Research on Different Energy Transition Pathway Analysis and Low-Carbon Electricity Development: A Case Study of an Energy System in Inner Mongolia

Abstract

To achieve carbon neutrality targets in the power sector, regions with rich coal and renewable energy resources are facing unprecedented pressure. This paper explores the decarbonization pathway in the power sector in Inner Mongolia, China, under different energy transition scenarios based on the Long-Range Energy Alternatives Planning System (LEAP) model. This includes renewable energy expansion, carbon capture and storage (CCS) applications, demand response, and economic regulation scenarios. Subsequently, a combination of the Logarithmic Mean Divisia Index (LMDI) and Slack-Based Measure Data Envelopment Analysis (SBM-DEA) model was developed to investigate the influencing factors and power generation efficiency in low-carbon electricity. The results revealed that this region emphasizes first developing renewable energy and improving the carbon and green electricity market and then accelerating CCS technology. Its carbon emissions are among the lowest, at about 77.29 million tons, but the cost could reach CNY 229.8 billion in 2060. We also found that the influencing factors of carbon productivity, low-carbon electricity structures, and carbon emissions significantly affected low-carbon electricity generation; their cumulative contribution rate is 367–588%, 155–399%, and −189–−737%, respectively. Regarding low-carbon electricity efficiency, the demand response scenario is the lowest at about 0.71; other scenarios show similar efficiency values. This value could be improved by optimizing the energy consumption structure and the installed capacity configuration.

1. Introduction

As global climate change worsens, achieving carbon neutrality is becoming increasingly important for all countries to implement green development. At present, 135 countries have pledged to achieve carbon-neutral goals by the mid-century, with 125 countries also aiming to achieve net-zero emissions goals before 2070 [1]. As the world’s largest carbon emitter, China has made a commitment to achieve a carbon emissions peak by 2030 and carbon neutrality by 2060 at the 75th UN General Assembly [2]. About 43% of greenhouse gas emissions in China come from coal-fired power plants (CFPPs), and electricity from CFPPs accounts for more than 60% of the total electricity supply [3]. Therefore, it is urgent to accelerate the green transformation of energy structures, combining with national policies and local government requirements to further increase the share of clean energy power generation.

China is the largest carbon emitter in the world and urgently requires a low-carbon electricity transition to improve energy conservation management and reduce carbon emissions [4]. To ensure a stable electricity supply, low-carbon electricity systems are needed to respond to the real-time changes in future power systems. The development of renewable energy technology, the construction of a low-carbon electricity market, and carbon reduction policy support are the main elements to drive the low-carbon electricity transition [5]. Owing to the ongoing improvements in wind power, solar power, energy storage, and other green technologies, gradually promoting CCS retrofitting technology and constantly enhancing the carbon market and green electricity certification, the development of low-carbon electricity is expected to accelerate further. The Inner Mongolia region is an important energy base in China. It not only has rich natural resources and fossil fuels but also undertakes the important task of power transmission and resource extraction. This region has implemented some low-carbon energy strategies to accelerate social modernization, promote energy transition, and achieve environmentally sustainable development. Inner Mongolia has given priority to expanding alternative energy sources such as wind energy, solar energy, and different energy storage types, continuously boosting the percentage of non-fossil fuel energy in end-use energy [6,7].

At present, many scholars have conducted an abundance of studies on the development of energy demand and power structure forecasting models. The energy and carbon emission forecasting models are divided into top-down and bottom-up models. The top-down frameworks rely on economic indicators, such as energy prices and economic elasticity, and show the overall relationship between energy production and consumption, making them suitable for developing energy strategies and assessing macroeconomic policies. The bottom-up model focuses on technical details, solves the lowest cost energy balance solution under the premise of energy system constraints, and has the advantages of high flexibility and high accuracy [8]. The LEAP model is a typical representative of the bottom-up approach, which is conducive to forecasting medium and long-term energy demand, energy planning, and CO₂ emissions of the energy system. The primary reason is that it reduces the difficulty of the model and flexibly constructs the model structure [9]. Li et al. constructed the LEAP-CHINA model to develop different scenarios to forecast China’s total energy demand, end-use industry subsectors, end-use energy subvarieties, and processing and conversion sectors [10]. Bhuvanesh et al. developed a low-carbon power generation plan in 2030 for the electricity expansion planning in India using the LEAP-EnergyPLAN model [11]. Zhang et al. used the LEAP model to build three scenarios to investigate the peak of CO₂ emissions and reduction pathways in five northwestern provinces in China [12]. Romain et al. developed an energy model to consider important factors that influence future energy plans and alternative pathway development in the Benin Republic [13]. El-Sayed et al. analyzed different scenarios of electricity generation and energy consumption between 2020 and 2050 using LEAP in Egypt [14]. Wang et al. used the LMDI model to explore the total energy consumption growth of three industries in Hunan, while the LEAP system was used to analyze the profound impacts of the different effects on the total energy consumption [15].

The decomposition analysis technique is a methodology to determine which factor affects carbon emissions and energy consumption. It includes structural decomposition analysis, index decomposition analysis, production theory decomposition analysis, and so on. From the perspective of the energy influencing factors, the LMDI model has become the most widely used decomposition method owing to its remarkable flexibility and ability to thoroughly analyze influencing factors at both national and regional levels. Ma et al. developed the LMDI method based on energy and CO₂ allocation Sankey diagrams to analyze the contributions of influencing factors to the growth of energy-related CO₂ emissions [16]. Mousavi et al. analyzed the influencing factors of carbon emissions on energy consumption in the industrial sector, drivers of carbon intensity in electricity generation, and key drivers of carbon emissions due to total fossil fuel combustion in Iran [17]. Paulo M et al. presented a temporal IDA-LMDI formulation to expressly include the effect of capacity factors and analyze the evolution of Latin America and the Caribbean’s aggregate carbon emission intensity of electricity [18]. Wang utilized the new generalized LMDI method to analyze the driving factors dominating China’s energy consumption over the period 1991–2011 [19].

Energy efficiency analysis methods are an important research field for scholars, so the energy efficiency measurement results are vital for evaluating energy allocation and growth. The efficiency evaluation methodologies generally comprise parametric and non-parametric approaches. Data envelopment analysis (DEA) is a typical non-parametric method based on linear programming, which can effectively tackle the typical problems of efficiency assessment, referring to multiple inputs and outputs. The DEA method is widely used in energy efficiency evaluation. Tone introduced the SBM-DEA model in 2001, which deals with the variation in input slack and output slack, and integrates the undesired output in the efficiency framework [20]. Some domestic and international scholars discussed energy efficiency and industrial sector efficiency in different regions of the world. Fong et al. explored the relationship between CO₂ emission and energy consumption on a sustainable development system for the 11 cities of the Guangdong–Hong Kong–Macao Greater Bay Area during 2010–2016 using the SBM-DEA model [21]. Shu et al. used the super SBM-DEA methodology to assess the EE of 168 economies globally from 2000 to 2017 [22]. Gökgöz et al. used the super SBM-DEA model to find and compare the efficiency scores of European countries and regions for the period of 2011–2015 [23]. Choi studied the steel production efficiency of Korean companies using the SBM-DEA model [24]. Huang et al. constructed a three-stage SBM-DEA model to evaluate the total factor energy efficiency in China during 2000–2012 from management and environmental perspectives [25].

Many researchers have used various models to assess sustainable energy transition pathways. The importance of regional energy transition planning is also becoming more prominent. The commonly used model for regional energy transition planning is the energy optimization model. Nyangchak analyzed renewable energy efficiency and influence factors in Qinghai Province using a combination of the LMDI model, the SBM-DEA model, and a field research approach [26]. Gang et al. conducted an in-depth analysis of energy-related carbon emissions of the different industries using the Kaya equation and carbon emissions among industries using the LMDI model in Shandong Province [27]. Xiao et al. explored strategies for advancing wind–solar–storage systems to help Jiangxi transition to a low-carbon energy structure using the LEAP model and different energy scenarios [9]. Li et al. analyzed regional carbon-emission efficiency differences in China using the LMDI and other methods, finding that energy efficiency, technological progress, and energy scale are key factors [28]. Cong et al. designed an intelligent framework of an energy system in which supplies from renewable energy, grid power, and the storage battery are optimized based on their lifecycle CO₂ emissions [29]. A comprehensive review of the current literature highlighted some notable research gaps: (1) To date, a considerable number of studies have focused on forecasting models and carbon reduction analysis for a national-level energy transition, but there is still a lack of research on regional energy transition pathway comparisons and low-carbon power systems for coal-dominant and renewable energy resource-rich regions. (2) While a significant amount of research has compared different energy transition pathways to select the best solution to achieve the “dual carbon” goal, there have been few discussions on comparative scenarios considering future demand response, tradable carbon emission (TCE) + green electricity certification (TEC), and low-carbon electricity technology improvement. (3) There are a few studies that systematically evaluate the influencing factors and power efficiency of low-carbon power across different energy transition scenarios, which is essential in constructing the future low-carbon electricity system and effectively reducing carbon emissions in energy systems.

The main purpose of this research was to explore the effectiveness of different energy transition scenarios for achieving carbon reduction in the power sector. This paper takes Inner Mongolia as an example to construct a multi-scenario energy transition using the LEAP model to analyze the carbon emission reduction potential, electricity cost change, power generation structure, and installed capacity amount. In addition, we discuss the influencing factors and power efficiency of low-carbon electricity across different energy transformation scenarios. This model offers theoretical guidance and policy suggestions for other resource-dependent regions across China for energy system decarbonization and the low-carbon electricity construction. The main contributions of this study are as follows: (1) This study is the first to integrate the deployment of renewable energy and storage systems, the acceleration of CCS technology, future demand response, and the TCE and TGC markets under the same framework; four energy transition scenarios were systematically compared. (2) The research methodology combines the LEAP, LMDI, and SBM-DEA models to establish a comprehensive “scenario analysis—influencing factors identification—efficiency evaluation” low-carbon electricity assessment system. (3) We select Inner Mongolia as the study area to provide a valuable reference for energy transition pathway analysis and low-carbon electricity planning in other regions with similar energy structures.

The paper is organized as follows: Section 1 summarizes the current research and literature review of energy transition pathway analysis. The methodology and model used in this research are introduced in Section 2. Section 3 describes the study areas, scenario setting, and data source. Section 4 shows the results and detailed discussion of the energy transition pathway and low-carbon electricity. The conclusion and suggestions are presented in Section 5.

2. Methodology

2.1. Analysis Framework

To evaluate and predict the evolution of the energy structure, carbon emissions, and cost changes for different transition pathways from 2030 to 2060 under the background of carbon neutrality, we constructed a comprehensive model to explore the development of low-carbon power in the energy system, represented in Figure 1. Firstly, this study predicted Inner Mongolia’s energy structure under different energy transition scenarios from 2030 to 2060 using the LEAP model, based on macro-parameters such as the terminal industry activity level, energy intensity, and overall planning. In addition, this paper integrated the LMDI model and SBM-DEA model to analyze the influencing factors and efficiency performance of low-carbon electricity in great depth. This enabled us to predict the carbon emission trends, construct a low-carbon electricity system, and formulate carbon reduction policies for different energy transitions. The low-carbon electricity referred to in this paper mainly includes wind power, solar power, energy storage, and coal-fired power with/without CCS.

2.2. LEAP Model

This study used the LEAP model developed by the Stockholm Environment Institute and Boston University; this model is closely integrated with various scenario analyses. It comprehensively considers population growth, industrial structure changes, low-carbon technology promotion, economic development, policy implementation, and other factors, which can predict energy supply, power structure, and pollutant emissions.

The LEAP-Inner Mongolia model framework proposed in this paper is based on the actual situation and the background of carbon neutrality. The model mainly includes the various analysis elements, such as key parameter setting, different energy transition assumptions, end-energy demand analysis, CO₂ emission environmental impact analysis, and electricity cost analysis, and it aims to provide precise analyses of the future different energy transition pathways in Inner Mongolia. It should be noted that the power generation sector mainly involves coal-fired power, wind energy, solar energy, and energy storage. The energy supply module refers to primary energy and secondary energy. The energy demand is divided into four sections: primary industry, secondary industry, tertiary industry, and residential life.

(a)

Energy demand

In this study, the energy demand model focused on four major sectors in Inner Mongolia: primary industry, secondary industry, tertiary industry, and the household sector. The total energy demand is calculated by Equation (1).

$$E{D}_{j,t}=\sum _i A{L}_{i,t}\times E{I}_{i,j,t}$$

(1)

where ED_j.t is the end-use energy demand for j-th energy in t-th year; AL_i,t is the activity level of the i-th sector in t-th year; and EI_i,j,t is the energy intensity associated with the j-th energy of the i-th sector in t-th year.

The activity level involves the local economic development, and the growth rate method was selected to assess the economic activity level, as shown in Equation (2).

$$F{A}_{i.t}=E{C}_i\times {\left(1+A{G}_i\right)}^{t-1}$$

(2)

where FA_i,t is the economic activity level of the i-th sector of t-th year; EC_i is the economic activity level of the i-th sector of the reference year; and AG_i is the annual growth ratio of the economic activity level for the i-th sector.

The processing conversion consumption refers to the energy loss during the conversion process and is calculated as the difference between the total energy input and energy output over a specified period. It is expressed as Equation (3)

$$E{S}_{j,t}={\displaystyle \frac{{ED}_{j,t}+E{L}_{j,t}}{{\eta }_{j,t}}}\times (1-{\eta }_{j,t})$$

(3)

where ES_j,t represents the total amount of energy consumption in conversion and processing of the j-th energy type in t-th year; EL_j,t is the j-th energy loss in t-th year; and η_j,t is the conversion efficiency of the j-th energy in t-th year.

The total energy demand includes the end-use sector energy demand, conversion consumption, and energy losses, as shown in Equation (4).

$$T{E}_{j,t}={ED}_{j.t}+E{S}_{j.t}+E{L}_{j.t}$$

(4)

where TE_j,t represents the total demand for j-th energy in t-th year.

(b)

Carbon emissions from energy

The total carbon emissions from energy system are estimated by Equation (5).

$$CE=T{E}_{j,t}\times \mathrm{E}{\mathrm{F}}_{j,t}$$

(5)

where CE is the total carbon emissions, and EF_j,t is the carbon emissions factors from the j sector in t-th year.

(c)

Cost analysis

For a better understanding of the changes in electricity system costs under different energy transition scenarios, according to the optimization LEAP model tool, this model aims to evaluate the energy system configuration with the lowest total net present value of system social cost during the entire calculation period by Equation (6).

$$\begin{array}{cc}\hfill C& ={\displaystyle \sum _{(\forall y\in Year,\forall t\in Technology)}} (P{G}_{t,y}\times VO{M}_{t,y}+I{C}_{t,y}\times C{I}_{t,y}+\hfill \\ & I{C}_{t,y}\times FO{M}_{t,y}+C{E}_{t,y}\times C{P}_y+G{E}_{t,y}\times G{P}_{t,y})\times (1+r{)}^{-(y-2020)}\hfill \end{array}$$

(6)

where C is the energy system cost of electricity unit converted into the base year during the entire research period; PG_t,y is the power generation by t-th technology in y-th year; VOM_t,y is the variable operation and maintenance costs by t-th technology in y-th year; IC_t,y is the total installed capacity by t-th technology in y-th year; CI_t,y is the capital investment cost by t-th technology in y-th year; FOM_t,y is the fixed operation and maintenance costs by t-th technology in y-th year; CE_t,y is the carbon emission in TCE market by t-th technology in y-th year; CP_t,y is the carbon price by t-th technology in y-th year; GE_t,y is the green electricity certification in TGC market by t-th technology in y-th year; GP_t,y is the green electricity certification price in y-th year; r is the discount rate.

2.3. LMDI Model

The LMDI model is an effective mathematical method for the factor decomposition analysis of energy-related carbon emissions. This method takes the logarithmic average weight of the overall change and accurately distributes the logarithmic increase in every driving factor to ensure that the sum of all factors’ contributions equals the overall change and the residual equals zero. Because the logarithmic average is path-independent, the decomposition results are still additive and consistent regardless of how the time interval is split or accumulated, and the cumulative error is not generated. At present, the LMDI method is widely used to analyze the impact of changes in carbon emissions and energy consumption.

Based on previous studies using the LMDI approach, this paper incorporated the Kaya identity into the LMDI decomposition framework. Because the traditional Kaya decomposition directly calculates the elasticity coefficient via a multiplication method and is susceptible to the generation of non-additive residuals, the various terms of the Kaya equation are used as driving factors in the LMDI. The logarithmic average weighted formula is used instead of the simple difference or elasticity coefficient method to calculate contributions, thereby effectively eliminating cross-term residuals. An in-depth analysis of the influencing factors in low-carbon electricity was then conducted for different scenarios, which are decomposed into six components: the low-carbon electricity structure, electricity intensity, energy security, energy intensity, carbon productivity, and carbon emissions. The influencing factors contributing to low-carbon electricity are systematically calculated by the Kaya identity as shown in Equation (7).

$$\begin{array}{c}{LE}_t=\left({\displaystyle \frac{{LE}_t}{{EG}_t}}\right)\times \left({\displaystyle \frac{{EG}_t}{{EP}_t}}\right)\times \left({\displaystyle \frac{{EP}_t}{{EC}_t}}\right)\times \left({\displaystyle \frac{{EC}_t}{{EI}_t}}\right)\times \left({\displaystyle \frac{{EI}_t}{C_t}}\right)\times C_t\\ =LG\times GP\times PC\times CI\times IC\times CC\end{array}$$

(7)

where LE_t is the low-carbon electricity amount in t-the year; EG_t is total electricity generation in t-the year; EP_t is the total energy production in t-the year; EC_t is total energy consumption in t-the year; EI_t is total energy investment cost in t-the year; and C_t is total carbon emissions in t-the year.

The abbreviations of LG, GP, PC, CI, IC, and CC correspond to low-carbon electricity structure, electricity intensity, energy security, energy intensity, carbon productivity, and carbon emissions, respectively. By applying the additive LMDI decomposition model, the variation in low-carbon development between the base year and target year is shown in Equation (8).

$$\mathsf{\Delta }LE=L{E}_t-L{E}_0=\mathsf{\Delta }{LE}_{LG}+\mathsf{\Delta }{LE}_{GP}+\mathsf{\Delta }{LE}_{PC}+\mathsf{\Delta }{LE}_{CI}+\mathsf{\Delta }{LE}_{IC}+\mathsf{\Delta }{LE}_{CC}$$

(8)

where ΔLE is the change in the amount of low-carbon electricity. ΔLE_LG, ΔLE_GP, ΔLE_PC, ΔLE_CI, ΔLE_IC, and ΔLE_CC represent the change in low-carbon electricity structure, electricity intensity, energy security, energy intensity, carbon productivity, and carbon emissions, respectively. The components of the additive decomposition of LMDI can be expressed by Equations (9)–(14).

$$\mathsf{\Delta }{LE}_{LG}=\sum {\displaystyle \frac{(L{E}_t-L{E}_0)}{(\mathrm{l}\mathrm{n}L{E}_t-\mathrm{l}\mathrm{n}L{E}_0)}}\times \mathrm{l}\mathrm{n}\left({\displaystyle \frac{{\mathrm{L}\mathrm{G}}_t}{{LG}_0}}\right)$$

(9)

$$\mathsf{\Delta }{LE}_{GP}=\sum {\displaystyle \frac{(L{E}_t-L{E}_0)}{(\mathrm{l}\mathrm{n}L{E}_t-\mathrm{l}\mathrm{n}L{E}_0)}}\times \mathrm{l}\mathrm{n}\left({\displaystyle \frac{{GP}_t}{{GP}_0}}\right)$$

(10)

$$\mathsf{\Delta }{LE}_{PC}=\sum {\displaystyle \frac{(L{E}_t-L{E}_0)}{(\mathrm{l}\mathrm{n}L{E}_t-\mathrm{l}\mathrm{n}L{E}_0)}}\times \mathrm{l}\mathrm{n}\left({\displaystyle \frac{{PC}_t}{{PC}_0}}\right)$$

(11)

$$\mathsf{\Delta }{LE}_{CI}=\sum {\displaystyle \frac{(L{E}_t-L{E}_0)}{(\mathrm{l}\mathrm{n}L{E}_t-\mathrm{l}\mathrm{n}L{E}_0)}}\times \mathrm{l}\mathrm{n}\left({\displaystyle \frac{{CI}_t}{{CI}_0}}\right)$$

(12)

$$\mathsf{\Delta }{LE}_{IC}=\sum {\displaystyle \frac{(L{E}_t-L{E}_0)}{(\mathrm{l}\mathrm{n}R{E}_t-\mathrm{l}\mathrm{n}R{E}_0)}}\times \mathrm{l}\mathrm{n}\left({\displaystyle \frac{{IC}_t}{{IC}_0}}\right)$$

(13)

$$\mathsf{\Delta }{LE}_{CC}=\sum {\displaystyle \frac{(L{E}_t-L{E}_0)}{(\mathrm{l}\mathrm{n}L{E}_t-\mathrm{l}\mathrm{n}L{E}_0)}}\times \mathrm{l}\mathrm{n}\left({\displaystyle \frac{{CC}_t}{{CC}_0}}\right)$$

(14)

2.4. SBM-DEA Model

The traditional DEA models primarily use radial or angular measurements, and radial models ignore slack variables. Meanwhile, angular models only consider efficiency for one orientation angle of input or output. Consequently, the measurement of the radial or angle DEA is inaccurate. According to these principles, a non-radial and non-angled SBM-DEA model based on relaxation variable measures is proposed. The SBM-DEA model can quickly incorporate relaxation variables into the model’s objective, which can effectively consider environmental factors in the measurement of the SBM-DEA model. Moreover, the SBM-DEA model avoids the effects of radial and angular differences. There are two important characteristics of the SBM-DEA model: (1) The results of efficiency measures are not affected by decision-making units (DMUs) in measuring input and output terms. (2) The difference between the efficiency value and each input–output is monotonically decreasing.

In the process of low-carbon electricity efficiency evaluation, this paper adopts the SBM-DEA model, primarily because it can solve complicated scenarios, and inputs and outputs are not required to change proportionally or in the same direction. In addition, the SBM-DEA model has the advantages of a non-radial and non-oriented approach, accurately assessing input and output inefficiencies independently and comparing performance in low-carbon electricity under different energy transition scenarios, to offer reasonable improvements for low-efficiency units. The initial efficiency value, input slack variable, desired output slack variable, and undesired output slack variable will be calculated by the SBM-DEA model.

The SBM-DEA model is shown as follows:

$$\begin{array}{cc}\hfill \underset{\rho ,\lambda ,s_x^{-},s_y^{+},s_b^{-}}{min}& \rho ={\displaystyle \frac{1-\frac{1}{m}\sum _{i=1}^m \frac{s_{x_i}^{-}}{x_{ik}}}{1+\frac{1}{s+s^{\prime }}\left(\sum _{r=1}^s \frac{s_{yr}^{+}}{y_{rk}}+\sum _{t=1}^{s^{\prime }} \frac{s_{b_t}^{-}}{b_{ik}}\right)}}\hfill \\ \hfill s.t.& {\displaystyle \sum _{j=1}^n} {\lambda }_j{x}_{ij}+s_{x_i}^{-}=x_{ik},i=1,\dots ,m\hfill \\ & {\displaystyle \sum _{j=1}^n} {\lambda }_j{y}_{rj}-s_{y_r}^{+}=y_{rk},r=1,\dots ,s\hfill \\ & {\displaystyle \sum _{j=1}^n} {\lambda }_j{b}_{tj}+s_{b_t}^{-}=b_{tk},t=1,\dots ,s^{\prime }\hfill \\ & {\lambda }_j\ge 0,j=1,\dots ,n\hfill \\ & s_{x_i}^{-}\ge 0,s_{y_r}^{+}\ge 0,s_{b_t}^{-}\ge 0\hfill \end{array}$$

(15)

where ρ is the low-carbon electricity efficiency value, and the range of the value is 0–1; j is the number of DMUs; n is the number of decision-making units; m is the number of input indicators; S and S′ represent the number of indicators of the desired output and undesired output; $s_{xi}^{-}$ is the input slack variable; $s_{yr}^{+}$ is the slack variable of the desired output; $s_{b_t}^{-}$ is the slack variable of the undesired output; λj is the intensity variable; x_ij, y_rj, and b_tj are the m-dimensional input vectors; x_ik, is the input variable; y_rk is the desired output variable; and b_tk is the undesired output variable. If the value of ρ = 1, the decision unit is valid, and the corresponding $s_{xi}^{-}$, $s_{yr}^{+}$, and $s_{b_t}^{-}$ are all 0. If the value of ρ is <1, it also shows that the decision unit is invalid.

In this paper, we evaluate the efficiency of low-carbon electricity by taking the energy consumption amount, power system investment, and installed capacity as input indicators, the low-carbon electricity generation amount as the desirable output, and the CO₂ emissions as the undesirable output, as shown in

Table 1. The variable description of SBM-DEA.

Items	Variable Type	Variable Name	Unit
Input variables		Energy Consumption	Million T
		Total cost	Billion CNY
		Installed capacity	Thousand MW
Output variables	Desired output	Low-carbon electricity	Billion kWh
	Undesired output	CO2 emission	Million T

.

3. Study Area and Scenario Setting

3.1. Study Area Overview

Inner Mongolia spans the three major regions of Northeast China, North China, and Northwest China and is a crucial energy base in China due to its abundant energy resources. Figure 2 presents the location of Inner Mongolia in China. Inner Mongolia’s energy output has shown an increasing trend each year; coal production accounts for more than 90% of the total energy production. Moreover, there is significant potential for renewable energy development, such as wind energy, solar energy, and bioenergy. By 2024, Inner Mongolia’s total electricity consumption reached 402 billion kWh, and renewable energy power generation accounted for about 23.5%. In addition, electricity transmission reached 315 billion kWh, accounting for 16% of the country’s total electricity transmission. The carbon emission amount of Inner Mongolia’s power sector is about 330 million tons, accounting for about 52.4% of the regional total carbon emission, which is significantly higher than the national average [30,31].

However, Inner Mongolia belongs to the inland undeveloped areas in China, and the carbon emission problem is more prominent than that of coastal developed areas. In response to the “dual carbon goal”, Inner Mongolia plans to achieve the carbon peak goal in 2030 and carbon neutrality in 2060. Therefore, power systems will gradually phase out CFPPs, promote CCS technology, improve renewable energy construction, and update energy storage facilities. Based on energy planning and current limitations, this study proposes different energy transition scenarios.

3.2. Scenario Setting

When discussing energy transition pathways, we need to fully consider the impacts of different energy pathways on carbon reduction. This paper constructs four energy transition scenarios under the context of the “dual carbon” goal, including policy adjustment scenarios, low-carbon technology development scenarios, and power conservation scenarios. Scenario analysis enables us to clearly understand the change in carbon emissions and low-carbon electricity from the energy system in Inner Mongolia across different energy transition pathways.

3.2.1. Developing Renewable Energy and Storage System Scenario

To address the issue of climate change problems, many nations have actively promoted the utilization of renewable energy and advocated for the withdrawal of coal-fired power. To achieve China’s dual carbon goals as soon as possible, it is necessary to promote a major transformation in the energy structure, accelerate the development of clean energy, and gradually phase out CFPPs. Inner Mongolia has an abundance of renewable energy resources, especially wind energy and solar energy. With priority given to the development of wind energy and solar energy, and the support of national energy policies, the scale of the wind power and photovoltaic industries has been steadily increasing.

In addition, energy storage systems can efficiently store excess renewable energy from power generation, thereby reducing dependence on coal-fired power. We consider energy storage as comprising long- and short-term storage. Long-term storage utilizes the hydrogen energy storage system, and short-term storage employs the Li battery energy storage system. The ratio of long-term and short-term energy storage is based on the results of existing research [30,32].

This paper refers to the “Carbon Neutrality Research Report of China before 2060” [33], Inner Mongolia 14th Five-Year Development Plan [34], and the research on the gradual withdrawal of coal power [35,36] to set up the scenarios for accelerating the development of renewable energy.

3.2.2. Developing CCS Technology Scenarios

Owing to the uncertainty and intermittency of renewable energy, retaining a portion of coal-fired power could mitigate the power grid deficits, and CCS technology could decrease CO₂ emissions from CFPPs. Several scholars have anticipated that fossil fuels will continue serving as the primary source of energy consumed in the long-term period. At the same time, the United States enacted Act 45Q to enhance carbon capture, utilization, and storage (CCUS) projects in 2018; this offers a great deal of financial support in the development of CCUS retrofitting technology [37]. The EU established the “CCS Directive” to develop a legal framework for the safe geological storage of CO₂ in 2009 [38].

The development of CCUS projects in the United States and other regions provides valuable reference for Inner Mongolia’s future energy transition. However, the development of CCS technology in Inner Mongolia is still in the pilot demonstration stage. In 2011, Inner Mongolia launched its inaugural CCS project, which effectively captures CO₂ produced during the coal-to-oil process and stores it underground, and it implemented a CCUS demonstration project in the field of iron and steel in 2022. In addition, the multinational energy company, China Datang Group, established the world’s first 100 MW-scale CCS demonstration project on CFPPs in the Inner Mongolia region. Furthermore, the Inner Mongolia 14th Five-Year Plan proposed the facilitation of CCUS projects in the coal-fired power sector [34]. The advancement of the demonstration project and environmental regulation confirms the viability of the implementation of CCS technology in Inner Mongolia’s energy system.

This paper sets up the CCS technology development scenarios with guidance from regional reports and current research, such as the Global CCS Report 2023 [39], China Energy Transition Outlook 2060 [40], and current research about CCS and the energy supply–demand pathway [7,41,42].

3.2.3. Developing Demand Response Scenarios

Demand response can drive the collaborative management of electricity load and power supply by altering consumer electricity consumption behaviors, balancing the electricity supply–demand relationship, and incentivizing consumers to reduce electricity use during periods of low power generation and increase it during peak times. Furthermore, demand response can guide users to modify their electricity consumption behaviors based on price-based or incentive-based signals, resulting in the total installed capacity being reduced, which could adjust the balance in the supply–demand relationship. When peaks or valleys occur in the power grid, the demand response side will determine the demand response quantity and release corresponding information to the demand response market [43]. At present, the electricity market has been incrementally implementing a series of demand response policies in China, aiming to develop the potential of demand response resources in various regions, incentivizing a diverse range of enterprises to actively take part in demand response, and constructing a systematic library of demand response resources. These policies will improve the flexibility and operational efficiency of power systems.

This paper sets up the demand response scenarios and refers to “Inner Mongolia Power Grid and Market Demand Response Side Implementation Rules” by the Inner Mongolia Energy Bureau and current research on demand response in western Inner Mongolia from 2030 to 2060 [44,45,46].

3.2.4. Developing TCE and TGC Market

The carbon price and green electricity play a crucial role in future energy transition pathways. The construction of the electricity market, TCE market, and TGC market is conducive to reducing carbon emissions and encouraging the high-carbon-emission power sector to proactively reduce its emissions. China’s carbon market is still in its infancy. However, the carbon markets in some countries are relatively mature, and a few European countries have achieved their carbon emissions targets by implementing a reasonable carbon trading market. In 2017, authorities identified electricity companies in Inner Mongolia. These companies were incorporated into the national carbon trading market. In 2024, the Inner Mongolia electricity market conducted large-scale green electricity trading for the first time, integrating it into the medium- and long-term trading system. At the same time, it implemented the Dual Control of Energy Consumption to Dual Control of Carbon Emissions Work Plan for the Pilot Transition to explore the establishment of a market-based mechanism to promote emission reduction.

This paper established the scenario of “TCE + TGC” by referring to the model of carbon prices and green electricity prices from current research [47,48,49].

The specific scenario setting is shown in

Table 2. Scenario settings and descriptions.

Items	Scenario Descriptions
Scenario 1 (S1): The renewable energy + storage system	The percentage of renewable energy and storage systems will increase to 30% in 2030 and will reach 90% in 2060.
Scenario 2 (S2): The CCS technology development	The CFPPs will gradually be equipped with CCS, and the CFPPs with CCS will reach from 10% to 90% between 2030 and 2060.
Scenario 3 (S3): The demand response	In total, 10% of electricity users actively participate in demand response, and the demand response is 8% of the maximum load of the whole society in 2030; 40% of electricity users actively participate in demand response, and the demand response is 20% of the maximum load of the whole society in 2060.
Scenario 4 (S4): The TCE and TGC scenarios	Implementation of carbon price and green electricity prices in the power sectors, with a carbon price of 100 CNY/ton in 2030 and 700 CNY/ton in 2060, and a green electricity price of 100 CNY/MWh in 2030 and 500 CNY/MWh in 2060.

.

3.3. Data Sources

This study adopts the combination of current research, government reports, and mathematical model results to select the data source. The initial population and economic indicators are from the Inner Mongolia 14th Five-Year Plan and Inner Mongolia Statistical Yearbook. The indicators of the urbanization rate, the proportion of the three industries, and other power parameters refer to the government report, Inner Mongolia Power 14th Five-Year Plan, and the International Energy Agency. In the context of “dual carbon”, the development parameters of Inner Mongolia are shown in

Table 3. Description of basic data under carbon neutrality policy.

Index	2020	2030	2040	2050	2060
The average annual growth rate of GDP	4.95%	5.32%	4.18%	3.35%	2.31%
The population (millions)	25.23	23.40	22.98	22.06	21.11
The urbanization rate	67.84%	69.96%	71.25%	75.41%	77.38%
The proportion of the three industries	11.84%: 39.56%: 48.59%	6.55%: 38.04%: 57.73%	5.54%: 36.53%: 59.48%	4.53%: 35.01%: 61.23%	3.50%: 33.50%: 63.00%

.

This study refers to the existing literature on the prediction of future energy production and the data on energy consumption of various power generation types in Inner Mongolia [6,7,41]. The carbon emission factor and cost parameters are derived from the current literature and research [7,32,50,51,52]. It includes wind power, solar power, energy storage, coal-fired power, CCS retrofitting, and so on. The economic parameters settings are shown in Appendix A,

Table A1. The economic parameters of various types of power generation.

Items	Cost	Unit	2020	2030	2040	2050	2060
Wind power	Investment cost	CNY/kW	8149	7608	7456	7302	6580
	Fixed OM cost	CNY/kW	310	290	270	250	230
	Variable OM cost	CNY/kWh	0.087	0.079	0.071	0.063	0.055
Solar power	Investment cost	CNY/kW	8000	7527	7082	6619	6160
	Fixed OM cost	CNY/kW	216	206.00	196	186	176
	Variable OM cost	CNY/kWh	0.060	0.051	0.043	0.036	0.030
Coal-fired power	Investment cost	CNY/kW	5620	5293	4985	4695	4421
	Fixed OM cost	CNY/kW	150	135	120	105	90
	Variable OM cost	CNY/kWh	0.19	0.16	0.13	0.10	0.07
Coal-fired power with CCS	Investment cost	CNY/kW	36,688	33,020	29,718	26,747	24,073
	Fixed OM cost	CNY/kW	229	200	180	162	145
	Variable OM cost	CNY/kWh	2.5	2.0	1.5	1.0	0.5
Storage (Li battery)	Investment cost	CNY/kW	1700	1500	1300	1100	900
	Fixed OM cost	CNY/kW	55	50	45	40	35
Storage (Hydrogen storage)	Investment cost	CNY/kW	4699	4514	4337	4167	4004
	Fixed OM cost	CNY/kW	105	95	85	75	65

. The carbon emission factors are shown in Appendix A,

Table A2. The carbon emission parameters of various types of power generation.

Items	Unit	2020	2030	2040	2050	2060
Wind	CO2g/kWh	7	6	5	4	3
Solar	CO2g/kWh	65	58	50	41	31
Coal-fired	CO2g/kWh	930	900	860	810	750
Coal-fired with CCS	CO2g/kWh	93	90	86	81	75
Li battery	CO2g/kWh	72	67	62	57	52
Hydrogen storage	CO2g/kWh	61	57	53	49	44

.

4. Results and Discussion

4.1. Total Electricity Demand Analysis

Based on the carbon neutrality assumption scenarios, the bottom-up methodology was used to build the LEAP-Inner Mongolia model to predict future power consumption. The prediction results are presented in 10-year periods, as shown in Figure 3. Under the carbon neutrality scenario, with the increase in regional GDP, different industrial structure adjustments, and population changes, the electricity generation demand will increase from 348.5 billion kWh in 2020 to 1410 billion kWh in 2060, an average annual increase of 4.56%. Owing to the gradual increase in electrification in construction, transportation, electricity transition, and other industrial sectors, the electricity demand will increase. As a major energy export province, the regional electricity production demand is gradually growing.

Because of Inner Mongolia’s energy, chemical, and metallurgical industries, among others, the expansion of production has driven a significant increase in electricity demand in the secondary industry. At the same time, stable economic growth has led to a continued expansion of electricity consumption in the tertiary industry, which is an important part of economic growth between 2040 and 2060. Due to the decline in the proportion of the primary industry, the growth of its electricity demand is relatively slow compared with other industries.

4.2. Power Generation and Installed Capacity Analysis

Figure 4 and Figure 5 show the different power type generations levels and installed capacities in different scenarios. The total power generation and low-carbon electricity show a growth trend. In all scenarios, wind power and coal-fired power are the main contributors to power generation. Coal-fired power generation exhibits a steady decrease; renewable energy electricity generation is constantly rising. By comparing the installed capacity and power generation, the total installed capacity in S3 was found to be slightly lower than in other scenarios. Furthermore, the lower the installed capacity, the higher the percentage of renewable energy power generation and coal-fired power generation with CCS, which may lead to better performance in low-carbon electricity production.

In S1, the majority of CFPPs will be phased out by 2060, and only 10% of coal-fired power generation will remain in 2060. Coal-fired power generation with CCS will reach 73.68 billion kWh. In terms of power structure changes, wind power and solar power will be the main methods of power generation, and the installed capacity will increase by 154.64 and 74.07 thousand MW by 2060, respectively. The energy storage scale will increasingly rise from 7.5 thousand MW in 2030 to 39.4 thousand MW in 2060, with a cumulative energy power increase of 346.98 billion kWh.

Considering the development of CCS technology, S2 pays more attention to establishing CFPPs with CCS technology to achieve net-zero emissions in the energy system by 2060. Coal-fired power with CCS will become the main source of power; its installation capacity will also reach 57.56 thousand MW in 2060, and the accumulated power generation will be 959.93 billion kWh from 2020 to 2060. Coal-fired power generation is expected to be phased out by 2060. The share of power generation composed of wind power, solar power, and energy storage in the total power generation will rise from 14.5% to 59.1%, and between 2020 and 2060, the total installed capacity of wind power and solar power will increase by almost 17 times.

In S3, the demand response strategy reduces peak load. This drives the continuous compression of the coal-fired power capacity and extends operation hours. At the same time, the installed capacity of renewable energy and energy storage decreases accordingly, making the total installed capacity of S3 the lowest among all scenarios. However, due to the incomplete development of low-carbon electricity, coal-fired power generation still accounts for a higher proportion in S3 than in other scenarios. In S4, the installed capacity and the share of different power generation units are similar to those in S2. Compared to S2, the coal-fired power capacity with CCS will decrease by about 32.5%, the energy storage capacity will expand almost two-fold, and most CFPPs will be retired by 2060, highlighting the important role of TCE and TGC markets in encouraging the development of renewable energy and discouraging coal-fired power generation.

4.3. Carbon Emission Analysis

As shown in Figure 6, total carbon emission shows a general downward trend in S1, S2, and S4. However, only S3 will reach the carbon peak target in 2040. The carbon emissions in S1 are the lowest between 2030 and 2040; the carbon emissions in S2 are the lowest between 2050 and 2060 and lower than those in S1. Most of the carbon emissions in the power sector in Inner Mongolia come from CFPPs. Due to the framework of the dual carbon target, carbon emissions are required to decrease by 2060, and the energy system of Inner Mongolia is under considerable pressure to meet this emission peak requirement.

The total carbon emissions in S1–S4 are 298.33 million tons, 306.63 million tons, 322.35 million tons, and 297.97 million tons in 2030, decreasing to 160.14 million tons, 77.29 million tons, 279.57 million tons, and 156.64 million tons, respectively, by 2060. S2 shows a significant downward trend in carbon emissions, contributing toward achieving the carbon-neutral goal by 2060. As Inner Mongolia has a high proportion of coal-fired power, the introduction of CCS technology scenarios can significantly improve the carbon reduction effectiveness in the regional energy system. This is followed by S1, where the total carbon emissions are only 15% higher than those in S2; thanks to Inner Mongolia having abundant wind and solar resources and the continued increase in operation hours, renewable energy is steadily replacing coal-fired power. Due to the lack of large-scale deployment of CCS technology, coal-fired power generation is still used as a flexible supplementary power source, and the rate of carbon emission reduction is slower than that in S2.

In S3, the total carbon emissions are reduced by only 12% over a 30-year period; although the installed capacity and total costs are lower than those in the other scenarios, the share of coal-fired power is still high, meaning that the total carbon emissions are 42.72% higher than in S1 and 72.35% higher than in S2. S4 has almost the same carbon emissions as S1; the primary reason is that the implementation of TCE and TGC policies not only increases the profits from renewable energy and supports coal-fired power with CCS but also accelerates the retirement of coal-fired power.

4.4. Total Power System Cost Analysis

The total power system costs represent the electricity production cost, including the investment cost, operations and maintenance costs, and environmental costs. The total power system costs in the four energy transition scenarios are presented in Figure 7. The cumulative total power system costs of all scenarios showed similar exponential growth. S2 has the highest cumulative cost among the scenarios, with CNY 7118 billion from 2020 to 2060, mainly due to the high cost of CCS retrofitting technology and Inner Mongolia’s strong reliance on fossil fuel. This is followed by S1, with 6588 billion CNY; the increase in the total cost is mainly driven by the large-scale installation of renewable energy, the high cost of energy storage, and some of the CFPPs being equipped with CCS technology. The total cost of S3 is the cheapest because the proportion of energy storage and CFPPs installed with CCS is relatively small.

In S1, the cumulative power system cost for wind power and solar power between 2020 and 2060 was the highest at CNY 2871 billion and CNY 1298 billion, respectively. This was followed by energy storage and CFPPs, with the cumulative cost of CNY 614 billion and CNY 1094 billion, respectively. In S2, because of the centralized development of CCS technology and the withdrawal of CFPPs, power generation will experience a fundamental change. The total cost of coal-fired power with CCS will reach CNY 1083 billion in 2060. This is followed by the cumulative cost of wind power and solar power at CNY 2283 billion and CNY 1109 billion, respectively. Compared to S2, the cumulative cost is approximately 12.5% higher.

In S3, the order of the cumulative total social cost for 2030–2060 is similar to wind, solar, and coal-fired power between 2030 and 2040 in S1. Because more than 20% of CFPPs complete CCS retrofitting projects, the cost of coal-fired power with CCS will be approximately equal to that of solar power in 2060. However, the total installed capacity in S3 is less than that in S1, and its cumulative cost is only about 75% of that in S1. In S4, due to the accelerated development of the carbon price and green electricity certificate market, the share of renewable energy and energy storage decreases by 11% compared with S1, but the cost is reduced by 15%; the share of CFPPs with CCS decreases by 15% compared to S2, and the cost is reduced by 23%.

Combining the carbon emission results for the different energy transition pathways, due to the limitation of carbon neutrality constraints and renewable energy goals, our findings indicate that CFPPs without CCS must be phased out first to achieve carbon neutrality and rapidly develop CCS technology and energy storage facilities in the long term, while the Inner Mongolia region strengthens demand response strategies to reduce the curtailment of wind and solar power rates and the installed capacity of CFPPs.

4.5. Low-Carbon Electricity Efficiency

The low-carbon energy efficiency in different scenarios during 2030 and 2060 is indicated in Figure 8. The average energy efficiency under the four scenarios is 0.850, 0.769, 0.706, and 0.847, respectively. S1 and S4 have similar energy efficiencies, and S3 has the lowest low-carbon electricity efficiency, with about a 29% room for improvement. The efficiency value of six DMUs reaches a value of 1, reflecting the optimal production level; the remaining ten DMUs are all below 1, among which S2 in 2040 (0.575), S2 in 2050 (0.554), and S3 in 2050 (0.573) demonstrate relatively low efficiency, showing significant improvement potential.

According to the efficiency results, the slack variables for low-carbon electricity are relatively large. Through the different slack variables shown in

Table 4. The slack variables of different energy transition scenarios from 2030 to 2060.

Item	Time	DMUs	ΔREEG	ΔREGP	ΔREPC	ΔRECI	ΔREIC
S1	2030	1	0	0	0	0	0
	2040	2	26.2	0	0.4	134.1	62.6
	2050	3	33.5	0	1.4	194.7	68.3
	2060	4	0	0	0	0	0
S2	2030	5	13.5	0	0	10.5	14.6
	2040	6	71.6	0	0	256.7	93.5
	2050	7	78.3	31.2	1.5	292.1	78.3
	2060	8	0	0	0	0	0
S3	2030	9	0	0	0	0	0
	2040	10	63.1	0	0.4	166.5	133.5
	2050	11	91.8	0	0.5	294.6	173.1
	2060	12	6.9	0	1.5	0	145.8
S4	2030	13	0	0	0	0	0
	2040	14	45.8	0	0	183.5	77.7
	2050	15	47.8	0	0	273.2	71.6
	2060	16	0	0	0	0	0

, the low-carbon electricity in several inefficient scenarios can be analyzed. In terms of energy consumption, the slack variables’ values in S2 and S3 are relatively large, indicating that coal consumption for coal-fired power with/without CCS is slightly higher. Although CCS technology significantly reduces carbon emissions, it also increases coal consumption to some extent, meaning that the utilization proportion of coal needs to be further reduced. Regarding power system costs, the majority of DMUs have reached the efficiency frontier, so there is limited room for cost optimization.

As for the installed capacity, S3 has the most room for optimization, primarily because of the simultaneous growth in the installed capacity and power generation, meaning that peak shaving and valley filling have greater potential for effectively reducing the installed capacity. This is followed by S1, mainly due to its low utilization hours for renewable energy and the idleness of the energy storage system during certain periods, resulting in an excessive installed capacity. Regarding low-carbon power generation, there is still room for improvement in all scenarios, especially S2. This shows that the power system has not yet effectively converted all resources into low-carbon power generation under the already invested conditions of energy consumption, total cost, and installed capacity. In terms of carbon emissions, the slack variables in S3 are higher than in the other scenarios, mainly due to the reliance on a large share of coal-fired power, so carbon emission reduction measures need to be further strengthened. Therefore, low-carbon electricity efficiency can be improved by optimizing energy consumption, reasonably controlling installed capacity, and reducing carbon emissions.

4.6. Low-Carbon Electricity Influencing Factors Analysis

Under the “dual carbon” goal, low-carbon electricity in Inner Mongolia will grow significantly, but the increments are different among the four scenarios.

Table 5. The decomposition results of low-carbon electricity over different scenarios and periods (unit: billion kWh).

Item	Time	ΔLELG	ΔLEGP	ΔLEPC	ΔLECI	ΔLEIC	ΔLECC	ΔLE
S1	2020–2030	72.47	91.11	37.44	−35.86	45.63	−27.45	183.33
	2030–2040	148.30	72.89	23.18	−117.22	150.15	−7.33	269.96
	2040–2050	169.34	135.27	60.63	−208.45	326.61	−131.47	351.94
	2050–2060	168.23	72.43	418.82	−386.33	614.11	−569.40	317.86
S2	2020–2030	50.31	83.11	19.85	−17.13	31.58	−25.04	142.67
	2030–2040	239.38	72.97	−26.59	−191.03	349.97	−7.34	437.36
	2040–2050	272.55	159.26	−50.71	−271.26	1043.60	−154.79	998.65
	2050–2060	68.24	84.05	169.24	−102.84	610.51	−660.78	168.42
S3	2020–2030	52.69	83.99	−37.21	44.99	−37.44	−25.31	81.71
	2030–2040	132.77	64.19	2.85	−94.01	119.32	−6.46	218.68
	2040–2050	151.58	119.69	−0.76	−158.60	246.66	−116.33	242.24
	2050–2060	164.33	64.70	197.15	−210.41	370.53	−508.68	77.62
S4	2020–2030	82.46	94.58	31.07	−29.96	51.78	−28.50	201.43
	2030–2040	187.10	80.15	−14.26	−132.57	260.47	−8.06	372.83
	2040–2050	191.74	152.88	5.59	−171.37	537.90	−148.59	568.15
	2050–2060	126.27	79.62	237.18	−166.22	774.86	−625.99	425.74

shows the outcomes of decomposing the influencing factors of low-carbon electricity generation using the LMDI method. The research results show that S2 has the highest increase in cumulative low-carbon electricity, at around 1747.10 billion kWh, followed by S1, at around 1123.09 billion kWh, and S3 has the lowest increase, with only 620.25 billion kWh. We investigated the driving factors behind low-carbon electricity development across different energy transition scenarios in Inner Mongolia between 2020 and 2060 using the LMDI decomposition model. With contributions from high to low orders, the promotion factors in low-carbon electricity are carbon productivity, low-carbon electricity structure, electricity intensity, and energy security. The carbon emission and energy intensity are the main suppressors of the increase in low-carbon electricity.

The relative contribution ratios for driving factors in low-carbon electricity during 2030–2060 are shown in

Table 6. The contribution ratio of different factors to the development of low-carbon electricity over different scenarios and periods (unit: %).

Item	Time	ΔLELG	ΔLEGP	ΔLEPC	ΔLECI	ΔLEIC	ΔLECC
S1	2020–2030	0.40	0.50	0.2	−0.2	0.25	−0.15
	2030–2040	0.55	0.27	0.09	−0.43	0.56	−0.03
	2040–2050	0.48	0.38	0.17	−0.59	0.93	−0.37
	2050–2060	0.53	0.23	1.32	−1.22	1.93	−1.79
S2	2020–2030	0.35	0.58	0.14	−0.12	0.22	−0.18
	2030–2040	0.55	0.17	−0.06	−0.44	0.80	−0.02
	2040–2050	0.27	0.16	−0.05	−0.27	1.05	−0.15
	2050–2060	0.41	0.50	1.00	−0.61	3.62	−3.92
S3	2020–2030	0.64	1.03	−0.46	0.55	−0.46	−0.31
	2030–2040	0.61	0.29	0.01	−0.43	0.55	−0.03
	2040–2050	0.63	0.49	0.00	−0.65	1.02	−0.48
	2050–2060	2.12	0.83	2.54	−2.71	4.77	−6.55
S4	2020–2030	0.41	0.47	0.15	−0.15	0.26	−0.14
	2030–2040	0.50	0.21	−0.04	−0.36	0.70	−0.02
	2040–2050	0.34	0.27	0.01	−0.30	0.95	−0.26
	2050–2060	0.30	0.19	0.56	−0.39	1.82	−1.47

. The carbon productivity is the largest positive contributing factor in all scenarios, suggesting that the improvement in carbon productivity is the core driving force in the growth of low-carbon electricity. The cumulative contribution percentage is 367% in S1, 569% in S2, 588% in S3, and 373% in S4. This contribution rate will continue to increase over time; this is because the cost of low-carbon power generation technology has decreased, and the large-scale development of clean energy has increased, significantly reducing the carbon emission intensity per unit of electricity generation and thereby improving carbon productivity. The contribution of low-carbon power structures ranks second, and the cumulative contribution percentage in S3 (399%) is significantly higher than that in the other scenarios (196% in S1, 158% in S2, and 155% in S4). The demand response moderately increases the share of renewable energy consumption and reduces the curtailment of wind power and solar power, thereby increasing the effective share of clean energy in the electricity structure. This contribution shows a slight fluctuation trend over the 30-year period. The contribution percentages for electricity density and energy security are similar, and both have a positive effect. Moreover, these two contributing factors both refer to energy production, mainly because Inner Mongolia has a high level of energy production, and it also pays more attention to the synergy between energy production and consumption.

During the period of 2020–2060, energy density and carbon emissions will have a negative impact on the development of low-carbon electricity, and the effect of carbon emissions will be stronger than that of energy density; the impacts will fluctuate in all scenarios. In S1–S4, the cumulative contributions of energy density to low-carbon electricity are −244%, −144%, −324%, and −120%, respectively. The cumulative contributions of carbon emission to low-carbon electricity are −234%, −427%, −737%, and −189%, respectively. The energy density measures energy consumption per unit of electricity. Its negative contribution indicates that the newly added low-carbon capacity has not been converted into a high-efficiency energy output to some extent, which is caused by low power generation efficiency and improper technology configuration. From the perspective of the carbon emission effect, with the expansion of low-carbon electricity and the installed capacity, the higher the negative value, the greater the indication that there is still some energy-intensive capacity in operation and that the growth rate of electricity demand is exceeding the decarbonization rate.

5. Conclusions, Policy Implications, and Future Work

5.1. Conclusions

This study evaluated the different energy transition pathways for the power sector in Inner Mongolia under the constraints of carbon neutrality using the LEAP model, setting up four energy transition pathways: renewable energy and storage systems, CCS technology, demand response, and TCE and TGC scenarios. In addition, we comprehensively explored the influencing factors and efficiency of low-carbon electricity in different scenarios. An LMDI and SBM-DEA coupling model was further constructed for the quantitative analysis of the contributions of low-carbon electricity structure, energy security, energy intensity, carbon productivity, and carbon emission to low-carbon electricity, evaluating the low-carbon electricity generation efficiency in the energy system. The main conclusions are as follows:

(1)

Regarding carbon emissions in the energy system, the contribution of carbon reduction in the energy system for each scenario is in the order of CCS technology scenarios > renewable energy and energy storage > TCE and TGC > demand response. According to the total cost comparison among different scenarios, combined with the carbon reduction potential, Inner Mongolia should focus on the synergistic effect of “renewable energy and storage system” and “TCE and TGC” in the short term. On the one hand, it has rich wind energy and solar energy resources; on the other hand, carbon prices and green electricity certification can increase the benefits of low-carbon electricity in reducing carbon emissions in the power system. With the decline in the cost of low-carbon electricity technology, Inner Mongolia should aim to achieve major decarbonization and the goal of being carbon-neutral by 2060. The deployment of CCS technology should be accelerated in the medium and long term because coal-fired power is still dominant in this region, and it is difficult to retire in the short term. Although the high share of renewable energy meets most of the electricity demand, coal-fired power is responsible for ensuring peak load regulation capacity and grid stability. In addition, the auxiliary role of demand response should be strengthened. It not only has the lowest cost but also prevents overcapacity and abandonment in solar and wind power.

(2)

Regarding influencing factors in low-carbon electricity, the low-carbon electricity structure, carbon productivity, energy intensity, and carbon emission are the most critical. The low-carbon electricity structure and carbon productivity are the main positive drivers in the growth of low-carbon electricity; the energy intensity and carbon emission are the main negative constraints, and electricity intensity and energy security have lower impacts. As the cost of low-carbon technology continues to decline, it will be deployed on a large scale to significantly reduce total carbon emissions. Under the cost constraints, the Inner Mongolia region should coordinate the development of both carbon emission reduction and the installed capacity of low-carbon electricity. From the perspective of power generation effectiveness in low-carbon electricity, the low-carbon electricity generation efficiency in each scenario is ranked as follows: renewable energy and energy storage are the highest, the TCE and TGC are second, and demand response is the lowest. In the process of improving efficiency, it is essential to balance the dynamic relationship between energy production and consumption and optimize the configuration of the installed capacity of different types of power generation. This is because Inner Mongolia has abundant fossil fuels and CFPPs, meaning that it would be difficult to retire all coal-fired power units in the long term.

5.2. Policy Implications

5.2.1. Strengthening Low-Carbon Electricity Trading System

Inner Mongolia has abundant fossil fuel energy, especially coal production, and has the task of achieving a significant electricity transition to achieve the dual carbon targets of the energy system. At present, it is difficult to achieve the large-scale retirement of CFPPs, strengthening the establishment of a tradable carbon price and a green certificate trading market in the power system, which will be conducive to the construction of a low-carbon power market and the realization of a sustainable energy system.

5.2.2. Focus on the Development of Storage Systems and CCS Retrofitting Technology

Due to the rich wind and solar energy resources in Inner Mongolia, the development of large-scale wind and photovoltaic power will be implemented in the future to ensure the stability of the power grid and improve the flexibility of the power system; it is still necessary to deploy energy storage and coal-fired power with CCS. However, since the costs of energy storage and CCS retrofitting technology are still relatively high, the transformation and technological update should be accelerated, and government finance and policy support should be provided.

5.2.3. Strengthen Public Awareness of Energy Saving

The results of the scenario analysis reveal that the demand response scenario has the lowest carbon reduction potential among other scenarios and the lowest energy system costs. Demand response significantly decreases coal-fired power capacity. At the same time, it enhances the power grid’s flexibility and decreases the curtailment of renewable energy. To accelerate the carbon reduction process, policymakers should prioritize public awareness of energy conservation and incentivize end-user participation in DR programs, enabling the faster decarbonization of regional energy systems.

5.3. Future Work

This study constructs the LEAP-Inner Mongolia model to forecast Inner Mongolia’s energy structure, carbon emissions, and energy system costs of different energy transition scenarios in the context of the carbon neutrality goal and evaluates the influencing factors and power efficiency of low-carbon electricity under different energy transition scenarios. However, this study has some limitations. Firstly, although this paper constructs different energy transition scenarios, it has not yet systematically assessed the feasibility and uncertainty of scenario parameters. Subsequent research will introduce sensitivity analysis and Monte Carlo simulation to quantify the impact of key parameters. In addition, due to the limitations in data and research, we cannot provide detailed predictions of energy demand and energy structure from the perspective of different end-use industries. We also do not consider the scenarios for electric vehicles participating in the electricity market and carbon market. Future work will explore the carbon peak and carbon neutrality paths for different industries in depth. Finally, we plan to apply frontier models to conduct horizontal comparisons of different transition paths, thereby enhancing the depth and breadth of energy system forecasting.