Optimization of residual coalbed methane extraction wells and analysis of development and emission reduction benefits in the Songzao mining area

Abstract

The utilization of coalbed methane (CBM) resources in closed coal mines is crucial for energy transition and green low-carbon development. This study systematically analyzed geological factors, CBM reserves, and economic factors to evaluate the potential of CBM extraction from the Songzao mining area of Chongqing. An evaluation indicator system was established, comprising 3 primary indicators, 13 secondary indicators, and 5 closed coal mines as the scheme layer. This system facilitated accurate quantification of feasibility evaluation weights for CBM extraction under complex geological conditions. The game theory was employed to refine the integration of subjective and objective weights, which ensured the objectivity and accuracy of the evaluation results. A feasibility evaluation model for CBM extraction from goafs in closed coal mines was constructed using the TOPSIS method, with results ranked as follows: Songzao coal mine > Fengchun coal mine > Shihao coal mine > Yuyang coal mine > Datong No. 1 coal mine. Furthermore, the trends in energy conservation and emission reduction benefits over the past eight years since the closure of the Songzao coal mine were analyzed, encompassing both CBM emissions and power generation. The findings highlight the necessity of CBM extraction from goafs in closed coal mines in China and demonstrate the significant potential of CBM resource development for reducing greenhouse gas emissions and promoting clean energy alternatives, providing a reference for theoretical research and practice.

Introduction

Environmental issues, such as the greenhouse effect, have prompted many countries to reduce their reliance on coal. This shift has led to a decline in coal production and consumption, resulting in the closure of numerous coal mines. In some countries, such as France and Germany, the coal industry has been completely phased out^1,2,3. Consequently, a significant number of mines have been closed, many of which contain substantial untapped resources. If these underground resources remain sealed without proper utilization, they will not only constitute a significant waste of valuable resources but may also trigger a series of environmental and safety issues^4,5,6. Coalbed methane (CBM), primarily composed of methane, is one of the most important energy sources. Methane, as a potent greenhouse gas, has a global warming potential more than 28 times greater than that of CO₂, making it the second most impactful greenhouse gas after CO₂⁷. However, methane is also a clean energy source, and its effective utilization can significantly contribute to the development of green energy. In 2023, China accounted for 18% of the global methane emissions, making it the largest contributor to CH₄ emissions⁸. Additionally, the number of closed coal mines in China continues to rise. The low recovery rate of coal causes residual CBM in abandoned coal mines, which may lead to hazards such as explosion, gas leakage, and air pollution^9,10,11. Therefore, it is crucial to implement preventive and control measures to mitigate these risks.

A series of studies have explored the CBM extraction from closed coal mines. Jerzy Hadro et al.¹² employed a novel method based on post-mining emission predictions to assess the resource potential of the Upper Silesian Coal Basin AMM. They provided valuable conclusions and recommendations for its future development. Sechman H et al.¹³ identified the causes, sources, and characteristics of the periodic variations in CH₄ and CO₂ concentrations in the near-surface area following the closure of coal mines in the Wałbrzych coal basin. Adam Duda et al.¹⁴ utilized a model developed by a French research institute to estimate CH₄ emissions from loosely consolidated, damaged, and over-exploited coal seams. C. Özgen Karacan¹⁵ focused on gas extraction from the abandoned Springfield coal mine in Indiana and conducted reservoir modeling studies to examine potential variations in gas emissions and seal leakage. Meng Zhaoping¹⁶ established an evaluation model for CBM resources in abandoned mines and applied it to assess the CBM resources in the goaf of Jinsheng Coal Mine. Liang Yunpei et al.¹⁷ reviewed methods for estimating sources and quantities of residual CBM in extraction areas, analyzed the mechanisms and monitoring techniques for residual methane emissions, and proposed strategies and challenges for methane reduction. Ma Cuimei et al.¹⁸ calculated the CH₄ emissions from coal mining, post-mining activities, abandoned mines, and open-pit mining in China from 2010 to 2016. They also analyzed the trends and composition of these emissions. Yang Han et al.¹⁹ examined the availability of CBM resources in the abandoned mining areas of Shanxi Province. Furthermore, they analyzed the necessity and feasibility of their development and utilization. Liu Wenge et al.²⁰ estimated the historical methane emission trends from Chinese coal mines using the T3 method, developed a primary factor analysis model for methane emissions, and projected methane emissions until 2060. Liu Xiaolei et al.²¹ proposed a “resource estimation + preliminary evaluation” framework for estimating and evaluating gas resources in abandoned coal mines. In summary, most studies on CBM extraction from closed mines have focused on resource potential estimation, gas concentration variation analysis, and emission prediction. However, a systematic evaluation framework for assessing the feasibility of CBM extraction remains lacking. Moreover, there is insufficient research on the energy-saving and emission-reduction benefits of CBM extraction from specific closed coal mines.

The Songzao mining area of Chongqing, a high-gas mining region, still has abundant CBM potential. However, these resources have not been effectively exploited, resulting in resource wastage and environmental concerns. Therefore, it holds significant theoretical and practical value to assess CBM resources in the goaf of the Songzao mining area of Chongqing and to evaluate extraction feasibility.

This study analyzed the geological conditions, goaf degree, and water accumulation in five coal mines within the Songzao mining area of Chongqing. It estimated the residual CBM resources in the extraction areas of these coal mines, as well as the construction costs and expected revenues from CBM extraction projects. A comprehensive evaluation model was established to assess the feasibility of CBM extraction from the five coal mines. Based on the evaluation results, the CBM emissions and power generation potential from the closed Songzao Coal Mine over an eight-year period were estimated. The findings provide a foundation for the efficient development of CBM in the goaf of this mining area and highlight the necessity of extracting CBM from abandoned coal mines.

Materials and methods

Project overview

The Songzao mining area, located in Qijiang District of Chongqing, is one of the largest anthracite production bases in Southwest China and serves as the primary thermal coal production hub in the region. The mining area was established in August 1958 and benefits from convenient transportation, with the Chongqing-Guizhou Railway and Expressway passing through it. The mining area features coal seams consisting of 6–14 layers, low permeability, and abundant CBM reserves. It is a mining area with soft, low-permeability, and complex geological conditions in China. The current research primarily focused on five specific mines: Songzao Coal Mine, Yuyang Coal Mine, Fengchun Coal Mine, Datong No.1 Coal Mine, and Shihao Coal Mine, all of which are classified as coal and gas outburst wells. The Datong No.1, Shihao, and Yuyang Coal Mines exploit the gently inclined coal seams in the west wing of the study area, while the Fengchun Coal Mine operates on a steeply inclined coal seam in the southeast wing, and Songzao Coal Mine extracts from an inclined coal seam in the northeast wing, as depicted in (Fig. 1).

Fig. 1

The relative locations of various coal mines in the Songzao mining area. (the figure drawn by adobe photoshop CC 2019).

Feasibility evaluation methods

Analytic hierarchy process (AHP), entropy method, TOPSIS and other multivariate decision analysis methods have been widely applied across various industries, effectively addressing the challenges of multi-factor and multi-level optimization. These methods have been utilized in applications such as optimizing offshore pipeline routing, assessing inundation risks in subway system, and conducting comprehensive evaluation of construction projects^22,23,24. Similarly, the optimization of CBM extraction from goafs in the Songzao mining area falls within the domain of complex decision analysis.

The feasibility evaluation of CBM extraction from goafs in the Songzao mining area involved a three-step procedure. First, it is essential to select appropriate evaluation indicators and construct a comprehensive evaluation indicator system. Second, the determination of indicator weights employed three primary methods. AHP, a subjective weighting method, depends on the expert judgment, while the entropy method is an objective weighting method influenced by the quality of data. However, the entropy method may exhibit some limitations in overall planning. To address these shortcomings, the game theory was used to integrate the weights derived from the AHP and entropy method. This integration effectively overcomes the limitations of any single evaluation method and enhances the accuracy of the results. Finally, the TOPSIS evaluation model was established based on the determined indicator weights. This model quantified the feasibility of CBM extraction by measuring how closely the evaluation indicator aligned with the positive ideal solution. The methodological flow is illustrated in (Fig. 2).

Fig. 2

Flow chart of feasibility evaluation of CBM extraction from goafs in the Songzao mining area. (the 3D image drawn by SketchUp Pro 2022).

Analytic hierarchy pocess

Establishment of judgment matrix

Each factor at the same hierarchical level is compared with each other to establish a judgment matrix. The quantitative relationships among these factors are presented in (Table 1).

Table 1 Principle of the 9-scale method.

Single ranking within the hierarchy

After completing the construction of the matrix, the feature vector is derived through appropriate calculations. This feature vector represents the weight value indicating the degree of influence of each factor on a particular factor from the previous level, which is referred to as the single ranking result.

Consistency test of the weight vector

The maximum eigenvalue $\:{\lambda\:}_{max}$ is first calculated, and then the test index is determined.

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

(1)

where n is the order of matrix.

A test index of CI = 0 indicates a pass in the consistency test; if CI ≠ 0, the random consistency ratio needs to be calculated.

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

(2)

where RI is the mean random consistency, with the value presented in (Table 2).

If CR ≤ 0.1, the weight vector passes the consistency test. If CR > 0.1, the weight vector fails this test, requiring adjustments to the judgment matrix and re-evaluation until CR ≤ 0.1.

Table 2 Random consistency index values.

Entropy method

(1) The original data for j indicators across i samples are used to construct the initial evaluation matrix.

$$B = \left( {\begin{array}{*{20}c} {b_{{11}} } & \ldots & {b_{{1j}} } \\ \vdots & \ddots & \vdots \\ {b_{{i1}} } & \cdots & {b_{{ij}} } \\ \end{array} } \right)$$

(3)

(2) The initial evaluation matrix is normalized using the range method to obtain matrix $\:P=\left({p}_{ij}\right)$.

For positive indicators, a larger value indicates a better impact on the evaluation results, the standardization formula is as follows:

$$\:{P}_{ij}=\frac{{b}_{ij}-min\left({b}_{i}\right)}{max\left({b}_{i}\right)-min\left({b}_{i}\right)}$$

(4)

Conversely, for the negative indicators, the standardized formula is:

$$\:{P}_{ij}=\frac{\text{max}\left({b}_{i}\right)-{b}_{ij}}{\text{max}\left({b}_{i}\right)-\text{min}\left({b}_{i}\right)}$$

(5)

(3) Based on the definition of information entropy, the information entropy acquisition formula for a set of data is:

$$e_{i} = - \frac{1}{{\ln n}}\mathop \sum \limits_{{j = 1}}^{n} \left( {P_{{ij}} \times {\text{ln}}P_{{ij}} } \right),\;{\text{where}}\left( {0 \le e_{i} \le 1} \right)$$

(6)

(4) The information utility value d_i and index weight W_i of each index can be calculated using Eqs. (7) and (8).

$$\:{g}_{i}=1-{e}_{i}$$

(7)

$$\:{W}_{i}=\frac{{g}_{i}}{\sum\:_{i=1}^{m}\:{g}_{i}}$$

(8)

Game theory

The game theory-based comprehensive weighting method combines the weights obtained through the AHP and the entropy method to minimize the range of subjective and objective weights. The specific calculation process is outlined below:

(1) Determine the comprehensive weight coefficient vector for each evaluation index$\:\:{W}_{i}^{T}$.

$$W_{i}^{T} = \beta _{1} V_{i}^{T} + \beta _{2} Z_{i}^{T} ,i = 1,...,{\text{m}}$$

(9)

where $\:{V}_{i}^{T}$ and $\:{Z}_{i}^{T}$ are the weight vectors of evaluation indexes derived from AHP and entropy method, respectively.

$\:{\beta\:}_{1}$ and $\:{\beta\:}_{2}$ are the coefficients representing subjective and objective weights, respectively.

To appropriately allocate the weights determined by both methods, a cross-programming model was introduced to optimize the two linear combination coefficients, $\:{\beta\:}_{1}$ and $\:{\beta\:}_{2}$.

$$min\left\| {\beta _{1} V_{i}^{T} + \beta _{2} Z_{i}^{T} } \right\|_{2} {\text{,}}\;{\text{i = 1,}}...{\text{,}}\;{\text{m}}$$

(10)

(2) The optimal first derivative condition is:

$$\:\left\{\begin{array}{c}{\beta\:}_{1}{V}_{i}^{T}{V}_{i}+{\beta\:}_{2}{V}_{i}{Z}_{i}^{T}={V}_{i}{V}_{i}^{T}\\\:{\beta\:}_{1}{Z}_{i}{V}_{i}^{T}+{\beta\:}_{2}{Z}_{i}{Z}_{i}^{T}={Z}_{i}{Z}_{i}^{T}\end{array}\right.$$

(11)

The system of linear equations can be obtained:

$$\:\left(\begin{array}{cc}{V}_{i}{V}_{i}^{T}&\:{V}_{i}{Z}_{i}^{T}\\\:{Z}_{i}{V}_{i}^{T}&\:{Z}_{i}{Z}_{i}^{T}\end{array}\right)\left[\begin{array}{c}{\beta\:}_{1}\\\:{\beta\:}_{2}\end{array}\right]=\left[\begin{array}{c}{V}_{i}{V}_{i}^{T}\\\:{Z}_{i}{Z}_{i}^{T}\end{array}\right]$$

(12)

TOPSIS method

(1) Weighted normalized decision matrix.

The weighted normalized decision matrix, denoted as matrix U, is obtained by multiplying the normalized matrix P with the weight W for each index, derived from the game theory-based comprehensive weighting method.

$$U = W_{m} *P_{{nm}} = \left( {\begin{array}{*{20}c} {W_{1} P_{{11}} } & {W...P...} & {W_{m} P_{{1m}} } \\ {W...P...} & {W...P...} & {W...P...} \\ {W_{1} P_{{n1}} } & {W...P...} & {W_{m} P_{{nm}} } \\ \end{array} } \right)$$

(13)

(2) Determining positive and negative ideal points q⁺ and q⁻.

In the decision matrix U, the vector formed by the maximum elements of each column represents the positive ideal point q⁺, while the vector formed by the minimum elements of each column represents the negative ideal point q⁻.

$$\left\{ {\begin{array}{*{20}c} {q^{ + } = \left( {q_{1}^{ + } ,q_{2}^{ + } ,...q_{n}^{ + } } \right),q_{i}^{ + } = max\left\{ {q_{{nm}} } \right\},i = 1,2,...m} \\ {q^{ - } = \left( {q_{1}^{ - } ,q_{2}^{ - } ,q_{n}^{ - } } \right),q_{i}^{ - } = min\left\{ {q_{{nm}} } \right\},i = 1,2,...m} \\ \end{array} } \right.$$

(14)

(3) Calculating the distance D⁺ and D⁻ from each definition point u_i to the ideal solution.

$$\:\left\{\begin{array}{c}{D}^{+}=\sqrt{\sum\:_{i=1}^{n}\:({q}_{nm}-{{q}_{m}}^{+}{)}^{2}}\\\:{D}^{-}=\sqrt{\sum\:_{i=1}^{n}\:({q}_{nm}-{{q}_{m}}^{-}{)}^{2}}\end{array}\right.$$

(15)

where a smaller D⁺ indicates a closer proximity the positive ideal solution, while a smaller D⁻ indicates a closer proximity to the negative ideal solution.

(4) Calculating the closeness C_i:

$$C_{i} = \frac{{D_{i}^{ - } }}{{\left( {D_{i}^{ + } + D_{i}^{ - } } \right)}},i = 1,2,...n$$

(16)

where the closeness degree C_i of the evaluation object reflects the extent to which its overall index approaches the positive ideal solution. The value of C_i ranges between 0 and 1.

Calculating of different methods

Estimation of CBM emission

A new calculation method, proposed by Di Chen, Kholod, Gang Lin et al.^25,26,27,28, was employed to estimate the CBM emissions from abandoned coal mines. This method classifies abandoned coal mines into two types: dry coal mines and flooded coal mines. It takes into account various factors, such as the initial methane flow at the time of abandonment and the duration of abandonment, to calculate CBM emissions from abandoned coal mines. The calculation formulas are as follows:

$$\:{E}_{i}={q}_{0}\times\:E{F}_{i}\times\:\rho\:$$

(17)

$$\:E{F}_{dry}={\left(1+{b}_{dry}\times\:{D}_{dry}\times\:t\right)}^{-\frac{1}{{b}_{dry}}}$$

(18)

$$\:E{F}_{flooded}=exp(-t\times\:{D}_{flooded})$$

(19)

where E_i is the CBM emissions from type i mines; q_₀ is the initial CBM from the goaf in abandoned coal mines (m³/year); EF_i is the dimensionless emission factor for abandoned coal mine; ρ is the density of CH₄ (measured at 25 °C and 1 atm pressure, with a value of 0.67 kg/m³); b is the dimensionless rate of decline over time (with a value of b_dry=2.017); D_i is the initial decline rate (1/year), where D_dry=0.302, D_flooded=0.672 ²⁹; t is the number of years since t₀.

Estimation of CBM power generation

CBM power generation not only mitigates greenhouse gas emissions and alleviates environmental pressure but also maximizes resource utilization. Furthermore, as a renewable energy source, it helps reduce reliance on fossil fuels and promotes the diversification and sustainable development of the energy structure. Given these advantages, calculations were made to estimate the potential power generation based on the anticipated methane emissions.

For the Songzao mining area, the CH₄ content was assumed to exceed 95%, with a calorific value greater than 8000 kcal/m³. The combustion value of 1 m³ of CBM was equivalent to 1.14 kg of standard coal. The theoretical power generation (denoted as d) ranged between 3.2 and 3.3 kWh, and it was assigned a value of d = 3.25 kWh in this study²⁶. The expected CBM power generation is given by Eq. (20).

$$\:W=E/\rho\:\times\:d(1-{\beta\:}_{2})$$

(20)

where d is the theoretical power generation from 1 m³ of CBM (kWh); E is the CBM emission (t); β₂ is the proportion of self-consumed electricity by the CBM generator set; W is the expected power generation from CBM (kWh).

Results and discussion

Feasibility evaluation

Establishment of evaluation system

Due to variations in mining conditions across different mines, the construction methods and techniques employed must be differentiated within the construction system. CBM resources serve as a critical index for assessing project feasibility and are a primary consideration in the development of CBM in goafs. Given the disparities highlighted, the economic viability of different coal mines also varies significantly. Therefore, economic conditions were incorporated into the evaluation system to ensure a comprehensive assessment of project feasibility and sustainability.

Based on the aforementioned analyses, an evaluation indicator system was established, comprising 3 primary indicators, 13 secondary indicators, and 5 closed coal mines as the scheme layer, as illustrated in (Fig. 3).

Fig. 3

Evaluation system for CBM extraction from goafs in coal mines.

Mining condition B₁

The efficiency of CBM extraction is directly related to the factors such as the stability of geological structure, the buried depth and dip angle of coal seams, the tendency of spontaneous combustion, and the area of goafs. A complex geological structure may result in an uneven distribution of CBM, thereby complicating the extraction process. While deeper coal seams facilitate the preservation of CBM, they also necessitate advanced extraction techniques. Variations in seam inclination influence the migration and accumulation patterns of CBM. According to current national standards, oxygen uptake (V_d) is classified into three grades: Grade I (V_d>0.7 cm³/g, prone to spontaneous combustion), Grade II (0.4 < V_d<0.7 cm³/g, spontaneous combustion), and Grade III (V_d < 0.4 cm³/g, not prone to spontaneous combustion)²⁹. Higher grades indicate an increased likelihood of spontaneous combustion, thereby elevating the associated risks.

Coalbed methane resources B₂

The primary factors influencing CBM resources include CBM emission, coal seam thickness, residual coal resources, and the abundance of CBM, which collectively provide a satisfactory resource base for extraction. However, water accumulation in goafs and confined water outflows in overlying strata may result in low gas production or gas extraction challenges during the extraction process³⁰. The distribution of unaccumulated space determines the mining extraction pathways and channels. The appropriate management of water accumulation can be achieved through drainage measures, while the unaccumulated spaces facilitate the extraction of CBM.

Economic conditions B₃

Construction cost C₁₂

In some coal mines within the mining area, significant water accumulation necessitates drainage prior to the commencement of CBM extraction. Consequently, the construction costs associated with CBM extraction encompasses four components: development well investment, ground construction investment, basic reserve fund, and drainage cost.

$$\:\:{P}_{C}={P}_{W}+{P}_{G}+{P}_{B}+{P}_{D}\:$$

(21)

where P_W is the development well investment. The cost of drilling and completing a well with a depth of 500 m is 1.4 million RMB. The cost increases by 1000 RMB for every 1-meter increase in well depth³¹.

P_G is the ground construction investment, with an approximate cost of about 10.08 million RMB/km². Each coalbed methane vertical well controls an area of 300 × 300 m², and the cost per well is calculated at 0.9 million RMB.

P_B is the basic reserve fund, which is calculated at 12% of the fixed assets, where fixed assets are the sum of development well and ground construction investments³².

P_D is the drainage cost, which is 0.61 million RMB/year.

Estimated output value C₁₃

Currently, the estimation of CBM resources in abandoned coal mines primarily employs the material balance method, material composition method, and monthly decline curve method^33,34,35. However, these models are most applicable to mining areas with comprehensive data or relevant field tests. Due to a lack of foundational data in the Songzao mining area, estimating resources using the aforementioned methods proves challenging. Consequently, the volume method was utilized to estimate the residual coal resources in the goaf.

The volumetric method is a commonly used static evaluation technique for assessing the amount of CBM resources³⁶.

$$\:Q=A\cdot\:h\cdot\:\rho\:\cdot\:S\cdot\:q$$

(22)

Q is the amount of CBM resources, m³; A is the coal seam distribution area, m²; h is the average thickness of the coal seam, m; ρ is the average density of the coal seam, t/m³; q is the average gas content of the coal seam, m³/t.

If the coal reserves within the known calculation range are available (expressed as M/t), Eq. (22) can be simplified as follows:

$$\:Q=M\cdot\:q$$

(23)

The residual gas content was predicted following AQ/T 1018–2006 Mine gas emission prediction method. For high metamorphic coal with a gas content exceeding 10.00 m³/t.r, the residual gas content can be selected according to Table 3; for high metamorphic coal with a gas content below 10 m³/t.r, the residual gas content is calculated using Eq. (24).

Table 3 Values for residual gas content.

$$\:{W}_{c}=\frac{{10.385e}^{-7.207}}{{W}_{0}}$$

(24)

where W_c is the residual gas content in the coal seam, m³/t. r; W₀ is the original gas content in the coal seam, m³/t. r.

The exploitable reserves for Datong No. 1, Shihao, Yuyang, Fengchun, and Songzao coal mines are 43.92, 27.52, 33.04, 19.40, and 52.42 million tons, respectively. The recovery rates for Datong No. 1, Shihao, and Yuyang coal mines are 80, 78.69, and 68.40%, respectively; however, data for Songzao and Fengchun coal mines are unavailable. Therefore, a recovery rate of 80% was assumed for these coal mines. The average yield of volatile matter (V_daf) in the main coal seams of the Songzao mining area is 9.38%, with residual gas content ranging from 4 to 6 m³/t. r. Based on these data, the estimated CBM resource volume in the goaf is between 158 million and 237 million m³. The respective proportions of CBM resources for Datong No. 1, Shihao, Yuyang, Fengchun, and Songzao coal mines are 22.27, 14.87, 26.46, 9.84, and 26.57%, respectively. The CBM resources for these coal mines are illustrated in (Fig. 4).

Fig. 4

Amount of CBM resources in the goaf.

Due to technical, economic, and geological constraints, the total extractable volume of CBM cannot be fully realized. Therefore, the extractable volume is estimated to be 70% of the geological resource volume³⁷. The formula for calculating the estimated output value is expressed as follows:

$$\:{I}_{V}=\:Q\cdot\:\eta\:\cdot\:\:{P}_{E}\cdot\:0.7$$

(25)

where Q is the amount of CBM resources in the goaf; η is the power generation efficiency of the generator, which is 2.5 kWh/m³; and P_E is the unit price, which is 0.577 RMB/kWh.

The quality of evaluation results primarily depends on the specific features of the indicators, such as their numerical values or developmental trends. Indicators whose numerical values increase or exhibit a positive developmental trend are considered positive indicators. Conversely, indicators whose numerical values decrease or show a negative developmental trend are considered negative indicators.

After establishing the evaluation system, a thorough analysis of each indicator’s nature is required to ensure the smooth progression of subsequent processes, as detailed in (Table 4).

Table 4 Properties and descriptions of indicators.

Determination of weights

According to the AHP, the maximum eigenvalue (λ_max) of the judgment matrix was calculated as 3.0183, with a consistency ratio (CR) of 0.017 (< 0.1). Therefore, the judgment matrix satisfied the consistency criteria. The judgment matrices and weights for primary and secondary indicators are presented in (Tables 5, 6, 7 and 8; Fig. 5).

Table 5 Judgment matrix for primary indicators.

Table 6 Judgment matrix for mining condition B₁.

Table 7 Judgment matrix for CBM resources B₂.

Table 8 Judgment matrix for economic conditions B₃.

Fig. 5

Indicator weights derived from the AHP.

Based on data from closed coal mines, formulas for construction costs, and estimated output value, the specific values for each indicator were obtained, as shown in (Table 9).

Table 9 Specific values for various evaluation indicators.

Furthermore, the indicator weights were calculated using the entropy method, with the results presented in (Table 10).

Table 10 Indicator weights.

Substituting $\:{\beta\:}_{1}$= 0.7344 and $\:{\beta\:}_{2}$= 0.2656, derived from the game theory, into the combined weighting formula for indicator weights yields the subjective, objective, and combined weights for each indicator, and the results are presented in (Table 11).

Table 11 Indicator weights derived from the game theory.

Figure 6 compares the weight distributions of secondary indicators obtained from the AHP, entropy method, and game theory.

Fig. 6

Comparison of weights of secondary indicators.

Establishment of TOPSIS model

Based on matrix P and combined weights W derived from the game theory, the weighted normalized decision matrix U is derived.

$$\:U=\left[\begin{array}{cccc}0.054&\:0&\:\dots\:&\:0.1399\\\:0&\:0.0008&\:\dots\:&\:0\\\:0.0151&\:0.032&\:\dots\:&\:0.1035\\\:0.0108&\:0.0356&\:\dots\:&\:0.0344\\\:0.0216&\:0.0101&\:\dots\:&\:0.0578\end{array}\right]$$

The corresponding positive and negative ideal solutions are obtained as follows:

q⁺={0.054,0.0356,0.038,0.0389,0.0525,…,0.1399}.

q⁻={0,0,0,0.01945,…,0}.

The distances to these solutions, along with the closeness for each evaluation indicator, are presented in (Table 12).

Table 12 Ideal solutions and closeness calculation results.

The results indicated that among the five coal mines in the Songzao mining area, the evaluation ranking is as follows: Songzao > Fengchun > Shihao > Yuyang > Datong No. 1.

Evaluation of energy saving and emission reduction

The above evaluation results determined that the goaf at the Songzao Coal Mine exhibited a high feasibility for CBM extraction. According to a report by the United States environmental protection agency (EPA)³⁸, CBM emissions from flooded coal mines almost cease within eight years after abandonment. Consequently, Songzao Coal Mine was selected as the research subject to estimate the CBM emissions over an eight-year post-closure period and the potential electricity generation from this methane.

Since Songzao coal mine is a flooded mine, only the relevant section of the formula applicable to flooded mines was employed for estimation. The absolute gas emission rate during the shutdown period is essentially stable at 114.37 m³/min. After the mine was closed, the gas extraction system was no longer operational. Considering the reduced overflow from rising water levels, the gas emission rate was estimated at 50%. Once the shaft was sealed, gas continued to flow into the underground passage, with an anticipated 20% natural overflow to the surface. The methane emissions, cumulative emissions, power generation, and cumulative power generation for the first eight years were calculated using Eqs. (17)–(20), and the results are illustrated in (Figs. 7 and 8).

Fig. 7

CH₄ emissions from Songzao coal mine.

Fig. 8

CH₄ power generation in Songzao coal mine.

Discussion

The comparison of the three methods for determining indicator weights indicated that the AHP plays a crucial role in the weighting process, while the entropy method is used for data adjustment and optimization. Among the 13 secondary indicators, the top three ranked by weight are the water-free space in the goaf C₁₀, estimated output value C₁₃, and water accumulation in the goaf C₉. This ranking indicates that water accumulation in the goaf significantly affects CBM extraction, while project revenue is directly related to economic benefits and commercial feasibility.

A combination of construction costs and estimated output value allows for the estimation of net development revenue. Data from the Datong No. 1 Coal Mine reveal a substantial goaf area and considerable CBM resources, yet the resource richness is relatively low. The number of wells in this coal mine is significantly higher than that in other coal mines, which increases construction costs beyond the estimated output value and ultimately leads to a low economic viability for development.

The results obtained using the TOPSIS method indicated that, among the five coal mines, Songzao Coal Mine has the highest weight, followed by Fengchun Coal Mine and Shihao Coal Mine. The geological conditions at Songzao coal mine are relatively stable, featuring favorable mining conditions, minimal water accumulation in the goaf, ample remaining space, and abundant CBM resources. This suggests that the higher-ranked coal mines possess goafs that are not completely submerged. Once the goaf is drained, the reduction in groundwater pressure diminishes the support provided by water. This leads to loosening, collapse, and deformation of the surrounding rock, which may result in further surface subsidence. For the reuse after drainage, it is necessary to reassess the stability of surrounding rock and the suitability of geological conditions. If extraction conditions are not met, additional support and reinforcement measures will be required. This will increase both workload and investment costs, thereby reducing the economic viability of coalbed methane extraction.

According to the calculation results of energy-saving and emission reduction benefits for Songzao Coal Mine, both annual emissions and annual power generation exhibit a decreasing trend over time. In the first four years, annual CH₄ emissions decreased from 2056.82 t to 273.94 t, while annual power generation declined from 9.4783 million kWh to 1.2624 million kWh, both reaching approximately 15% of their initial values. However, in the following four years, the rate of decline for these two indicators gradually slowed. Conversely, cumulative emissions and cumulative power generation increased rapidly in the first four years, where cumulative emissions rose from 2056.82 t to 3917.57 t and cumulative power generation increased from 9.4783 million kWh to 18.053 million kWh. In the following four years, the data for these two indicators stabilized.

In the early stages following coal mine closure, rapid release of internal mine pressure, significant change in coal seam structure, and fluctuations in groundwater levels facilitate the quick release of CBM originally stored within the coal seams. In the following years, downhole conditions gradually stabilized, pressure balanced, and the CBM release rate decreased. Effective capture and utilization of released CBM resources would not only support the diversified development of energy sources but also significantly contribute to environmental protection and resource recycling. Such efforts would facilitate the mining industry’s transition towards a greener, more sustainable development model through the implementation of circular management of mine waste³⁹.

This study presented a research method for assessing residual CBM in the Songzao mining area that demonstrates broad applicability. Unlike emission prediction methods that primarily focus on the relationship between geological factors and CBM emissions or those based solely on data algorithms^40,41, this study developed a more comprehensive approach. The feasibility evaluation aims to identify coal mines suitable for CBM extraction, while the estimation of CBM emissions and power generation concentrates on assessing the post-mining benefits. These two components are complementary and mutually supportive. This methodology can be extended to other regions with similar geological characteristics. However, practical applications require careful alignment with local conditions, comprehensive investigations of underground geology and CBM characteristics, and the selection of scientifically sound estimation methods for CBM resources. Additionally, accurate calculations of construction costs and expected output values are essential. It is also important to fully consider factors such as national conditions, policies, regulations, and environmental protection requirements. Relevant evaluation indicators should be appropriately introduced and adjusted to ensure the practicability of research methods and the accuracy of evaluation results.

Conclusions

To facilitate the efficient development and utilization of CBM from the goafs in closed coal mines, this study investigated five closed coal mines in the Songzao mining area of Chongqing. The main conclusions are as follows:

• (1) An evaluation indicator system for the feasibility of CBM extraction from the goafs in closed coal mines was established based on the contextual background of the Songzao mining area of Chongqing. The weights obtained from AHP and entropy method were integrated through game theory. The analysis revealed that the three indicators with the highest weights—water-free space in the goaf, estimated output value, and water accumulation in the goaf exerted the most significant impact on the results. Furthermore, by employing the TOPSIS method, a feasibility evaluation model for CBM extraction in abandoned coal mines was developed. The feasibility evaluation results demonstrated the following ranking: Songzao Coal Mine > Fengchun Coal Mine > Shihao Coal Mine > Yuyang Coal Mine > Datong No. 1 Coal Mine.

• (2) Based on the evaluation results, the Songzao Coal Mine, exhibiting the highest feasibility for CBM extraction, was selected as the research subject. An estimation of CBM emissions and power generation over the past eight years following coal mine closure was conducted. The results indicated that, in the first four years, methane emissions decreased to approximately 15% of the initial emissions, while cumulative power generation increased by 50%. In the subsequent four years, the stabilization of underground structural pressure led to a significant reduction in the rate of methane emissions. The annual emissions decreased from the initial 2056.82 to 18.63t, while the cumulative power generation stabilized at around 1928.09 kWh. These observations clearly demonstrate the significant effectiveness of CBM extraction in energy conservation and emission reduction.

• (3) This study preliminarily verified the feasibility of CBM extraction in the Songzao mining area of Chongqing and estimated associated carbon emissions and power generation. However, some limitations exist within the research. The evaluation indicators fail to adequately address the complexities of CBM extraction, and challenges in data collection resulted in the use of assumed values for some data. Prior to coal mine closure, it is recommended to conduct comprehensive underground data collection, prepare a detailed closure report, and install an underground monitoring system for on-site monitoring in real time. This will enhance the convenience and accuracy of data acquisition in future research. Additionally, future studies need to further improve the evaluation system and deepen the evaluation methods. Key indicators such as environmental risks and social impacts should be included to achieve a more accurate assessment of CBM extraction.