LCA and Scenario Analysis of Building Carbon Emission Reduction: The Influencing Factors of the Carbon Emission of a Photovoltaic Curtain Wall

Abstract

The problem of global warming has become a major global concern, and reducing greenhouse gas emissions is crucial to mitigate its effects. Photovoltaic power generation is clean, low-carbon energy. Photovoltaic products can convert solar energy into electricity, reducing CO₂ emissions to an extent. This paper introduces the life cycle evaluation theory to assess the carbon emissions of photovoltaic curtain walls. PVsyst software allows for the simulation and calculation of power generation under different influencing factors, which provides valuable information about the carbon reduction potential of photovoltaic curtain walls. The evaluation of carbon emissions and their influencing factors using grey correlation analysis further enhances the understanding of the benefits and limitations of photovoltaic curtain walls. According to the results of grey correlation analysis, this paper concludes that the degree of various influencing factors on carbon emission of a photovoltaic curtain wall under different scenarios in descending order is as follows: orientation, location, inclination, shadow occlusion, and seasonal changes. The research findings of this paper provide a theoretical reference for the future development and application of photovoltaic curtain walls. By demonstrating the carbon reduction potential of this technology, this study contributes to promoting the adoption of photovoltaic curtain walls as a sustainable solution to mitigate the effects of global warming.

1. Introduction

Global warming has become the focus of the world’s attention in recent years. It is an almost irreversible trend. One of the main causes of global warming is the massive emission of greenhouse gases. In 2009, the “Buildings and Climate Change” report of the United Nations Environment Programme (NNEP) said that global buildings emit 8.6 billion tons of CO₂ every year, and emissions will increase to 203.0156 billion tons by 2030 [1]. In the United States, the construction industry ranks third in greenhouse gas (GHG) emissions [2]. In the European Union, the construction industry is the largest energy consumer, accounting for approximately 36% of greenhouse gas emissions and 40% of the total energy consumption [3]. To cope with global warming, China has put forward the goal of “peak carbon dioxide emissions, carbon neutrality” [4].

According to data from the China Building Energy Consumption Research Report in 2021, the total energy consumption of China’s construction industry in 2018 was 2.147 billion tce, accounting for 46.5% of China’s total energy consumption. In 2018, the total carbon emissions of China’s construction industry were 4.93 billion tCO₂, accounting for 51.2% of China’s total energy carbon emissions [5]. Therefore, accelerating the process of energy conservation and emission reduction in the construction industry is a key link to realizing the goal of “peak carbon dioxide emissions and carbon neutrality” in China.

Series of photovoltaic products have different energy properties in the process of production and use. The production of photovoltaic products is a process of energy consumption. However, photovoltaic products can convert solar energy into electric energy in the use process. From the carbon emissions perspective, photovoltaic power generation is a clean and low-carbon energy source [6]. Therefore, the application of photovoltaic products is a process of energy production.

Some of the research mentioned [7,8,9], such as the use of life cycle assessment and environmental assessment models, can provide valuable insights into the carbon emissions of buildings and the factors that influence them. However, as pointed out, more detailed research is needed to better understand the specific factors contributing to carbon emissions in different aspects of the construction industry. Furthermore, current research on photovoltaic curtain walls, such as Wu et al. [10], Li et al. [11], and Carvalho et al. [12], is still in its infancy, many of who evaluate its performance and economy but rarely involve relevant research on the factors related to carbon emission reduction throughout its life cycle. In addition, there is a lack of relevant research on influencing factors of the carbon reduction effect of photovoltaic buildings. Therefore, a new research idea is to study the carbon emission reduction potential of photovoltaic curtain walls under different scenarios.

This study aims to obtain the carbon reduction potential of photovoltaic curtain walls and quantitatively analyze the impact of various factors on carbon emissions. The study results provide an objective assessment of the influence of each factor on the carbon emissions of photovoltaic curtain walls. Furthermore, the research findings could also help inform decision-making around adopting photovoltaic curtain walls as a carbon reduction strategy. In this paper, the life cycle assessment theory and Donghe carbon emission analysis and calculation software developed by Southeast University are introduced into the carbon emission calculation of photovoltaic curtain walls. To obtain the carbon reduction of photovoltaic curtain walls, this paper simulated and calculated the power generation under different influencing factors using PVsyst 7.2 software. Then the paper assessed and sorted the carbon emission and its influencing factors in the whole life cycle of photovoltaic curtain walls by grey association analysis.

2. Literature Review

2.1. The Carbon Emissions of the Construction Industry

In recent years, due to the relative epidemic improvement, the economy has begun to recover. However, it has also brought about an increase in the cost of electricity and other energy costs. The construction industry, as a “big household” of electricity consumption and carbon dioxide emissions, urgently needs to be transformed. The following are some studies that have analyzed the factors affecting carbon emissions in China’s construction industry. Zhang Guorui et al. [13] calculated the carbon emissions of buildings in the operation stage and used the STIRPAT model to analyze the factors that influence building carbon emissions. Xu Jinjun et al. [14] used the life cycle assessment method to analyze the whole life cycle carbon emissions of building concrete and identified the key parameters affecting carbon emissions. Zhao Yu et al. [15] identified the factors influencing carbon emissions of prefabricated buildings and analyzed their importance and relationships using the DEMATEL-ISM method. In other regions, B. Steen, N. Itsubo, and A. Inaba et al. assessed the carbon emissions of specific buildings [16]. Sharrard et al. [17] developed a layered LCA model to calculate the energy consumption and environmental impact of the U.S. construction industry. In Li’s research [18], the environmental impact of materials is calculated by IO-LCA, and the environmental impact of the construction phase is calculated by the process analysis method. Some of the research mentioned, such as the use of life cycle assessment and environmental assessment models, can provide valuable insights into the carbon emissions of buildings and the factors that influence them. However, as pointed out, more detailed research is needed to better understand the specific factors contributing to carbon emissions in different aspects of the construction industry.

2.2. Photovoltaic Curtain Wall

It is encouraging to see the development and application of photovoltaic curtain walls to reduce carbon emissions in the construction industry. The studies referenced by Chen et al. [19], Rustu Eke et al. [20], Niccolò Aste et al. [21], and N. Aste et al. [22] provide important insights into the performance and effectiveness of these systems. Chen’s [19] findings suggest that photovoltaic curtain walls not only provide electricity but can also improve the indoor thermal environment through natural ventilation. Rustu Eke’s [20] study evaluated the performance of a 40 KWP photovoltaic curtain wall system in Turkey, examining seasonal changes in performance parameters related to solar data and meteorological parameters. Niccolò Aste’s [21] research compared the performance of different photovoltaic technologies under temperate climate conditions and studied the influence of climatic parameters on energy production. Finally, N. Aste et al. [22] evaluated the long-term performance of a pilot photovoltaic power plant at Politecnico di Milano and carried out an economic analysis based on real historical data. These studies contribute to the growing body of knowledge concerning photovoltaic curtain walls and their potential to reduce carbon emissions in the construction industry. However, it is clear from the literature that the research on photovoltaic curtain walls is still in its early stages and focuses mainly on evaluating their performance and economics. There is a need for further research to understand the factors related to carbon emission reduction throughout the life cycle of these systems. Xie Y.X. et al. [23] quantified the carbon reduction effect of a photovoltaic building renovation project in northern China, while Wei Jiang et al. [24] transformed existing rural houses in a severely cold area of China into zero-energy buildings through overall passive energy-saving transformation and technical innovation. Pengfei Si et al. [25] explored the suitability of solar photovoltaic buildings for typical office buildings in cold plateau areas, while Mattia Manni et al. [26] investigated the minimum building form to maximize solar energy utilization and minimize carbon emissions. Despite these efforts, there is still a lack of research on the influencing factors of the carbon reduction effect of photovoltaic buildings, indicating the need for further investigation in this area.

2.3. Influencing Factors of Carbon Emission of Photovoltaic Curtain Wall

In this paper, the influence factors of carbon emission of photovoltaic curtain walls with high frequency are found in

Table 1. Selection of factors influencing carbon emissions of BIPV.

Literature	Lighting Inclination	Installation Orientation	Height	Installation Area	Shadow Occlusion	Light Intensity	Geographical Location	Season Changes
Xie Y.X. [23]	$\surd$	$\surd$		$\surd$		$\surd$		$\surd$
Manni [26]						$\surd$	$\surd$
Lobaccaro [27]				$\surd$	$\surd$	$\surd$	$\surd$	$\surd$
Lam [28]			$\surd$	$\surd$		$\surd$	$\surd$
Pulido-arcas [29]	$\surd$	$\surd$		$\surd$				$\surd$
Si P.F. [25]						$\surd$	$\surd$	$\surd$
Jiang W. [24]	$\surd$	$\surd$	$\surd$	$\surd$	$\surd$		$\surd$
Shirazi [30]	$\surd$	$\surd$	$\surd$	$\surd$	$\surd$	$\surd$	$\surd$	$\surd$

through literature analysis.

According to the literature review, among the relevant studies on carbon emission factors of photovoltaic curtain walls, installation orientation, lighting inclination, shadow occlusion, geographical location, season, and other factors appear more frequently. So it is determined that the factors that affect the carbon reduction effect of photovoltaic curtain walls in this study are lighting inclination, installation orientation, season changes, geographical location, and shadow occlusion.

2.4. Research on Building Carbon Emission Based on Life Cycle Assessment (LCA)

Indeed, research on carbon emissions based on life cycle assessment (LCA) has made significant progress [31,32]. The LCA method is commonly employed to calculate the carbon emissions associated with a product throughout its whole life cycle, including raw materials, production, use, and disposal. In recent studies, Grossi et al. [33] used One Click LCA and Athena Impact Estimator for Buildings to evaluate the life cycle of an all-electric laboratory at Concordia University in Montreal. This indicates the application of LCA in assessing the environmental impact of specific buildings. Additionally, Xue et al. [34] proposed an improved life cycle calculation framework to evaluate the carbon emissions of prefabricated concrete columns during the construction phase. The framework utilized a new calculation method based on the common calculation method of carbon emission coefficient to estimate the carbon emissions of construction workers. This highlights efforts to refine and expand the scope of LCA methodologies. While the research on carbon emissions based on LCA has formed a relatively complete system [35], there is still a lack of studies specifically focused on the carbon emissions of photovoltaic buildings using LCA. Consequently, introducing the LCA method into the assessment of carbon emissions from photovoltaic buildings and even enhancing the traditional LCA method itself represents a promising research direction for the future. By considering the entire life cycle of photovoltaic buildings, from manufacturing the photovoltaic components to their installation, operation, and eventual decommissioning, researchers can gain a comprehensive understanding of the environmental impact and carbon emissions associated with these structures. This information is crucial for optimizing the design, construction, and operation of photovoltaic buildings to minimize their carbon footprint and promote sustainable energy solutions.

3. Materials

3.1. Scope Definition

Combined with ISO 14044:2006 (Environmental management—Life cycle assessment—Requirements and guidelines) [36] and the whole life cycle carbon emission characteristics of photovoltaic curtain walls and the existing LCA methods, the research scope of this paper is as follows [37,38,39]:

• The factory building, production equipment, construction workers, and transport vehicles involved in the photovoltaic curtain wall production are not within the scope of this study;
• As China’s standards on the value of carbon emission factors are not perfect, only the common energy sources in the construction process with specified carbon emission factors are considered;
• The data list of devices such as inverters, DC confluence boxes, and others have little impact on the carbon reduction effect; therefore, this part is not within the scope of this study;
• Due to the small area of 1 kWp photovoltaic curtain wall, only 6 m², the installation stage of the photovoltaic curtain wall was not considered when calculating the carbon emission of photovoltaic curtain wall during its whole life cycle;
• The paper assumes the functional unit is 1 kWp, representing the installed capacity of photovoltaic modules. With current technology, the lifespan of photovoltaic modules is about 25 years.

In this paper, life cycle assessment theory, Donghe carbon emission analysis, and calculation software are introduced to calculate the carbon emission of photovoltaic curtain walls. To attain the photovoltaic curtain wall carbon reduction effect, PVsyst software simulation is used to calculate the photovoltaic curtain wall under different influencing factors of generating capacity, followed by using the grey correlation method to analyze the life cycle of the photovoltaic curtain wall carbon emissions and influencing factors of evaluation and sorting. To sum up, this paper’s research scope is shown in Figure 1.

3.2. Data Source

Research on life cycle carbon emissions is based on a large amount of inventory data, so the reliability and accuracy of data sources and how to deal with them after they are obtained are particularly important. Based on a clear system boundary, real scene and background data can be obtained by the unit of process flow to avoid repetition or omission.

The data sources and statistical methods of this paper are primarily as follows:

• Investigate photovoltaic enterprises. Detailed real data can be collected by directly investigating photovoltaic enterprises. This involves examining relevant reports and consulting industry experts to obtain information on mainstream production technologies employed in the country under study. By accessing primary data from the industry, researchers can acquire accurate and specific information;
• Read relevant domestic and foreign literature and industry reports; after extraction, integration, and analysis, take the average data of domestic literature as the calculation basis to make the data applicable;
• When real data is difficult to obtain, representative background data consistent with the product supply chain can be obtained from LCI databases such as CLCD and GaBiEcoinvent [40];
• When both scene and background data are unavailable, inventory data of similar processes or products can be used instead.

There is no exact logical correlation between the input-output data of photovoltaic curtain wall production, recovery, and disposal stages [38]. For the convenience of calculation, the data have been sorted according to the component accounting functional unit of 1 kWp.

3.2.1. Production Stage of Raw Materials and Battery Components

According to the relevant literature [41,42,43,44], photovoltaic module production is mainly divided into the industrial silicon production stage, polysilicon production stage, silicon wafer production stage, cell production stage, and battery modules production stage. Other materials required for producing 1 kWp photovoltaic curtain walls, such as frame and joints, are less, and so have little impact on the results of this study. Therefore, the materials required for installing a photovoltaic curtain wall are not considered in this paper.

(1)

Industrial silicon production stage.

Industrial silicon refers to the purity of 99% silicon material. Silicon ore is refined into industrial silicon using carbon in large electric arc furnaces.

(2)

Polysilicon production stage.

Polysilicon refers to silicon with a purity of about 6 N, which is used in solar cell production. This production is the most important intermediate product of polysilicon photovoltaic modules. In this process, industrial silicon is purified in a furnace by repeatedly pouring and blowing in chemicals until it finally solidifies. There are many ways to produce polysilicon, but the preferred method is the improved Siemens process. This process is often used to produce polysilicon in China. The process consumes a lot of electricity during production.

(3)

Silicon wafer production stage.

This is an important process in producing photovoltaic modules because it also requires a large amount of energy consumption. A mechanical wire cutter is used to cut polysilicon into a number of silicon wafers, each 180 μm thick.

(4)

Cell production stage.

The production of photovoltaic cells involves many steps, such as surface treatment, dopant diffusion, junction formation, coating, etc. Many types of chemicals and other raw materials are consumed in the production process, including pole printing strips, nitrogen, oxygen, and argon.

(5)

Battery modules production stage.

During the battery module production phase, individual cells are interconnected in series and parallel modes to produce the battery module. There are 20–40 cells in a battery module, which requires using other materials such as EVA, aluminum frames, etc.

To sum up, the silicon elements in each stage of the process flow as shown in Figure 2.

According to the flow direction of silicon element in the above photovoltaic curtain wall battery module, the data of the photovoltaic curtain wall production stage and recycling stage are summarized in

Table 2. Input and output data of the whole process of photovoltaic curtain wall production.

	Material Energy Input		Material Output
	Kind	Quantity	Kind	Quantity
Industrial silicon	Silica	27.85 kg	Industrial silicon	10.20 kg
	Charcoal	4.27 kg
	Petroleum coke	7.20 kg
	Scraps of wood	4.08 kg
	Bituminous coal	1.68 kg
	Recycled water	1836.2 kg
	Electricity	119.28 kWh
Poly silicon	Industrial silicon	10.20 kg	Poly silicon	7.61 kg
	Water coolant	1642.24 kg
	Electricity	875.15 kWh
Silicon wafer	Polysilicon	7.61 kg	Silicon wafer	240.72 tablets
	Steel wire	8.71 kg
	Electricity	175.73 kWh
Battery cell	Silicon chip	240.72 tablets	Battery cell	1.02 kWp
	Fresh water	1261.27 kg
	Electricity	126.48 kWh
Battery module	Cell	1.02 kWp	Battery module	1 kWp
	Copper	0.49 kg
	Fresh water	348,610 kg
	EVA film	7.06 kg
	TPT backplane	3.64 kg
	Tempered glass	62.22 kg
	Organic silica gel	40 kg
	Aluminum sash	13.12 kg
	Electricity	58.5 kWh

[41,42,43,44]:

3.2.2. Transportation Stage

Assuming that the supplier address and installation address of the photovoltaic curtain wall are both in the same city, and the distance between the two places is assumed to be 50 km by road truck diesel. According to the data [38], the weight of 1 kWp photovoltaic curtain wall is 85.06 kg.

3.2.3. Recovery and Disposal Stage

As the photovoltaic curtain wall has been newly advocated in recent years, almost all photovoltaic curtain walls have not reached the final stage of their life cycle. So it is difficult to collect the recycling methods and data of photovoltaic curtain walls in the later stage. After consulting the materials and referring to the papers of domestic and foreign research scholars [45,46], the specific data of main material consumption, energy consumption, and material recovery in this process are determined in

Table 3. Input and output data of photovoltaic curtain wall recycling and disposal stage.

Energy Consumption	Material Consumption	Output	Quantity	Processing Mode
Electricity consumption 128.24 kWh	Waste battery panel 1 kWp	Silicon chip	217 tablets	Recycle
		Steel	0.229 kg	Recycle
		Aluminum alloy	13.55 kg	Recycle
		Plastic	0.776 kg	Incineration
		EVA	4.26 kg	Fill in waste
		Glass	48.98 kg	Recycle
		TPT	3.26 kg	Incineration
		Sealed silica gel	0.265 kg	Incineration

.

4. Methods

4.1. The LCA Carbon Emissions Calculation Method

For the convenience of calculation, the carbon emissions in the life cycle of photovoltaic curtain walls can be divided into four branches [41]: the production stage of raw materials, the production stage of component systems, the transportation process, and the recycling and disposal stage. Therefore, the LCA carbon emission calculation model of the photovoltaic curtain wall is the summation function of these four carbon emission sub-models [41], as shown in Formula (1).

$$C=C_{CS}+C_{ZS}+C_{YS}+C_{HS}$$

(1)

In this formula, C represents the total carbon emission of the photovoltaic curtain wall in the whole life cycle, kg. If the value of C is negative, it indicates that the product can reduce carbon dioxide emissions in its whole life cycle, thus contributing to energy conservation and emission reduction in the construction industry; C_CS represents carbon emissions in the production stage of raw materials, kg; C_ZS represents carbon emissions in the production stage of component systems, kg; C_YS represents carbon emissions in the transportation process, kg; C_HS represents carbon emissions in the recycling and disposal stage, kg.

(1)

The production stage of raw materials.

$$C_{CS}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{M}_{\mathrm{i}}}\times F_{\mathrm{i}}$$

(2)

In this formula, M_i is the consumption of the i material, kg; F_i is the carbon emission factor of the i material, kgCO₂/kg. The carbon emission factor is derived from the IPCC emission factor database EFDB 2021 [47].

(2)

The production stage of component systems.

$$C_{ZS}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{M}_{\mathrm{i}}}\times F_{\mathrm{i}}$$

(3)

(3)

The transportation process.

$$C_{YS}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{P}_{\mathrm{i}}}\times D_{\mathrm{i}}\times T_{\mathrm{i}}$$

(4)

In this formula, P_i is the weight of the i material during transportation, t; D_i is the transportation distance of the i material in the transportation process, km; T_i is the carbon emission factor of the i material transportation mode in the transportation process, kgCO₂/tkm.

(4)

The recycling and disposal stage.

$$C_{ZS}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{W}_{\mathrm{i}}}\times F_{\mathrm{i}}-{\displaystyle \sum _{\mathrm{j}=1}^{\mathrm{n}}{E}_{\mathrm{j}}\times K_{\mathrm{j}}}$$

(5)

In this formula, W_i is the amount of i waste to be treated, kg; F_i is the carbon emission factor of the i waste, kgCO₂/kg; E_j is the amount of the j material recovered, kg; K_j is the carbon emission factor of the jth material, kgCO₂/kg.

(5)

The carbon reduction calculation method of photovoltaic curtain wall

Using PVsyst software, the power generation of a photovoltaic curtain wall under different influencing factors was simulated and calculated to obtain the carbon reduction effect of the photovoltaic curtain wall, as shown in Formula (6).

$$C_{ZS}={\displaystyle \sum _{\mathrm{i}=1}^{\mathrm{n}}{E}_{\mathrm{i}}}\times R_{\mathrm{e}}$$

(6)

In this formula, E_i is the power consumption in this stage, kwh; R_e is the carbon emission factor of electricity, CO₂/kwh.

According to the previously established carbon reduction effect model and the collected data, the corresponding calculation is done using Donghe carbon emission analysis and calculation software 2.0. The calculation standard is ISO 14044:2006 [36].

4.2. Grey Correlation Analysis Model of Carbon Emission of Photovoltaic Curtain Wall under Different Influencing Factors

Grey association analysis is mainly applied to the analysis process in scientific research, which is aimed at the imperfect system information and the difficulty in determining the primary and secondary relationship among various influencing factors. It is a method for inferring the correlation degree of curves between sequences according to the similarity of geometric shapes. The main idea is to measure the correlation degree between data by studying the correlation degree between data to assist decision-making.

4.2.1. PVsyst Software Parameter Settings

First, select the location of the project. As shown in Figure 3, entering the latitude and longitude of an area can automatically locate the region (for this paper, Xuzhou was chosen as the project reference site). Then import the Meteonorm8.0 weather database as shown in Figure 4. The final task is to configure the initial solution system. The photovoltaic module was selected, and the input planned power was 1 kWp. Poly silicon cell module produced by Jiangsu Linyang Company (Qidong, China) was selected for the photovoltaic module, and GW700-XS made by Goodwe Company was selected for the inverter. Specific parameters are shown in Figure 5.

4.2.2. Simulate the Power Generation of Photovoltaic Curtain Wall under the Influence of Different Factors

(1)

Influence of installation orientation.

By changing the orientation angles of −90, 0, 90, and 180, that is, the orientation of installation is due east, due south, due west, and due north, the annual power generation is obtained and multiplied by the carbon emission coefficient of 0.7035 kgCO₂/kWh of the East China Power grid, the annual carbon reduction is obtained. Finally, the carbon emission of a photovoltaic curtain wall over its full life cycle can be obtained from Formula (1).

(2)

Influence of lighting inclination angle.

Since the photovoltaic curtain wall is to be installed on the facade of the building, the variation range of its inclination angle is limited. Assume that the variation range of the inclination angle is 60 to 90 degrees [30]. So in the software—main parameters—orientation, the orientation angle is kept at 0, and the optimization is based on the annual total radiation by changing the lighting inclination angles of 60, 70, 80, and 90 to obtain the annual power generation.

(3)

Influence of season changes.

In the software main parameters orientation, keep the orientation angle at 0 and the lighting inclination angle at 90, and optimize based on the total annual radiation. Multiply the carbon emission factor of the East China power grid by 0.7035 kgCO₂/kWh to get the annual carbon reduction, and attain the life cycle carbon emission of photovoltaic curtain wall via Formula (1).

(4)

Influence of geographical location.

China is rich in solar energy resources, but the sunshine duration varies in different regions. According to the average annual sunshine duration, China can be roughly divided into five categories [48]. Therefore, Harbin, Beijing, Shanghai, and Hong Kong, the typical cities in the five regions, are selected as the installation locations. Based on annual total radiation, the annual power generation is obtained, which is multiplied by the carbon emission factor of the power grid to obtain the annual carbon reduction. Then, the life cycle carbon emission of the photovoltaic curtain wall is obtained by Formula (1).

(5)

Influence of shadow occlusion.

In the software-main parameters orientation, keep the azimuth angle at 0 and the lighting inclination angle at 90 and optimize it based on the annual total radiation. In the solar irradiation trajectory, change the shelter’s height to 0 m, 10 m, 20 m, and 30 m, respectively, to obtain the annual power generation.

4.2.3. Grey Association Analysis Model

The carbon reduction effect of photovoltaic curtain walls may be related to many factors. The correlation degree between different factors and the carbon reduction effect of a photovoltaic curtain wall can be obtained by grey correlation analysis. The specific process of grey relational analysis is as follows [41]:

(1)

Determine comparison sequences $X_{\mathrm{i}}=\left\{X_{\mathrm{i}\left(1\right)},X_{\mathrm{i}\left(2\right)},\dots ,X_{\mathrm{i}\left(n\right)}\right\}$ (i = 1, 2, …, m)

In this formula, m is the number of indicators; n is the dimension of the sequence;

(2)

Determine comparing sequences: X₀. Choose the maximum (minimum) value of each index or the appropriate reference value according to the purpose to be achieved to form the comparing sequence, $X_0=\left\{X_{0\left(1\right)},X_{0\left(2\right)},\dots ,X_{0\left(n\right)}\right\}$;

(3)

Normalizing (dimensionless) the parent sequence and the contrast sequence

$$X_{\mathrm{i}\left(k\right)}^{\prime }=\frac{X_{\mathrm{i}}\left(k\right)}{\frac{1}{m}{\displaystyle \sum _{k=1}^m{X}_{\mathrm{i}}\left(k\right)}}$$

(7)

In this formula, k represents the Kth dimension in the comparison sequence;

(4)

Calculate the grey association coefficient ζ_i (k)

$${\xi }_{\mathrm{i}}\left(k\right)=\frac{{\min }_{\mathrm{i}}{\min }_k\left|X_0\left(k\right)-X_{\mathrm{i}}\left(k\right)\right|+\rho {\max }_{\mathrm{i}}{\max }_k\left|X_0\left(k\right)-X_{\mathrm{i}}\left(k\right)\right|}{\left|X_0\left(k\right)-X_{\mathrm{i}}\left(k\right)\right|+\rho {\max }_{\mathrm{i}}{\max }_k\left|X_0\left(k\right)-X_{\mathrm{i}}\left(k\right)\right|}$$

(8)

In this formula, ρ represents the resolution factor and the value range is [0, 1], usually taking the value of 0.5;

(5)

Calculate the grey association degree between the parent sequence and the contrast sequence $ɤ$

$$ɤ\left(X_0,X_{\mathrm{i}}\right)=\frac{1}{n}{\displaystyle \sum _{k=1}^n{\xi }_{\mathrm{i}}}\left(k\right)$$

(9)

(6)

Analyzing and sorting the influence of comparison sequence according to the value of grey association degree.

5. Results

5.1. LCA Carbon Emission of Photovoltaic Curtain Wall

5.1.1. Industrial Silicon Production Stage

Based on previously collected data and carbon emission characterization factors specified in the IPCC emission factor database EFDB 2021, inputting the materials and energy required for industrial silicon production in photovoltaic curtain walls into the Donghe carbon emission calculation software, the carbon emissions of the industrial silicon production stage in photovoltaic curtain wall production are shown in

Table 4. Carbon emissions in industrial silicon stage.

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Electricity	kWh	119.28	0.7035	kgCO2/kWh	83.91
Petroleum coke	kg	7.2	3.1553	kgCO2/kg	22.72
Charcoal	kg	4.27	3.185	kgCO2/kg	13.6
Bituminous coal	kg	1.68	0.744	kgCO2/kg	1.25
Scraps of wood	kg	4.08	0.1609	kgCO2/kg	0.66
Recycled water	kg	1836.2	0.000194	kgCO2/kg	0.36
Silica	kg	27.85	0.76	kgCO2/kg	21.17
					143.67

.

According to the output result, the carbon emission in the production stage of industrial silicon is 143.67 kgCO₂/kWp.

5.1.2. Polysilicon Production Stage

Similarly, input the materials and energy consumed in the production of polycrystalline silicon in photovoltaic curtain walls into the Donghe carbon emission calculation software, and obtain the carbon emissions of the polycrystalline silicon production stage in photovoltaic curtain wall production as shown in

Table 5. Carbon emissions in the polysilicon stage.

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Electricity	kWh	875.15	0.7035	kgCO2/kWh	615.67
Industrial silicon	kg	1642.24	0.000168	kgCO2/kg	0.28
Water coolant	kg	10.2	0.76	kgCO2/kg	7.8
					623.75

.

It can be seen from Table 5 that the carbon dioxide emission in the polysilicon production stage is 623.75 kgCO₂/kWp, which is about 4.34 times the carbon emission in the industrial silicon production stage.

5.1.3. Silicon Wafer Production Stage

Input the materials and energy consumption during the silicon wafer production stage, and obtain the carbon emissions of this stage, as shown in

Table 6. Carbon emissions in silicon wafer production stage.

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Electricity	kWh	175.73	0.7035	kgCO2/kWh	123.63
Steel wire	kg	8.71	2.208	kgCO2/kg	19.23
Polysilicon	kg	7.61	0.76	kgCO2/kg	5.78
					148.64

.

It can be seen from Table 6 that the carbon dioxide emission in the silicon wafer production stage is 148.64 kgCO₂/kWp.

5.1.4. Cell Production Stage

During the cell production stage, the input data and carbon emissions output are shown in

Table 7. Carbon emissions in the production stage of battery cells.

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Electricity	kWh	126.48	0.7035	kgCO2/kWh	88.98
Fresh water	kg	1261.27	0.000168	kgCO2/kg	0.21
					89.19

.

It can be seen from Table 7 that the carbon dioxide emission in the silicon wafer production stage is 88.19 kgCO₂/kWp.

5.1.5. Battery Modules Production Stage

The input and output data for the cell module production stage are shown in

Table 8. Carbon emissions in the production stage of battery modules (including Al substructure).

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Electricity	kWh	58.5	0.7035	kgCO2/kWh	41.15
TPT backplane	kg	3.64	2.62	kgCO2/kg	9.54
Fresh water	kg	348,610	0.000168	kgCO2/kg	58.57
Aluminum sash	kg	13.12	2.6	kgCO2/kg	34.11
EVA film	kg	7.06	2.62	kgCO2/kg	18.5
Copper	kg	0.49	10.87	kgCO2/kg	5.33
Organic silica gel	kg	40	1.83	kgCO2/kg	73
Tempered glass	kg	62.22	0.08668	kgCO2/kg	5.39
					245.59

. The carbon emission data in the table includes the carbon emission of the aluminum frame in its frame structure.

It can be seen from Table 8 that the carbon dioxide emission in the production stage of the battery module is 245.59 kgCO₂/kWp. In summary, the carbon emissions of the entire production stage of a photovoltaic curtain wall are 1250.84 kgCO₂/kWp.

5.1.6. Carbon Emission of Photovoltaic Curtain Wall during Transportation Stage

According to the data, the carbon emission factor of light diesel truck transportation is 0.2461 kgCO₂/t × km, and the carbon emission of photovoltaic curtain wall transportation stage is 0.08506 × 50 × 0.2461 = 1.05 kgCO₂/kWp. At this stage, the photovoltaic curtain wall will be completed. The impact of carbon emissions is very small, and the uncertainty is high.

5.1.7. Carbon Emissions in Photovoltaic Curtain Wall Recycling and Disposal Stage

According to the data, the carbon emission factor of light diesel truck transportation is 0.2461 kgCO₂/t × km, and the carbon emission of photovoltaic curtain wall transportation stage is 0.08506 × 50 × 0.2461 = 1.05 kgCO₂/kWp. At this stage, the photovoltaic curtain wall will be completed. The impact of carbon emissions is very small, and the uncertainty is high.

(1)

Abandoned disposal.

In the disposal stage of photovoltaic curtain wall waste, burning plastics, EVA, TPT, sealing silicone, etc., will emit carbon dioxide. The carbon emissions during this stage are shown in

Table 9. Carbon emissions from waste disposal.

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Electricity	kWh	128.24	0.7035	kgCO2/kWh	90.22
TPT	kg	3.26	2.62	kgCO2/kg	8.54
EVA	kg	4.26	2.62	kgCO2/kg	11.16
Plastic	kg	0.776	6.79	kgCO2/kg	5.27
Sealed silica gel	kg	0.265	1.83	kgCO2/kg	0.49
					115.68

.

(2)

Recycling.

In the recycling stage of a photovoltaic curtain wall, the carbon emissions that can be reduced by recycling silicon wafers, steel, aluminum alloy, glass, etc., are shown in

Table 10. Total amount of carbon emissions recovered.

Name	Unit	Quantity	Factor Coefficients	Factor Units	Carbon Emissions (kgCO2)
Aluminum alloy	kg	13.55	20.92	kgCO2/kg	283.47
Glass	kg	48.98	0.176	kgCO2/kg	35.07
Steel	kg	0.229	1.06	kgCO2/kg	0.24
					318.78

.

To sum up, the carbon emission of incineration and landfill wastes is 115.68 kgCO₂/kWp, and the carbon emission of 318.78 kgCO₂/kWp can be reduced by recycling aluminum, glass, and other materials. Although some energy and materials are consumed, and some carbon dioxide is emitted, some carbon emissions are also saved in the recycling process. According to the formula, the carbon dioxide emission at this stage is reduced by 203.1 kg. Therefore, without considering the electricity generation of a photovoltaic curtain wall, the carbon emissions of the entire life cycle of a photovoltaic curtain wall are 1048.79 kgCO₂/kWp.

5.2. Grey Association Analysis of Carbon Emissions in the Life Cycle of a Photovoltaic Curtain Wall under Different Influencing Factors

5.2.1. Influence of Installation Orientation

Under the scenario that only installation orientation changes, the power generation and life-cycle carbon emissions of photovoltaic curtain walls are shown in

Table 11. Influence of installation orientation on power generation of a photovoltaic curtain wall.

Installation Orientation	Annual Power Generation/(kWh)	Carbon Reduction/(kgCO2)	LCA Carbon Emissions/(kgCO2)
South	726	510.74	538.29
West	599	421.40	627.63
North	365	256.78	792.25
East	597	419.99	629.04

.

5.2.2. Influence of Lighting Inclination Angle

Under the scenario that only installation lighting inclination angle changes, the power generation and life-cycle carbon emissions of photovoltaic curtain walls are shown in

Table 12. Influence of lighting inclination angle on power generation of a photovoltaic curtain wall.

Inclination/(°)	Annual Power Generation /(kWh)	Carbon Reduction/(kgCO2)	LCA Carbon Emissions/(kgCO2)
60	1052	740.08	308.95
70	962	676.77	372.26
80	853	600.09	448.94
90	726	510.74	538.29

.

5.2.3. Influence of Season Changes

Under the scenario that only season changes, the monthly energy generation of photovoltaic curtain walls is shown in Figure 6, and the power generation and life-cycle carbon emissions of photovoltaic curtain walls are shown in

Table 13. Influence of season on carbon reduction of a photovoltaic curtain wall.

Season	Quarterly Power Generation/(kWh)	Carbon Reduction/(kgCO2)	LCA Carbon Emissions/(kgCO2)
Spring	193.7	136.27	125.99
Summer	104.1	73.23	189.02
Autumn	211.9	149.07	113.19
Winter	216.3	152.17	110.09

.

5.2.4. Influence of Geographical Location

Under the scenario that only geographical location changes, the power generation and life-cycle carbon emissions of photovoltaic curtain walls are shown in

Table 14. Influence of geographical location on carbon reduction of a photovoltaic curtain wall.

Location	Annual Power Generation/(kWh)	Carbon Reduction/ (kgCO2)	Life Cycle Carbon Emissions/(kgCO2)
Harbin	738	519.18	529.85
Beijing	715	503.00	546.02
Shanghai	676	475.57	573.46
Hong Kong	647	455.16	593.87

.

5.2.5. Influence of Shadow Occlusion

Under the scenario that only shadow occlusion changes, the power generation and life-cycle carbon emissions of photovoltaic curtain walls are shown in

Table 15. Influence of shading on carbon reduction of a photovoltaic curtain wall.

Shadow Shelter	Annual Power Generation/(kWh)	Carbon Reduction/(kgCO2)	Life Cycle Carbon Emissions/(kgCO2)
0	1001	704.20	344.83
10	913	642.30	406.73
20	644	453.05	595.98
30	600	422.10	626.93

.

5.2.6. Grey Association Analysis

The specific steps of grey correlation analysis based on the above data are as follows:

(1)

Determine the comparison sequence

To unify the data units, they are all converted into carbon emissions in a quarter as follows:

• Orientation X₁ = (134.57, 156.91, 198.01, 157.26); )
• Inclination X₂ = (77.24, 93.07, 112.24, 134.57);
• Season X₃ = (125.99, 189.02, 113.19, 110.09);
• Shadow X₄ = (141.61, 148.64, 154.09, 160.25);
• Location X₅ = (86.21, 101.68, 148.99, 156.73);

Choose the maximum values of five influencing factors as the mother sequence: $X_0^{\prime }$ = (141.61, 189.02, 198.01, 160.25);

(2)

Standardize the indicators according to Formula (7);

(3)

The correlation coefficient calculated according to Formula (8) is shown in

Table 16. Grey association coefficient of influencing factors of carbon emission of a photovoltaic curtain wall.

	Orientation LCA Carbon Emissions/(kgCO2)	Inclination LCA Carbon Emissions/(kgCO2)	Season LCA Carbon Emissions/(kgCO2)	Shadow LCA Carbon Emissions/(kgCO2)	Location LCA Carbon Emissions/(kgCO2)
Factor change 1	1.00	0.95	0.68	0.66	0.80
Factor change 2	0.68	0.55	0.47	0.69	0.47
Factor change 3	0.92	0.81	0.44	0.68	1.00
Factor change 4	0.97	0.41	0.76	0.70	0.42

. To show the correlation coefficient between each influencing factor and the carbon emission of a photovoltaic curtain wall more intuitively, the data results are converted into a correlation coefficient heat map shown in Figure 7.

(4)

According to Formula (9), the correlation degree of each influencing factor is calculated and sorted, as shown in

Table 17. Grey association degree and correlation sequence of influencing factors of carbon emission of a photovoltaic curtain wall.

Evaluation Items	Degree of Relevance	Rank
Orientation LCA Carbon Emissions/(kgCO2)	0.89	1
Location LCA Carbon Emissions/(kgCO2)	0.68	2
Inclination LCA Carbon Emissions/(kgCO2)	0.68	3
Shadow LCA Carbon Emissions/(kgCO2)	0.67	4
Season LCA Carbon Emissions/(kgCO2)	0.59	5

.

6. Discussion

Compared to other studies that solely focus on the life cycle assessment of photovoltaic buildings within a single scenario, this study delves deeper into the realm of carbon reduction within the construction industry by exploring the immense potential of photovoltaic curtain walls. By considering different scenarios, the study aims to unveil the diverse range of opportunities for reducing carbon emissions. Employing grey relational analysis, the research investigates the varying impacts of installation orientation, lighting inclination angle, seasonal changes, geographical location, and shadow occlusion on the carbon reduction capabilities of photovoltaic curtain walls. Through this comprehensive examination, the study seeks to provide valuable insights for achieving sustainable and environmentally friendly construction practices. The following findings can be extracted from the research results:

In the production stage of photovoltaic curtain wall, the polysilicon production stage and battery module production stage have more carbon emissions, which are 623.75 kgCO₂/kWp and 245.59 kgCO₂/kWp, accounting for 59.52% and 23.43% of the total carbon emissions in the production stage, respectively. Therefore, reasonably improving the production process of these two stages can effectively reduce the CO₂ emissions in the production process of photovoltaic curtain walls. For example, in the production stage of polysilicon, the carbon emission of photovoltaic curtain walls can be reduced to a great extent by continuously improving the production process of photovoltaic modules and reducing power consumption at this stage. In the production stage of battery modules, carbon emissions mainly come from the consumption of water resources, so reducing water consumption as much as possible at this stage can effectively reduce the carbon emissions of photovoltaic curtain walls. In addition, the carbon emission of photovoltaic curtain walls in the recovery and disposal stage is negative, indicating that the recovery of photovoltaic curtain walls can bring certain environmental benefits and reduce the environmental damage caused by its production stage. While photovoltaic curtain walls can greatly reduce greenhouse gas emissions by converting solar energy into electric energy during operation, thus reducing the impact of the construction industry on the environment. Under each scenario of this study, the maximum reduction of carbon emissions through power generation can reach 70.61%.

Without considering the electricity generation of photovoltaic curtain walls, the carbon emissions of the entire life cycle of photovoltaic curtain walls are 1048.79 kgCO₂/kWp. And then, as per energy generation, carbon reduction per year is about 500 to 700 kgCO₂/kWp. So from the whole life cycle perspective, the amount of carbon emissions that photovoltaic curtain walls can reduce by generating electricity can offset the carbon emissions from production to recycling in about two years. With current technology, the lifespan of photovoltaic modules is about 25 years. So, the LCA carbon emission of a photovoltaic curtain wall is about −16,450 kgCO₂/kWp to −11,450 kgCO₂/kWp. In other words, the photovoltaic curtain wall can save about 11,450 to 16,450 kg of carbon emissions from production to recycling by generating electricity.

In addition, the CO₂ emission of photovoltaic curtain walls can be reduced to different degrees by changing the installation orientation, lighting inclination, seasonal factors, geographical location, and shadow occlusion. Among them, the change in photovoltaic curtain wall installation orientation caused the most significant change in carbon emissions, followed by geographical location, lighting inclination, and shadow occlusion. Seasonal changes caused by carbon emissions are very small. Under the scenario of changing the orientation of photovoltaic curtain walls, the installation of photovoltaic curtain walls on the south facade can reduce the maximum carbon emission. This indicates that photovoltaic curtain walls installed on the south facade of the building can fully absorb solar energy and convert it into electric energy to effectively mitigate the current trend of global warming. Under the scenario of changing the lighting inclination of a photovoltaic curtain wall, the life-cycle carbon emission decreases as the lighting inclination decreases. According to the results of the grey correlation analysis, the correlation degree between installation orientation, geographical location, lighting inclination, shadow occlusion, and seasonal change and the carbon emission of a photovoltaic curtain wall is more than 0.5. Therefore, they are all important factors affecting the carbon emission of photovoltaic curtain walls. The correlation degree of installation orientation is the highest at 0.89, the most important factor. The second is the geographical location, lighting inclination, shadow occlusion, and seasonal change. So the reasonable design of photovoltaic curtain wall installation direction, geographical location, lighting inclination, and reduction of nearby shadow occlusion can bring huge environmental benefits.

7. Conclusions

This study evaluates the carbon emissions and influencing factors of the whole life cycle of photovoltaic curtain walls through gray correlation analysis and clarifies the following points:

(1)

The most carbon emission stage in the whole life cycle of 1 kWp photovoltaic curtain wall is the polysilicon production stage, reaching 623.75 kgCO₂, followed by the battery module production stage of 245.59 kgCO₂; Throughout the whole production process of photovoltaic curtain wall, carbon emissions mainly come from the consumption of energy such as electricity;

(2)

During the process of installation inclination from 90 to 60, the carbon emission of photovoltaic curtain wall in the whole life cycle gradually decreased from 538.29 kgCO₂ to 308.95 kgCO₂;

(3)

The LCA carbon emission of a photovoltaic curtain wall is about −16,450 kgCO₂/kWp to −11,450 kgCO₂/kWp. In other words, the photovoltaic curtain wall can save about 11,450 to 16,450 kg of carbon emissions from production to recycling by generating electricity;

(4)

Among the influencing factors of seasonal change, summer is the season with the highest total radiation, but its quarterly carbon emission is as high as 189.02 kgCO₂/kWp/ quarter, which is the highest among the four seasons. Indicating that there is a big difference between the effective radiation amount obtained by photovoltaic curtain wall plane in summer and the radiation amount in the environment;

(5)

The carbon reduction effect of photovoltaic curtain walls installed in buildings in northern cities is more obvious than that in southern cities;

(6)

According to the grey correlation analysis, it can be seen that the correlation degree between installation orientation, geographical location, installation inclination, shading and seasonal change, and carbon emission of photovoltaic curtain walls all exceed 0.5, so they are all important factors affecting carbon emission of photovoltaic curtain walls. The highest correlation degree of installation orientation is 0.89, the most important factor. Secondly are geographical location, installation inclination, shade, and seasonal change.