Retrofit Analysis of Exterior Windows for Large Office Buildings in Different Climate Zones of China

Abstract

In the energy-saving retrofit of existing buildings, investors are particularly concerned about the energy-saving performance of exterior windows and the payback period of additional costs. This study evaluates representative cities in four different climate zones in China to simulate the energy consumption of large office buildings after replacing different glass windows and conducting energy-saving and economic feasibility assessments. The research method includes the following steps: First, a baseline model of large office buildings in four cities was established using AutoBPS and OpenStudio. Then, the baseline and retrofit models of replacing glass windows were simulated using the EnergyPlus V9.3.0 to obtain multiple hourly energy consumption results. The commercial electricity and gas prices in the four cities were adjusted to calculate the total cost within 20 years after replacing different types of windows. Using the discounted payback period (DPP), net present value (NPV), and profitability index (PI) as evaluation indicators, a feasibility analysis was conducted in the four regions to evaluate the economic feasibility of replacing building windows. The simulation results show that considering economic feasibility and meeting energy-saving standards, it is more economical to choose windows with moderate U-value and SHGC value in the four regions than to choose windows with the smallest U-value and SHGC value, and that both energy savings and economic benefits are closely related to building age, with older buildings (especially those in Changsha and Shenzhen) showing greater benefits. Furthermore, the optimal window types in the four cities determined in this study can recover the investment cost within the window life, with Harbin (SC), Beijing (C), Changsha (HC), and Shenzhen (HW) with the payback period of 6.60, 15.66, 10.16, and 11.42 years, respectively. The research model established in this study provides a useful evaluation path for selecting windows for the energy-saving retrofit of large office buildings in cities in different climate zones and provides data support for the decision making of energy-saving retrofit investors.

1. Introduction

Global warming is accelerating with the continuous expansion of human habitats and energy consumption. The net anthropogenic emissions of carbon dioxide (CO₂) must approach zero by mid-century (2050) to stabilize the global mean temperature at the level targeted by international efforts [1]. Decarbonizing the building sector is extremely important to mitigating climate change as the sector contributes 40% of the overall energy consumption and 36% of the total greenhouse gas emissions in the world [2]. Approximately 50% of the energy consumption in the building sector is attributed to the heating, ventilation, and air conditioning (HVAC) of 20 years in this study systems [3]. The energy consumption of HVAC systems depends on the design of the building envelope [4], and windows account for up to 60% of the heat losses in the building, making them the least energy-efficient component [5]. The energy efficiency of the same type of window varies in different climate zones, highlighting the importance of selecting appropriate windows to achieve energy-saving goals in buildings.

China’s building energy efficiency work started in the 1980s. The mandatory building energy-efficiency design standards continue to improve the energy-saving rate, and the requirements for the thermal performance of the envelope structure also continue to improve. Isolating thermal conduction is an effective solution for window thermal management [6]. According to the market survey, the window types in the existing buildings are single-glazing, double-glazing, and multiple-glazing. At present, single-layer glass windows have been withdrawn from the market [7]. Double-layer or multi-layer glass windows generally reduce energy consumption by adding gas, vacuuming, coating, and other ways to isolate heat conduction [8], and the thermal insulation performance of windows has improved. With the global promotion of zero-carbon buildings, functional windows such as aerogel glass windows [9], electrochromic (EC) windows [10], phase change material (PCM) windows [11], hydrogel glass windows [12], and photovoltaic glass windows [13] have been developed. These windows possess energy-saving and other advantages by incorporating relevant functional materials into the glass. For example, high-transparency siliconized cellulose aerogel (SiCellA) has been developed from wood pulp, which can be inserted into gaps between multi-pane glasses, providing superior thermal insulation performance. The window’s transparency can be adjusted by modifying the material composition, and the manufacturing cost is around USD 1 per square foot [14].

Among building energy consumption simulation software, the EnergyPlus software is the most commonly used software [15,16,17]. Through simulation analysis by the EnergyPlus software, many high-performance windows have better energy-saving rates than traditional windows. Vacuum gas-chromic smart windows were found to exhibit excellent thermal insulation performance across varying climate conditions [18]; PCM windows integrated with 3 mm solid-phase change materials outperform clear double-glazed windows, saving up to 17.2%, 14.0%, and 5.8% energy for HVAC systems in warm, mixed, and cold climates, respectively, resulting in the overall energy savings of 9.4%, 6.7%, and 3.2% for the entire building [19]. Another study explored the energy performance of thermochromic smart windows and radiant coolers in a medium-sized office building using EnergyPlus simulations, reporting reductions in the total energy consumption of 10.6% and 23.0%, respectively [20]. Other researchers have focused on integrating advanced materials into window systems. For instance, windows incorporating solid-solid phase change materials (PCMs) and silica aerogels were evaluated using the EnergyPlus software [21]. In Bangladesh, researchers’ simulation using the EnergyPlus software revealed that semi-transparent CdTe-based building-integrated photovoltaic (BIPV) windows could reduce electricity consumption by 30–61% across various climate conditions compared to conventional windows [5]. In southern Italy, photochromic windows (PCWs) were evaluated against traditional low-emissivity (Low-E) windows and transparent double-glazed windows using the EnergyPlus software, and the study concluded that PCWs could achieve annual energy savings of up to 9.3% compared to transparent windows and 4.1% compared to Low-E windows [22]. Despite the ongoing research and promotion of these functional windows, the mainstream market for building windows in China still primarily consists of gas-filled or low-emissivity (Low-E) coated double-pane or multi-pane windows. This highlights a gap between research innovations and market adoptions and the need for further studies to bridge this divide.

Different types of windows have varying economic costs and benefits in different cities [7]. Therefore, the selection of windows for the retrofit of existing buildings is also diversified. For instance, when renovating public buildings such as government schools, cost considerations may favor using coated windows with SHGC values ranging from 0.4 to 0.6, which offer shorter payback periods [23]. In tropical climates, windows retrofitted with glass material with a lower U-value can provide better overall thermal transfer value (OTTV) for the building envelope subsystem. Reducing energy consumption translates into energy cost savings for buildings with retrofitted windows that can offset the installation costs of new glass materials [24]. In the hot summer and cold winter regions of China., the optimized variable transparency shape-stabilized phase change material (VTSS-PCM) window, the VTSS-PCM window, reduced the total annual unfavorable heat transfer (TAHT) by 30.14% and the total annual cost by 28.39% [25]. In the hot, arid climate, double-pane windows with either air or argon as the insulator gas exhibited the shortest payback period [26]. In a residential building, the operating costs of PVC windows are over 30% lower than in the case of wooden windows and almost 20% lower compared to aluminum windows [27]. In cold climates, the application of the low-emissivity window film on the outward-facing surface of the inner pane of the double-glazed windows helped to reduce heat loss through the windows in winter and unwanted heat gains in summer by almost 36% and 35%, respectively. This resulted in a 6% reduction in the building’s annual energy consumption for heating purposes. However, the relatively high price of the films and the low price of district heating resulted in a rather long payback period of around 30 years [28]. The economic costs and benefits of window glass are also affected by the direction in which it is installed. In the study of 30 double-glazing configurations in the Indian composite climate zone, researchers found that based on the reduction in annual net cooling and heating costs, the optimal window orientation order is southeast, southwest, south, northeast, northwest, north, east, west [8]. In hot climates, large windows can be installed on the north side of the building, while a small window is recommended for the south side to minimize the building’s energy consumption [29]. Therefore, in the energy-saving retrofit of existing buildings, if the budget is limited, you can prioritize the replacement of the window with the best orientation.

The research above shows that the continuous emergence of novel window technologies and materials has promoted global carbon reduction efforts. However, when retrofitting existing building windows, regional variations and long-term costs need to be further studied because the economic benefits of window technology vary from region to region, and detailed cost-benefit analysis is carried out for specific local climate and economic conditions. Assessing the economic benefits between the energy savings and replacement costs of advanced window technologies to understand their true economic impact over the building’s life cycle is critical to making informed decisions about window retrofitting. The purpose of this study was to investigate the effects of different window types on building energy consumption and economic payback periods in four different climate zones in China. This paper utilizes EnergyPlus to establish energy simulation models for large office buildings in these four climate zones. Furthermore, it examines various window options based on their prices and associated energy consumption costs, ultimately identifying the most cost-effective window selection for office buildings in each climate zone. This research provides valuable insights into the economic implications of window retrofitting, emphasizing the importance of region-specific analyses.

2. Methodology

The research workflow of this paper is shown in Figure 1. The first step was establishing baseline energy models for large office buildings in different climate zones. According to the “Uniform Standard for the design of civil buildings (GB 50352-2019) [30] in China, China contains five major climate zones. In this study, large office buildings in Harbin (severe cold), Beijing (cold), Changsha (hot summer and cold winter), and Shenzhen (hot summer and warm winter) were selected as research objects. The baseline building models for new buildings were generated using AutoBPS. AutoBPS is a building energy modeling tool developed by Hunan University in Changsha, China [31]. AutoBPS can generate prototype building energy models using OpenStudio and EnergyPlus for different combinations of climate zone, year built, and building type [32]. AutoBPS can modify various parameter scenarios and generate different building models according to different specifications and standards [33,34], and can be used for retrofit analysis [35], demand response analysis [36], photovoltaic generation analysis [37], and thermal storage system analysis [38].

The second step was to develop the retrofit energy models for each window replacement scenario for each climate zone. Based on market research and inquiries, five different window replacement scenarios were selected for each climate zone. The window construction in EnergyPlus was modified to simulate the energy consumption of each window replacement scenario of each climate zone. The hourly energy consumption data of the baseline and retrofit models were used for the retrofit analysis.

The third step was to perform an economic analysis of five window replacement scenarios for each climate zone. The annual energy cost of the building was calculated based on city-specific electricity and natural gas prices, combined with the cost of the glass windows. The total cost of the building over 20 years in different climatic zones was then calculated, considering both the capital value and the time value of money. An economic performance analysis model was established to reflect the economic feasibility of window replacement. The feasibility of optimal window selection in the four regions was analyzed using DPP, NPV, and PI. The total 20-year cost of the building in this paper referred to the sum of the building’s energy cost and the cost of windows, ignoring other costs such as building maintenance cost.

2.1. Climate Zones and Weather Data

In China, a major climate classification is for the thermal design of buildings, mainly concerned with the conduction of heat gain/loss through the building envelope and the corresponding thermal insulation issues [39]. It has five major climatic types: severe cold, cold, hot summer and cold winter, mild and hot summer, and warm winter.

The energy consumption of buildings in mild climate zones was less sensitive to the building envelope system. Therefore, four cities were selected to represent the other four climate zones, including Harbin (severe cold), Beijing (cold), Changsha (hot summer and cold winter), and Shenzhen (hot summer and warm winter). Their specific information is shown in

Table 1. Geographical coordinates of the regions.

City	Thermal Climate Zone		Longitude	Latitude	HDD18	CDD26	Solar Energy Resource Area	Climatic Regions
	GB50176-2016 [40]	ASHRAE Standard 169-2021 [41]
Harbin	1C	7	126.53	45.8	5032	14	II	Severe cold (SC)
Beijing	2A	4	116.4	39.9	2699	94	II	Cold (C)
Changsha	3B	3	112.93	28.23	1466	230	III	Hot summer and cold winter (HC)
Shenzhen	4A	2	114	22.53	223	374	III	Hot summer and warm winter (HW)

. Geographical coordinates of the regions.

The weather data in the EnergyPlus Weather (EPW) format were obtained from the official website (https://energyplus.net/weather accessed on 20 July 2024) for the cities.

2.2. Generation of Baseline Energy Models for Large Office Buildings

The case study large office prototype building has a total floor area of 43,330.86 m², with a north-south orientation, comprising twenty above-ground floors and one basement floor. The length is 57.83 m, and the width is 35.68 m. The floor-to-floor height is 3.96 m, and the total building height is 81.7 m. The building has a rectangular geometry with a window–wall ratio of 0.3881. As shown in Figure 2, the top, first, and basement floors were modeled in detail, while the middle floor was created with a multiplier of 18 to represent the other 18 floors to speed up the simulation. The geometrical specifications of the large office building remain consistent across the four regions studied.

The heating, ventilation, and HVAC systems consist of chillers and cooling towers for cooling, the water-cooled centrifugal chiller nominal capacity is 1.17 MW and natural gas boilers for space heating and service hot water systems. The operational schedules of HVAC systems, occupants, lights, and plug loads are established according to the Chinese Standard GB50189-2015 [42].

Table 2. Major parameters of the building envelope and internal loads.

Parameters	Harbin	Beijing	Changsha	Shenzhen
Exterior wall heat transfer coefficient (W/m2·K)	0.35 (≤0.43)	0.45 (≤0.50)	0.60 (≤0.60)	0.80 (≤0.80)
Roof heat transfer coefficient (W/m2·K)	0.25 (≤0.35)	0.40 (≤0.45)	0.40 (≤0.40)	0.50 (≤0.50)
Window U-value/SHGC	2.208/0.217 (≤2.3/-)	2.262/0.422 (≤2.4/≤0.48)	2.493/0.356 (≤2.6/≤0.40)	2.888/0.316 (≤3.0/≤0.35)
People density (person/m2)	0.05 (0.1)	0.05 (0.1)	0.05 (0.1)	0.05 (0.1)
Equipment power density (W/m2)	15 (15)	15 (15)	15 (15)	15 (15)
Lighting power density (W/m2)	8.93 (0.90)	8.93 (0.90)	8.93 (0.90)	8.93 (0.90)
Cooling/heating temperature setpoint (°C)	26/20 (26/20)	26/20 (26/20)	26/20 (26/20)	26/20 (26/20)
Chiller COP	5.40 (5.40)	5.50 (5.50)	5.60 (5.60)	5.70 (5.70)
Boiler heating efficiency	0.9 (0.9)	0.9 (0.9)	0.9 (0.9)	0.9 (0.9)

Note: The value in () is the standard value according to the Chinese Standard GB50189-2015.

presents the major parameters of the building envelope and internal loads, including the U-factor and SHGC of exterior windows.

Table 3. Window composition of the baseline model for the four cities.

City	Glass Type
Harbin	REF C TINT HI 6 mm glass + 13 mm air + 6 mm clear glass
Beijing	ECREF-2 COLORED 6 mm glass + 13 mm argon gas + 6 mm clear glass
Changsha	REF D TINT 6 mm glass + 13 mm air + 6 mm clear glass
Shenzhen	REF B CLEAR LO 6 mm glass + 13 mm argon gas + 6 mm clear glass

Note: the order of glass type marking is from outer layer to inner layer; REF B is titanium coating; REF C is pewter coating; REF D is tin-oxide coating; TINT is tinted with inorganic materials to increase absorption; LO, HI is low- and high-transmittance coating, respectively; ECREF is electrochromic glass that darkens by becoming more reflective; COLORED is the colored (darkest) state of electrochromic glass.

shows the window composition of the baseline model for the four cities.

In order to compare the authenticity of the baseline model, eight public building windows were collected in four different climate zones from other existing research papers. Their window U-values and SHGC values were selected, and similar values were found in the EnergyPlus parameter file as simulation parameters. The specific parameters are shown in

Table 4. Collected cases’ window parameter information.

Code	City	Building Types	U-Value (W/(m2·K))	SHGC (-)	Solar Energy Resource Area	Climatic Regions
Case1 [39]	Harbin	office	2.556	0.609	II	(SC)
Case2 [43]	Shenyang	office	5.067	0.4	II	(SC)
Case3 [39]	Beijing	office	2.888	0.316	II	(C)
Case4 [44]	Seoul	daycare centers	3.835	0.768	II	(C)
Case5 [45]	Changsha	office	5.894	0.861	III	(HC)
Case6 [46]	Wuxi	four-star resort hotel	3.835	0.768	III	(HC)
Case7 [47]	Phoenix	office	2.511	0.704	III	(HW)
Case8 [39]	Hongkong	office	4.513	0.781	III	(HW)

.

The main goal of this study is to evaluate the energy-saving potential and economic feasibility of building window retrofits in China’s climate region. Although the window parameters of Seoul and Phoenix are not directly applicable to building energy analysis in China, their wide performance range (including high SHGC value and low U-value windows) provides a data reference for this study to expand the comparative analysis of window performance. To ensure the applicability of the study, the main simulation results and conclusions are based on window parameters that meet the Chinese design standards.

2.3. Window Type Selection

The shape coefficient is the ratio of the exterior area of a building to the volume enclosed by the exterior area. The shape coefficient of the case study building is 0.11 m⁻¹. Based on the window–wall ratio and shape coefficient of the building, the exterior window requirement of the Chinese Standard GB55015-2021 [48] for the four climate zones is listed in

Table 5. Exterior window requirement of GB55015-2021 for the case study building.

Parameters	Harbin	Beijing	Changsha	Shenzhen
U-value (W/(m2·K))	≤2.1	≤2.0	≤2.2	≤2.5
SHGC (-)	-	≤0.40	≤0.35	≤0.30

. Ten different glass windows were selected from the EnergyPlus database, and their U-value and SHGC are shown in

Table 6. U-value and SHGC values for glazed windows.

Name	Glass Type	U-Value ((W/(m2·K))	SHGC	Harbin	Beijing	Changsha	Shenzhen
W1	6 low-transmittance Low-E + 13 air + 6 transparent	1.761	0.568	X
W2	3 low-transmittance Low-E + 6 air + 3 clear + 6 air + 3 clear	1.525	0.472	X
W3	6 low-transmittance low-film66 + 13 air + 6 transparent + 13 air + 6 transparent	1.22	0.363	X	X
W4	6 low-transmittance low-film66 + 6 air + 6 transparent + 6 air + 6 transparent	1.712	0.358		X
W5	6 tri-silver Low-E + 12 air + 6 transparent	1.63	0.296		X	X
W6	6 tri-silver low-film66 + 13 air + 6 transparent + 13 air + 6 transparent	1.22	0.254			X	X
W7	6 tri-silver low-film33 + 13 air + 6 clear + 13 air + 6 clear	1.196	0.154	X	X		X
W8	6 highly transparent heat reflective + 13 air + 6 transparent	2.416	0.234				X
W9	6 colored heat absorption + 13 air + 6 transparent	1.761	0.169			X	X
W10	6 tri-silver low-film66 + 6 air + 6 clear + 6 air + 6 clear	1.712	0.265			X

Note: all glass and inter-layer air thickness units are “mm”; film66 and film33 are coated polyester films with a nominal visible transmittance of 66 percent and 33 percent; “X” represents the selection.

. Figure 3 shows the U-value and SHGC of the ten glass windows and the requirements of the four climate zones. As shown in Table 5, for the Harbin region, window types W1, W2, W3, and W6 were chosen for energy simulation; for the Beijing region, window types W3, W4, W5, and W7 were chosen; for the Changsha region, window types W5, W6, W9, and W10 were chosen; and for the Shenzhen region, window types W6, W7, W8, and W9 were chosen. The energy consumption simulations were carried out accordingly for each region. This study did not consider window shading devices (such as roller shades or blinds) and their opening and closing schedules. Future research could further refine the impact of shading devices on energy consumption.

This study did not consider window shading devices (such as roller shades or blinds) and their opening and closing schedules. Future research could further refine the impact of shading devices on energy consumption.

2.4. Cost and Energy Costs

2.4.1. Initial Total Investment of Window

In China, the price of windows includes glass and labor. The maintenance cost of windows during use is low, and the recycling efficiency of windows after scrapping is not high. Therefore, this paper does not consider the maintenance cost of windows during use and the recycling efficiency after scrapping. Through the investigation of 6 glass manufacturers, the unit area prices of ten types of windows in the four cities were obtained, as shown in

Table 7. Price per unit area of windows.

Name	Harbin (CNY/m2)	Beijing (CNY/m2)	Changsha (CNY/m2)	Shenzhen (CNY/m2)
W1	412	-	-	-
W2	455	-	-	-
W3	479	527	-	-
W4	-	540	-	-
W5	-	544	522	-
W6	590	-	603	847
W7	-	680	-	1103
W8	-	-	-	906
W9	-	-	624	1059
W10	-	-	550	-

Note: “-” indicates no query.

. The window price in Harbin is 412–610 CNY/m², 527–680 CNY/m² in Beijing, 522–624 CNY/m², and 786–1103 CNY/m² in Shenzhen. The total area of windows in this building is 5929.56 square meters. According to the price data in Table 7, the total cost is the initial investment cost of the window retrofit.

2.4.2. Total Energy Cost

The research object of this paper is public buildings, the main energy sources of which are electricity and natural gas.

Table 8. Electricity tariffs for commercial and industrial use in four cities.

City	Critical Peak Period (CNY/kWh)	On Peak Period (CNY/kWh)	Mid-Peak Period (CNY/kWh)	Off-Peak Period (CNY/kWh)
Harbin	1.3294	1.1181	0.7660	0.4139
Beijing	1.3029	1.1572	0.8333	0.5499
Changsha	1.4576	1.2223	0.7813	0.3403
Shenzhen	1.4259	1.1462	0.7615	0.2770

Note: data from China State Grid’s latest electricity price in August 2024.

presents the industrial and commercial time-of-use electricity tariffs for the four cities, as obtained from the official websites of their respective provincial and municipal development and reform commissions. The four regions have different off-peak hours, mid-peak hours, on-peak hours, and critical peak hours.

In northern China, natural gas is the main heating energy source; the heating season is from November 15 to March 15 each year. The calorific value of natural gas is about 35.53 MJ/m³, and

Table 9. Industrial natural gas prices in four cities.

City	Prices (CNY/m3)
Harbin	3.48
Beijing	2.45
Changsha	3.853
Shenzhen	4.3

shows the natural gas prices in the four cities queried on the official website of each city by China’s Development and Reform Commission.

The total energy cost M is calculated by Equation (1):

$$C_{energy}={\displaystyle \sum _{i=1}^{8760}{P}_h}\times Q_h$$

(1)

where

• P_h is the energy price corresponding to moment h;
• Q_h is the energy consumption corresponding to the moment h, calculated from the hour-by-hour electric energy consumption and hour-by-hour natural gas energy consumption generated by EnergyPlus.

In this study, the energy-saving potential of the baseline, the collected cases, and the retrofit case with the lowest energy cost were analyzed, after which the energy-saving rate could be obtained using Equation (2).

$$Energy-saveingrate=\left(1-{\displaystyle \frac{E{C}_y}{E{C}_y}}\right)\times 100\%$$

(2)

where

• EC_y is the annual energy consumption cost for the retrofit cases;
• EC_y is the annual energy consumption cost for the baseline or collected cases.

2.5. Economic Indicators

There are many evaluation indexes of project economic benefit, such as payback period, return on investment, net present value of investment, etc. Each index reflects the economy of the project from an important angle. Whether the time value of money is considered in the evaluation index system of investment projects, the evaluation index of economic effect can be divided into static index and dynamic index. Static evaluation indicators include static investment payback period and investment return rate, and the calculation of indicator data does not consider the time value of funds. The calculation of dynamic evaluation indicators needs to consider the time value of funds, which are generally net present value, internal rate of return, investment present value rate, and so on. This study conducted a feasibility analysis in four regions using DPP, NPV, and PI as evaluation metrics to assess the economic viability of window replacement in buildings.

2.5.1. Cumulative Discounted Cash Flow Payback Period (DPP)

After replacing the windows, the total operating cost of the building (C_total) includes the initial investment cost (C_initial) and the lifetime energy cost, calculated over a 20-year service life. For details, see Equations (3) and (4).

$$C_{initial}=P_w\times S_w$$

(3)

$$C_{total}=C_{initial}+C_{energy}\times T$$

(4)

where

• Pw = price of ten different glass windows;
• Sw = total window area of the building;
• T = 20 years.

A simple payback period (PP) can serve as a basic index for evaluating a project. However, it does not account for the time value of money, may not accurately reflect the project’s actual cash flow, and cannot truly represent the real financial situation. Therefore, the DPP, which is the time required to accumulate discounted cash flows to return the initial investment, is used. We used Formula (5) to calculate the annual cash inflow amount, calculated the discounted cash flow using Equation (6), and determined the investment return time using Equations (7) and (8); DPP occurs in the year when V_t passes from negative to positive.

$${\mathrm{C}}_{\mathrm{t}}=E{C}_y-E{C}_y$$

(5)

$$PV={\displaystyle \frac{C_t}{{(1+i)}^t}}$$

(6)

when t = 0

$$V_0=-C_{initial}$$

(7)

when t = 1, 2, 3 …T

$$V_t=V_0+{\displaystyle \frac{C_t}{{(1+i)}^t}}$$

(8)

where

• V_t = unrecovered initial investment;
• C_t = net cash inflow during the period t, That is the energy cost that can be saved by replacing windows in each city;
• i = discounted rate;
• t = period in which cash inflows are related.

2.5.2. Net Present Value (NPV)

NPV is commonly used in projects to assess the benefits and costs of projects [49]. If the NPV of a project is positive, it indicates that the project can recover the investment. The greater the NPV, the earlier the investment cost recovery and the faster the return on investment [37,50]. NPV can be calculated by Equation (9).

$$NPV={\displaystyle \sum _{t=1}^T\frac{C_t}{{(1+i)}^t}}-C_{initial}$$

(9)

where

• C_t = net cash inflow during the period t; t is the number of time periods;
• i = discounted rate.

2.5.3. Profitability Index (PI)

The profitability index (PI) is a ratio of cash inflow’s present value and cash outflow’s present value from a project. PI measures the financial attractiveness of a project, with higher values corresponding to more attractive projects [51], and can be calculated using Equation (10).

$$PI={\displaystyle \frac{{\displaystyle \sum _{t=1}^T\frac{C_t}{{(1+i)}^t}}}{C_{initial}}}$$

(10)

where

• C_t = net cash inflow during the period t;
• t = the number of time periods.

3. Simulation Results

Considering the different climate zones in China, this paper established baseline energy models for large office buildings in four climate zones. Then, retrofit models were developed in each climate zone, considering five scenarios. Finally, this paper performed an economic analysis of five window replacement scenarios using three economic indicators.

3.1. Baseline Simulation Results

The energy consumption of the baseline and collected cases across four cities was simulated using the EnergyPlus software. It was difficult to quantitatively analyze the simulation results due to the lack of real data. The baseline was set based on the Chinese design standards, the Changsha prototype building model, and information on case windows collected from the literature. Governments require the Chinese design standards in any climate zone. The reasonableness of the prototype buildings was verified by comparing the total energy consumption represented by the prototype buildings with the local statistical yearbook [52]. The case window information was real data rather than hypotheses. Therefore, the results of the baseline were reasonable. The results for monthly electricity consumption are presented in Figure 4, while Figure 5 shows the monthly natural gas consumption. As observed in Figure 4, with the exception of Case 3 in Beijing, where monthly electricity consumption closely mirrors that of the baseline, the electricity consumption in all the other cases significantly exceeds the baseline. This indicates that public buildings constructed in compliance with newer standards consume less electricity, and replacing exterior window materials with more efficient options contributes to energy savings.

In Figure 5, the monthly natural gas consumption of Harbin Case 1, Shenzhen Case 7, and Shenzhen Case 8 is lower than the baseline. Referring to Table 4, it is notable that the SHGC values for these cases are higher than those of the baseline and other cases within the same climate zone. We hypothesize that this is due to the specific climate conditions in Harbin, which is located in a severely cold zone with high heating demand in winter. A higher U-value generally leads to greater heat loss, but the increased SHGC in Case 1 may allow for higher solar heat gains during the day, partially offsetting the heat loss and reducing heating demand, particularly in winter. Similarly, Shenzhen, located in a subtropical climate zone where cooling is the primary energy demand, benefits from windows with higher SHGC values, as they increase solar heat gain and reduce the cooling load. Figure 6 illustrates the annual energy consumption across the four cities. With the exception of Case 3 in Beijing, where energy use is slightly lower than the baseline, the baseline scenario has the lowest total energy consumption.

3.2. Retrofit Case Simulation Results

The annual energy cost of each retrofit case was calculated based on the energy price and electricity price structure (see Table 8 and Table 9, for specific data). The results are summarized in

Table 10. Cost of energy consumption per square meter per year.

Name	Harbin (RMB/m2)	Beijing (RMB/m2)	Changsha (RMB/m2)	Shenzhen (RMB/m2)
W1	100.14	-	-	-
W2	98.70	-	-	-
W3	97.38	87.80	-	-
W4	-	88.06	-	-
W5	-	88.42	87.57	-
W6	99.89	-	87.25	103.81
W7	-	87.55	-	95.07
W8	-	-	-	102.80
W9	-	-	87.02	101.36
W10	-	-	89.99	-

Note: “-” indicates no simulation.

, which lists the annual energy costs of the retrofit options in the four cities. The window type with the lowest annual energy cost was selected for the four different regions from Table 7 for the next analysis.

After simulating the retrofit model with replaced windows, the energy-saving effect was evaluated and compared with the baseline and collection cases, and the energy-saving potential was analyzed. The results shown in Figure 7 show that retrofitting exterior windows can produce significant energy-saving effects in both the baseline and collection cases. In Changsha (HC) and Shenzhen (HW), replacing exterior windows with high-performance materials can significantly save energy, with Changsha (HC) achieving a reduction of 10.95% and Shenzhen (HW) achieving a higher reduction of 20.01%.

3.3. Economic Feasibility Analysis

In previous studies [33,53], the service life of windows was often set at 20 years, and related research, such as window retrofit, was carried out based on this. So, the window life cycle is assumed to be 20 years in this study. The economic viability of optimal window selection is assessed using three key financial metrics: the DPP, NPV, and PI. Table 7 presents the retrofit costs for various window options, excluding the minimum maintenance costs and the residual value at disposal.

Using Equations (3) and (4) and the data provided in Table 3 and Table 4, the total cost of the renovated building over the 20-year service life of the windows was calculated for different window options across the four cities. The results, summarized in

Table 11. Total 20-year cost for different windows in four cities.

Harbin	Total Cost (CNY/m2)	Beijing	Total Cost (CNY/m2)	Changsha	Total Cost (CNY/m2)	Shenzhen	Total Cost (CNY/m2)
W1	2414.73	W3	2283.02	W5	2273.47	W6	2923.1
W2	2429.09	W4	2301.12	W6	2348.08	W7	3004.37
W3	2426.69	W5	2312.35	W9	2364.31	W8	2961.97
W6	2587.9	W7	2430.94	W10	2349.74	W9	3086.23

, identify the most cost-effective window selections for each city: Harbin (W1), Beijing (W3), Changsha (W5), and Shenzhen (W6).

This study adopts a discount rate of 3.576% per year based on the data published by the International Monetary Fund on 25 September 2024. This discount rate ensures that the economic analysis reflects the current financial conditions in China. The results of the DPP were calculated using Equations (5)–(7), with detailed values presented in

Table 12. The DPP of the optimal window selections in the four different cities.

City	Selected Window	Retrofitting Buildings	DPP (Year)
Harbin	W1	Baseline	>20
		Case1	>20
		Case2	6.60
Beijing	W3	Baseline	>20
		Case3	>20
		Case4	15.66
Changsha	W5	Baseline	>20
		Case5	10.16
		Case6	15.66
Shenzhen	W6	Baseline	>20
		Case7	14.14
		Case8	11.42

. Similarly, the NPV of the optimal window selection was determined using Equation (8) and is summarized in

Table 13. The NPV of the optimal window selections in the four different cities.

City	Selected Window	Retrofitting Buildings	Returns Within the Service Life (RMB)	Initial Investment (RMB)	NPV (RMB)
Harbin	W1	Baseline	38.60 × 104	244.30 × 104	−205.70 × 104
		Case1	182.10 × 104		−62.20 × 104
		Case2	64.97 × 104		405.34 × 104
Beijing	W3	Baseline	114.50 × 104	312.49 × 104	−197.98 × 104
		Case3	58.30 × 104		−254.19 × 104
		Case4	439.85 × 104		127.36 × 104
Changsha	W5	Baseline	44.11 × 104	309.52 × 104	−265.41 × 104
		Case5	620.91 × 104		311.38 × 104
		Case6	443.41 × 104		133.89 × 104
Shenzhen	W6	Baseline	72.07 × 103	502.23 × 104	−495.03 × 104
		Case7	762.71 × 104		260.48 × 104
		Case8	921.16 × 104		418.93 × 104

. For the baseline scenario, the DPP for window replacements exceeds the service life of the windows, resulting in negative NPV values. In the case models, the results vary across different cities and building types. In Harbin and Beijing, for buildings constructed before 2015, while the replacement of high-performance windows offers energy-saving potential, the high initial costs lead to unfavorable economic outcomes. Specifically, in Cases 1 and 3, the DPP exceeds the service life of the windows, and the NPV remains negative. In Changsha and Shenzhen, high-performance window replacements in pre-2015 buildings demonstrate both energy-saving and economic feasibility. The payback periods are within the service life of the windows, with values of 10.16 years (Case 5), 15.66 years (Case 6), 14.14 years (Case 7), and 11.42 years (Case 8), respectively. These results highlight the greater economic viability of window retrofits in warmer climates with higher cooling energy demands.

According to the results of the NPV, the retrofitting buildings with positive values were screened out and further analyzed through the PI. The PI is calculated using Equation (9), as shown in

Table 14. The PI of the optimal window selections in the four different cities.

City	Window	Retrofitting Buildings	PI
Harbin	W1	Case2	2.66
Beijing	W3	Case4	1.41
Changsha	W5	Case5	2.01
		Case6	1.43
Shenzhen	W6	Case7	1.52
		Case8	1.83

. From the results in Table 14, it can be seen that the PI values of the retrofit models screened by the NPV are all more than one, indicating that the replacement of high-performance windows has good profitability.

Based on the evaluation of the above three economic indicators, Harbin (W1), Beijing (W3), Changsha (W5), and Shenzhen (W6) have energy-saving potential and are economically feasible in the energy-saving retrofit of buildings constructed before 2015 in the four cities.

4. Discussion

This study evaluates the energy efficiency and economic feasibility of replacing exterior windows with high-performance materials in public buildings in four cities. The analysis results indicate that window characteristics are a key determinant of the effectiveness of such energy efficiency retrofits and vary significantly between climate zones, which is consistent with previous studies of the same type [54]. The key findings, window selection recommendations, and future research directions are highlighted below.

(1)

Window characteristics and retrofit effectiveness

Buildings constructed before implementing modern energy standards (e.g., GB50189-2015) typically realize more significant energy savings and economic benefits when retrofitted with high-performance windows. This trend is particularly evident in Changsha and Shenzhen, where older buildings with lower base thermal performance realized significant energy consumption reductions, resulting in shorter payback periods and NPV. The longer the time lag between building construction and implementing the latest standards, the lower the performance of their original windows, and the more significant the energy savings resulting from the retrofit.

In contrast, the existing windows with lower U-values and SHGC values had lower economic returns, even though some energy savings can be realized. For example, in Harbin and Beijing, the high cost of replacing otherwise relatively efficient windows resulted in payback periods that exceeded the useful life of the windows, leading to negative net present values. In such cases, alternative strategies such as improved wall or roof insulation may be more effective.

(2)

Window selection strategies for climate zones

This study’s important contribution is providing actionable guidelines for exterior window selection for different window characteristics and climate zones. For older buildings in hot summer climate zones (e.g., Changsha and Shenzhen), windows with high SHGC can significantly improve cooling energy efficiency, consistent with the performance characteristics of the Phoenix data. In contrast, in cold climate zones (e.g., Harbin), windows with low heat transfer coefficients (U-values) (similar to the Seoul data) are more effective in reducing heating energy consumption. For new construction, window selection should be carefully weighed against alternative retrofit strategies to ensure affordability.

This recommendation for window selection by climate zone and window characteristics provides additional refinement to the existing energy efficiency retrofit studies and directly aligns with the goal of this study to optimize exterior window retrofit strategies.

(3)

Policy implications

The findings emphasize the importance of developing targeted retrofit programs and climate-responsive policies. In regions with a large share of older buildings, such as Changsha and Shenzhen, energy efficiency retrofits can be promoted through phased incentives (e.g., window replacement subsidies or loan discounts) to maximize energy and economic benefits. In regions with predominantly new buildings, promoting other retrofit measures, such as insulation improvements, may be more appropriate. Policymakers should also consider window retrofit guidelines for different climate zones to optimize energy savings and affordability.

(4)

Limitations and future research directions

There are some limitations to this study. For example, the effects of other envelope factors, such as wall or roof insulation, were not considered, and these factors may work together with window retrofits to affect the overall energy savings. In addition, although international data were used to extend the performance characteristics, the main analysis was still dominated by the Chinese context, and the international applicability of the study’s findings needs to be further explored. Future research could delve deeper into the synergistic effects of integrated retrofit strategies for windows, walls, and roofs and the retrofit effects under different global standards.

Internal shading devices (e.g., curtains and shades) and external shading devices (e.g., roller shades and blinds) play an important role in building energy performance. According to Raimundo et al. [55], these shading devices are effective in reducing solar heat gain in the summer, thus reducing cooling demand and assisting heating in the winter by optimizing solar gain. In addition, the study by Steen Englund et al. [56] verified the significant impact of shading device use and opening/closing strategies on building energy consumption through actual measurements and simulations, especially in buildings with high variations in occupancy behavior. These studies show that the proper configuration and operational management of shading devices can significantly improve energy efficiency in different climatic conditions and building types. Therefore, future research should further focus on shading device design optimization and intelligent control strategies to maximize building energy efficiency.

5. Conclusions

This study investigated the energy efficiency and economic feasibility of high-performance window replacement in different climate zones and building types. The results highlight the importance of tailoring retrofit strategies to a building’s specific climate, economy, and built climate, economy, and environment. The following are the main conclusions drawn from the analysis:

(1)

Economic evaluation beyond U-values and SHGC values

The economic performance of window replacement is not only affected by the thermal transfer coefficient (U-value) and SHGC, although these indicators are critical. Rather than pursuing the minimum U-value and SHGC values, the study identified a balance that both meets energy conservation standards and optimizes economic outcomes. Moderate U-values and SHGC values are generally more cost-effective, especially in the climate zones studied.

(2)

Economic feasibility in Changsha and Shenzhen

High-performance windows are both energy-efficient and economically viable in Changsha and Shenzhen. The discounted payback period in these cities is within the useful life of the window, and the NPV is positive. These results show that window retrofits can achieve dual goals under specific climate conditions and building characteristics: improving sustainability and achieving a return on investment.

(3)

Policy recommendations for older buildings

Retrofit programs targeting older buildings have the greatest potential to achieve significant energy savings and economic returns. In areas such as Changsha and Shenzhen, where much of the building stock predates modern standards, policymakers could prioritize retrofitting these buildings. Incentive mechanisms, such as subsidies or financial support for window replacement, could encourage building owners to adopt energy-efficient upgrades, helping to align personal economic interests with broader sustainable development goals.