet the envelope performance requirements. Instead of an isolated individual performance analysis of WWR, thermal transmittance, and optical performance, a hierarchical analysis of window design variables is proposed.

Therefore, this study analyzed the effects of a thermal performance element (U-value) and windows’ optical performance elements (VLT, SHGC) on the buildings’ heating and cooling loads through energy simulations of windows, which significantly impact building energy loads. This study aims to propose energy-efficient window plans and offer flexibility in façade design to enhance building energy performance.

2. Materials and Methods

2.1. Experimental Method

This study conducted an experiment to analyze the effects on windows’ performance composition and buildings’ heating and cooling loads. We used ANSI, an interface using EnergyPlus, an engine developed by DOE, as an energy analysis tool. DesignBuilder is a typical third-party program of EnergyPlus verified with the building envelope test of ANSI/ASHRAE Standard 140-2017 (aka BESTEST) [33] and was analyzed equivalently to the existing EnergyPlus testing results [34].

This study limited the scope of the experiment to offices in South Korea. Solar energy in window planning significantly affects buildings with a high frequency of use in the daytime [9,35]. Notably, for offices with high cooling and lighting loads, solar heat gain produces a considerable difference in energy consumption, thereby serving as a key energy performance factor [35].

The window-to-wall ratio (WWR) also affects building energy consumption [10,36]. Higher WWRs result in a more significant impact [36,37]. Therefore, this experiment aims to identify the influence of each WWR as a glazing performance design variable.

The experiment sequence is as shown in (Figure 1). The experiment was conducted twice, and the method used is as follows.

(1)

First, the impact of each performance element (U-value, VLT, SHGC) on building load was analyzed by fixing the WWR of the mass at 50% and changing the windows’ performance composition. The window area was modeled so that all four sides have the same ratio for each direction.

(2)

Based on the experimental results above, the second experiment was conducted with a focus on the U-value and SHGC, which are performance elements with a great impact on energy load. This study analyzed the influence of each variable on energy load and performance differences according to the WWR. This is to provide a rational window plan applying the U-value and SHGC.

2.2. Simulation Verification Method

The simulation model of this study was developed in the form of a tower mass with high energy consumption but also high preference. It is a 15-story tower with a floor area of 20 m × 20 m.

Table 1. Modeling image and setting of simulation.

	Model Setting
	Building type	Office
	Hight	15th floor
	Floor height	4 m
	Floor space	400 m2
	Window-wall ratio	40%
		50%
		60%
		70%

shows the exterior and the form of the model. There were four types of WWR (40%, 50%, 60%, 70%), which were set equally in each direction.

Table 2. Simulation setting conditions.

Parameter	Value
Site	Seoul, Republic of Korea
Dimension	Floor area	400 m2
	Height	60 m
Construction	Wall	Thickness	300 m
		R-value	2.847 m2K/W
		U-value	0.351 W/m2K
	Slab	Thickness	200 m
		R-value	3.893 m2K/W
		U-value	0.257 W/m2K
	Roof	Thickness	330 m
		R-value	2.886 m2K/W
		U-value	0.346 W/m2K
Indoor condition	Temperature	Heating	22 °C
		Cooling	24 °C
	Occupancy density	0.111 people/m2
	Metabolic factor	0.90
	Power density	11.77 W/m2
	Lighting	5.0 W/m2–100 lux
	Ventilation fresh air	10.0 ℓ/s·person
	HVAC	Fan Coil Unit (4-pipe), Air cooled
Operation schedule	Weekdays		7:00~19:00
	Weekends		Off
Weather data	Seoul TMY2

presents the simulation input conditions to analyze the building’s heating and cooling loads. The simulation’s internal heating, ventilation, and infiltration conditions were set according to UK standards (UK NCT).

The simulation was performed using Design Builder, and the TMY2 (Typical Meteorological Year) weather data of Seoul provided by TRNSYS was applied to the experiment. The TMY2 weather data have the form of the test reference year (TRY) and include horizontal solar radiation additionally spread to the surface and solar radiation directly applied to the vertical surface data. The system finds the typical month of each year for 30 years (1961–1990) and then combines the data of the next 12 months [31]. The average monthly values of key weather factors are presented in

Table 3. Weather data factors and monthly averages.

Month	Dry Bulb Temp. (°C)	Dew Point Temp. (°C)	Relative Humidity (%)	Global Radiation (kWh/m2)	Wind Speed (m/s)
Jan.	−1.29	−7.84	60.40	37.20	2.35
Feb.	0.98	−7.24	54.38	44.14	2.55
Mar.	6.87	−1.76	55.42	60.83	2.77
Apr.	13.04	2.84	53.49	74.67	2.72
May.	18.27	10.23	62.36	81.97	2.76
Jun.	22.82	15.67	66.09	85.90	2.07
Jul.	25.29	20.30	74.33	67.50	2.19
Aug.	24.84	19.97	74.31	65.91	1.56
Sep.	20.92	14.80	69.32	52.41	1.83
Oct.	15.48	7.74	61.31	49.69	1.78
Nov.	6.20	−0.68	62.61	33.85	2.14
Dec.	0.99	−5.65	59.81	39.54	2.25

.

2.3. Configuring Window Performance for Experimental Models

To analyze the heating and cooling loads of an office building according to the influence of the U-value, VLT, and SHGC, we conducted an experiment by combining each of the window performance variables. Units of 0.5 W/m²K subdivided the U-value from 0.5 to 3.0 W/m²K, and VLT and SHGC were subdivided into 7 types by units of 0.1% from 0.2 to 0.8%.

(1)

The first experiment was on the impact of each element, and all elements, except one experimental element, were fixed at the median value to test the change in each performance element.

Table 4. Window performance configuration settings for Experiment A.

Case	Window Performance
	U-Value (W/m2K)	VLT (%)	SHGC (%)
A-1	0.5	0.5	0.5
	1.0
	1.5
	2.0
	2.5
	3.0
A-2	2.0	0.2	0.5
		0.3
		0.4
		0.5
		0.6
		0.7
		0.8
A-3	2.0	0.5	0.2
			0.3
			0.4
			0.5
			0.6
			0.7
			0.8

shows the setup of this experiment.

(2)

Based on the experimental results above, the second experiment was conducted with a focus on the U-value and SHGC, which greatly impact heating and cooling loads. VLT was fixed at 0.5%, and the loads according to the change in the U-value and SHGC were identified to analyze the correlation between the performance elements and the loads.

Table 5. Window performance configuration settings for Experiment B.

Case	Window Performance
	U-Value (W/m2K)	VLT (%)	SHGC (%)
B-1	0.5	0.5	0.2
			0.3
			0.4
			0.5
			0.6
			0.7
			0.8
B-2	1.0
B-3	1.5
B-4	2.0
B-5	2.5
B-6	3.0		0.2
			0.3
			0.4
			0.5
			0.6
			0.7
			0.8

shows the setup of this experiment.

3. Results and Discussion

3.1. Analysis of the Influence of Each Window Performance Factor

The first experiment consisted in identifying the impact of each element. The change was applied to only one element, while the other two were fixed at the median value for the experiment. The experimental results of WWR 50% are presented in Figure 2, Figure 3 and Figure 4.

Higher U-values lead to less cooling load and more heating load. Consequently, the sum of heating and cooling loads also increases. Higher VLTs lead to less cooling load and more heating load. The sum of heating and cooling loads shows a decrease. Finally, higher SHGCs lead to more cooling load and less heating load. The sum of cooling and heating loads also increases.

As for the changes in heating and cooling loads of each element, the U-value changed by 3.31%, VLT by 1.28%, and SHGC by 22.13%, proving that the change from SHGC is the biggest.

Table 6. Change rate of heating and cooling load according to performance factor (%).

	U-Value	VLT	SHGC
Cooling load change rate (%)	−20.31	−3.12	+76.37
Heating load change rate (%)	+165.19	+4.34	−55.28
Heating and cooling load change rate (%)	+3.31	−1.28	+22.13

shows the rates of change in each element’s cooling load, heating load, and heating and cooling loads.

The change in the sum of the U-value heating and cooling loads is relatively low, but the individual change in cooling load and heating load was high. The change in the heating load of the U-value was 165.19%, indicating a massive impact on the heating load. Conversely, for VLT, the effect on each heating and cooling load was small at 3.31%, thereby having an insignificant impact on total heating and cooling loads. SHGC shows a 22.13% change in heating and cooling loads, indicating that it has the greatest impact on all elements. The difference in cooling load was 76.37%, and the change in heating load was 55.28%, indicating that the effect on cooling load is more remarkable.

As a result, SHGC is the performance element that has the greatest influence on heating and cooling loads, followed by the U-value and VLT.

3.2. Annual Heating and Cooling Load According to U-Value and SHGC

Figure 5, Figure 6 and Figure 7 show the results of Experiment B when WWR is 50%, providing cooling load, heating load, and heating and cooling loads according to the changes in the U-value and SHGC. The results show that the cooling load increases with a higher SHGC and lower U-value. Heating load increases with a lower SHGC and higher U-value. Therefore, there must be a low SHGC and high U-value to increase the efficiency of the cooling load. Conversely, the heating load must have a high SHGC and low U-value. Each element is required to have conflicting attributes to reduce the load. Moreover, an inversion occurs between U-value = 0.5 W/m²K and U-value = 3.0 W/m²K with SHGC = 0.6% of heating and cooling loads.

High SHGC indicates that more solar energy is entering inside, and a low U-value suggests that higher thermal performance keeps the solar energy that entered inside from escaping. Therefore, this indicates that solar energy inflow has increased after SHGC = 0.6%, creating a greenhouse effect combined with the U-value with good thermal performance and increasing heating and cooling loads.

3.3. A Hierarchical Analysis of U-Value and SHGC using Window Area Ratio

The following inversion sections vary by WWR. Figure 8 shows heating and cooling loads calculated considering the U-value and SHGC of each WWR. Here, there are the sections of inversion. Inversion occurs at SHGC = 0.7% for WWR 40%, at SHGC = 0.6% for WWR 50%, at SHGC = 0.5% for WWR 60%, and after SHGC = 0.5~0.4% for WWR 70%. Higher WWRs lead to lower SHGCs in the inversion sections. Therefore, SHGC must be considered as more when WWR is higher to reduce the load. It is necessary to select the optimal combination by considering both the U-value and SHGC of windows according to the WWR.

Figure 9 shows heating and cooling loads according to the U-value and SHGC of each WWR, as in Figure 8. Generally, higher U-values are more disadvantageous in terms of energy efficiency. Therefore, heating and cooling loads of ‘U-value = 3.0 W/m²K’ must be the biggest. However, as shown in Figure 9, higher U-values were more advantageous regarding energy efficiency in high-value SHGC combinations. This was even more apparent with higher WWRs. For example, heating and cooling loads were the smallest in WWR 70% and U-value = 3.0 W/m²K at SHGC = 0.8%, and the loads were the biggest in U-value = 0.5 W/m²K when the thermal performance was rather good.

Table 7. Heating/cooling load per floor area and minimum value according to U-value and SHGC by window area ratio (unit: kWh/m²).

WWR	SHGC (%)	U-Value = 0.5	U-Value = 1.0	U-Value = 1.5	U-Value = 2.0	U-Value = 2.5	U-Value = 3.0
40%	SHGC = 0.2	113.3	116.0	118.9	121.9	124.8	127.8
	SHGC = 0.3	117.2	119.2	121.5	124.1	126.6	129.1
	SHGC = 0.4	121.9	123.1	124.8	126.9	128.9	131.1
	SHGC = 0.5	127.4	127.7	128.8	130.3	131.8	133.5
	SHGC = 0.6	133.5	132.9	133.2	134.0	135.1	136.4
	SHGC = 0.7	140.0	138.6	138.1	138.3	138.9	139.7
	SHGC = 0.8	147.2	144.8	143.5	143.0	143.1	143.5
50%	SHGC = 0.2	115.1	118.2	121.7	125.4	129.0	132.6
	SHGC = 0.3	120.6	122.4	125.1	128.1	131.0	134.1
	SHGC = 0.4	127.2	127.8	129.4	131.7	133.9	136.4
	SHGC = 0.5	135.0	134.2	134.7	136.0	137.6	139.5
	SHGC = 0.6	143.6	141.4	140.7	141.0	141.8	143.1
	SHGC = 0.7	153.2	149.3	147.4	146.7	146.8	147.5
	SHGC = 0.8	163.6	158.0	154.8	153.1	152.4	152.3
60%	SHGC = 0.2	117.1	120.4	124.4	128.7	133.0	137.4
	SHGC = 0.3	124.2	125.8	128.5	131.9	135.3	138.9
	SHGC = 0.4	132.9	132.6	133.9	136.2	138.7	141.5
	SHGC = 0.5	143.3	140.9	140.7	141.7	143.2	145.2
	SHGC = 0.6	154.9	150.2	148.2	147.8	148.3	149.5
	SHGC = 0.7	167.6	160.6	156.8	155.0	154.5	154.8
	SHGC = 0.8	181.4	172.0	166.3	163.0	161.3	160.7
70%	SHGC = 0.2	119.3	122.7	127.1	132.0	137.0	142.0
	SHGC = 0.3	128.2	129.2	131.9	135.6	139.4	143.6
	SHGC = 0.4	139.2	137.6	138.5	140.7	143.4	146.5
	SHGC = 0.5	152.7	148.0	146.6	147.2	148.6	150.7
	SHGC = 0.6	167.4	159.6	155.9	154.5	154.6	155.7
	SHGC = 0.7	183.5	172.6	166.5	163.2	161.9	161.8
	SHGC = 0.8	200.6	186.7	178.1	173.0	170.1	168.8

SHGC unit: %, U-value unit: W/m²K.

presents the results of Experiment B, marking the sum of minimum heating and cooling loads at each SHGC in shading. Heating and cooling loads are increasing along with the U-value regardless of VLT, but in combination with SHGC, the loads decreased at the conversion point of SHGC and then increased again. Therefore, U-values of minimum cooling and heating loads vary depending on the SHGC. Moreover, there is a difference in the converted values of SHGC depending on the WWR. Higher WWRs lead to lower SHGCs in the inversion sections. In other words, higher WWRs indicate a greater impact on solar energy, which is why SHGC must be considered.

3.4. Sinter and Discussion

Around 60–70% of a building’s heating and cooling loads originate from the outside, with windows and doors in offices contributing up to 40% of heat loss [24,38]. Since office buildings have extensive areas due to designs that pursue a wide view, the increased solar heat inflow in summer causes high cooling loads and poor thermal comfort [27,38]. Solar radiation energy through windows has different effects on building energy consumption in winter and summer, which is why windows are a complicated design element. Previous studies are proposing strategies to reduce energy consumption by improving the envelope performance of windows [11,14,16,24]. While the effectiveness of glazing as a strategy for enhancing window performance has been verified, verification of each element’s individual performance has not been conducted [39,40]. Instead of analyzing individual elements, this study conducted an experiment to ascertain performance with predetermined window sets. This approach has limitations in providing objective indicators for the efficiency of each element and the application of glazing design.

The present study analyzed the impact of each glazing element—U-value, VLT, and SHGC—on window planning for energy reduction, differentiating the variable values to discern individual performances. The study distinguished itself from prior research by establishing an integrated hierarchy for each element after identifying individual performances. Further, by conducting experiments on the energy performance of glazing variables according to the WWR, a significant design element influencing energy consumption, the study holds significance by conducting a hierarchical analysis between window glass design variables. The results of this study are as follows. First, SHGC had the greatest impact on building energy loads among window performance elements, followed by the U-value and VLT. Yoon et al. [35] also analyzed the effect of glazing performance on building energy performance and discovered that SHGC and VLT have a stronger effect than the U-value on building energy in buildings with high cooling and lighting loads. This is similar to the results of this study.

This study added the impact of WWR to the analysis. The results of analyzing the change and implications of window performance elements according to WWR are displayed in

Table 8. Change rate and influence of each performance indicator on the window area ratio.

	WWR 40%			WWR 50%			WWR 60%			WWR 70%
	U-Value	VLT	SHGC	U-Value	VLT	SHGC	U-Value	VLT	SHGC	U-Value	VLT	SHGC
Change rate (%)	4.80	2.12	17.37	3.31	1.28	22.13	1.27	0.92	26.66	1.31	0.71	31.03
Influence (%)	19.76	8.74	71.50	12.40	4.78	82.83	4.41	3.18	92.41	3.95	2.14	93.91

. Higher WWRs led to less impact of the U-value and VLT, whereas the impact of SHGC increased to 93.91% in WWR 70%. The effect of VLT was insignificant. Since VLT is an element that affects glare and lighting load, the impact in terms of energy performance may have turned out to be small [30,35,37].

Among the performance elements, SHGC showed the most significant change and impact in all WWRs, followed by the U-value and VLT. Therefore, the effect of SHGC was greatest in all WWRs, with higher WWRs resulting in a more pronounced impact. This result indicates that SHGC must be thoroughly considered to improve energy performance in building design.

Second, the two performance elements have different energy impacts and require opposite conditions for reducing loads. According to the change in the U-value and SHGC, the experiment showed that the U-value was disadvantageous to the cooling load and advantageous to the heating load. In contrast, SHGC was advantageous to the cooling load and disadvantageous to the heating load. This result was equally found in other studies as well.

Choi et al. [9] conducted an energy simulation on the interactions of building envelope design factors and discovered that the infiltration air change rate (ACR) and window insulation (WDI) had a considerable influence on the heating load, and SHGC had a great impact on the cooling load. Shin et al. [41] also analyzed that the decrease in the U-value led to a 63% decrease in office heating load and a 30% increase in cooling load, thereby reducing the total heating and cooling loads by 4%. SHGC led to a 6–18% increase in heating load and a 5–12% decrease in cooling load, thereby reducing energy by 16%. Moreover, Lee et al. [37], Raji et al. [42], and Zhao et al. [43], who experimented on the impact of glazing performance according to climate features, discovered that the U-value must be considered in polar regions where heating is required, whereas SHGC must be regarded in hot climates where cooling is needed.

Finally, after examining the heating and cooling loads according to the U-value and SHGC, certain sections exhibited an inversion. As shown in Table 7, loads are increasing along with the U-value, but in combination with SHGC, the loads decreased at the conversion point of SHGC and then increased again. In other words, in combination with SHGC and the U-value, when the U-value is lower than a certain level, it creates a greenhouse effect and increases heating and cooling loads.

Yoon et al. [35] suggested that the U-value is inversed in 1.2 W/m²K. However, this study additionally conducted the WWR experiment and discovered that the glazing standards vary depending on the WWR. Furthermore, an inversion occurs based on SHGC instead of the U-value. This is also proved equally by Yoon et al. [44]. SHGC, where inversion occurs, decreases as WWR increases.

The heat causes inversion, and internal heat accumulated indoors due to solar radiation cannot escape to the outside due to the U-value, thereby increasing the cooling load [24,35]. Thus, more extensive window areas lead to more solar radiation inflow, which reduces the SHGC in the inversion sections and increases the impact of the SHGC. Therefore, it is necessary to consider adequate combinations of the U-value and SHGC to regulate solar radiation inflow and prevent the greenhouse effect.

One thing to note in this study is that buildings with higher WWRs are more greatly affected by SHGC and that there is a need for an optimal combination of the U-value and SHGC to increase the effect of load reduction. Lower U-values lead to less load, but the minimum cooling and heating loads appear differently in combination with SHGC. For example, in WWR 50% and SHGC = 0.5%, the load is smallest when U-value = 1.5 W/m²K but smallest when U-value = 1.0 W/m²K in SHGC = 0.6%. Kim et al. [44] also revealed that building energy demand increases proportionately to SHGC, and buildings with higher WWRs show a higher rate of increase. Thus, it was found that maintaining SHGC at 0.4% or lower is beneficial for energy performance. This intention to provide the optimum SHGC is equivalent to the attempt of this study. However, this study did not limit the scope to SHGC but also revealed the complex relationship between the U-value and SHGC.

Window design is intricately related to various design variables, such as glazing, WWR, envelope area, height, scale, and volume. This study conducted experiments with only glazing and WWR and presented the interaction between SHGC and the U-value but faced limitations in clarifying the relationship between the variables. Moreover, the most energy-efficient combination of the glazing elements can be achieved by adopting a low SHGC. However, the glass becomes darker as the SHGC decreases, complicating the maintenance of views. Therefore, glazing cannot be selected solely on the basis of energy performance. Apart from energy performance, other evaluation criteria, such as indoor environment comfort and occupants’ preferences, must be considered to propose an appropriate range for glazing. Further research is currently underway, and subsequent studies will include these considerations to present a clear relationship between the variables.

4. Conclusions

In this study, we conducted an energy simulation experiment on office buildings with different WWRs to propose a window plan to improve energy performance. This study analyzed the impact of window design variables (U-value, VLT, SHGC, and WWR) and established a hierarchy among them. The analysis showed that, among the glazing elements, SHGC had the most significant impact, followed by the U-value and VLT. SHGC showed an increasing influence with larger WWRs: 71.5% for 40% WWR, 82.83% for 50% WWR, 92.41% for 60% WWR, and 93.1% for 70% WWR. Additionally, for buildings with larger WWRs, considering the optical performance rather than just reducing the U-value is crucial for enhancing the energy load reduction effect. This was verified through experiments on heating and cooling loads based on the U-value and SHGC, confirming their interaction and the importance of reflecting an optimal combination of these elements in window planning. Korea’s existing guidelines only specify that the U-value of windows facing the outside air should be less than 1.5 W/m²K. The results of this study are significant as they provide evidence for enhancing window plans primarily focused on thermal performance.

However, it is essential to acknowledge the limitations of the experimental results, which are confined to office buildings in south Korea. In addition, the simulation was conducted using a model with fixed directions, size, and window area ratio, and analyses of solar radiation inflow and lighting load were excluded. As the experiment was conducted on an office with a large lighting load, this can be pointed out as a clear limitation of this study. To address these limitations, future research will be undertaken to develop specific window design techniques through a more comprehensive analysis, incorporating greater model diversity (region, climate, form, size, etc.).