Research on Environmental Evaluation Index of Carbon-Based Power Generation Formats Under the “Dual Carbon Goals”

Abstract

As a major source of carbon emissions, the carbon-based power generation industry requires a scientifically robust environmental performance evaluation system to facilitate its green transition and sustainable development. Focusing on unique transition dynamics across four carbon-based power generation formats, this study compares environmental dimension indicators across typical ESG evaluation frameworks and proposes an innovative evaluation index model of environmental performance based on common metrics, with a particular emphasis on their contribution potential to the “Dual Carbon Goals”. The framework’s core innovation lies in its Dual Carbon-focused indicator system, which evaluates three critical indicators overlooked by mainstream ESG methodologies. It extends to include upstream/downstream processes, addressing gaps in current evaluation systems. The findings reveal that core environmental issues, such as climate change, pollution emissions, and resource utilization, exhibit broad commonality in ESG evaluations. Among the assessed indicators, carbon emission intensity carries the highest weight, underscoring its centrality in each power generation sector’s efforts to align with the Dual Carbon Goals. Furthermore, the analysis demonstrates that underground coal gasification combined cycle power generation has a relatively favorable environmental performance, ranking slightly below natural gas combined cycle but above shale gas combined cycle power generation. In contrast, traditional coal-fired power generation exhibits significantly poorer environmental outcomes, highlighting both the efficacy of technological upgrades in reducing emissions and the urgent need for transitioning away from conventional coal-based power.

1. Introduction

In 2024, China’s total carbon emissions were 12.603 billion tons [1], of which the power industry emitted 5.640 billion tons [2], accounting for 44.75%. With the proposal of the “Dual Carbon Goals”, the power sector is committed to green transformation. However, in the short term, the power industry remains China’s largest carbon emission sector, with carbon emissions accounting for about 40% [3,4]. The China Power Industry Annual Development Report 2024 [5] points out that in 2023, the carbon emission intensity (CO₂ emission per unit power generation) of China’s power industry was 540 g/kWh, of which the carbon emission intensity of thermal power was 821 g/kWh. According to relevant predictions, to achieve carbon neutrality, the carbon emission intensity of the power industry needs to be reduced to 50 g/kWh [6]. Fossil energy power generation, represented by coal-fired power, is a top priority for emission reduction and carbon reduction in the power industry. In addition to the above impact on the “Dual Carbon Goals”, fossil energy power generation also has other environmental problems, such as pollutant emissions and excessive resource consumption [7]. Therefore, under the “Dual Carbon Goals”, the research on the ESG environmental index impact evaluation of different power generation formats is of great significance.

ESG evaluation refers to the overall evaluation of the sustainable development capabilities, opportunities, or impacts of a given enterprise or organization, emphasizing the measurement of enterprise value from three dimensions: Environment (E), Social (S), and Governance (G), and quantifying corporate social responsibility. Cucchiella et al. (2017) proposed the Multi-Criteria Decision Analysis (MCDA) theory by establishing indicators such as waste asset utilization rate, waste gas emissions, and environmental protection expenditures [8]. Zhang Sa (2017) proposed that it is necessary to adjust measures to local conditions, closely combine policy orientation and actual needs, focus on the sustainable development of the ecological environment, and give ESG a deeper meaning [9]. Qi (2018) established a systematic performance evaluation system for coal mine enterprises based on the Analytic Hierarchy Process (AHP) and Fuzzy Comprehensive Evaluation (FCE), organically combining the environmental evaluation of coal mine enterprises with safety performance and work performance [10]. Yuan Jiahai et al. (2018) cross-subdivided Environment, Social, Governance with Pressure, State, and Response to build an ESG performance evaluation system for large power generation listed companies [11]. The application results show that environmental performance has the greatest impact on the overall enterprise. Zaharova et al. (2021) proposed that including greenhouse gas emission intensity as an environmental performance evaluation indicator helps assess enterprise environmental risks [12]. Wang Xiaomeng (2021) believes that the ESG evaluation system is an effective management tool to support the implementation of the “Dual Carbon Goals”, and the evaluation results will provide direction for enterprises to implement emission reduction and sustainable development [13]. Liu et al. (2022) used the super-efficient MinDS model to analyze the environmental performance of the power industry and pointed out that it is necessary to optimize the factor combination and improve the monitoring mechanism to improve the environmental performance level [14]. In summary, the discussion on ESG has made diversified progress, but the selection of key indicators in the environmental dimension, the construction of evaluation systems, and the research on key emission control formats are still rare.

For the power generation sector with the largest carbon emission volume, the environment is a key issue supporting its development. Changes in ecological and environmental conditions dominate the direction and path of subsequent development of power plants and mid-downstream production and marketing entities [15]. The environmental evaluation of different power generation formats represented by coal-fired power will effectively quantify the performance of each format in climate change, environmental pollution, resource consumption, and other aspects and identify the current contribution potential and environmental protection effectiveness of power generation modes to the “Dual Carbon Goals”. Moreover, existing studies have shown that in power generation enterprises, the performance of environmental performance has the greatest impact on their comprehensive ESG evaluation, and the relative closeness between the two is highly consistent [11]. In general, environmental indicators are the core of ESG evaluation for power generation formats, so this paper only discusses the ESG environmental dimension evaluation of different power generation formats, and the social dimension and governance dimension are not considered for the time being.

Combining the existing literature and based on ESG theoretical practices, this study innovatively constructs an environmental evaluation index model, emphasizes the contribution potential of each format to the “Dual Carbon Goals”, uses the Analytic Hierarchy Process (AHP) method to assign index weights, and carries out evaluation calculations on different power generation formats.

The research advances current practice by developing a technology-specific evaluation model that captures unique transition dynamics across four carbon-based power generation formats. The framework’s core innovation lies in its Dual Carbon-focused indicator system, which evaluates three critical indicators (carbon capture difficulty, carbon emission reduction cost, and carbon storage potential) overlooked by mainstream ESG methodologies. The carbon emission intensity is used as a core metric that aligns with widely accepted international frameworks like the GHG Protocol, ensuring compatibility with global carbon accounting practices. By incorporating upstream/downstream processes, the model mirrors lifecycle assessment (LCA) approaches used in international ESG evaluations, improving cross-border benchmarking and addressing gaps in current evaluation systems.

The methodology demonstrates superior applicability for transition planning. The deterministic scoring system provides clearer differentiation than fuzzy evaluation methods, accurately quantifying performance gaps between formats. Through theoretical research and empirical analysis, it aims to explore the environmental impact performance of different power generation formats, hence providing a decision-making basis for power generation enterprises under the emission reduction orientation and promoting the transition of carbon-based power generation formats to a more green, low-carbon, and sustainable path.

2. Methodology

2.1. Power Generation Formats

This paper selects four power generation formats, including traditional and emerging power generation formats, as samples for ESG environmental dimension performance evaluation. The overview of each power generation format is shown in

Table 1. Overview of sample power generation operations.

Format	Fuel Mining	Fuel Transportation	Energy Conversion and Utilization
Underground Mining–Coal-fired Unit	Conventional underground mining	Road/Railway	Ultra-supercritical generation unit
Underground Coal Gasification–Combined Cycle Unit	Underground coal gasification	None	Coal is gasified underground into combustible gas, which is used to generate electricity in a combined cycle unit
Imported Natural Gas–Combined Cycle Unit	Conventional natural gas	LNG ship + Pipeline	Natural gas is used to generate electricity in a combined cycle unit
Imported Shale Gas–Gas Cycle Unit	Shale gas imported from North America	LNG ship + pipeline

.

2.2. Analytic Hierarchy Process (AHP)

The qualitative and quantitative indicators in the environmental dimension evaluation system are uniformly mathematically quantified, and a numerical value (comprehensive index) is calculated to reflect the weight of the evaluation indicator in the overall evaluation system. The 1–9 scale method proposed by Santy is used to compare the indicators at each level, two by two [16], and judge the relative importance, as shown in

Table 2. Importance scale.

Element	Scale	Interpretation
aij (element in the judgment matrix)	1~9	Relative importance of i to j; the larger the value, the higher the importance
	Reciprocal of 1–9	Relative importance of j to i

.

After the judgment matrix is constructed, the root method is used to solve the matrix to complete the weight assignment of each level of units. The formula is as follows:

$${\omega }_i=\sqrt[n]{{\prod }_{j=1}^n{a}_{ij}} ;(i,j=\mathrm{1,2},3,\cdots ,n)$$

(1)

where ω_i is the processed n-dimensional vector, n is the order of the matrix, and a_ij is the element of the judgment matrix.

Normalize the vector ω_i so that the sum of the elements is 1:

$${\xi }_i={\displaystyle \frac{{\omega }_i}{{\sum }_{i=1}^n{\omega }_i}}$$

(2)

The feature vector can be obtained, which is the weight vector of the indicators at this level (ξ₁, ξ₂, ξ₃, …, ξ_n)^T.

Then, multiply the weights of each single-level indicator by the weights of the upper-level indicators to obtain the overall weight vector of the evaluation system (ξ₁, ξ₂, ξ₃, …, ξ_n)^T:

$${\xi }_j={\sum }_{i=1}^n{\xi }_i{\xi }_j^i;(i=\mathrm{1,2},3,\cdots ,n;j=\mathrm{1,2},3,\cdots ,m)$$

(3)

where ξ_j is the weight of the underlying indicator, ${\xi }_j^i$ is the weight of the lower-level indicator relative to the upper-level indicator weight ξ_i, and m is the number of underlying indicators.

In addition, each judgment matrix needs to be tested for consistency using the following formula to check whether there are logical loopholes in the judgment matrix:

$${\lambda }_{max}={\displaystyle \frac{1}{n}}{\sum }_{i=1}^n{\displaystyle \frac{{\left(\mathrm{D}\xi \right)}_i}{{\xi }_i}}$$

(4)

$$CI={\displaystyle \frac{{\lambda }_{max}-n}{n-1}}$$

(5)

$$CR={\displaystyle \frac{CI}{RI}}$$

(6)

where λ_max is the maximum eigenvalue of matrix D, n is the order of the matrix, CI is the consistency index, RI is the random consistency index, and CR is the consistency ratio.

When CI < 0.1, it indicates that the consistency level of the judgment matrix is within the acceptable range, and the consistency test is passed. At this time, the corresponding feature vector can be used as the weight vector. Among them, the value of RI depends on the order of the matrix, and the specific values can be referred to

Table 3. RI values of matrix order 1–10.

n	1	2	3	4	5	6	7	8	9	10
RI	0	0	0.52	0.90	1.12	1.26	1.36	1.41	1.46	1.49

.

To ensure the rationality of the combined weight distribution and the reliability of the single-level consistency test results, it is necessary to carry out the overall consistency test, which can be carried out through the following formula:

$${CR}_{\mathrm{t}\mathrm{o}\mathrm{t}\mathrm{a}\mathrm{l}}={\displaystyle \frac{{\sum }_{i=1}^n{{\xi }_{i}CI}_i}{{\sum }_{i=1}^n{{\xi }_{i}RI}_i}}$$

(7)

3. Construction of the Evaluation System

3.1. Index Selection

The environmental dimension indicators proposed by different agencies are compared. The MSCI (Morgan Stanley Capital International) ESG evaluation system is composed of 10 secondary indicators and 35 tertiary indicators, among which the environmental dimension is mainly composed of four themes: climate change, natural capital, pollutants and waste, and environmental opportunities. The environmental indicators of FTSE Russell cover five aspects: supply chain environment, pollution and resources, climate change, biodiversity, and water security. The Thomson Reuters ESG environment category consists of 3 secondary indicators—resource use, low carbon emissions, and innovation—and 61 tertiary indicators. RobecoSAM Smart ESG (Dow Jones) evaluates the environment from the aspects of environmental reports, operational ecological efficiency, and environmental policies. In China, the SynTao Green Finance ESG evaluation system has a high degree of awareness, and its environmental indicators have three secondary indicators: environmental negative events, environmental disclosures, and environmental management. The environmental indicators of the Sino-Securities ESG evaluation system combine the enterprise operation and management concept and have five projects under it: green products, business objectives, external certification, internal management system, and violation events. The environmental dimension indicators of the China Securities ESG evaluation system focus on event disclosures, including themes such as environmental management, pollution and waste, environmental opportunities, climate change, and natural resources. The China Bond ESG evaluation system evaluates environmental performance through green development, ecological protection, pollution prevention and control, environmental management, resource use, and other aspects.

These evaluation systems maintain common attention to core issues, such as climate change, pollution emissions, and resource use, and reflect the differentiated characteristics that international institutions focus on regarding supply chain management; Chinese domestic systems emphasize policy compliance, reflecting the phased characteristics of ESG development in different market environments, which can be applied in different scenarios and purposes. However, overall, they are still the principles or indicators common to multiple industries, with insufficient pertinence, and the attention to the “Dual Carbon Goals” is not comprehensive or specific enough and cannot fully or accurately reflect the specific practice performance of the power generation format supporting the “Dual Carbon Goals”. Therefore, the environmental indicators are selected by emphasizing the impact evaluation of the power generation format on the environment under the “Dual Carbon Goals”, and the environmental index evaluation system of the power generation format is constructed under the “Double Carbon Goals”, as shown in

Table 4. Evaluation indicators for environmental dimensions of the power generation industry.

First-Level Index	Second-Level Index	Interpretation, Occurrence Link, Selection Reason	Attribute	Correlation
Contribution Potential to “Dual Carbon Goals”	Carbon Emission Intensity	CO2 emission per unit power generation, characterizing the CO2 emission level of the power generation format, directly reflecting the size of the contribution potential to the “Dual Carbon Goals”, covering the whole process.	Quantitative	Negative
	Carbon Capture Difficulty	Mainly depends on the concentration and pressure of emitted CO2. Generally, the higher the concentration and pressure, the easier it is to capture, mainly in the energy combustion power generation and transportation links.	Qualitative	Negative
	Carbon Emission Reduction Cost	Expenditure for reducing unit CO2 emissions. High costs will affect the development of CCUS source–sink matching.	Quantitative	Negative
	Carbon Sequestration Potential	Potential volume of CO2 sequestration, an important factor in CCUS source–sink matching.	Qualitative	Positive
	Waste Energy Utilization Opportunities	Utilization of waste heat and waste gas, saving other energy consumption and reducing carbon emissions, occurring in the mining or power generation link.	Qualitative	Positive
	Reuse Potential of Abandoned Assets (Underground)	Utilization of underground structural space and facilities, which can simultaneously recover solidified energy and carbon.	Qualitative	Positive
	Reuse Potential of Abandoned Assets (Ground)	Reuse of some large ground devices. Some power generation formats do not have the use scenario of underground abandoned assets but have the reuse potential of ground abandoned assets.	Qualitative	Positive
	Renewable Energy Compatibility Potential	Ability to integrate and develop renewable energy, such as solar energy and wind energy, and achieve multi-dimensional collaborative carbon reduction.	Qualitative	Positive
Pollutant Emissions	Wastewater Discharge Intensity	Wastewater discharge per unit power generation.	Quantitative	Negative
	Dust Emission (PM10)	Dust emission per unit power generation, mainly occurring in the coal transportation link.	Quantitative	Negative
	NOX and Sulfide (SOX, HS) Emissions	NOX and SOX contained in waste gas.	Quantitative	Negative
	COD Emissions	Organic pollutants contained in wastewater. The higher the COD content, the more serious the organic pollution.	Quantitative	Negative
	Waste Solid Emissions	Emissions of coal gangue, fly ash, tailings, etc.	Quantitative	Negative
Resource Consumption	Energy Consumption Intensity	Energy consumption per unit power generation (including electricity, coal, oil, and gas).	Quantitative	Negative
	Water Consumption	Water consumption per unit power generation.	Quantitative	Negative
	Water Saving Capacity	Water saving proportion of water consumption per unit power generation, reflecting the water saving and circular utilization capacity of the power generation format.	Quantitative	Positive
	Land Use	Land occupation area.	Quantitative	Negative

.

3.2. Structure Model of Hierarchical Evaluation

Taking the impact performance of different power generation formats on the ESG environmental dimension as the target layer (A), comprehensively selecting three main factor indicators of the impact of different power generation formats on the environment as the criterion layer (B), and carrying out gradual evaluation of the criterion layer, decompose the main factors in the criterion layer (B) into 17 key indicators as the index layer (X). Based on the Analytic Hierarchy Process, construct a hierarchical structure model for evaluating the impact performance of different power generation formats on the ESG environmental dimension, as shown in Figure 1.

3.3. Relative Importance Analysis

(1)

Value analysis of the A-B judgment matrix

The three first-level indicators, including the contribution potential to “Dual Carbon Goals”, pollutant emissions, and resource consumption, are selected as the criterion layer of this analysis. Among them, the “Dual Carbon Goals” contribution potential refers to the direct or indirect reduction in carbon emissions, reflecting the effectiveness in supporting the “Dual Carbon Goals”. In power enterprises that take low-carbon emission reduction as a long-term priority, if the power generation format performs poorly in the “Dual Carbon Goals” contribution potential, even if other pollutant emissions are low and the resource consumption is small, the implementation of the format model will be highly controversial or affected, so this indicator is relatively the most important in the criterion layer.

Although the pollution problem cannot be completely solved, it has been greatly improved compared with the past, and each format will basically be equipped with auxiliary recovery measures to reduce the impact on the environment, so the importance of this indicator is second. While ensuring sustainable development, reasonable and effective resource consumption is acceptable. With the increasing integration of renewable energy resources into carbon-based power generation and the implementation of water recycling technologies, the importance of resource consumption in the criterion layer is considered to be lower than that of environmental pollution.

The value distribution of the criterion layer A-B judgment matrix is as shown in

Table 5. Value distribution of criterion layer A-B judgment matrix.

A	B1	B2	B3
B1	1	7/3	7/2
B2	3/7	1	5/3
B3	2/7	3/5	1

.

(2)

Value analysis of the B1-X judgment matrix

Carbon emission intensity refers to the CO₂ emission per unit power generation, which is governed by technology difficulty in carbon capture, emission reduction cost, and sequestration potential. Generally, the higher the CO₂ emission concentration, the lesser the capture difficulty. According to different power generation processes, the suitable CO₂ capture methods and implementation costs are different. Therefore, the importance of the emission reduction cost is considered to be relatively high. The carbon sequestration potential reflects the volume of CO₂ that can be used for geological storage or other carbon fixation methods after the power generation format captures CO₂. The three are all important factors for subsequent CCUS and entering the carbon source and sink matching ranking, and CCUS is widely regarded as a key technical system indispensable for achieving the goal of carbon neutrality and has accumulated many practical application cases in the industrial field. Therefore, the importance of carbon capture difficulty and sequestration potential is equally important.

Co-produced energy utilization opportunities are other forms of energy accompanied by resource consumption or energy production. The utilization of this part of energy can reduce the consumption of main energy and indirectly reduce CO₂ emissions. For example, in the underground mining–IGCC unit power generation format, part of the waste heat can be used for heating if market demand exists. However, this part of the energy is minor, so the importance of this indicator is weak.

The potentially reusable abandoned assets in mining–power generation formats can be categorized into two parts: ground structure and underground structure. Comparing the two, the underground structure has a larger space and greater utilization potential. Currently, there are more research activities focusing on the utilization of abandoned mines for energy storage and carbon sequestration. Therefore, the reuse potential of abandoned assets in the underground structure part is slightly more important.

According to the analysis of the latest 2023 CO₂ Emission Report released by the IEA (International Energy Agency) [17,18], the wide application of renewable energy, such as hydropower, wind power, and solar power, has effectively reduced the global CO₂ emissions in the past five years. It can be seen that the compatibility potential with renewable energy on the existing power generation format helps with overall sustainable development, so the importance of this indicator is strong.

The value distribution of the index layer B1-X judgment matrix is as shown in

Table 6. Value distribution of criterion layer B1-X judgment matrix.

B1	X1	X2	X3	X4	X5	X6	X7	X8
X1	1	3	6/5	3	6	5	5	4
X2	1/3	1	1/2	9/8	8/3	7/3	7/3	4/3
X3	5/6	2	1	2	5	4	4	3
X4	1/3	8/9	1/2	1	7/3	7/4	7/4	5/3
X5	1/6	3/8	1/5	3/7	1	2/5	1/2	1/2
X6	1/5	3/7	1/4	4/7	5/2	1	6/5	4/5
X7	1/5	3/7	1/4	4/7	2	5/6	1	3/4
X8	1/4	3/4	1/3	3/5	2	5/4	4/3	1

.

(3)

Value analysis of the B2-X judgment matrix

Wastewater, waste gas, and waste solids are the more common indicators for measuring pollutant emissions in the current energy industry ESG environmental evaluation system. Wastewater and waste gas emissions are waste liquids and gases with pollutants discharged by the power generation format during energy consumption and production activities, which flow into the environment and directly or indirectly affect the balance of the natural ecosystem through relevant media, and are the main pollutants discharged by the power generation format. COD emissions characterize the content of organic pollutants in wastewater. The higher the content, the greater the harm to the natural environment, so the importance of this indicator is slightly higher than that of the wastewater discharge intensity indicator. Nitrogen oxides and sulfur oxides are the main substances of air pollution. Their gaseous deposits can form acid rain, which will destroy the soil structure; affect plant growth; cause crop death, land desertification, and other corrosive disasters; and migrate with monsoons and air currents, rapidly expanding the scope of disasters, so the importance of this indicator is the highest. Waste solid emissions are the solid environmental problems of the power generation format, which are highly harmful, large in quantity, and difficult to handle. The accumulation of a large number of solid wastes not only occupies land and endangers the soil but also pollutes water bodies and the atmosphere, thereby affecting environmental health in a wide range. Therefore, the importance of this indicator is relatively high.

The value distribution of the index layer B2-X judgment matrix is as shown in

Table 7. Value distribution of criterion layer B2-X judgment matrix.

B2	X9	X10	X11	X12	X13
X9	1	2/3	1/3	2/5	1/3
X10	3/2	1	2/3	1/2	2/3
X11	3	3/2	1	3/2	5/4
X12	5/2	2	2/3	1	4/5
X13	3	3/2	4/5	5/4	1

.

(4)

Value analysis of the B3-X judgment matrix

Energy consumption intensity reflects the use of resources such as coal, electricity, oil, and gas in the power generation format, which is comprehensive and has relatively high importance. Water use covers all links of fuel extraction, processing, and actual power generation in the power industry, and it is a veritable large water user, especially the underground mining–coal-fired unit power generation format, whose production water consumption accounts for 38–40% of industrial water consumption [19]. The water consumption in each link is closely related to energy consumption and CO₂ emissions [20,21], so the importance of the water consumption indicator is medium. The water resource recycling rate is the reverse performance of the water consumption index of the power generation format, reflecting the water saving capacity, but for the characteristics of each power generation format, the water saving volume is small, so the importance of this indicator is low. The construction of power infrastructure often requires the occupation of a large amount of land resources, resulting in environmental footprints, thereby having an adverse impact on biodiversity in the region. Therefore, the importance of this indicator is medium to high.

The value distribution of the index layer B3-X judgment matrix is as shown in

Table 8. Value distribution of criterion layer B3-X judgment matrix.

B3	X14	X15	X16	X17
X14	1	7/2	5	3
X15	2/7	1	3/2	2/3
X16	1/5	2/3	1	1/2
X17	1/3	3/2	2	1

.

4. Index Weighting and Scoring

4.1. Matrix Calculation and Consistency Test

The root method is used to solve the judgment matrix to obtain the characteristic vector results. After consistency verification, the weight distribution of each level of indicators is finally determined. The detailed weight distribution is shown in

Table 9. Judgment matrix calculation results.

Matrix	Weight Vector ξ	λmax	CI	RI	CR	Consistency Test
A-B	(0.5815, 0.2581, 0.1604) T	3.0012	0.0006	0.52	0.0012	Passed
B1-X	(0.3011, 0.1181, 0.2318, 0.1079, 0.0392, 0.0647, 0.0596, 0.0775) T	8.0749	0.0107	1.41	0.0076	Passed
B2-X	(0.0917, 0.1487, 0.2838, 0.2254, 0.2503) T	5.0439	0.0110	1.12	0.0098	Passed
B3-X	(0.5459, 0.1483, 0.1030, 0.2028) T	4.0070	0.0023	0.90	0.0026	Passed

.

4.2. Calculation Results

After determining the weights of each single-level indicator, set the weights of irrelevant indicators to zero. According to Formulas (3) and (7), the characteristic vector of the index layer (X) at the target layer (A) can be obtained. After the overall consistency test, the index weight distribution results of the ESG environmental dimension evaluation system are obtained. The specific values are shown in

Table 10. The overall evaluation index weight of environmental dimension.

		B	B1	B2	B3	ξA(Xi)
	ξB(Xi)		0.5815	0.2581	0.1604
X
X1			0.3011	0	0	0.1751
X2			0.1181	0	0	0.0687
X3			0.2318	0	0	0.1348
X4			0.1079	0	0	0.0628
X5			0.0392	0	0	0.0228
X6			0.0647	0	0	0.0376
X7			0.0596	0	0	0.0347
X8			0.0775	0	0	0.0451
X9			0	0.0917	0	0.0237
X10			0	0.1487	0	0.0384
X11			0	0.2838	0	0.0733
X12			0	0.2254	0	0.0582
X13			0	0.2503	0	0.0646
X14			0	0	0.5459	0.0876
X15			0	0	0.1483	0.0238
X16			0	0	0.1030	0.0165
X17			0	0	0.2028	0.0325

.

After calculation, the overall consistency index CI_A(X) is 0.0094, the overall random consistency index RI_A(X) is 1.2533, and the overall consistency ratio CR_A(X) is 0.0075 < 0.1, so the overall consistency test is passed.

4.3. Scoring Criteria

The constructed environmental index evaluation system for carbon-based power generation formats is used to evaluate the whole process environmental performance of each format. Among them,

The format of “underground mining–coal-fired unit” is named as PC;

The format of “underground coal gasification–combined cycle unit” is marked as UCGCC;

The format of “imported natural gas–combined cycle unit” is marked as NGCC;

The format of “imported shale gas–gas + steam combined cycle unit” is recorded as SNGCC.

The evaluation system uses a 10-point system. The scores of each indicator are evaluated according to the previous analysis combined with the actual operational performance of each format. The point of each index is assigned as follows: good (9–10), moderate-to-good (7–8), moderate (5–6), poor-to-moderate (3–4), and poor (1–2). The scoring forms were distributed to five experts in related fields, including one professor, one associate professor, two senior engineers, and one research associate. Assigned points by the experts are collected, and average assigned scores are calculated in

Table 11. Scoring of indicators.

First-Level Index	Second-Level Index	UCGCC	PC	NGCC	SNGCC
Contribution Potential to “Dual Carbon Goals”	Carbon Emission Intensity	3	4	2	7
	Carbon Capture Difficulty	3	5	5	8
	Carbon Emission Reduction Cost	2	5	4	7
	Carbon Sequestration Potential	2	4	2	8
	Waste Energy Utilization Opportunities	4	5	4	0
	Reuse Potential of Abandoned Assets (Underground)	5	3	5	1
	Reuse Potential of Abandoned Assets (Ground)	8	5	8	0
	Renewable Energy Compatibility Potential	3	7	3	7
Pollutant Emissions	Wastewater Discharge Intensity	4	6	4	4
	Particle Emission (PM10)	1	8	4	8
	NOX and Sulfide (SOX, HS) Emissions	2	2	2	6
	COD Emissions	2	6	5	8
	Waste Solid Emissions	1	8	5	9
Resource Consumption	Energy Consumption Intensity	2	4	4	6
	Water Consumption	3	6	5	6
	Water Saving Ability	7	3	6	3
	Land Occupation	2	8	2	8

.

In terms of carbon emission intensity, after UCGCC improves the proportion of effective components of syngas, it will be lower than the high emission of PC combustion. NGCC performs the best due to clean and low-carbon fuel, and SNGCC is slightly higher than NGCC due to mining technology and transportation methods. Existing studies have shown that the UCGCC carbon emission reduction costs and capture costs have outstanding advantages, being only about half of those of NGCC power generation.

In terms of carbon capture difficulty, UCGCC has the lowest capture difficulty due to the high concentration of CO₂ in the syngas in the power generation link. PC has a medium capture difficulty due to the low concentration of CO₂ in the coal-fired flue gas. NGCC and SNGCC have the lowest concentrations before capture and the greatest capture difficulties.

UCGCC has the strongest carbon storage potential and can be directly used for geological sequestration or chemical utilization (such as urea and methanol production) without complicated pretreatment. The CO₂ of PC is insufficient in purity and difficult to use directly in chemistry, relying on depleted reservoirs. NGCC and SNGCC have limited potential due to low CO₂ concentrations and need to cooperate with other industrial projects (such as cement plants) or sequestration projects to achieve cross-industry utilization of CO₂.

In terms of waste heat utilization, NGCC and SNGCC can use wellhead pressure energy for power generation and fracturing fluid thermal energy recovery during resource extraction, use LNG cold energy for power generation and electric drive compressors to match renewable energy during storage and transportation, and use the waste heat of steam exhaust for heating. UCGCC recovers waste heat from the produced syngas for heating and recovers waste heat from the combined cycle. PC can only use low-concentration mine gas and has no significant waste energy recovery in the storage and transportation processes.

In terms of the reuse of abandoned assets, the underground goaf formed by UCGCC coal mining has a high porosity and can be directly transformed into a CO₂ storage tank or a gas storage tank. However, compared with PC, the underground space formed by UCGCC is deeper and with higher stress, making it more difficult to use than shallow underground coal mining. The production wells of SNGCC can be used as CO₂ injection wells for enhanced production. The depleted reservoirs in the NGCC format can be used for CO₂ sequestration, but they need to cooperate with oil extraction. For ground facilities, PC has a strong versatility due to its compatible equipment (boiler and turbine) and large space; NGCC (SNGCC) has a high reuse flexibility relying on modular equipment and hydrogen energy compatibility; UCGCC is limited by the specialty of syngas-fed unit, with high transition costs, but can be made into mobile units.

The compatibility with renewable energy is the best for NGCC and SNGCC, with fast response to the command from grid operation and flexible participation in the energy internet [22]. Some equipment, such as compressors in the extraction and transportation sector, can be driven by green energy, and the flexibility of gas turbines can provide the strongest support for the grid to accommodate green energy. UCGCC can use part of green power for gasification equipment, and the power generation link can mix biomass, but the rigidity of the gasification process limits its flexibility. PC equipment has the worst compatibility due to the long start-up duration of the boiler.

Underground mining of coal requires coal washing and power generation cooling cycles, resulting in more wastewater discharge. A considerable part of shale gas extraction in the market requires a large amount of water for hydraulic fracturing, and the mixed fracturing fluid is discharged. UCGCC gasification discharges wastewater, and NGCC discharges little wastewater in the whole cycle. In terms of dust emissions, PC coal transportation, loading and unloading, and boiler emissions are the highest. UCGCC combustion produces a small amount of dust, and its gasification and SNGCC fracturing dust emissions are relatively limited. NGCC combustion has the lowest dust emissions due to no ash content. As for NOx/SOx emissions, PC coal combustion produces high-concentration pollutants. UCGCC needs to carry out desulfurization and denitrification treatment. SNGCC fracturing equipment emissions are relatively low, and NGCC has the best control effect with low-nitrogen combustion technology and SCR (Selective Catalytic Reduction). In terms of COD, although the fracturing fluid of SNGCC has a high organic matter concentration, the recovery rate can reach 70%. The coal chemical wastewater of UCGCC has a high COD content, and the PC mine water contains high-concentration organic matter. In terms of waste solid generation, PC has the largest output of fly ash and desulfurized gypsum. UCGCC has less ash residue. SNGCC fracturing sand can be partially reused, and NGCC has no waste solid generation due to no ash content in combustion.

PC has the highest energy consumption due to its having the lowest thermal efficiency and a high energy consumption in coal mining and transportation. UCGCC ranks second due to additional energy consumption in the gasification. SNGCC has higher energy consumption than NGCC due to high energy consumption in fracturing and reservoir treatment. NGCC has the lowest energy consumption in the whole cycle. UCGCC realizes a high water saving rate through gasification wastewater reuse. SNGCC relies on fracturing fluid recovery, reaching 70%, but the fracturing still needs to consume a large amount of fresh water, and the comprehensive water saving proportion is limited by the extraction process. The cooling water of the PC unit has great potential for circular utilization, but the overall water consumption of the coal-fired process is large. NGCC has a small water saving potential due to low water consumption. PC mining areas and power plants cover a large area, and UCGCC, NGCC, and SNGCC only occupy land by power generation facilities based on the previous assumptions.

5. Evaluation Results

The scores of each indicator are shown in Table 11, and the evaluation results are shown in

Table 12. Evaluation results of ESG environmental indicators of power generation.

First-Level Index	Second-Level Index	UCGCC	PC	NGCC	SNGCC
Contribution Potential to “Dual Carbon Goals”	Carbon Emission Intensity	0.875	0.525	1.401	1.226
	Carbon Capture Difficulty	0.412	0.275	0.137	0.137
	Carbon Emission Reduction Cost	1.078	0.539	0.404	0.404
	Carbon Sequestration Potential	0.439	0.314	0.126	0.188
	Waste Energy Utilization Opportunities	0.091	0.068	0.137	0.137
	Reuse Potential of Abandoned Assets (Underground)	0.150	0.188	0.075	0.113
	Reuse Potential of Abandoned Assets (Ground)	0.104	0.277	0.173	0.173
	Renewable Energy Compatibility Potential	0.270	0.135	0.361	0.361
Pollutant Emissions	Wastewater Discharge Intensity	0.118	0.095	0.213	0.095
	Dust Emission (PM10)	0.269	0.038	0.346	0.307
	NOX and Sulfide (SOX, HS) Emissions	0.293	0.147	0.440	0.366
	COD Emissions	0.349	0.116	0.524	0.407
	Waste Solid Emissions	0.517	0.065	0.581	0.517
Resource Consumption	Energy Consumption Intensity	0.350	0.175	0.613	0.525
	Water Consumption	0.143	0.071	0.166	0.143
	Water Saving Ability	0.099	0.099	0.050	0.116
	Land Occupation	0.260	0.065	0.260	0.260
Total Score		5.820	3.193	6.006	5.476

.

6. Discussion

The environmental performance evaluation of carbon-based power generation formats reveals critical insights into their sustainability trade-offs and transition potential. Among the four formats assessed—PC, UCGCC, NGCC, and SNGCC—each demonstrates distinct advantages and challenges across carbon management, resource efficiency, and system flexibility, painting a nuanced picture for energy transition strategies.

UCGCC emerges as a technologically compelling compromise between traditional coal and cleaner gas systems. Its syngas optimization enables carbon emissions significantly lower than PC, though still higher than gas-based alternatives. The true differentiation lies in carbon capture economics: UCGCC’s high-pressure flue gas stream allows capture costs lower than PC, while its underground cavities offer potential geological storage, positioning it as a transitional solution for coal-reliant regions. Such advantages would be more pronounced under the incentive mechanism of carbon emission trading policy [23]. However, this promise is tempered by the gasification process’s inherent rigidity, which limits renewable energy integration compared to the NGCC and SNGCC. The modular infrastructure of natural gas-based formats also presents greater adaptability for hydrogen blending and grid-balancing roles in renewable-heavy energy systems.

The difference in environmental performance is obvious when examining emissions profiles. PC’s challenges persist across all metrics, from coal washing wastewater to fly ash generation, while UCGCC demonstrates measurable improvements through gasification wastewater recycling and reduced solid waste. Yet even these advances pale against NGCC’s emissions performance, where low-nitrogen combustion and SCR systems achieve near-zero NO_X/SO_X emissions without desulfurization byproducts. The high water intensity associated with reservoir treatment in SNGCC reflects the ongoing environmental trade-offs of unconventional gas extraction. The environmental impact of pollutant emissions is heterogeneous in time and space. If the environmental sensitivity of the regions where the emission sources are located is considered, the advantages of low-emission formats in some environmentally fragile regions would be more pronounced.

Infrastructure repurposing potential reveals another dimension of the energy transition. UCGCC’s deep coal provides potential storage capacity, but unknown practical risks exist, as few field tests have been conducted. To compare, SNGCC’s readily convertible production wells and CO₂ injection operation offer certainty. This contrasts with PC’s surface assets, where conventional power plant equipment offers broad compatibility, including Carnot Batteries.

Five experts in industry research were invited to assign scores, and there is inevitable subjective dependence in point assignment from the experts. Major dispersed scores are in “X5: waste energy utilization opportunities”, “X6: reuse potential of abandoned assets (underground)”, and “X7: reuse potential of abandoned assets (ground)”. Further communication with the consulted experts indicates that the major concern is that the “opportunity” and “potential” might refer to the technical availability but not necessarily the economic feasibility. This is where the divergence originated. As presented in Section 3.3, the relative importance of X5–X7 is much smaller than X1–X4. Therefore, the evaluation results still bear academic value. In the future, it is suggested to expand the pool of consulted experts and introduce objective weighting methods to verify or supplement the AHP results, hence mitigating the issue of subjectivity.

These findings carry significant implications for differentiated decarbonization policies. Regions with stranded coal assets might prioritize UCGCC development to leverage existing mining infrastructure and geological storage potential, while gas-importing economies could accelerate NGCC deployment as a bridge technology. The analysis underscores the fact that no single format delivers a perfect solution. Strategic deployment should match regional resources, existing infrastructure, and renewable integration timelines. Regulatory frameworks should evolve beyond uniform carbon pricing to technology-specific incentives that recognize UCGCC’s capture advantages and NGCC/SNGCC’s grid flexibility contributions, ensuring the energy transition maintains both environmental integrity and system reliability.

7. Conclusions

Combining the existing literature and based on ESG theoretical practices, this paper innovatively constructs an environmental evaluation index model. The model emphasizes the contribution potential to the “Dual Carbon Goals”, uses the Analytic Hierarchy Process to analyze and calculate the influencing factors, assigns index weights, and draws the following main conclusions:

(1)

Through reviewing the existing ESG evaluation systems, it is found that core issues such as climate change, pollution emissions, and resource use have universality in environmental performance evaluation.

(2)

On the basis of common issues, characteristic indicators focusing on the “Dual Carbon Goals” are proposed, the evaluation to the impact of the power generation format on the environment under the “Dual Carbon Goals” is emphasized, and an ESG environmental index evaluation system for carbon-based power generation formats is constructed, which consists of 3 first-level indicators of “Dual Carbon Goals” contribution potential, pollutant emissions, and resource consumption and 17 second-level indicators such as carbon emission intensity and carbon emission reduction cost.

(3)

The relative importance of indicators at all levels is analyzed in detail, a judgment matrix is established, and the comprehensive weights of influencing factors are calculated and determined. The carbon emission intensity and carbon emission reduction cost are the main influencing factors of the evaluation system, with weights reaching 0.1751 and 0.1348, respectively, and the comprehensive weights of other influencing factors are all below 0.1.

(4)

Environmental evaluation is carried out on each format. UCGCC with highly effective components has a better overall environmental performance, slightly lower than NGCC and higher than SNGCC, while PC performs poorly. For regions where coal remains the major energy resource, the UCGCC format not only has great competitive potential in carbon emission reduction and environmental impact, but it also has great advantages in coupling CCUS technology.

(5)

The framework’s emphasis on common ESG indicators creates a foundation for joint research or policy harmonization. Collaborative projects with different organizations could standardize evaluation methodologies, enabling comparative studies of transitional technologies across regions. The model’s Dual Carbon-centric design complements global initiatives like the Paris Agreement, offering a template for regions with rich fossil fuel resources reliant on carbon-based power.