Sustainability and Material Flow Analysis of Wind Turbine Blade Recycling in China

Abstract

Many decommissioned wind turbines (WTs) present significant recycling management challenges. Improper disposal wastes resources and generates additional carbon emissions, which contradicts the Sustainable Development Goals (SDGs). This study constructs a sine cosine algorithm (SCA)–ITransformer–BiLSTM deep learning prediction model, integrated with dynamic material flow analysis (DMFA) and a multi-dimensional Energy–Economy–Environment–Society (3E1S) sustainability assessment framework. This hybrid approach systematically reveals the spatiotemporal evolution patterns and circular economy value of WTs in China by synthesizing multi-source heterogeneous data encompassing policy dynamics, technological advancements, and regional resource endowments. Results demonstrate that China will enter a sustained wave of WT retirements post-2030, with an annual decommissioned capacity exceeding 15 GW. By 2050, new installations and retirements will reach a dynamic equilibrium. North and Northwest China are emerging as core retirement zones, accounting for approximately 50% of the national total. Inner Mongolia and Xinjiang face maximum recycling pressures. The recycling of decommissioned WTs could yield approximately CNY 198.5 billion in direct economic benefits and reduce CO₂ equivalent emissions by 4.78 to 8.14 billion tons. The 3E1S framework fills critical gaps in quantifying the comprehensive benefits of equipment retirement, offering a theoretically grounded and practically actionable paradigm for the global wind industry’s circular transition.

1. Introduction

To achieve the goal of limiting global temperature rise to 2 degrees Celsius, as specified in the Paris Agreement [1,2], China declared its commitment to reaching a carbon peak [3] by 2030 and achieving carbon neutrality by 2060 [4]. As a significant contributor to carbon emissions, the power generation industry has swiftly implemented relevant policies and vigorously promoted clean energy generation, including solar and wind energy [5]. Since 2020, the installed capacity of wind power has grown at an average annual rate of 19.1% [6]. In 2023, China witnessed a remarkable increase of 79.37 GW in its newly added wind power installation capacity, which comprised 67.84% of the global new installation capacity [6]. However, the sharp increase in the quantity of installed wind turbines (WTs) also poses significant challenges to their management during the decommissioning phase. The average operational duration of WTs is around 20 years [7], and the first batch of turbines installed in China will gradually reach the decommissioning stage. In 2020, China’s decommissioned WTs generated approximately 900 tons of solid waste [8], and it is predicted that by 2050, the solid waste generated by decommissioned WTs in China will reach 17.2 million tons [9]. During the decommissioning phase of WTs, if not handled properly, it will not only cause additional carbon emissions, rendering wind power, a clean energy source, no longer “clean”, but also result in a significant waste of resources and irreversible ecological damage, which not only fails to align with the circular economy concept but also goes against the requirements of sustainable development [10]. Consequently, immediate measures for the recovery and exploitation of decommissioned WTs are a pressing need, and the spatiotemporal evolution patterns of decommissioned WTs must be fully grasped. This study constructs a sine cosine algorithm (SCA)–ITransformer–BiLSTM deep learning model and a multi-dimensional coupled sustainable comprehensive benefit evaluation framework encompassing Energy–Economy–Environment–Society (3E1S), clarifying the capacity of decommissioned WTs in China, including those before 2050, and conducting a comprehensive assessment of the sustainable and overall advantages of reclaiming and making use of decommissioned WTs across various provinces. This provides a complete analytical framework as well as a specific research model for predicting the quantity of decommissioned WTs, managing recycling, promoting circular utilization, and evaluating the sustainable and comprehensive benefits in China and even globally to promote the safe and orderly handling of decommissioned WTs and achieve sustainable development.

2. Literature Review

2.1. Prediction of the Quantity of Decommissioned WTs

To predict the quantity of decommissioned WTs, it is necessary to first forecast the future installed capacity of wind power [7]. Most existing studies utilize linear regression [11] or rely on forecasts from authoritative institutions as methods for predicting wind power installed capacity [12]. Based on this, scholars both domestically and internationally have carried out forecasts of the number of decommissioned WTs, whether on a regional or a national scale. Lefeuvre et al. [13] have forecasted the global quantity of decommissioned WTs by 2050, specifically highlighting Asia and Europe as key regions for turbine decommissioning. Ortegon et al. [14], assuming that all WTs have a lifespan of 20 years from installation to decommissioning, have predicted the quantity of decommissioned WTs in the United States by 2030. From the perspective of material flow, Liu and Barlow [9] have forecasted the global number of decommissioned WTs by 2050. By introducing material flow analysis (MFA), Tazi et al. [15] calculated the quantity of decommissioned WTs in France by 2020. Lichtenegger et al. [16], using stochastic modeling, have predicted the number of decommissioned onshore and offshore WTs in EU countries by 2050. By comprehensively applying the fixed-time method, regression models, and Weibull models, Johst et al. [7] have forecasted the quantity of decommissioned WTs in Germany before 2050. Combining linear equations, Weibull functions, and dynamic material flow analysis (DMFA), Chen et al. [12] have predicted the amount of decommissioned WTs in Guangdong Province by 2050. In summary, the existing research exhibits the following limitations. (1) There exists a shortage of models that can accurately predict the installed capacity of wind power. (2) The lifespan uncertainty of decommissioned WTs has not been adequately addressed in current studies. (3) The wind power installation data utilized in prior research are inconsistent with the latest datasets and development trends, leading to significant deviations between analytical results and real-world scenarios. (4) There remains a paucity of investigations on decommissioned WT capacities, whether on a regional or a national scale in China.

2.2. Material Recycling of Decommissioned WTs

Existing research on material recycling from decommissioned WTs primarily focuses on wind turbine blade recycling (Details of WTs recycling are provided in Appendix B), with a decommissioned 1 MW wind turbine generating approximately 8 to 13 tons of blade waste [17,18]. Wind turbine blades mainly consist of fiber-reinforced polymers, including glass fiber, carbon fiber, and hybrid fibers, as well as thermosetting resins like epoxy resin or polyester [7]. Generating fibers calls for a considerable quantity of mineral resources and energy. Therefore, recycling and extracting fibers from decommissioned WT blades will significantly reduce the consumption of mineral resources and energy [17] and can also reduce external dependence on energy resources, ensuring national energy resource security. Currently, the methods of blade recycling mainly include direct landfill, secondary use after light processing [19,20], mechanical crushing, pyrolysis, combustion, and chemical dissolution [21]. Direct landfills will cause serious resource waste and ecological environmental pollution. This is contrary to the concept of the circular economy and does not align with the demands of sustainable development. Secondary use after light processing fails to maximize the benefits of decommissioned blades. Mechanical crushing has problems such as the decreased performance of recycled materials and low recycling utilization rates [22]. Pyrolysis recycling produces high carbon emissions [18]. Chemical dissolution is considered to be the recycling method with the highest recovery rate and the best environmental and economic benefits [23,24]. Although existing studies have clarified the types and quantities of recyclable resources in decommissioned WTs and evaluated the advantages and disadvantages of various recycling methods, there remains a notable gap in the research that employs specific recycling methodologies to systematically quantify recyclable resources from decommissioned WTs at national or other administrative regional scales.

2.3. Benefits Assessment of Recycling Decommissioned WTs

Existing studies have evaluated the advantages of decommissioned WTs in multiple aspects such as energy, economy, environment [25], and technology. The impact of each disposal option on the environment was discussed from the perspective of energy consumption [18], concentrating on processing energy usage, recycling gains, and trends in blade technology. By constructing a financial performance model, the impact of different recycling technologies on the economics of decommissioned WTs was analyzed [24]. The findings of the research suggest that the utilization of chemical dissolution methods for material recycling can yield significant economic benefits. The circular economy of decommissioned WT recycling was analyzed by calculating the product circularity index [26], and the environmental benefits were evaluated by determining the carbon footprint using life-cycle assessment (LCA) [5]. The main recycling technologies for decommissioned WTs were systematically summarized [17], and their impact on the performance of core recycling materials such as fibers was described. At the same time, considering the characteristics and performance of recycled fibers, the feasibility of their recycling and reuse was discussed. Current research predominantly assesses the benefits of decommissioned WTs through singular energy or economic dimensions, while neglecting the holistic integration of economic, environmental, and social factors required for a complete evaluation of the multi-dimensional sustainability benefits derived from recycling decommissioned WTs.

2.4. Gaps in Existing Research

An analysis of the existing literature reveals the following gaps in current research:

(1)

There is a lack of accurate prediction models. Currently, research regarding the projection of wind power’s installed capacity is almost non-existent. Most existing studies rely on simple linear planning or prediction data from authoritative institutions for future wind turbine installed capacity data, which cannot fully represent the uncertainty of the spatial and temporal distribution of newly added WTs.

(2)

There is a lack of a comprehensive evaluation system for recycling benefits. Existing research only evaluates the benefits of decommissioned WTs from a single dimension, failing to highlight the sustainable and comprehensive benefits of recycling decommissioned WTs.

(3)

There is a lack of research on decommissioned WTs in China. As the world’s leader in wind power installed capacity, China urgently requires assessment research regarding the recycling and utilization of decommissioned WTs. Nevertheless, currently, there is no research clarifying the national- and provincial-level spatial and temporal distribution of decommissioned WTs in China, or research clarifying the number of recyclable resources that can be generated from the recycling and utilization of decommissioned WT blades in the country, and no research that clarifies the sustainable comprehensive benefits of the multi-dimensional coupling of decommissioned WT recycling and utilization in China.

2.5. Contributions

In response to the aforementioned gaps, the innovations and contributions of this paper include the following:

(1)

This study pioneers the application of an intelligent optimization algorithm-integrated deep learning framework to wind turbine installed capacity growth prediction, where the novel coupling of the SCA-ITransformer-BiLSTM architecture achieves dual enhancement in predictive performance. (1) It significantly improves accuracy through multi-scale feature extraction. (2) It demonstrates superior robustness against multi-source uncertainties, including policy fluctuations, market dynamics, and technological disruptions, as evidenced by 23.4% lower error variance in cross-validation scenarios compared to conventional approaches. The hybrid model’s bidirectional temporal dependency modeling and adaptive optimization mechanisms effectively address the nonlinear interdependencies between renewable energy deployment and socio-economic drivers.

(2)

This study introduces multiple factors to accurately measure the number of decommissioned WTs and recyclable resources and materials. (1) By combining DMFA with the Weibull distribution function, a refined prediction of the decommissioned WT capacity at the national and provincial level in China was conducted, depicting the spatiotemporal dynamic evolution of decommissioned WTs. Meanwhile, variables and factors such as technological progress, policy changes, and development trends were fully considered to adjust the prediction model, ensuring that the forecast outcomes were closer to the actual situation. (2) Based on many references in the literature, this study conducted a quantitative analysis of the quantity and worth of recoverable resources derived from decommissioned WTs, filling the relevant research gap. This will provide a more intuitive and comprehensive perspective for stakeholders such as the government and enterprises in assessing the comprehensive and sustainable advantages of recycling decommissioned WTs.

(3)

A sustainable comprehensive benefit evaluation framework for decommissioned WTs, integrating 3E1S, has been constructed. The multi-dimensional assessment avoids the limitations of a single-dimensional assessment of the benefits of recovering decommissioned WTs. This will provide a more comprehensive and objective perspective and judgment criteria for different interest groups, such as governments, enterprises, and the public, to view and assess the recycling of decommissioned WTs.

(4)

In light of the current situation of decommissioned WTs, from diverse directions, like time scale, the legal system, and stakeholders, and supported by a wealth of data and literature, multiple targeted and feasible suggestions have been put forward. The aim is to promote the safe and orderly recycling and utilization of decommissioned WTs, thereby realizing the development concept of the circular economy and achieving the SDGs.

The following parts of this paper are structured as follows. In Section 3, the specific methods and analytical frameworks employed in this research are elaborated upon in detail. Section 4 showcases the empirical investigation conducted. Section 5 sums up the key findings and offers relevant policy recommendations.

3. Theoretical Model and Analytical Framework for Predicting the Capacity of Decommissioned WTs in China and Evaluating Sustainable Comprehensive Benefits

As shown in Figure 1, the decommissioned WT capacity prediction model constructed in this study comprises two parts: wind power installed capacity prediction and decommissioned WT capacity calculation. The sustainable comprehensive benefit evaluation model for decommissioned WT recycling includes the calculation of the number and market value of recyclable substances, the construction and calculation of multi-dimensional evaluation indicators, and further analysis.

3.1. Installed Capacity Prediction of Wind Power Based on the SCA-ITransformer-BiLSTM Model

SCA, a meta-heuristic optimization algorithm, is founded on trigonometric functions [27]. It enables a broader search within the parameter space of the model, aiming to find a more optimal combination of parameters [28]. When adjusting parameters, a single ITransformer [29,30] or BiLSTM [31] might rely on traditional random search or gradient-based optimization methods, which are prone to becoming stuck in local optima. However, SCA, through its unique periodic variations in sine and cosine functions, can achieve a better balance between exploration and exploitation capabilities during the search procedure, enabling it to escape local optima and find more suitable parameters for ITransformer and BiLSTM, thereby enhancing the model’s performance.

When predicting the installed capacity of wind power, an integrated deep-learning prediction model considering multiple factors was built. Firstly, historical data on wind power installed capacity were used as the dependent variable, while GDP, policy support, and other factors were used as independent variables. Expected data, such as GDP and policy support, were obtained through a time series analysis model (ARIMA) [32]. Secondly, historical data were inputted into the SCA-ITransformer-BiLSTM model for learning and training (the model’s performance is shown in Appendix A, Figure A1). After iterative training, an ideal prediction model was derived, which predicted wind power’s installed capacity in China before 2050.

3.2. Calculating the Capacity of Decommissioned WTs, as Well as the Quantity and Value of Recycled Resources in China

This study assumes that the capacity of decommissioned WTs follows a Weibull distribution [11,32]. Based on this assumption, the DMFA [33] method is employed to calculate the capacity of decommissioned WTs in China [10].

$$Outflow{(t)}_s={\displaystyle \sum _{t^{\prime }}inflow(t-t^{\prime })}\overline{\omega }(t^{\prime })$$

(1)

$$\overline{\omega }(t^{\prime })=\left({\displaystyle \frac{\beta }{T}}\right)\left({\displaystyle \frac{t^{\prime }}{T}}\right){\beta }^{-1}exp{\left[{\displaystyle \frac{t^{\prime }}{T}}\right]}^{\beta }$$

(2)

where $\overline{\omega }(t^{\prime })$ represents the probability density of the specific Weibull distribution function, $t^{\prime }$ denotes the lifespan of the wind turbine unit in years $t$, $T$ signifies the average lifespan of the wind turbine unit, and $\beta$ indicates the shape factor of the Weibull curve. $Outflow{(t)}_s$ represents the decommissioned WT capacity in province $S$ in year $t$, and $inflow$ represents the newly installed capacity of WTs. The units of both $Outflow{(t)}_s$ and $inflow$ are GW, assuming $\beta$ = 2.2 [11].

Most existing studies suggest that the operational lifespan of WTs is around 20 years [33]. Compared with the US and European countries, the development of China’s wind power industry commenced relatively late. CWEA statistics show that by 2008, China’s cumulative wind power installed capacity was around 12 GW, while that of the United States during the same period was 25 GW. In 2009, China’s newly installed wind power capacity exceeded 10 GW for the first time. During the year 2012, the National Development and Reform Commission (NDRC) and the Ministry of Finance issued a notice promoting the implementation of opinions regarding the development of the wind turbine industry to guide its orderly and healthy progress. Since that time, the wind power industry in China has embarked on a high-speed path of high-quality development.

Owing to the absence of comprehensive regulations and industry norms, fans deployed before 2012 inevitably suffered from poor performance and quality issues. Therefore, this study assumes that fans deployed before 2012 have a serviceable life of 15 years, while those deployed after 2012 have a serviceable life of 20 years. Based on these differences, the first scenario assumes that the average serviceable life of fan units is 15 years. In the second scenario, the average serviceable life of fan units is set at 20 years. Considering possible technological advancements in the future, a third scenario is added, where the service life of fans is assumed to be 25 years. These hypothetical scenarios are consistent with relevant foreign research [11,19].

Based on the computed quantity of decommissioned WTs on a national and provincial scale, and by integrating the composition proportion and cost of recyclable materials obtained from decommissioned WTs mentioned in the relevant studies [18,24], we can calculate the aggregate quantity and worth of recyclable resources at both the national and provincial tiers.

$$W_{(\mathrm{WT})}=R_w\times C_{m,g}$$

(3)

$$A_{(\mathrm{WT})}=R_w\times C_{a,g}$$

(4)

where $W_{(\mathrm{WT})}$ represents the weight of the blades, and the unit of $W_{(\mathrm{WT})}$ is tons. $R_w$ indicates the capacity of decommissioned WTs, and the unit of $R_w$ is GW. $C_{m,g}$ represents the weight of the blades in each GW of decommissioned WTs, and the unit of $C_{m,g}$ is ton/GW. $A_{(\mathrm{WT})}$ represents the area of the blades of decommissioned WTs, and the unit of $A_{(\mathrm{WT})}$ is m². $C_{a,g}$ represents the floor area per ton of blades, and the unit of $C_{a,g}$ is m²/GW.

$$W_z=W_{(\mathrm{WT})}\times M_z\times R_i$$

(5)

$$V_z=W_z\times P_z$$

(6)

where $W_z$ represents the weight of a certain resource, and the unit of $W_z$ is tons. $M_z$ represents the proportion of a certain resource in wind turbine blades, $V_z$ represents the value of a certain resource, and the $P_z$ represents the unit price of a certain resource. The unit of $V_z$ is USD, and $P_z$ is USD/ton. When calculating the weight and area of decommissioned WT blades, reference was made to data on area, power, weight, and efficiency from the relevant literature [18,24]. It is assumed that the recyclable blade weight of a 1 MW decommissioned WT ranges from 8 to 13 tons, with a proportion of 70% for fibers and 30% for resins [17]. In terms of recycling methods, this study selected chemical dissolution recycling, which is the most environmentally friendly and economically beneficial method, as the hypothetical recycling method. $R_i$ represents the recyclability rates of fibers or resins. The recyclability rates of fibers and resins are 95% and 50%, respectively [24]. The proportion of glass fiber in the recycled fiber is 43.7% and that of carbon fiber is 56.3%. The recycling proportions are referenced from the wind power installed capacity data of the National Energy Administration of China and the China Electricity Council, as well as the literature [24].

3.3. Constructing an Evaluation Framework for the 3E1S Benefits of Decommissioned WT Recycling

To conduct a thorough assessment of the comprehensive benefits associated with the recycling of decommissioned WTs, this research formulates 12 evaluation indicators. These indicators are developed based on an extensive review of the literature and derived from the 3E1S framework, as presented in

Table 1. Elements related to the recycling and utilization of decommissioned WTs in the existing literature.

Essential Factor	Research Elements Mentioned in the Existing Literature
	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Energy conservation			√		√				√	√
Recovering energy consumption	√	√		√	√				√	√		√	√	√
Energy recovery intensity					√		√
Energy payback period					√		√
Direct economic benefits	√		√			√	√	√	√				√	√
Recovery cost	√		√			√	√	√	√		√		√	√
Greenhouse gas emissions	√	√		√	√	√			√	√		√
Land area conservation					√				√		√	√			√
Resource conservation	√			√	√						√		√	√	√
Land acidification					√							√
Industrial water conservation					√							√
Carcinogenic toxicity to humans					√							√
Stratospheric ozone depletion												√
Marine eutrophication												√
Laws and regulations	√	√	√	√	√

Note: 1 = [34], 2 = [7], 3 = [16], 4 = [17], 5 = [18], 6 = [8], 7 = [19], 8 = [24], 9 = [35], 10 = [26], 11 = [11], 12 = [36], 13 = [21], 14 = [22], and 15 = [12]. The “√” in the table indicates that the corresponding “Research Elements Mentioned in the Existing Literature” has been studied or considered in relation to the “Essential Factor” in the existing literature.

. The set of 12 indicators is designed to establish a comprehensive evaluation system, enabling a holistic appraisal of the sustainable and overall benefits of recycling decommissioned WTs across different provinces.

3.3.1. Energy Dimension

• E11: Primary energy saving

This indicator represents the primary energy saved during the material production process by using recycled blade materials from decommissioned WTs instead of new raw materials.

$$E_z=(E_i-E_j)\times Q$$

(7)

where $E_z$ represents the total amount of primary energy saved by using recycled materials instead of virgin materials, $E_i$ represents the amount of energy required to produce an equivalent amount of virgin materials, $E_j$ represents the amount of energy required to recycle each unit of material, and $Q$ represents the quantity of recycled materials (measured in weight or other units). The units of $E_z$, $E_i$, and $E_j$ are GJ. The data are referenced from the relevant literature [18].

• E12: recovery energy consumption

This indicator represents the total amount of energy directly consumed in each step of the entire recycling process of wind turbine blades, including collection, transportation, dismantling, sorting, processing, and remanufacturing.

$$E_q={\displaystyle \sum W_z\times S_z}$$

(8)

where $E_q$ represents the total energy consumption, $W_z$ represents the weight of a certain resource, and $S_z$ represents the total life-cycle production energy consumption of a certain resource. The unit of $E_q$ is GJ, $S_z$ is GJ/ton, and $W_z$ is tons. The data are referenced from the relevant literature [18].

• E13: recycling energy intensity

This indicator reflects the energy consumption per unit of the recycling benefit within the wind turbine blade recycling operation. A lower value indicates the higher energy effectiveness of the recycling procedure.

$$\mathrm{REI}={\displaystyle \frac{E_v}{C_v}}$$

(9)

where $\mathrm{REI}$ represents the intensity of energy recovery, $E_v$ represents unit energy consumption, and $C_v$ represents unit revenue. The data are referenced from the relevant literature [18]. The unit of $\mathrm{REI}$ is GJ/USD, $E_v$ is GJ, and $C_v$ is USD.

• E14: energy recovery period

This metric stands for the time needed for the wind turbine blade to generate energy equal to the total energy consumed during its whole life cycle since it starts to produce energy.

$$\mathrm{EPBT}={\displaystyle \frac{E_{LCA,n}-E_{LCA,m}}{E_{out,s}}}$$

(10)

$\mathrm{EPBT}$ indicates the energy recovery period, $E_{LCA,n}$ represents the complete life-cycle energy consumption of wind turbine blades without recovery, and $E_{LCA,m}$ represents the complete life-cycle energy consumption under the condition of wind turbine blade recovery, with data referenced from the relevant literature [18]. $E_{out,s}$ indicates the annual power generation of WTs in a certain province, with data sourced from Xiaozhuanfeng (Xiaozhuanfeng. https://meteo.agrodigits.com/login?redirect=/app/energy/photovoltaic (accessed on 27 April 2025)). The unit of $\mathrm{EPBT}$ is years, the units of $E_{LCA,n}$ and $E_{LCA,m}$ are GJ, and the unit of $E_{out,s}$ is GJ/year.

3.3.2. Economic Dimension

• E21: direct economic benefits

This indicator represents the direct economic value of resources obtained from recycling decommissioned WT blades.

$$V_d={\displaystyle \sum W_z}\times P_z$$

(11)

$V_d$ represents the direct economic value of resources obtained from the recycling of decommissioned WT blades, $W_z$ indicates the weight of a certain resource and the $P_z$ unit price of a certain resource. The unit of $V_d$ is USD, $P_z$ is USD/ton, and $W_z$ is tons. The data are sourced from the relevant literature [24].

• E22: recovery cost

This indicator refers to the total cost incurred during the recycling process of decommissioned WT blades, encompassing various stages such as collection, transportation, dismantling, and disposal. Here, transportation distances are assumed for three different scenarios based on different provinces: 250 km, 500 km, and 750 km. Labor costs are referenced from the per capita average annual income data of each province, which are officially released by the National Bureau of Statistics (NBS), while other expenses are referenced from the literature [24].

$${\mathrm{Cost}}_z=C_g+C_t+C_h$$

(12)

$$C_g=C_o+C_m+C_e+C_q$$

(13)

${\mathrm{Cost}}_z$ represents recycling costs, $C_g$ represents fixed costs (assuming they are the same for each province), $C_t$ represents transportation costs in a certain province, and $C_h$ represents labor costs in a certain province. $C_o$ represents collection fees, $C_m$ represents management fees, $C_e$ represents fixed asset depreciation costs, and $C_q$ represents equipment maintenance costs. The units of the above variables are all USD.

3.3.3. Environmental Dimension

• E31: carbon dioxide emission reduction

This indicator represents the decrease in carbon dioxide emissions. The decrease is due to the recycling and reuse of materials from decommissioned WTs blades. This situation contrasts with the direct use of virgin materials during production and application.

$$C_r=C_i\times W_{z,s}$$

(14)

$C_r$ represents the amount of carbon dioxide emission reduction achieved by recycling and reusing materials, $C_i$ represents a reduction in carbon dioxide emissions achieved by recycling unit materials, and the $W_{z,s}$ represents the weight of decommissioned WT blades in a certain province. The data are referenced from the relevant literature [26]. The unit of $C_r$ is t CO₂ eq, $C_i$ is t CO₂ eq/ton, and $W_{\mathrm{z}}$ is tons.

• E32: land area savings

This indicator refers to the land area saved by recycling decommissioned WT blades compared to direct landfills.

$$S_a=W_{z,s}\times C_{a,g}$$

(15)

$S_a$ represents the amount of land area saved, $W_{z,s}$ represents the weight of decommissioned WT blades in a certain province, and $C_{a,g}$ represents the area per ton of decommissioned WT blades. The unit of $S_a$ is m², $C_{a,g}$ is m²/ton, and $W_{z,s}$ is tons. The data are referenced from the relevant literature [36].

• E33: resource savings

This indicator represents the total amount of resources saved by recycling materials (including glass fiber, resin, and carbon fiber) from decommissioned WT blades rather than producing them from scratch.

$$S_j=W_{z,s}\times C_{s,g}$$

(16)

$S_j$ indicates the amount of resource conservation, $W_{z,s}$ represents the weight of decommissioned WT blades in a certain province, and $C_{s,g}$ indicates the quantity of recyclable materials per ton of decommissioned WT blades. The units of $S_j$ and $W_{z,s}$ are tons, and the unit of $C_{s,g}$ is ton/ton.

• E34: terrestrial acidification

This indicator represents the degree of land acidification caused by directly landfilling or discarding decommissioned WT blades without proper recycling.

$$T_a=W_{z,s}\times T_{a,g}$$

(17)

$T_a$ indicates the reduction in terrestrial acidification values due to recycling activities, $W_{z,s}$ represents the weight of decommissioned WT blades in a specific province, and $T_{a,g}$ represents the reduction in terrestrial acidification values for each ton of decommissioned WT blades recycled. The values are referenced from the relevant literature [36]. The unit of $T_a$ is kg SO₂ eq and $T_{a,g}$ is kg SO₂ eq/ton.

3.3.4. Social Dimension

• E41: human carcinogenicity toxicity

This indicator shows that recycling and disposing of decommissioned WT blade components can reduce the emission of harmful substances to humans, thus decreasing the risk of cancer among people.

$$\mathrm{HT}=W_{z,s}\times S_{h,g}$$

(18)

$\mathrm{HT}$ represents the mass of chemical substances with human carcinogenic toxicity reduced per ton of recycled blades, $W_{z,s}$ represents the weight of decommissioned WT blades in a certain province, $S_{h,g}$ indicates the reduction in HT achieved by recycling each ton of WT blades. The values are referenced from the relevant literature [36]. The unit of $\mathrm{HT}$ is kg 1,4-DB eq and $S_{h,g}$ is kg 1,4-DB eq/ton.

• E42: industrial water conservation amount

This indicator represents the industrial water consumption saved during the reproduction process due to the resources obtained from the recycling and reuse of decommissioned WTs.

$$S_w=W_{z,s}\times S_{w,g}$$

(19)

$S_w$ indicates the amount of industrial water saved, $W_{z,s}$ represents the weight of decommissioned WT blades in a certain province, and $S_{w,g}$ indicates the amount of industrial water saved by recycling each ton of decommissioned WT blades. The values are referenced from the relevant literature [36]. The unit of $S_w$ is m³, $S_{w,g}$ is m³/t.

4. Empirical Analysis Results

This study focuses on decommissioned WTs at the national level and across various provinces, municipalities directly under the Central Government, and autonomous regions in China. Taiwan, Hong Kong, and Macao are excluded due to the unavailability of the relevant data. Firstly, based on historical wind power data on installed capacity from 2012 to 2023 (sourced from annual reports of the Wind Energy Professional Committee of the China Renewable Energy Society), the SCA-ITransformer-BiLSTM deep learning model was integrated by incorporating several factors, inclusive of historical policy-support gauges (derived from the Chinese government website’s entry statistics on “wind turbines”) and GDP figures (provided by the NBS), to project wind power’s installed capacity figures from 2024 to 2050. Secondly, by integrating the Weibull distribution function with the dynamic material flow approach and using the predicted installed capacity data, it analyzes the spatiotemporal evolution of decommissioned WTs in China and its administrative divisions under three different scenarios. Moreover, based on these analysis results, it calculates the quantity and economic value of the recoverable resources. Finally, establishing a comprehensive set of indicators, it assesses the sustainable benefits across the dimensions of 3E1S on both national and provincial tiers in China.

4.1. Predicted Installed Capacity of Wind Power

As shown in Figure 2, under both prediction scenarios, the magnitude of wind power installation in China will exceed 1000 GW by 2030. In 2040, under the deep learning prediction scenario, the scale of wind power installation in China will exceed 2000 GW, while the “Wind Energy Declaration” (Beijing Declaration on Wind Energy. https://baijiahao.baidu.com/s?id=1680510870626804448&wfr=spider&for=pc (accessed on 27 April 2025)) prediction scenario predicts a slower growth rate, with the installed capacity exceeding 2000 GW around 2045. By 2050, under the deep learning prediction scenario, China’s wind power installed capacity will arrive at 3176 GW, while the “Wind Energy Declaration” prediction scenario predicts an installed capacity of approximately 2400 GW.

Looking at the provinces (Figure 3 and Figure 4), Inner Mongolia will always maintain the highest installed capacity. It is expected that by 2030, the key year for “carbon peaking,” Inner Mongolia, Xinjiang, Gansu, Hebei, Shandong, etc., will rank among the top five in terms of expected wind power installed capacity. Among them, Inner Mongolia will reach around 175 GW, Xinjiang, Gansu, and Hebei will each exceed 60 GW, and provinces such as Shandong will also approach 60 GW, having a cumulative installed capacity close to 57 GW. By 2050, the expected wind power installed capacity in Inner Mongolia will reach around 548 GW, which is close to China’s cumulative installed capacity as of 2023. Xinjiang and Gansu will still rank second and third, with expected cumulative installed capacities of about 252 GW and 189 GW, respectively, while Henan will jump to fourth place, having a cumulative installed capacity of approximately 188 GW, and Hebei, ranking fifth, will have a capacity of about 178 GW. Beijing and Xizang, among other places, will maintain very low installed capacity levels both in 2030 and 2050.

The reasons for this distribution can be roughly attributed to the following aspects: (1) The allocation of wind energy resources. In 2023, the average wind speed at a height of 70 m in Inner Mongolia, Xinjiang, and Gansu was 6 m/s, with some areas even reaching 7–8 m/s, which far exceeds the national average wind speed of 5.5 m/s [37]. China Energy News, 2023-04-10(013)). (2) The robust demand for electricity. Although the wind energy resources in Shandong and Hebei are not as abundant as those in Inner Mongolia and other regions, Shandong and Hebei ranked second and sixth in electricity consumption in 2023, respectively. The strong electricity demand has driven the development of more economically and environmentally friendly clean energy sources such as wind power. (3) It is due to environmental protection policies. Xizang is among the regions in China that possess the most abundant wind energy resources and has one of the lowest installed wind power capacities, which should be attributed to the consideration of protecting the Qinghai–Xizang Plateau as an ecological safety barrier.

4.2. Prediction of Decommissioned WT Capacity

This study considered three scenarios of wind turbine retirement: 15-year, 20-year, and 25-year lifespans, representing scenario 1, scenario 2, and scenario 3, respectively. From a national perspective, the capacity of decommissioned WTs will reach 86 to 212 GW by 2030, approximately 1.08 × 10⁶ to 2.64 × 10⁶ t in weight, and the direct land occupation will reach 5.20 × 10⁶ to 1.27 × 10⁷ m². By 2040, the capacity of decommissioned WTs will reach 389 to 797 GW, approximately 5.04 × 10⁶ to 1.03 × 10⁷ t in weight, and the direct land occupation will reach 2.34 × 10⁷ to 4.78 × 10⁷ m². By 2050, the capacity of decommissioned WTs will be approximately 1012 to 1745 GW, approximately 1.29 × 10⁷ to 2.22 × 10⁷ t in weight, and the direct land occupation will be approximately 6.08 × 10⁷ to 1.05 × 10⁷ m².

By the year 2030, Inner Mongolia, Xinjiang, Hebei, Qinghai, and Shandong will rank among the top five provinces for cumulative decommissioned WT capacity (Figure 5, Figure A2 and Figure A3). Both Inner Mongolia and Xinjiang are projected to have decommissioned WT capacities exceeding 20 GW, with Inner Mongolia at approximately 26 GW and Xinjiang at around 20 GW. Hebei, Qinghai, and Shandong are expected to have about 13 GW each. By 2040, Inner Mongolia, Xinjiang, Gansu, Hebei, and Henan will still rank in the top five for cumulative decommissioned WT capacity. Inner Mongolia is projected to reach approximately 120 GW, Xinjiang around 65 GW, while Gansu, Hebei, and Henan are expected to have about 45 GW each. By 2050, Inner Mongolia and Xinjiang are still expected to occupy the top two spots for cumulative decommissioned WT capacity, with Gansu, Henan, and Hebei taking the third, fourth, and fifth positions, respectively. During this period, Inner Mongolia’s cumulative decommissioned WT capacity will reach 280 GW, Xinjiang will have about 137 GW, and both Gansu and Henan will each have around 100 GW, while Hebei will have approximately 94 GW.

From a regional perspective (Figure 6, Figure A4 and Figure A5), North China, Northwest China, and East China rank in the top three in terms of decommissioned WT capacity. By 2030, the cumulative decommissioned WT capacity in North China will be approximately 106 GW, while that in Northwest China will be approximately 59 GW, accounting for about 27% of the total. The cumulative decommissioned WT capacity in East China will be approximately 37 GW, accounting for about 17% of the total. By 2050, the cumulative decommissioned WTs capacity in North China will remain at about 27%, with a cumulative decommissioned capacity of approximately 466 GW. The proportion in Northwest China will decrease to about 23%, with a cumulative decommissioned capacity of approximately 404 GW. East China will remain at about 17%, with a cumulative decommissioned capacity of approximately 288 GW.

4.3. Quantity and Economic Value of Recycled Resources from Decommissioned WTs

The existing research on the recycling of decommissioned WTs primarily focuses on the recycling process of wind turbine blades [17]. Therefore, this study also takes decommissioned WT blades as the research object for recycling and utilization. Wind turbine blades are predominantly made up of fibers and resins. Based on the calculation formula constructed in Section 3, specific recycling results for each part are obtained. The results show that recycling decommissioned WT blades alone can recover abundant resources and generate hundreds of millions of yuan in economic value. By 2050, a total of 1.23 × 10⁷ to 2.12 × 10⁷ tons of various resources can be recycled, including 3.30 × 10⁶ to 5.70 × 10⁶ tons of glass fibers, 3.77 × 10⁶ to 6.50 × 10⁶ tons of carbon fibers, and 5.22 × 10⁶ to 9.01 × 10⁶ tons of resins. The direct economic benefit is approximately CNY 198.5 billion.

4.4. Multi-Dimensional Comprehensive Benefit Evaluation of Retired Blade Recycling and Utilization

Starting from 3E1S, and based on a vast amount of the reference literature, this study constructs 12 evaluation indicators. Calculating the specific results of each indicator clarifies the specific multi-dimensional advantages of recycling decommissioned WT blades in various provinces by 2050 (as shown in Figure 7; detailed data for each province under the three scenarios can be found in Appendix A,

Table A1. Province-level specific returns up to 2050 under scenario 1 (WT lifespan: 15 years).

Province	Primary Energy Savings (GJ)	Recovery Energy (GJ)	Recovery Energy Intensity (GJ/USD)	Energy Payback Period (Years)	Direct Economic Gain (USD)	Terrestrial Acidification (kg SO2 eq)	Recovery Costs (USD)	Carbon Dioxide Emission Reduction (t CO2 eq)	Land Area Savings (m2)	Industrial Water Savings (m3)	Human Toxicity Potential (kg 1,4-DB eq)	Resource Savings (t)
Shandong	1.70 × 107	2.19 × 107	4.65 × 101	1.44 × 10−2	1.02 × 1010	1.09 × 106	9.78 × 108	4.17 × 108	5.37 × 106	4.63 × 108	2.34 × 109	1.09 × 106
Jiangsu	1.68 × 107	2.19 × 107	4.65 × 101	1.65 × 10−2	1.02 × 1010	1.09 × 106	9.81 × 108	4.18 × 108	5.38 × 106	4.64 × 108	2.35 × 109	1.09 × 106
Hebei	1.77 × 107	2.30 × 107	4.65 × 101	3.17 × 10−2	1.07 × 1010	1.14 × 106	1.03 × 109	4.39 × 108	5.65 × 106	4.87 × 108	2.47 × 109	1.14 × 106
Zhejiang	4.83 × 106	6.30 × 106	4.65 × 101	2.62 × 10−2	2.93 × 109	3.12 × 105	2.82 × 108	1.20 × 108	1.54 × 106	1.33 × 108	6.74 × 108	3.12 × 105
Anhui	5.98 × 106	7.79 × 106	4.65 × 101	1.65 × 10−2	3.62 × 109	3.87 × 105	3.48 × 108	1.49 × 108	1.91 × 106	1.65 × 108	8.35 × 108	3.87 × 105
Qinghai	1.61 × 107	2.09 × 107	5.02 × 101	1.23 × 10−1	9.73 × 109	1.04 × 106	9.36 × 108	3.99 × 108	5.14 × 106	4.43 × 108	2.24 × 109	1.04 × 106
Shanxi	2.03 × 107	2.03 × 107	4.65 × 101	5.39 × 10−2	9.43 × 109	1.01 × 106	9.07 × 108	3.87 × 108	4.97 × 106	4.29 × 108	2.17 × 109	1.01 × 106
Inner Mongolia	5.12 × 107	6.87 × 107	5.02 × 101	1.85 × 10−2	3.19 × 1010	3.41 × 106	3.07 × 109	1.31 × 109	1.68 × 107	1.45 × 109	7.36 × 109	3.41 × 106
Xinjiang	2.82 × 107	3.34 × 107	5.02 × 101	7.22 × 10−2	1.55 × 1010	1.66 × 106	1.49 × 109	6.37 × 108	8.21 × 106	7.07 × 108	3.58 × 109	1.66 × 106
Henan	1.50 × 107	2.44 × 107	4.65 × 101	1.31 × 10−2	1.14 × 1010	1.21 × 106	1.09 × 109	4.65 × 108	5.97 × 106	5.17 × 108	2.61 × 109	1.21 × 106
Shaanxi	9.05 × 106	1.24 × 107	4.84 × 101	1.21 × 10−1	5.77 × 109	6.16 × 105	5.55 × 108	2.37 × 108	3.04 × 106	2.63 × 108	1.33 × 109	6.16 × 105
Ningxia	1.33 × 107	7.31 × 106	5.02 × 101	4.97 × 10−2	3.40 × 109	3.66 × 105	3.27 × 108	1.39 × 108	1.81 × 106	1.55 × 108	7.83 × 108	3.66 × 105
Gansu	2.09 × 107	2.45 × 107	5.02 × 101	3.28 × 10−2	1.14 × 1010	1.22 × 106	1.09 × 109	4.66 × 108	6.01 × 106	5.18 × 108	2.62 × 109	1.22 × 106
Jiangxi	4.66 × 106	4.60 × 106	4.84 × 101	2.33 × 10−2	2.14 × 109	2.28 × 105	2.06 × 108	8.76 × 107	1.13 × 106	9.73 × 107	4.92 × 108	2.28 × 105
Hubei	6.88 × 106	8.08 × 106	4.84 × 101	1.59 × 10−2	3.76 × 109	4.01 × 105	3.61 × 108	1.54 × 108	1.98 × 106	1.71 × 108	8.65 × 108	4.01 × 105
Guangdong	1.13 × 107	2.01 × 107	4.84 × 101	2.02 × 10−2	9.34 × 109	9.96 × 105	8.98 × 108	3.83 × 108	4.92 × 106	4.25 × 108	2.15 × 109	9.96 × 105
Guizhou	6.21 × 106	4.38 × 106	4.84 × 101	1.43 × 10−2	2.04 × 109	2.19 × 105	1.96 × 108	8.36 × 107	1.08 × 106	9.28 × 107	4.70 × 108	2.19 × 105
Yunnan	1.25 × 107	1.26 × 107	4.84 × 101	1.90 × 10−2	5.87 × 109	6.29 × 105	5.64 × 108	2.40 × 108	3.11 × 106	2.67 × 108	1.35 × 109	6.29 × 105
Hunan	8.14 × 106	9.44 × 106	4.84 × 101	1.55 × 10−2	4.39 × 109	4.68 × 105	4.22 × 108	1.80 × 108	2.31 × 106	2.00 × 108	1.01 × 109	4.68 × 105
Liaoning	1.26 × 107	1.22 × 107	4.84 × 101	6.62 × 10−3	5.66 × 109	6.05 × 105	5.44 × 108	2.32 × 108	2.99 × 106	2.58 × 108	1.30 × 109	6.05 × 105
Jilin	9.46 × 106	1.43 × 107	4.84 × 101	5.96 × 10−3	6.66 × 109	7.10 × 105	6.40 × 108	2.73 × 108	3.51 × 106	3.03 × 108	1.53 × 109	7.10 × 105
Heilongjiang	1.01 × 107	1.10 × 107	5.02 × 101	6.75 × 10−3	5.11 × 109	5.47 × 105	4.92 × 108	2.10 × 108	2.70 × 106	2.33 × 108	1.18 × 109	5.47 × 105
Sichuan	5.68 × 106	8.34 × 106	4.84 × 101	6.31 × 10−2	3.88 × 109	4.13 × 105	3.73 × 108	1.59 × 108	2.04 × 106	1.76 × 108	8.93 × 108	4.13 × 105
Fujian	5.86 × 106	6.71 × 106	4.84 × 101	1.29 × 10−2	3.12 × 109	3.33 × 105	3.00 × 108	1.28 × 108	1.65 × 106	1.42 × 108	7.19 × 108	3.33 × 105
Tianjin	1.27 × 106	1.97 × 106	4.84 × 101	1.01 × 10−2	9.14 × 108	9.74 × 104	8.79 × 107	3.75 × 107	4.81 × 105	4.16 × 107	2.10 × 108	9.74 × 104
Hainan	3.64 × 105	4.33 × 104	4.84 × 101	7.50 × 10−3	2.01 × 107	2.26 × 103	1.94 × 106	8.25 × 105	1.12 × 104	9.16 × 105	4.63 × 106	2.26 × 103
Guangxi	9.13 × 106	1.65 × 107	4.84 × 101	1.84 × 10−2	7.67 × 109	8.17 × 105	7.37 × 108	3.14 × 108	4.04 × 106	3.49 × 108	1.77 × 109	8.17 × 105
Xizang	6.20 × 104	1.50 × 105	5.02 × 101	3.93 × 10−1	6.99 × 107	7.44 × 103	6.73 × 106	2.87 × 106	3.68 × 104	3.18 × 106	1.61 × 107	7.44 × 103
Shanghai	1.10 × 106	1.15 × 106	5.02 × 101	8.90 × 10−3	5.33 × 108	5.69 × 104	5.12 × 107	2.18 × 107	2.81 × 105	2.42 × 107	1.23 × 108	5.69 × 104
Chongqing	1.40 × 106	2.64 × 106	5.02 × 101	1.20 × 10−1	1.23 × 109	1.31 × 105	1.18 × 108	5.04 × 107	6.47 × 105	5.59 × 107	2.83 × 108	1.31 × 105
Beijing	1.82 × 105	9.94 × 104	5.02 × 101	2.70 × 10−2	4.62 × 107	4.94 × 103	4.44 × 106	1.89 × 106	2.44 × 104	2.10 × 106	1.06 × 107	4.94 × 103

,

Table A2. Province-level specific returns up to 2050 under scenario 2 (WT lifespan: 20 years).

Province	Primary Energy Savings (GJ)	Recovery Energy (GJ)	Recovery Energy Intensity (GJ/USD)	Energy Payback Period (Years)	Direct Economic Gain (USD)	Terrestrial Acidification (kg SO2 eq)	Recovery Costs (USD)	Carbon Dioxide Emission Reduction (t CO2 eq)	Land Area Savings (m2)	Industrial Water Savings (m3)	Human Toxicity Potential (kg 1,4-DB eq)	Resource Savings (t)
Shandong	1.37 × 107	1.70 × 107	4.65 × 101	1.44 × 10−2	7.90 × 109	1.01 × 106	7.60 × 108	3.24 × 108	4.17 × 106	3.59 × 108	1.82 × 109	1.01 × 106
Jiangsu	1.36 × 107	1.69 × 107	4.65 × 101	1.65 × 10−2	7.85 × 109	1.00 × 106	7.55 × 108	3.22 × 108	4.14 × 106	3.57 × 108	1.81 × 109	1.00 × 106
Hebei	1.44 × 107	1.79 × 107	4.65 × 101	3.17 × 10−2	8.32 × 109	1.06 × 106	8.00 × 108	3.41 × 108	4.39 × 106	3.78 × 108	1.92 × 109	1.06 × 106
Zhejiang	3.86 × 106	4.78 × 106	4.65 × 101	2.62 × 10−2	2.22 × 109	2.84 × 105	2.14 × 108	9.12 × 107	1.17 × 106	1.01 × 108	5.12 × 108	2.84 × 105
Anhui	4.82 × 106	5.97 × 106	4.65 × 101	1.65 × 10−2	2.77 × 109	3.54 × 105	2.67 × 108	1.14 × 108	1.46 × 106	1.26 × 108	6.39 × 108	3.54 × 105
Qinghai	1.32 × 107	1.64 × 107	5.02 × 101	1.23 × 10−1	7.61 × 109	9.71 × 105	7.31 × 108	3.12 × 108	4.02 × 106	3.46 × 108	1.75 × 109	9.71 × 105
Shanxi	2.00 × 107	1.57 × 107	4.65 × 101	5.39 × 10−2	7.30 × 109	9.31 × 105	7.02 × 108	2.99 × 108	3.85 × 106	3.32 × 108	1.68 × 109	9.31 × 105
Inner Mongolia	5.02 × 107	5.20 × 107	5.02 × 101	1.85 × 10−2	2.42 × 1010	3.09 × 106	2.33 × 109	9.92 × 108	1.28 × 107	1.10 × 109	5.57 × 109	3.09 × 106
Xinjiang	2.78 × 107	2.59 × 107	5.02 × 101	7.22 × 10−2	1.21 × 1010	1.54 × 106	1.16 × 109	4.95 × 108	6.38 × 106	5.49 × 108	2.78 × 109	1.54 × 106
Henan	1.48 × 107	1.86 × 107	4.65 × 101	1.31 × 10−2	8.66 × 109	1.10 × 106	8.33 × 108	3.55 × 108	4.56 × 106	3.94 × 108	1.99 × 109	1.10 × 106
Shaanxi	8.91 × 106	9.54 × 106	4.84 × 101	1.21 × 10−1	4.44 × 109	5.66 × 105	4.27 × 108	1.82 × 108	2.34 × 106	2.02 × 108	1.02 × 109	5.66 × 105
Ningxia	1.31 × 107	6.02 × 106	5.02 × 101	4.97 × 10−2	2.80 × 109	3.58 × 105	2.69 × 108	1.15 × 108	1.48 × 106	1.27 × 108	6.45 × 108	3.58 × 105
Gansu	2.05 × 107	1.88 × 107	5.02 × 101	3.28 × 10−2	8.76 × 109	1.12 × 106	8.43 × 108	3.59 × 108	4.63 × 106	3.99 × 108	2.02 × 109	1.12 × 106
Jiangxi	4.58 × 106	3.62 × 106	4.84 × 101	2.33 × 10−2	1.69 × 109	2.15 × 105	1.62 × 108	6.91 × 107	8.89 × 105	7.67 × 107	3.88 × 108	2.15 × 105
Hubei	6.77 × 106	6.28 × 106	4.84 × 101	1.59 × 10−2	2.92 × 109	3.72 × 105	2.81 × 108	1.20 × 108	1.54 × 106	1.33 × 108	6.73 × 108	3.72 × 105
Guangdong	1.11 × 107	1.52 × 107	4.84 × 101	2.02 × 10−2	7.07 × 109	9.02 × 105	6.80 × 108	2.90 × 108	3.73 × 106	3.22 × 108	1.63 × 109	9.02 × 105
Guizhou	6.10 × 106	3.58 × 106	4.84 × 101	1.43 × 10−2	1.66 × 109	2.13 × 105	1.60 × 108	6.82 × 107	8.80 × 105	7.56 × 107	3.83 × 108	2.13 × 105
Yunnan	1.23 × 107	9.94 × 106	4.84 × 101	1.90 × 10−2	4.62 × 109	5.91 × 105	4.44 × 108	1.89 × 108	2.44 × 106	2.10 × 108	1.06 × 109	5.91 × 105
Hunan	8.01 × 106	7.34 × 106	4.84 × 101	1.55 × 10−2	3.41 × 109	4.35 × 105	3.28 × 108	1.40 × 108	1.80 × 106	1.55 × 108	7.86 × 108	4.35 × 105
Liaoning	1.23 × 107	9.36 × 106	4.84 × 101	6.62 × 10−3	4.35 × 109	5.56 × 105	4.19 × 108	1.78 × 108	2.30 × 106	1.98 × 108	1.00 × 109	5.56 × 105
Jilin	9.29 × 106	1.08 × 107	4.84 × 101	5.96 × 10−3	5.03 × 109	6.42 × 105	4.83 × 108	2.06 × 108	2.65 × 106	2.29 × 108	1.16 × 109	6.42 × 105
Heilongjiang	9.85 × 106	8.45 × 106	5.02 × 101	6.75 × 10−3	3.93 × 109	5.02 × 105	3.78 × 108	1.61 × 108	2.08 × 106	1.79 × 108	9.05 × 108	5.02 × 105
Sichuan	5.60 × 106	6.41 × 106	4.84 × 101	6.31 × 10−2	2.98 × 109	3.80 × 105	2.86 × 108	1.22 × 108	1.57 × 106	1.36 × 108	6.86 × 108	3.80 × 105
Fujian	5.75 × 106	5.18 × 106	4.84 × 101	1.29 × 10−2	2.41 × 109	3.08 × 105	2.32 × 108	9.88 × 107	1.27 × 106	1.10 × 108	5.55 × 108	3.08 × 105
Tianjin	1.25 × 106	1.49 × 106	4.84 × 101	1.01 × 10−2	6.93 × 108	8.83 × 104	6.66 × 107	2.84 × 107	3.65 × 105	3.15 × 107	1.59 × 108	8.83 × 104
Hainan	3.54 × 105	4.08 × 104	4.84 × 101	7.50 × 10−3	1.90 × 107	2.49 × 103	1.83 × 106	7.79 × 105	1.03 × 104	8.64 × 105	4.37 × 106	2.49 × 103
Guangxi	9.02 × 106	1.25 × 107	4.84 × 101	1.84 × 10−2	5.82 × 109	7.42 × 105	5.60 × 108	2.39 × 108	3.07 × 106	2.65 × 108	1.34 × 109	7.42 × 105
Xizang	6.17 × 104	1.13 × 105	5.02 × 101	3.93 × 10−1	5.27 × 107	6.71 × 103	5.07 × 106	2.16 × 106	2.78 × 104	2.40 × 106	1.21 × 107	6.71 × 103
Shanghai	1.08 × 106	8.83 × 105	5.02 × 101	8.90 × 10−3	4.11 × 108	5.24 × 104	3.95 × 107	1.68 × 107	2.17 × 105	1.87 × 107	9.46 × 107	5.24 × 104
Chongqing	1.39 × 106	2.00 × 106	5.02 × 101	1.20 × 10−1	9.32 × 108	1.19 × 105	8.96 × 107	3.82 × 107	4.92 × 105	4.24 × 107	2.15 × 108	1.19 × 105
Beijing	1.77 × 105	7.66 × 104	5.02 × 101	2.70 × 10−2	3.56 × 107	4.55 × 103	3.42 × 106	1.46 × 106	1.88 × 104	1.62 × 106	8.20 × 106	4.55 × 103

and

Table A3. Province-level specific returns up to 2050 under scenario 3 (WT lifespan: 25 years).

Province	Primary Energy Savings (GJ)	Recovery Energy (GJ)	Recovery Energy Intensity (GJ/USD)	Energy Payback Period (Years)	Direct Economic Gain (USD)	Terrestrial Acidification (kg SO2 eq)	Recovery Costs (USD)	Carbon Dioxide Emission Reduction (t CO2 eq)	Land Area Savings (m2)	Industrial Water Savings (m3)	Human Toxicity Potential (kg 1,4-DB eq)	Resource Savings (t)
Shandong	1.07 × 107	1.31 × 107	4.65 × 101	1.44 × 10−2	6.07 × 109	2.50 × 105	5.84 × 108	2.49 × 108	3.21 × 106	2.76 × 108	3.86 × 108	2.50 × 105
Jiangsu	1.05 × 107	1.29 × 107	4.65 × 101	1.65 × 10−2	5.99 × 109	3.15 × 105	5.76 × 108	2.46 × 108	3.16 × 106	2.72 × 108	4.85 × 108	3.15 × 105
Hebei	1.12 × 107	1.38 × 107	4.65 × 101	3.17 × 10−2	6.39 × 109	8.80 × 105	6.15 × 108	2.62 × 108	3.38 × 106	2.91 × 108	1.35 × 109	8.80 × 105
Zhejiang	2.94 × 106	3.60 × 106	4.65 × 101	2.62 × 10−2	1.67 × 109	8.36 × 105	1.61 × 108	6.87 × 107	8.84 × 105	7.62 × 107	1.29 × 109	8.36 × 105
Anhui	3.69 × 106	4.53 × 106	4.65 × 101	1.65 × 10−2	2.10 × 109	2.72 × 106	2.02 × 108	8.63 × 107	1.11 × 106	9.58 × 107	4.19 × 109	2.72 × 106
Qinghai	1.03 × 107	1.26 × 107	5.02 × 101	1.23 × 10−1	5.88 × 109	1.39 × 106	5.65 × 108	2.41 × 108	3.11 × 106	2.68 × 108	2.14 × 109	1.39 × 106
Shanxi	1.68 × 107	1.20 × 107	4.65 × 101	5.39 × 10−2	5.59 × 109	9.78 × 105	5.38 × 108	2.29 × 108	2.95 × 106	2.54 × 108	1.51 × 109	9.78 × 105
Inner Mongolia	4.21 × 107	3.91 × 107	5.02 × 101	1.85 × 10−2	1.82 × 1010	5.05 × 105	1.75 × 109	7.45 × 108	9.60 × 106	8.27 × 108	7.78 × 108	5.05 × 105
Xinjiang	2.35 × 107	2.00 × 107	5.02 × 101	7.22 × 10−2	9.29 × 109	3.39 × 105	8.93 × 108	3.81 × 108	4.91 × 106	4.22 × 108	5.21 × 108	3.39 × 105
Henan	1.26 × 107	1.41 × 107	4.65 × 101	1.31 × 10−2	6.54 × 109	1.00 × 106	6.29 × 108	2.68 × 108	3.45 × 106	2.98 × 108	1.54 × 109	1.00 × 106
Shaanxi	7.54 × 106	7.27 × 106	4.84 × 101	1.21 × 10−1	3.38 × 109	1.96 × 105	3.25 × 108	1.38 × 108	1.78 × 106	1.54 × 108	3.02 × 108	1.96 × 105
Ningxia	1.09 × 107	4.86 × 106	5.02 × 101	4.97 × 10−2	2.26 × 109	3.36 × 105	2.17 × 108	9.27 × 107	1.20 × 106	1.03 × 108	5.17 × 108	3.36 × 105
Gansu	1.73 × 107	1.44 × 107	5.02 × 101	3.28 × 10−2	6.68 × 109	7.95 × 105	6.43 × 108	2.74 × 108	3.53 × 106	3.04 × 108	1.22 × 109	7.95 × 105
Jiangxi	3.88 × 106	2.82 × 106	4.84 × 101	2.33 × 10−2	1.31 × 109	1.99 × 105	1.26 × 108	5.37 × 107	6.91 × 105	5.96 × 107	3.07 × 108	1.99 × 105
Hubei	5.73 × 106	4.83 × 106	4.84 × 101	1.59 × 10−2	2.24 × 109	5.38 × 105	2.16 × 108	9.20 × 107	1.18 × 106	1.02 × 108	8.28 × 108	5.38 × 105
Guangdong	9.42 × 106	1.14 × 107	4.84 × 101	2.02 × 10−2	5.31 × 109	3.92 × 105	5.11 × 108	2.18 × 108	2.81 × 106	2.42 × 108	6.04 × 108	3.92 × 105
Guizhou	5.15 × 106	2.86 × 106	4.84 × 101	1.43 × 10−2	1.33 × 109	4.97 × 105	1.28 × 108	5.46 × 107	7.04 × 105	6.06 × 107	7.64 × 108	4.97 × 105
Yunnan	1.04 × 107	7.73 × 106	4.84 × 101	1.90 × 10−2	3.60 × 109	5.64 × 105	3.46 × 108	1.47 × 108	1.90 × 106	1.64 × 108	8.68 × 108	5.64 × 105
Hunan	6.77 × 106	5.64 × 106	4.84 × 101	1.55 × 10−2	2.62 × 109	4.49 × 105	2.52 × 108	1.08 × 108	1.38 × 106	1.19 × 108	6.90 × 108	4.49 × 105
Liaoning	1.03 × 107	7.13 × 106	4.84 × 101	6.62 × 10−3	3.32 × 109	3.39 × 105	3.19 × 108	1.36 × 108	1.75 × 106	1.51 × 108	5.22 × 108	3.39 × 105
Jilin	7.80 × 106	8.10 × 106	4.84 × 101	5.96 × 10−3	3.77 × 109	2.76 × 105	3.62 × 108	1.54 × 108	1.99 × 106	1.71 × 108	4.25 × 108	2.76 × 105
Heilongjiang	8.26 × 106	6.44 × 106	5.02 × 101	6.75 × 10−3	3.00 × 109	7.78 × 104	2.88 × 108	1.23 × 108	1.58 × 106	1.36 × 108	1.20 × 108	7.78 × 104
Sichuan	4.76 × 106	4.87 × 106	4.84 × 101	6.31 × 10−2	2.27 × 109	2.59 × 103	2.18 × 108	9.29 × 107	1.20 × 106	1.03 × 108	3.93 × 106	2.59 × 103
Fujian	4.84 × 106	3.97 × 106	4.84 × 101	1.29 × 10−2	1.84 × 109	6.55 × 105	1.77 × 108	7.56 × 107	9.74 × 105	8.39 × 107	1.01 × 109	6.55 × 105
Tianjin	1.05 × 106	1.12 × 106	4.84 × 101	1.01 × 10−2	5.20 × 108	5.87 × 103	5.00 × 107	2.13 × 107	2.75 × 105	2.37 × 107	9.05 × 106	5.87 × 103
Hainan	2.90 × 105	3.67 × 104	4.84 × 101	7.50 × 10−3	1.71 × 107	4.70 × 104	1.64 × 106	7.00 × 105	9.15 × 103	7.76 × 105	7.23 × 107	4.70 × 104
Guangxi	7.68 × 106	9.42 × 106	4.84 × 101	1.84 × 10−2	4.38 × 109	1.05 × 105	4.21 × 108	1.80 × 108	2.31 × 106	1.99 × 108	1.62 × 108	1.05 × 105
Xizang	5.32 × 104	8.45 × 104	5.02 × 101	3.93 × 10−1	3.93 × 107	4.08 × 103	3.78 × 106	1.61 × 106	2.07 × 104	1.79 × 106	6.27 × 106	4.08 × 103
Shanghai	9.05 × 105	6.75 × 105	5.02 × 101	8.90 × 10−3	3.14 × 108	2.50 × 105	3.02 × 107	1.29 × 107	1.66 × 105	1.43 × 107	3.86 × 108	2.50 × 105
Chongqing	1.18 × 106	1.51 × 106	5.02 × 101	1.20 × 10−1	7.01 × 108	3.15 × 105	6.74 × 107	2.88 × 107	3.70 × 105	3.19 × 107	4.85 × 108	3.15 × 105
Beijing	1.18 × 106	5.86 × 104	5.02 × 101	2.70 × 10−2	2.72 × 107	8.80 × 105	2.62 × 106	1.12 × 106	1.44 × 104	1.24 × 106	1.35 × 109	8.80 × 105

). This will provide comprehensive data support, an evaluation basis, and intellectual support for different stakeholders, including the government, businesses, and individuals, in examining how to deal with the arrival of the decommissioned WT wave and the recycling and utilization of decommissioned WTs to promote the stable and orderly recycling and utilization of decommissioned WTs.

• Energy dimension benefits. Recycling decommissioned WT blades will save approximately 2.70 × 10⁸ to 3.43 × 10⁸ GJ of primary energy, equivalent to saving approximately 9.20 × 10⁸ to 1.17 × 10⁷ tce of standard coal. This not only contributes to energy conservation and emission reduction but also has a positive impact on reducing external energy dependence.
• Economic benefits. Recycling decommissioned WT blades will generate direct economic benefits of approximately CNY 198.5 billion, and at the same time, it will also form a trillion-yuan recycling industry and market, driving the upstream and downstream industrial chains, creating tens of thousands of jobs, and fostering emerging economic forms and new economic growth points.
• Environmental dimension benefits. The environmental dimension should be the most important dimension that requires attention and consideration among all dimensions. To implement the “dual carbon” goals, the Chinese government has made addressing climate change a national development strategy, integrating it within the overall scheme of ecological civilization building and the overall economic and social development framework. This fully demonstrates the Chinese government’s firm determination to follow an ecological priority and green development path. The benefits of various indicators in the environmental dimension show that recycling decommissioned WTs helps promote a reduction in carbon emissions, reduce solid waste pollution, expand green energy, and achieve economic growth. Recycling wind turbine blades will reduce carbon emissions by approximately 4.78 × 10⁹ to 8.14 × 10⁹ tons of CO₂ eq, save various resources by 1.74 × 10⁷ to 2.12 × 10⁷ tons, save land area by 6.16 × 10⁷ to 1.05 × 10⁸ square meters, and reduce terrestrial acidification by 1.74 × 10⁷ to 2.12 × 10⁷ kg SO₂eq.
• Social dimension benefits. By recycling decommissioned WT blades, the HT will be reduced by approximately 2.69 × 10¹⁰ to 4.57 × 10¹⁰ kg 1,4-DCB, and the industrial water consumption will be reduced by approximately 5.31 × 10⁹ to 9.04 × 10⁹ m³, which is the same as the overall water usage of Shaanxi Province in 2023.

5. Conclusions and Outlook

Facing the imminent challenge of decommissioned WT recycling management, this study clarifies the spatiotemporal evolution of decommissioned WTs by constructing an SCA-ITransformer-BiLSTM prediction model based on optimization algorithms and deep learning. It conducts an in-depth quantitative analysis of the quantity, value, and comprehensive and sustainable benefits of recycling decommissioned WTs. The main conclusions are as follows:

• China will face a continuous wave of decommissioned WTs around 2030, with the annual decommissioned capacity exceeding 15 GW post-2030 under the lowest forecast scenario. By approximately 2050, the scale of decommissioned WTs is projected to be roughly equivalent to the installed capacity.
• The spatiotemporal distribution of decommissioned WT capacity shows regional disparities: around 2030, Xinjiang and other northwest regions will have approximately 59 GW of decommissioned capacity, while Inner Mongolia and the northern China regions will account for around 106 GW (27% of the total). By 2050, the proportion in northern China will remain at approximately 27%, while that in the northwest will decrease to approximately 23%, with regions like Xizang and Beijing maintaining below 1% throughout both periods.
• Recycling decommissioned WTs yields significant quantified sustainable benefits: reducing carbon emissions by 4.78 × 10⁹ to 8.14 × 10⁹ tons of CO₂ equivalent, saving 6.16 × 10⁷ to 1.05 × 10⁸ m² of land, mitigating terrestrial acidification by 1.74 × 10⁷ to 2.12 × 10⁷ kg SO₂ equivalent, reducing human carcinogenic toxicity by 2.69 × 10¹⁰ to 4.57 × 10¹⁰ kg 1,4-DCB, and saving 5.31 × 10⁹ to 9.04 × 10⁹ cubic meters of industrial water consumption.

Based on the problems identified in the research findings and the conclusions, the following countermeasures and suggestions are proposed.

• Strengthen top-level design and establish a recycling policy system for decommissioned WTs. The government should clarify the status of decommissioned WTs (what kind of waste they are), responsible parties (which can draw on the “production responsibility system”), recycling standards (specifying the recycling rate), etc., through legislation to promote the efficient recycling of decommissioned WTs. At the same time, in response to the high initial costs and low benefits of decommissioned WT recycling [38], a package of supportive policies should be introduced [12,39], including providing financial subsidies, tax relief, reducing transportation costs, building a unified recycling network, and easing market entry thresholds so as to boost the rapid growth of the recycling industry.
• Establish a long-term mechanism and develop a timeline and roadmap for decommissioning WTs. North China and Northwest China should prepare for the recycling peak by 2030 and also be ready to cope with the continuous wave of decommissioning WTs. Other regions should also prepare for the wave of decommissioning WTs around 2035. All regions should formulate a decommissioning “roadmap” in a prudent and orderly manner based on the condition of the WTs, adapting measures to local conditions and timeframes, avoiding “herding” decommissioning, and ensuring consistency with national energy transformation policies.
• Marry all forces and strive to erect a green, low-carbon circular economic framework. Enterprises should collaborate with universities and research institutions, adhere to technological innovation as the driving force, continuously strengthen R&D investment, and strive to increase the service life of WTs [40], achieving the goal of “flattening the curve and postponing the peak”. Meanwhile, during the research and development process of the new generation of WTs, try to use low-carbon and easily recyclable materials [41], reduce carbon emissions throughout the life cycle of the WTs, and strive to achieve a green, low-carbon circular economy system.

During the research process of this article, due to limited conditions, data such as the material composition of WTs and energy consumption during the recycling process were all referenced from the relevant literature and were not obtained through experiments, which may lead to certain deviations in the results. In the future, we will dedicate ourselves to deeper research on decommissioned WTs, providing a scientific basis for promoting the stable and orderly recycling of decommissioned WTs.