Spatiotemporal Characteristics and Influencing Factors of Renewable Energy Production Development in Ningxia Hui Autonomous Region, China (2014–2021)

Abstract

Promoting the development of low-carbon renewable energy is crucial for meeting the growing energy demand, reducing dependence on fossil fuels, and controlling carbon dioxide emissions. Clarifying the spatiotemporal characteristics of regional renewable energy production and its influencing factors will help optimize the spatial layout of renewable energy production and provide a solid theoretical basis for coordinating the development of all aspects of renewable energy production. Using panel data from 22 districts and counties in Ningxia from 2014 to 2021, this study employed the spatial Gini coefficient, Moran’s I index, standard deviational ellipse, and geographical detector to analyze the spatiotemporal evolution patterns and influencing factors of renewable energy production development in Ningxia. The results indicate that renewable energy production in Ningxia exhibits significant spatial agglomeration and autocorrelation. Temporally, renewable energy production shows a spatial expansion trend characterized by dynamic agglomeration patterns. The coupling degree between renewable energy generation and the spatial distribution of power production is relatively high, with notable regional disparities. Urbanization level, urban population, per capita GDP, and industrial SO₂ emissions have a positive impact on renewable energy production, while energy intensity and environmental regulation show insignificant effects. To further promote the development of renewable energy, Ningxia should strengthen power infrastructure construction at the county level, enhance the radiating and driving effects of high-value areas on surrounding cities and counties, optimize the spatial layout of power facilities based on the agglomeration trajectories of renewable energy production, integrate multiple types of renewable energy to improve overall generation efficiency and system stability, and encourage local enterprises to increase technological and economic investments in renewable energy, thereby advancing sustainable energy transition and achieving high-quality development in resource-based regions.

1. Introduction

To reduce greenhouse gas emissions and strive to limit the temperature rise to below 1.5 °C, countries that signed the Paris Agreement have taken various actions to address climate change [1]. Since 2000, the rapid growth in global carbon dioxide (CO₂) emissions has been primarily driven by the increasing energy demand in developing countries, making decarbonization more challenging in these regions compared to developed countries [2]. Facing the pressure and challenges of carbon reduction, China has proposed the goals of achieving carbon peaking by 2030 and carbon neutrality by 2060 [3]. China’s greenhouse gas emissions mainly originate from the combustion of fossil fuels such as coal, oil, and natural gas. Therefore, increasing renewable energy generation is a crucial pathway to reducing regional CO₂ emissions [4,5]. Over the past two decades (2001–2010 and 2011–2020), China’s installed renewable energy capacity increased by 2.05 times and 2.24 times, respectively, while renewable energy generation grew by 1.96 times and 1.81 times, respectively [6,7,8]. However, regions with high renewable energy generation in China do not align with regions with high electricity consumption [9]. Non-hydro renewable energy production is mainly concentrated in the northwest, while consumption is primarily in the southeast, indicating that the spatial separation between production and consumption has not fundamentally changed [10]. According to the China Electric Power Statistical Yearbook (National Energy Administration, 2020), in 2019, the per capita renewable energy generation in western China was 3801 kWh, 7.9 times that of eastern China [11]. This highlights that while increasing renewable energy generation, it is essential to optimize the spatial layout of renewable energy production to reduce resource waste caused by the mismatch between supply and demand centers.

In recent years, with the rapid development of renewable energy, many scholars have conducted extensive research and discussions on renewable energy production and development from various disciplines and perspectives. First, to enhance technological innovation and efficiency optimization in renewable energy production, many scholars have focused on optimizing power system design from an engineering perspective to improve the overall operational efficiency of photovoltaic and wind power plants. Due to the high variability of renewable energy generation, related technologies face challenges in meeting demand or managing surplus electricity. Consequently, some studies have focused on analyzing the spatiotemporal characteristics and complementarity of wind and solar power plants [12,13,14,15]. Regarding issues in renewable energy consumption, research has primarily analyzed the impact of key factors such as power balance, power regulation performance, transmission capacity, and load levels on renewable energy generation capacity [16]. Concerning the spatial distribution in renewable energy system optimization models, existing literature has mainly examined the influence of spatial concentration [17] and attempted to establish energy system optimization models to smooth power supply and improve renewable energy utilization. Other studies have developed supply–demand models based on spot electricity prices to capture the complete spatial dependency structure of wind power and analyze the spatial dependence of wind power plants [18]. To address the numerous obstacles encountered in accelerating the deployment of renewable energy plants, some studies have measured innovation value and evaluated new technological approaches to optimize wind power plant deployment, emphasizing the growing importance of geographic diversity [19]. Additionally, some scholars argue that optimizing the spatial utilization of renewable energy requires integrating spatial planning with energy system planning [20]. Such studies can enhance the flexibility of renewable energy site selection, alleviate land-use constraints, and improve energy productivity and economic sustainability.

With the widespread application of advanced models such as deep learning, scholars have combined satellite data and deep learning to explore the expansion of power plant spatial distribution in power systems. Yuehong Chen used satellite data and deep learning methods to reveal the expansion of photovoltaic power stations in China over the past decade. The study found that the area of photovoltaic power stations in China increased more than 600-fold from 2010 to 2022, with the highest annual growth rate of 53% in western China [21]. Kruitwagen, L., et al. provided a global inventory of commercial, industrial, and utility-scale photovoltaic installations using remote sensing images, machine learning, and large-scale cloud computing infrastructure [22]. Francisco Flores made the first attempt to assess how introducing hourly resolution impacts the results of integrated assessment models, particularly focusing on the Global Change Analysis Model (GCAM) to evaluate the effects of renewable energy variability on long-term decarbonization strategies [23]. Beyond production technology, studies have shown that institutional settings significantly influence spatial distribution and associated institutional costs. For example, A. Pechan compared different pricing scenarios to evaluate the impact of subsidy schemes and market design on the spatial distribution of wind energy installations. The results indicated that under grid constraints, the more spatially dispersed the expansion of wind power, the better its performance in terms of total cost and the share of wind power in final demand [24]. Furthermore, when explaining the spatiotemporal expansion patterns of wind energy and improving the coordination of multi-level energy policies, it is necessary to consider policies and initiatives at the state, county, and municipal levels [25]. These studies aim to rationally allocate photovoltaic and wind resources by incorporating spatial dimensions, thereby achieving efficient utilization of renewable energy and reducing the operational costs of energy systems.

To improve the economics and sustainability of renewable energy production, a significant body of literature has explored the complex relationships between energy flows and economic and ecological sustainability from both the production and consumption sides. From the production side, the locations and trade patterns of global basic materials may change due to the heterogeneity of renewable electricity. Therefore, some scholars have defined and assessed the “renewable energy pull effect”, investigating the green relocation of energy-intensive industries in countries with different renewable energy resource endowments and the resulting economic consequences [26,27]. Additionally, to enhance the ability to cope with multi-source economic complexities, some studies have explored the progressive synergy between rare earth elements and renewable energy, leveraging this synergy to improve economic complexity, which in turn promotes green economic and technological development while reducing CO₂ emissions and addressing pervasive climate change [28]. From the consumption side, existing literature presents mixed results on the relationship between renewable energy consumption and economic growth, as the impact of renewable energy consumption on economic growth can be positive, negative, or insignificant. As energy is a fundamental input for almost all economic activities and drives production, transportation, and technological progress, some scholars argue that increased renewable energy consumption is positively correlated with GDP and associated with higher economic output [29]. Other studies have found that the relationship between renewable energy consumption and economic growth depends on the level of renewable energy use, and developing countries need to exceed a certain threshold of renewable energy consumption to achieve positive economic growth through renewable energy investments [30].

To address increasingly prominent environmental challenges, it is also worth exploring whether the accelerated development of renewable energy positively feeds back into ecological sustainability. Eco-friendly innovative technologies are widely regarded as a key means to achieve carbon neutrality goals [31]. Therefore, some studies have focused on the synergistic relationship between renewable energy consumption, green technological innovation, and environmental regulation. Some scholars argue that green innovation technologies and renewable energy have a bidirectional relationship, while environmental regulations and renewable energy have a unidirectional causal relationship [32]. Other studies have used environmental regulation and energy efficiency as mediating factors to demonstrate that renewable energy and green technological innovation positively contribute to sustainable development in both the short and long term [33]. Building on this, Aamir Javed added economic growth, trade openness, and urbanization as factors to analyze their impact on ecological sustainability [34].

Energy issues are not only engineering problems but also socioeconomic problems [35,36], and energy geography issues represent the overlap and interaction between energy systems and human socioeconomic activities in geographical space [1]. Therefore, it is essential to pay attention to the geographical spatiality of energy issues. Current research primarily focuses on the impact of technological changes on the distribution and spatial expansion of renewable energy production facilities. However, in-depth analysis of their spatiotemporal distribution characteristics remains insufficient, and the spatial correlation of variables has not been fully considered. Particularly from the perspective of geography’s temporal and spatial attributes, research on the spatiotemporal development patterns of renewable energy is scarce, and the exploration of how socioeconomic factors influence the spatial development layout and industrial structure of renewable energy at different temporal and spatial scales remains inadequate. This not only limits our understanding of the dynamics of renewable energy distribution but also hinders the ability to formulate comprehensive planning and management strategies from a spatiotemporal perspective. Moreover, previous studies have mainly focused on economically developed regions, with less attention paid to underdeveloped regions, especially in western China. The development of renewable energy in different regions must consider their real-world contexts and determine the order of development rather than blindly pursuing renewable energy expansion [9]. Exploring the spatial distribution characteristics and dynamic evolution patterns of renewable energy in underdeveloped regions from both temporal and spatial dimensions will help objectively grasp the current status and distribution features of regional renewable energy development, thereby providing targeted policy recommendations for renewable energy development across China.

As the first national comprehensive demonstration zone for new energy, the Ningxia Hui Autonomous Region (hereinafter referred to as Ningxia) boasts abundant wind and solar resources. In recent years, through large-scale, intensive, and park-based development, significant progress has been made in wind and solar energy development. In 2022, Ningxia’s installed new energy capacity reached 30.41 million kilowatts, ranking ninth in the country. New energy accounted for over 50% of the total installed power capacity in the region, surpassing thermal power to become the largest power source in the autonomous region. Ningxia was also the fourth province in China, after Qinghai, Hebei, and Gansu, to exceed 50% new energy in its installed capacity. Moreover, it was the first provincial grid in China where wind and solar power output exceeded the total electricity consumption of the region. Additionally, the utilization rate of new energy reached 98.01% in 2022, ranking first in northwest China and first in the country for renewable energy power consumption capacity. The scale of new energy storage in operation reached approximately 1.154 million kilowatts, ranking second in the country. In 2022, the electricity transmitted from new energy sources reached 16.3 billion kWh, accounting for 17.25% of the total electricity transmitted from the region.

Based on this, this study analyzed the locational distribution and industrial spatial agglomeration characteristics of renewable energy in various cities and counties of Ningxia from 2014 to 2021, as well as the spatial autocorrelation patterns, to demonstrate the significant spatial dependency and agglomeration characteristics of renewable energy production. Using the geographical detector method, this study analyzed the economic, technological, environmental, and social factors influencing renewable energy production, revealing the driving factors’ impact on renewable energy production. The aim was to construct a rational spatial layout for renewable energy development, optimize the allocation of renewable energy resources, and provide a scientific basis for formulating relevant policies.

2. Materials and Methods

2.1. Study Area and Data Sources

Considering the research objectives and data availability, the study area is the Ningxia Hui Autonomous Region (Figure 1). This study used panel data from five cities in Ningxia from 2014 to 2021 for empirical analysis. The main core data and socioeconomic statistics were sourced from the China Energy Statistical Yearbook (National Bureau of Statistics, 2015–2022), China Electric Power Statistical Yearbook (National Energy Administration, 2015–2022), Ningxia Statistical Yearbook (Ningxia Bureau of Statistics, 2015–2022), and statistical yearbooks and bulletins from 5 prefecture-level cities (Yinchuan, Shizuishan, etc.), accessed through the Ningxia Statistical Information Network [37,38,39,40,41,42,43,44].

2.2. Variables

Renewable energy generation was considered the dependent variable reflecting renewable energy production capacity (including wind and solar energy, excluding hydropower in this study). For the independent variables (

Table 1. Variable explanations.

Variable	Symbol	Definition	Unit
Renewable energy production	REP	Renewable energy generation	10,000 kWh
Per capita GDP	GDP	GDP divided by population	Yuan
Energy intensity	EI	Energy consumption divided by GDP	Tce/10,000 Yuan
Industrial sulfur dioxide emissions	SO2	Sulfur dioxide emissions divided by GDP	kg/10,000 Yuan
Environmental regulation	ER	Industrial pollution control investment divided by GDP (industrial added value)	%
Urbanization level	UL	Urbanization rate	%
Urban population	UP	Urban population	Persons
Annual sunshine hours	SH	Annual sunshine hours	Hours
Average wind speed	WS	Average wind speed	m/s

), eight explanatory variables based on economic, technological, environmental, and natural meteorological factors—namely, per capita GDP, industrial sulfur dioxide emissions, urbanization level, urban population, energy intensity, environmental regulation, annual sunshine hours, and average wind speed—were used to explore the spatial interaction relationships of renewable energy production.

In this study, we conducted an in-depth empirical analysis from multiple perspectives, including industrial agglomeration level, industrial structure, regional spatial distribution characteristics, and related influencing factors.

2.3. Research Methods

(1) Spatial Gini Coefficient

The spatial Gini coefficient (SGC) [45,46] is an indicator used to measure the degree of industrial agglomeration, reflecting the equilibrium of industrial spatial distribution within a region, as shown in Equation (1):

$$\mathrm{S}\mathrm{G}\mathrm{C}={\sum }_{i=1}^n{({\displaystyle \frac{P_i}{{\sum }_{i=1}^n{P}_i}}-{\displaystyle \frac{Q_i}{{\sum }_{i=1}^n{Q}_i}})}^2$$

(1)

where P_i is the renewable energy generation of province i, and Q_i is the total power generation of province i. The spatial Gini coefficient ranges from 0 to 1. A higher Gini coefficient indicates a higher degree of spatial agglomeration in renewable energy production, while a lower coefficient indicates a more uniform spatial distribution.

(2) Standard Deviational Ellipse (SDE)

The standard deviational ellipse (SDE) is a typical method for analyzing the directional characteristics of spatial distribution. The standard deviational ellipse (SDE) method is used to study the spatial characteristics of renewable energy production. SDE reflects the spatial distribution characteristics of the research object, such as dispersion, agglomeration, and evolutionary trends [47]. This method has been widely applied in spatial assessments such as carbon emissions [48], energy consumption [49], energy intensity [50], and waste management [51]. It calculates the standard distance of a series of points along the major and minor axes and presents the results in the form of an ellipse on a map. By observing the length and direction of the ellipse axes, the dynamic trends of economic activities can be more intuitively described. The parameters of SDE include the centroid, major and minor axis standard deviations, and azimuth angle. The equations are as follows:

Centroid X, Y:

$${\overline{X}}_{\mathrm{w}}={\displaystyle \frac{{\sum }_{i=1}^n{w}_i{x}_i}{{\sum }_{i=1}^n{w}_i}}{,\overline{Y}}_{\mathrm{w}}={\displaystyle \frac{{\sum }_{i=1}^n{w}_i{y}_i}{{\sum }_{i=1}^n{w}_i}}$$

(2)

$${\overline{X}}_{\mathrm{i}} {={\mathrm{x}}_{\mathrm{i}}-\overline{X}}_{\mathrm{w}}{, \overline{y}}_{\mathrm{i}}={\mathrm{y}}_{\mathrm{i}}-{\overline{Y}}_{\mathrm{w}}$$

(3)

Rotation:

$$\theta =\mathit{arctan}{\displaystyle \frac{\left({\sum }_{i=1}^n{w}_i^2{\overline{x}}_i^2-{\sum }_{i=1}^n{w}_i^2{\overline{y}}_i^2\right)+\sqrt{\begin{array}{c}{\left({\sum }_{i=1}^n{w}_i^2{\overline{x}}_i^2-{\sum }_{i=1}^n{w}_i^2{\overline{y}}_i^2\right)}^2+4{\sum }_{i=1}^n{w}_i{\overline{x}}_i^2{\overline{y}}_i^2\\ \end{array}}}{2{\sum }_{i=1}^n{w}_i^2\overline{x}\mathrm{i}\overline{y}\mathrm{i}}}$$

(4)

X, Y standard distance:

$${\mathsf{\sigma }}_{\mathrm{x}} =\sqrt{{\displaystyle \frac{{\sum }_{i=1}^n{(w_i{\overline{x}}_{i}cos\theta -w_i{\overline{y}}_{i}sin\theta )}^2}{{\sum }_{i=1}^n{w}_i^2}}}$$

(5)

$${\mathsf{\sigma }}_{\mathrm{y}} =\sqrt{{\displaystyle \frac{{\sum }_{i=1}^n{(w_i{\overline{x}}_{i}sin\theta -w_i{\overline{y}}_{i}cos\theta )}^2}{{\sum }_{i=1}^n{w}_i^2}}}$$

(6)

where (x_i, y_i) are the latitude and longitude coordinates of province i; w_i is the regional weight; ($\overline{X}$_w, $\overline{Y}$_w) are the coordinates of the centroid; ($\overline{xi}$_i, $\overline{y}$_i) are the coordinates of each research object relative to the centroid; θ is the azimuth angle of the SDE, representing the angle between the north direction of the ellipse and the major axis; and σ_x and σ_y are the standard deviations of the major and minor axes of the ellipse, respectively.

Using the SDE method, the spatial agglomeration characteristics and evolutionary trends of renewable energy production in Ningxia were analyzed. By comparing the regional changes of SDE over time, the dynamic process of industrial spatial agglomeration can be comprehensively described [52]. When the SDE area decreases, it indicates that the spatial distribution of renewable energy production tends to be agglomerated; when the SDE area increases, it indicates diffusion; if the SDE area does not change significantly, it indicates that the spatial distribution of renewable energy production is relatively stable [45].

(3) Spatial Autocorrelation Analysis

Spatial autocorrelation analysis describes the spatial distribution patterns of phenomena, revealing the spatial agglomeration and heterogeneity interaction mechanisms of research objects [53]. A substantial body of scholarship has employed spatial autocorrelation models to investigate energy systems from multiple perspectives: Wang, S. analyzed the spatiotemporal evolution and driving factors of energy-related CO₂ emissions in China through an energy consumption lens, revealing significant east–west disparities in emission intensity [54]. Zhu et al. integrated Moran’s I indices with spatial panel econometric models to decode the spatial interdependence between air pollution patterns and renewable energy technological innovation, identifying threshold effects in policy interventions [55]. Ma et al. conducted a provincial-level spatial dependency analysis of renewable energy innovation in China, demonstrating path-dependent clustering characteristics along major economic corridors [56]. Quito. et al. extended this analytical framework to 43 European countries (1990–2019), establishing statistically significant spatial autocorrelations between energy efficiency, renewable penetration, and financial development trajectories [57]. This analysis method is divided into global and local autocorrelation levels. Global autocorrelation is used to evaluate the similarity of renewable energy production observations in adjacent spatial regions, thereby determining whether the entire industry exhibits spatial autocorrelation trends. However, when the study area is extensive, global autocorrelation may overlook spatial heterogeneity within the industry and fail to accurately reflect local spatial correlations within geographic units. Therefore, when exploring the spatial agglomeration characteristics of renewable energy production in Ningxia, the global spatial autocorrelation method was used for preliminary judgment. The Getis-Ord General G clustering method [58] and local indicators of spatial association (LISA) cluster maps [59] were employed to identify statistically significant (p < 0.05) high/low-value clusters across intra- and inter-city/county spatial units.

The equations are as follows:

$$\mathit{Global}\mathit{Moran}’s I={\displaystyle \frac{n{\sum }_{i=1}^n{\sum }_{j=1}^n{w}_{ij}\left(x_i-\overline{x}\right)(x_j-\overline{x})}{{\sum }_{i=1}^n{\sum }_{j=1}^n{w}_{ij}\ast {\sum }_{i=1}^n{(x_i-\overline{x})}^2}}$$

(7)

$$\mathit{Local}\mathit{Moran}’s I={\displaystyle \frac{n(x_i-\overline{x}){\sum }_{j=1}^n{w}_{ij}(x_j-\overline{x})}{{\sum }_{i=1}^n{(x_i-\overline{x})}^2}}$$

(8)

where x_i and x_j are the renewable energy generation of cities i and j; $\overline{x}$ is the average renewable energy generation of all cities; and w_ij is the binary adjacency spatial weight matrix (if city i is adjacent to city j, w_ij = 1; otherwise, w_ij = 0). The Moran’s I index ranges from −1 to 1. When Moran’s I > 0, the spatial distribution of renewable energy production shows an agglomeration pattern, meaning high (low) values tend to cluster near other high (low) values; when Moran’s I < 0, the spatial distribution shows dispersion, meaning high (low) values tend to cluster near low (high) values; when Moran’s I = 0, the spatial distribution of renewable energy production is random [60]. This study used global and local Moran’s I indices to analyze the spatial autocorrelation patterns and regional evolution characteristics of renewable energy production in Ningxia.

(4) Geographical Detector Analysis

The geographical detector is a set of statistical methods used to detect spatial heterogeneity and reveal the driving forces behind it. Its core idea is based on the assumption that if an independent variable has a significant impact on a dependent variable, the spatial distributions of the independent and dependent variables should be similar [61,62]. The geographical detector can be applied to study areas ranging from national to township scales, and it can achieve statistical precision comparable to larger sample sizes with fewer than 30 samples. In recent years, these methodologies have been extensively applied in energy and environmental studies [63,64]. For instance, Yu, B. employed the geodetector model to investigate the spatial heterogeneity of renewable energy generation efficiency mechanisms across 30 Chinese provinces, revealing significant regional disparities driven by policy and resource endowments [65]. Yangyang Zhang analyzed the spatiotemporal evolution of carbon emissions in the Bohai Bay region (43 cities) using the standard deviation ellipse method, explored their spatial clustering characteristics through LISA models, and further applied the geodetector framework to quantify the spatiotemporal heterogeneity of driving factors, revealing significant interaction effects between industrial structure and energy intensity [66]. When exploring the influencing factors of renewable energy production in the context of spatiotemporal dimensions, it is necessary to consider not only resource endowments but also the impact of economic, social, and ecological factors. Therefore, this study used the geographical detector to quantitatively analyze the influencing factors of renewable energy production.

The calculation model for detecting the influencing factors of renewable energy production is as follows:

$$P_{D,G}=1-{\displaystyle \frac{1}{{{\sigma }^2}_G}} {\sum }_{i=1}^m{n}_{D,i} {\sigma }_{D,i}^2$$

(9)

where P_D,G is the detection power value of factor D; ${{\sigma }^2}_G$ represents the variance of renewable energy production at the primary regional level (e.g., provincial level); ${\sigma }_{D,i}^2$ represents the variance of renewable energy production at the secondary regional level (e.g., city level); $n_{D,i}$ represents the number of renewable energy production units in the secondary region; m represents the number of secondary regions; and q_D,G ranges from 0 to 1, with a higher q_D,G value indicating a greater influence of factor G on renewable energy production D.

3. Results

3.1. Spatial Agglomeration and Spatial Autocorrelation Characteristics of Renewable Energy Production

Temporal Statistics: The total renewable energy production in Ningxia and the scale of various spatial Gini coefficients exhibited a composite evolutionary characteristic of continuous growth and fluctuating changes.

The spatial Gini coefficients (Figure 2) for wind power and solar photovoltaic power generation (non-hydro renewable energy, hereinafter referred to as renewable energy) production in Ningxia (compared with thermal power generation) ranged between 0 and 0.2. This indicates that renewable energy production exhibited spatial agglomeration characteristics. Although the overall degree of spatial agglomeration was relatively low, it gradually increased over time. Additionally, there were significant differences and fluctuations in the spatial agglomeration levels between wind power and photovoltaic power, indicating an evolutionary trend in agglomeration characteristics.

The spatial agglomeration degree of wind power showed a gradual upward trend, increasing from 0.053 in 2014 to 0.219 in 2021. After 2016, it surpassed other renewable energy sources, exhibiting the highest spatial agglomeration degree among all energy types. Photovoltaic power generation showed significant fluctuations, with a declining trend from 2014 to 2017, reaching its lowest value of 0.071 in 2017, and then gradually increasing to 0.114 by 2021. The spatial agglomeration degree of photovoltaic power generation was relatively significant in the initial stage. With the scale development, it steadily increased from 2017 to 0.114 in 2021, which was lower than that of wind power.

In contrast, thermal power generation remained stable and relatively dispersed, which aligns with the distribution status and characteristics of thermal power generation in Ningxia in recent years. The average spatial Gini coefficients for total renewable energy generation, wind power, and photovoltaic power were 0.132, 0.154, and 0.128, respectively, indicating a transition from a dispersed state to an agglomerated state.

Spatial Evolution: Renewable energy production in Ningxia exhibited dual characteristics of spatial agglomeration and spatial diffusion, as well as spatial autocorrelation characteristics.

Using ArcGIS 10.8, the global Moran’s indices for the total renewable energy production and its components in Ningxia from 2014 to 2021 were calculated (

Table 2. Global Moran’s indices for renewable energy production in Ningxia from 2014 to 2021.

Year	Renewable Energy Generation			Wind Power			Solar Power
	z-Value	p-Value	Moran’s I	z-Value	p-Value	Moran’s I	z-Value	p-Value	Moran’s I
2014	5.776	0.000	0.824	5.649	0.000	0.804	5.321	0.000	0.752
2015	5.538	0.000	0.787	5.577	0.000	0.792	5.281	0.000	0.735
2016	5.731	0.000	0.816	5.735	0.000	0.816	5.632	0.000	0.796
2017	5.645	0.000	0.800	5.583	0.000	0.785	5.671	0.000	0.804
2018	5.529	0.000	0.780	5.417	0.000	0.757	5.467	0.000	0.775
2019	5.586	0.000	0.790	5.445	0.000	0.763	5.505	0.000	0.779
2020	5.547	0.000	0.784	5.098	0.000	0.707	5.834	0.000	0.825
2021	5.534	0.000	0.785	5.283	0.000	0.744	5.449	0.000	0.769

). The global Moran’s indices for 2014–2021 were all greater than 0 and passed the 1% significance test, indicating significant spatial positive correlation and obvious spatial agglomeration characteristics for all types of renewable energy production in Ningxia. The average global Moran’s I indices for total renewable energy generation, wind power, and photovoltaic power were 0.796, 0.771, and 0.779, respectively, with the spatial autocorrelation of total renewable energy generation being the most significant. This spatial interdependence and correlation further strengthened the spatial agglomeration effect of renewable energy production.

To further test the clustering between cities and determine whether renewable energy production exhibits high-value or low-value clustering characteristics in space, the Getis-Ord General G indices for renewable energy production in Ningxia from 2014 to 2021 were calculated (

Table 3. Getis-Ord general G indices for renewable energy production in Ningxia from 2014 to 2021.

	Renewable Energy Generation			Wind Power			Solar Power
Year	z-Value	p-Value	General G	z-Value	p-Value	General G	z-Value	p-Value	General G
2014	4.869	0.000	0.072	5.006	0.000	0.081	4.373	0.000	0.070
2015	4.572	0.000	0.067	4.735	0.000	0.070	4.202	0.000	0.063
2016	4.637	0.000	0.068	4.717	0.000	0.071	4.431	0.000	0.064
2017	4.367	0.000	0.064	4.225	0.000	0.063	4.723	0.000	0.068
2018	4.264	0.000	0.065	4.098	0.000	0.063	4.555	0.000	0.067
2019	4.316	0.000	0.065	4.144	0.000	0.064	4.565	0.000	0.066
2020	4.258	0.000	0.064	4.050	0.000	0.068	4.630	0.000	0.065
2021	4.367	0.000	0.066	4.250	0.000	0.069	4.455	0.000	0.065

).

From 2014 to 2021, the observed values of General G were all greater than the expected value of 0.048, and all passed the 1% significance test, with positive z-scores, indicating significant spatial positive correlation and a trend of high-value clustering for all types of renewable energy production in Ningxia.

To visually display the local spatial correlation of the 22 districts and counties in Ningxia, LISA (local indicators of spatial association) maps for renewable energy production were generated using ArcGIS 10.8 at the 1% significance level (Figure 3, Figure 4, Figure 5 and Figure 6), where red, blue, pink, and light blue areas represent high–high (H-H), low–low (L-L), high–low (H-L), and low–high (L-H) clusters, respectively. Gray and white areas indicate no significance and no data, respectively.

From the LISA maps of renewable energy generation, it can be observed that the high-value clusters have moved southward from the three districts of Yinchuan City in the north to Wuzhong City and continue to expand, with Lingwu City, Qingtongxia City, and Hongsipu District consistently under Wuzhong in the H-H cluster. Zhongning County has also been in the high-value cluster since 2016. The L-L clusters are consistently located in Yuanzhou District, Jingyuan County, and Longde County of Guyuan City, with Huinong County and Dawukou County of Shizuishan City joining the low-value clusters since 2017. There are no H-L or L-H clusters. For wind power, the H-H clusters show a similar spatial distribution to the total, moving southward from the three districts of Yinchuan and expanding to Wuzhong, with Yanchi County, Lingwu, and Litong District separating from the high-value clusters since 2019. Additionally, Zhongning has been in the high-value cluster since 2016. The L-L clusters are relatively stable in the districts and counties of Shizuishan and Yuanzhou, Longde, and Jingyuan, with Yuanzhou separating from the low-value clusters since 2018. Since 2020, L-H clusters have been concentrated in Lingwu. There are no H-L clusters. The high-value areas for photovoltaic power generation show significant changes, with H-H clusters concentrated in Zhongning, Shapotou Qingtongxia, Hongsipu, and Tongxin from 2014 to 2016. From 2017 to 2019, the high-value areas moved to Yinchuan, and since 2020, Zhongning, Shapotou, Qingtongxia, Hongsipu, and Tongxin have returned to the high-value clusters. The L-L clusters are consistently concentrated in Yuanzhou, Pengyang County, Longde, and Jingyuan. The L-H cluster was only concentrated in Lingwu in 2014, and since 2016, Lingwu has become a high-value cluster. There are no H-L clusters.

Comparing the LISA maps for different years, it can be observed that the H-H and L-L clusters show a dynamic trend of gradually spreading to the central region, indicating that the issue of uneven resource distribution has been gradually alleviated. The global and local spatial autocorrelation tests show that renewable energy production in Ningxia has significant spatial autocorrelation, and the driving mechanisms of the spatial distribution of renewable energy production need further study. For districts and counties with relatively high renewable energy development and concentrated distribution in H-H clusters, it is necessary to improve the regulation capacity and flexibility of the power system by constructing transmission infrastructure, promote the rapid development of renewable energy storage, and facilitate high-level consumption of renewable energy. For L-L cluster districts and counties, due to resource endowment factors, the potential for renewable energy development is limited, and policy encouragement is needed to reduce the non-technical costs of renewable energy development through a series of measures. For L-H clusters, where areas with low renewable energy development are surrounded by areas with high development, the latter should provide technical, financial, and policy support to the former to break through bottlenecks. Additionally, the LISA map for thermal power generation in Ningxia from 2014 to 2021 shows that the high-value clusters for thermal power generation are consistently concentrated in Yinchuan, while the low-value clusters are concentrated in Guyuan.

3.2. Dynamic Agglomeration-Style Spatial Expansion Characteristics of Renewable Energy Production

Over time, renewable energy production in Ningxia exhibits spatial expansion characteristics characterized by dynamic agglomeration patterns.

(1) Coupling Analysis of Power Plant Layout and Renewable Energy Generation LISA.

Using the 91 Satellite Map, remote sensing images of Ningxia for three key years (2014, 2018, and 2021) were downloaded, with a spatial resolution of 3 m. The specific geographic locations of photovoltaic and wind power plants were identified and mapped, and their coverage areas were estimated. ArcGIS was employed to overlay the spatial layout of renewable energy power plants with the LISA maps of total renewable energy generation (Figure 7). This coupling analysis aimed to explore the spatiotemporal evolution patterns of renewable energy production in Ningxia.

Comparing the spatial locations and coverage area changes of photovoltaic and wind power plants across the three years reveals distinct expansion trends. From 2014 to 2018, the number and scale of photovoltaic power plants increased significantly, with large-scale expansions observed in Huinong, Xingqing, Lingwu, Zhongwei (excluding Haiyuan), and other areas. In contrast, wind power plants showed fewer additions but larger scales, primarily concentrated in Shapotou, Hongsipu, and Yanchi. From 2018 to 2021, the growth rate of photovoltaic plants slowed in Shizuishan, with sporadic large-scale additions in Dawukou and Pingluo. However, Lingwu, Zhongning, Haiyuan (adjacent to Shapotou), Hongsipu, and Tongxin continued to experience robust expansion, marked by both large-scale and numerous new plants. Smaller, scattered additions were observed in Xiji and Pengyang. In wind power development, Lingwu and Hongsipu saw substantial scale expansion of existing wind farms despite no significant increase in their number. Meanwhile, large-scale wind farms were newly added in Haiyuan and Tongxin. These spatial distribution patterns reflect varying spatiotemporal expansion strategies across regions: some prioritize stable deployment of large centralized plants, others focus on small distributed plants, and a few exhibit minimal growth.

By integrating temporal evolution and spatial distribution patterns, the coupling analysis of power plant layouts and LISA maps demonstrated a high overall alignment between renewable energy generation and production spaces, albeit with regional disparities. For example, areas with concentrated power plant distribution largely overlapped with high-value clusters. However, by 2021, single-type renewable energy plants (e.g., wind farms in Haiyuan and Tongxin) were insufficient to fully elevate regional production levels to high-value clusters. Similarly, dispersed small-scale photovoltaic plants in Huinong remained in low-value clusters. These findings highlight the need for diversified energy type deployment to enhance efficiency and regional balance. Additionally, Yanchi, with no significant plant expansion, shifted out of high-value clusters by 2021, while Litong transitioned from a high-value cluster to being surrounded by high-value areas due to stagnant expansion. These shifts may relate to indirect socioeconomic influences, which will be analyzed later. Overall, the coupling analysis validates the feasibility and rationality of the research methodology.

(2) Analysis Using the Standard Deviational Ellipse (SDE) Method

The spatial center of gravity for renewable energy generation in Ningxia shifted significantly, indicating coexisting agglomeration and diffusion trends alongside expanding geographic influence. Wind power dominated, but photovoltaic generation showed rapid growth, likely due to government support for solar projects. Since 2014, the SDE coverage of total renewable energy generation shifted from a “northeast–southwest” orientation, with the northeast boundary moving from Pingluo to Helan and the southwest boundary extending from Tongxin to Haiyuan. The western boundary expanded from Zhongning to Shapotou, while the eastern boundary gradually extended southward within Yanchi. By 2021, renewable energy production was concentrated in Yinchuan, Wuzhong, and Zhongwei (Figure 8). Wind power SDE coverage (Figure 9) expanded notably northward, with the northern boundary shifting from Pingluo to Yinchuan and the southern boundary extending from Hongsipu to Haiyuan. By 2021, wind power production was concentrated in Yinchuan, Wuzhong, and Zongwei. Photovoltaic SDE (Figure 10) exhibited a gentler “southwest–northeast” shift, with minimal area changes. By 2021, photovoltaic production was similarly concentrated in Yinchuan, Wuzhong, and Zhongwei. Thermal power SDE (Figure 11) stabilized after expanding southward in 2015, with production concentrated in Yinchuan and Wuzhong.

The center of gravity for renewable energy production SDE shifted within Litong. From 2014 to 2015, it moved 10.49 km southwest; from 2015 to 2017, 14.89 km southeast; and from 2018 to 2021, it shifted southward, then northward, and finally 6.61 km southwest. Wind power SDE trajectories mirrored total renewable energy trends: 18.13 km southwest from northern Litong (2014–2015), 28.24 km south to Hongsipu (2015–2017), and 7.66 km west after oscillations (2018–2021). Photovoltaic SDE diverged, moving from the Hongsipu–Litong border to the Qtongxia–Litong border: 16.35 km north (2014–2015), 16.51 km northeast (2015–2017), and 9.60 km southwest (2018–2021). In contrast, thermal power centers remained within Lingwu, likely linked to the development of the Ningdong Energy and Chemical Base.

3.3. Analysis of Influencing Factors on Renewable Energy Production

Using the geographical detector method for factor detection, the q-values of influencing factors on renewable energy production were ranked as follows: UL (urbanization level), UP (urban population), GDP (per capita GDP), SO₂ (industrial sulfur dioxide emissions), SH (annual sunshine hours), EI (energy intensity), WS (average wind speed), and ER (environmental regulation). Among these, EI, ER, and WS had p-values greater than 5%, indicating insignificant effects on renewable energy production (

Table 4. Factor detection results of influencing factors on renewable energy production.

Factor	GDP	UL	UP	EI	SO2	ER	SH	WS
q statistic	0.4548	0.5997	0.4946	0.2820	0.3010	0.1220	0.2999	0.2728
p value	0.0025	0.0000	0.0000	0.0524	0.0396	0.3926	0.0403	0.0599
q-value rank	3	1	2	6	4	8	5	7

). The results revealed that urbanization level had the strongest influence on renewable energy production. Higher urbanization rates increased energy demand, and urban migration and industrial agglomeration further amplified the need for renewable energy. Urban population ranked second, as population growth escalated energy consumption and environmental pollution, driving local renewable energy development. Per capita GDP, representing regional economic development, was closely tied to renewable energy production capacity. Industrial SO₂ emissions (q ≈ 0.3) also showed moderate influence, as controlling SO₂ emissions—a key pollutant from coal combustion—increased operational costs for coal-fired plants, incentivizing policymakers to promote renewable energy.

Energy intensity (EI) showed a marginally insignificant p-value (>5%), but its increase may still promote renewable energy production in regions with high EI, as these areas are more likely to transition from fossil fuels to renewables for emission reduction. Environmental regulation (ER) also showed insignificant results, contradicting expectations, likely due to inconsistent and insufficient industrial pollution control investments in the study area. Annual sunshine hours (SH) and average wind speed (WS), though critical for assessing solar and wind resources, yielded insignificant results due to minimal regional variations in these metrics.

To further validate the spatial agglomeration characteristics of renewable energy and their relationship with urbanization, population–economy scale, and environmental pollution, we constructed bar charts using growth rate data for urbanization level, urban population, per capita GDP, and industrial SO₂ emissions in 2014, 2018, and 2021. These charts were overlaid with LISA maps of renewable energy generation (Figure 12), visualizing trends and regional disparities in socioeconomic and environmental indicators.

High-value clusters consistently exhibited higher urbanization levels, urban populations, and per capita GDP compared to low-value clusters, aligning with geographical detector results. Zhongwei and Wuzhong demonstrate rapid socioeconomic growth: by 2021, Zhongwei achieved urbanization (7.5%), urban population (7.1%), and per capita GDP (35.9%), while Wuzhong reached 6.4%, 7.3%, and 44.9%, respectively. Both cities also reduced industrial SO₂ emissions by 70.3% and 68.6%, exceeding other regions. As industrial hubs, Zhongwei and Wuzhong prioritize industrial upgrades, resource-efficient enterprises, and technological innovation, fostering renewable energy growth.

Yinchuan, the administrative and economic center of Ningxia, has a large urban population and a tertiary-dominated economy, with emerging high-tech industries such as new materials and renewable energy processing. Renewable energy plants are densely clustered in the Ningdong Energy and Chemical Base, a key industrial base driving Yinchuan’s industrial optimization and economic transformation. Lingwu’s sustained high-value cluster status aligns with its industrial dynamism.

In contrast, although Shizuishan’s urbanization level is second only to Yinchuan, it has consistently remained in low-value clusters despite not having the lowest values in other provincial indicators. As shown in the spatial distribution maps of renewable energy power plants mentioned earlier, Shizuishan relied primarily on small-scale photovoltaic plants with almost no wind farms by 2021, reflecting its relatively low renewable energy resource abundance, which constrained overall renewable energy production. Additionally, as a resource-depleted city, Shizuishan’s fiscal revenue historically depended heavily on the coal industry. With coal resources exhausted, sluggish revenue growth has imposed financial pressure on investments in emerging industries like renewable energy. The decline of the coal sector has slowed population growth and exacerbated outmigration, while the development of renewable energy—a time- and process-intensive industry—struggles to meet short-term employment demands. The long-standing coal-dependent industrial structure has created significant inertial resistance to Shizuishan’s transition, as phasing out traditional industries and nurturing new ones require time and policy support, hindering rapid renewable energy expansion. Furthermore, the environmental degradation caused by coal mining necessitates substantial funding for restoration and governance, potentially diverting resources from renewable energy development and limiting its pace and scale.

Nonetheless, this crisis presents a transformative opportunity: resource depletion compels Shizuishan to seek new economic drivers. Renewable energy, as a clean and strategic emerging industry with broad prospects and market potential, has become a critical direction for Shizuishan’s transition.

Located in a mountainous region with relatively limited land resources, Guyuan faces constraints in the construction and development of renewable energy projects. The lower urbanization level and smaller urban population contribute to underdeveloped urban infrastructure and public services, resulting in limited support for renewable energy production and subdued market demand. These factors not only prolong the return on investment for renewable energy projects but also dampen investor enthusiasm. Additionally, Guyuan’s relatively singular industrial structure and lower economic development level have led to limited financial investment in renewable energy, causing slow progress in project development, insufficient technological innovation capacity, and hindered industrialization of renewable energy. Nonetheless, it is noteworthy that while Guyuan remains in low-value clusters, their spatial scope is gradually shrinking. Confronted with these challenges, Guyuan should actively seek breakthroughs by promoting the synergistic development of renewable energy with other industries. For instance, integrating distributed wind and solar projects with ecotourism, Beautiful Village initiatives, and other livelihood improvement programs could simultaneously drive economic growth and enhance social benefits, achieving a win-win scenario for both economic and socioeconomic outcomes.

4. Discussion

In July 2012, the National Energy Administration officially designated Ningxia as China’s first national comprehensive demonstration zone for new energy [67]. This milestone not only revitalized the development of Ningxia’s new energy industry but also marked a new phase of growth in renewable energy. The renewable energy non-hydro power consumption ratio (i.e., the proportion of renewable energy in total electricity consumption) reflects the development level of renewable energy and its contribution to regional green and low-carbon energy transitions. To further analyze the spatiotemporal characteristics and influencing factors of renewable energy production in Ningxia from 2014 to 2021, this study compared Ningxia with neighboring provinces (autonomous regions) in Northwest China—Qinghai, Gansu, Inner Mongolia, Shaanxi, and Xingjiang—as well as national clean energy demonstration zones using renewable energy non-hydro power consumption ratios from 2015 to 2021 [68] (2015 marks the first year of China’s annual renewable energy development monitoring and evaluation reports) (Figure 13).

(1) Comparison with the National Average

All provinces and the national average showed fluctuating but rising trends in renewable energy non-hydro power consumption ratios from 2015 to 2021. The national ratio increased from 5% in 2015 to 13.7% in 2021, indicating significant progress in renewable energy utilization and grid integration. Notably, Ningxia consistently ranked among the top regions in China from 2015 onwards, maintaining a ratio 8–13 percentage points above the national average. This underscores Ningxia’s leading role in advancing renewable energy development nationwide.

(2) Analysis of Renewable Energy Non-Hydro Power Consumption Ratios

As national energy demonstration zones, Ningxia, Qinghai, and Gansu have achieved remarkable success in renewable energy utilization. Ningxia’s high consumption ratio stems from rapid growth in wind and solar power, robust grid infrastructure, efficient dispatch management, and strong policy support. This progress reflects a structural shift in energy consumption from fossil fuels to renewables, reducing carbon emissions and environmental pollution. In recent years, Ningxia’s proactive energy transition and green development strategies have further elevated renewable energy’s share in its energy mix.

Qinghai Province benefits from abundant wind and solar resources but faces challenges in balancing energy consumption growth with structural adjustments. Gansu, despite achievements, struggles with renewable energy integration due to complex energy consumption patterns and geographical constraints. Inner Mongolia, Shaanxi, and Xingjiang have shown gradual improvements in consumption ratios but lag in stability and growth rates. To enhance renewable energy development, these regions should strengthen grid interconnections, increase project investments, and optimize energy structures.

(3) Regional Disparities and Ningxia’s Advantages

① Resource Advantages: Located in Northwest China, Ningxia boasts abundant solar and wind resources. Leveraging deserts, Gobi regions, and wastelands, it has established large-scale renewable energy bases, enhancing local consumption and cross-regional transmission. The utilization of coal mining subsidence areas, mountainous terrain, and barren lands further supports renewable energy development. ② Policy Support: Ningxia’s government prioritizes renewable energy through fiscal subsidies, tax incentives, and land allocation, fostering a favorable environment for project implementation. ③ Technological Innovation: Ningxia adopts cutting-edge solar PV and wind power technologies to improve generation efficiency and grid integration. ④ Industrial Chain Integration: A well-established industrial chain—encompassing equipment manufacturing, R&D, project construction, and operations—reduces costs and enhances market competitiveness. ⑤ Market Demand: Rising national demands for sustainability and energy transition position Ningxia as a key renewable energy producer with significant market potential. As a national demonstration zone, Ningxia’s progress in renewable energy integration sets a benchmark for China’s broader energy transformation.

5. Conclusions

To address the long-standing issue of spatial mismatch between renewable resource-rich regions and electricity load centers in China, this study focused on underdeveloped areas with abundant desert, Gobi, and marginal lands, using Ningxia’s renewable energy production as a case to quantify its spatiotemporal evolution patterns and explore underlying influencing factors. Based on empirical results and theoretical analysis, the following conclusions are drawn:

(1) In terms of the spatiotemporal evolution characteristics of renewable energy production, the total renewable energy production and spatial Gini coefficients in Ningxia exhibit a composite evolutionary pattern of continuous growth and fluctuations. Renewable energy production demonstrates dual characteristics of spatial agglomeration and diffusion, alongside spatial autocorrelation. According to spatial Gini coefficient analysis, renewable energy production shows significant spatial agglomeration, with wind power exhibiting a stronger and rising agglomeration trend compared to photovoltaic power. According to Moran’s I index results, globally, renewable energy production displays stable spatial autocorrelation, while locally, production clusters into H-H (high–high) and L-L (low–low) agglomerations, indicating positive spatial autocorrelation. Neighboring regions with similar development levels form hierarchical spatial structures. Few regions fall into L-H (low–high) clusters, and no H-L (high–low) clusters are observed. Overall, spatial agglomeration and diffusion coexist, with H-H and L-L clusters dominating and gradually expanding, while some regions (e.g., Guyuan) lag in renewable energy development.

(2) Over time, the renewable energy production in Ningxia has exhibited spatial expansion characteristics driven by dynamic agglomeration patterns. Coupling power plant spatial distribution with LISA maps of generation volumes demonstrates a high overall spatial coupling between renewable energy generation and production spaces, albeit with regional disparities. Reliance on single-type renewable energy plants (e.g., wind or solar) is insufficient to comprehensively enhance total production. According to standard deviational ellipse (SDE) results, wind and photovoltaic power generation centers are concentrated in Yinchuan–Wuzhong–Zhongwei, while total renewable energy production shows decentralized development, expanding its geographic influence compared to thermal power.

(3) The factor detection results of the geographical detector indicate that different influencing factors have varying degrees of impact on renewable energy production. Urbanization level, urban population, per capita GDP, industrial SO₂ emissions, and annual sunshine hours positively influence renewable energy production, while energy intensity, environmental regulation, and average wind speed show insignificant effects. Socioeconomic factors—such as population, economic development level, urbanization level, and technological advancement—may reshape power generation structures and spatiotemporal distribution patterns by altering traditional thermal power generation through mechanisms like renewable energy demand, development space, and policy interventions. Although natural factors like annual sunshine hours and wind speed directly affect renewable energy generation, their impact may diminish in significance due to the scale of the study area and the increasing maturity of renewable energy technologies.

Based on these findings, this paper proposes the following policy recommendations: ① Tailored Regional Planning: Considering resource and environmental constraints, cities and counties should develop renewable energy plans tailored to local conditions. Strengthening county-level power infrastructure will enhance renewable energy dispatch and transmission capabilities while amplifying the radiation-driven effects of the Yinchuan–Wuzhong–Zhongwei cities on neighboring regions. ② Industrial Transformation in Shizuishan: Shizuishan should leverage its transition phase to upgrade traditional industries, optimize industrial structures, and promote high-end, intelligent, and green development through renewable energy initiatives. ③ Distributed Photovoltaic Development in Guyuan: Guyuan should prioritize distributed and rooftop photovoltaic installations based on resource endowments to advance localized power generation and regional energy transition. ④ Optimized Power Facility Layout: Governments should align infrastructure planning with spatial agglomeration trends and shifts in generation centers (e.g., renewable vs. thermal power), ensuring coordinated deployment of diverse renewable energy plants. ⑤ Technological and Economic Investment: Despite abundant renewable resources in western regions, technological lag [56] necessitates increased enterprise investments in innovation to overcome bottlenecks and accelerate low-carbon transitions.

Against the backdrop of accelerating energy transition, future research should continuously monitor dynamics across all stages of renewable energy production and investigate the driving mechanisms behind spatiotemporal development patterns. While this study focused on the power production side of renewable energy in a single province, the transmission and consumption phases are equally critical. To avoid extreme spatial distribution of renewable energy, subsequent studies should holistically evaluate trade-offs between production costs and environmental impacts, enhance urban energy sustainability, improve energy demand elasticity, and develop integrated urban renewable energy systems [69].