Front. Energy Res. Frontiers in Energy Research Front. Energy Res. 2296-598X Frontiers Media S.A. 1645081 10.3389/fenrg.2025.1645081 Original Research Character and inspiration of coalbed methane extraction from underground boreholes Wang et al. 10.3389/fenrg.2025.1645081 Wang Bo 1 2 Visualization Funding acquisition Writing - review and editing Writing - original draft Conceptualization Methodology Zhang Xianshang 3 * Supervision Methodology Writing - original draft Investigation Funding acquisition Formal Analysis Writing - review and editing Ma Qianqian 1 4 Formal Analysis Methodology Resources Investigation Writing - review and editing Wang Zhen 1 * Writing - review and editing Investigation Methodology Visualization Wen Guangcai 1 * Funding acquisition Project administration Formal Analysis Writing - review and editing Investigation Supervision Long Qingming 1 Formal Analysis Methodology Investigation Writing - review and editing Liu Jun 1 Methodology Investigation Writing - review and editing Formal Analysis Yang Huiming 1 Methodology Investigation Writing - review and editing Formal Analysis Cao Jie 1 Writing - review and editing Investigation Methodology Formal Analysis Dai Linchao 1 Methodology Investigation Writing - review and editing Formal Analysis Xu Xuezhan 1 5 Investigation Methodology Writing - review and editing Formal Analysis Zhu Moran 1 Investigation Writing - review and editing Formal Analysis Methodology State Key Laboratory of Coal Mine Disaster Prevention and Control, China Coal Technology Engineering Group Chongqing Research Institute, Chongqing, China School of Mechanics and Civil Engineering, China University of Mining and Technology, Jiangsu, China Institute of Disaster Rock Mechanics, Liaoning University, Shenyang, China School of Mines, China University of Mining and Technology, Xuzhou, China School of Safety Science and Engineering, Anhui University of Science & Technology, Huainan, Anhui, China Correspondence: Xianshang Zhang, zhangxianshang@sina.com; Zhen Wang, cumtwangzhen@126.com; Guangcai Wen, wgc139@126.com 02 01 2026 2025 13 1645081 11 06 2025 10 11 2025 17 11 2025 Copyright © 2026 Wang, Zhang, Ma, Wang, Wen, Long, Liu, Yang, Cao, Dai, Xu and Zhu. 2026 Wang, Zhang, Ma, Wang, Wen, Long, Liu, Yang, Cao, Dai, Xu and Zhu https://creativecommons.org/licenses/by/4.0/ This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Coalbed methane (CBM) has become an important unconventional gas source in China, yet quantitative understanding of underground borehole extraction remains limited. Based on field data from the Wangpo coal mine, this study investigates the relationships among methane concentration, vacuum pressure, gas production rate, and cumulative extraction behavior. Results reveal that daily production rates follow a power-law–type decline consistent with pore–cleat–controlled gas flow, and the cumulative extraction curve can be divided into four characteristic stages corresponding to shifts in gas-migration mechanisms. The rate integral–derivative function exhibits a strong linear correlation with mass-balance time, underscoring its potential as a predictive tool for production forecasting. A clear threshold effect of vacuum pressure is identified, beyond which methane concentration improves significantly, while longer boreholes enhance gas capture due to increased exposed coal area. However, drilling length, cost, and geological heterogeneity must be balanced to maintain efficiency. The findings highlight the need for intelligent vacuum-pressure regulation based on real-time methane monitoring to optimize extraction stability and utilization efficiency. These insights offer practical guidance for CBM borehole management and process optimization.

 gas production vacuum pressure methane concentration underground borehole coalbed drainage The authors declare that financial support was received for the research and/or publication of this article. This work was supported by the National Natural Science Foundation of China (Grant Nos. 52204261, 52274150), the Natural Science Foundation of Chongqing, China (Grant Nos. CSTB2023NSCQ-MSX0577, CSTB2022NSCQ-JQX0014), and the Enterprise self-supporting projects (2024ZDYF18). section-at-acceptance Advanced Clean Fuel Technologies Introduction

Coalbed methane (CBM), a byproduct of coal formation, serves as both a clean and high-efficiency fuel and a potential source of gas hazards in underground coal mines (Ji et al., 2024; Pan et al., 2014; Flores, 1998). China possesses abundant CBM resources, with approximately 30.05 × 1012 m3 of CBM located at depths of less than 2000 m (Xu et al., 2023a). The extraction of CBM can mitigate the gas-related risks associated with coal mining, while also providing a source of clean energy to enhance energy sustainability, thereby contributing to efforts to restrain the greenhouse effect (Pan et al., 2014; Xu et al., 2023b).

Currently, there exist two methods of CBM extraction technology. One method involves surface extraction of CBM, which utilizes technology akin to that of natural gas development. Over the past decade, more than 12 billion m3 of methane has been extracted annually from ground wells, significantly contributing to domestic natural gas supply (Paper, 2024). Conversely, the other method is underground extraction through boreholes, which facilitates the control and mitigation of underground gas-related disasters. The total volume of underground methane extracted has been comparable to that extracted from the surface on an annual basis. However, the methane concentration in the extracted gas from underground sources had been highly uncontrollable and unpredictable, leading to low utilization rates, particularly during the later stages of underground CBM extraction. This was due to the challenge of effectively utilizing low-concentration methane, which was often deemed impractical or uneconomical to utilize and was consequently wasted (Pan et al., 2014). Hence, effective management and operation of underground boreholes were crucial for enhancing methane concentration levels,yet significant gaps remain in understanding the fundamental extraction characteristics and production behavior of these boreholes.

For the sake of maximizing methane extraction and utilization, many researchers have conducted deep research on the mechanism of CBM migration (Wang et al., 2024; Zhou H. et al., 2014). The gas migration process within coalbeds can be categorized into two primary processes: gas desorption and diffusion (Zhang et al., 2021; Guo H. et al., 2024), gas seepage (Wei et al., 2024; Zhao et al., 2018), occurring within fractures and pores of varying scales (Guo X. et al., 2024; Liu et al., 2024). The decrease in gas concentrations and pressure facilitated gas diffusion and seepage (Wang et al., 2023), respectively, with the driving forces being interchangeable based on gas adsorption principles and coal pore characteristics (Lin et al., 2023). Gas diffusion and seepage behaviors in coal collectively regulated CBM production, each playing distinct roles in different extraction stages (Si et al., 2019). A theoretical conversion model was developed to assess the predominant influence of diffusion and seepage on CBM production, based on the contribution of various methane forms to total production (Liu et al., 2018). Specifically, the gas flow process within coalbeds can alter reservoir pressure and coalbed methane production rates (Ye et al., 2014), holding significant implications for the prevention and management of coalbed methane disasters.

The methane production profile exhibited diverse trends as time increases progressed due to the varying roles of different influencing factors in gas drainage over time and space, particularly in ground wells. Chen et al. (2024) conducted an analysis on the 636 vertical wells to investigate the impact of different variables on CMB productivity. They concluded that the highly productive wells were significantly influenced by geological conditions, predominantly situated in regions with high gas content, moderate cover depth, high permeability, and elevated water head based on grey correlation analysis. Li C. et al. (2024) assessed and evaluated the porosity, permeability, adsorption constants, working fluid filtration loss, and the reservoir modification methods on gas production, considering the actual distribution of gas reservoir seepage characteristics. Gas injection has been identified as an effective method to enhance CH4 production (Chu et al., 2024); however, CO2 injection may damage the coalbed and reduce its permeability, while N2 injection can effectively restore permeability (Yu et al., 2023). A comprehensive understanding of the fluid loss, Langmuir adsorption equation, steady diffusion law, and Darcy’s law of seepage contributes to the accuracy of the well productivity evaluation model (Li C. et al., 2024; Chu et al., 2024). Upadhyay et al. (2024) simulated the complex flow phenomena including multi-component adsorption, two-phase flow, pressure-dependent diffusion and time-dependent desorption of coalbed methane to assess the production performance of coalbed methane wells.

The prediction of coalbed methane production plays a crucial role in assessing the production performance of coalbed methane wells. This aspect had attracted significant attention from researchers who had delved into various productivity prediction models and methods, particularly focusing on ground wells. Multiple method existed for predicting the productivity of ground wells, each characterized by specific principles, usage conditions, advantages, and disadvantages. Among these methods, numerical simulation stand out one of the most widely utilized approaches (Chu et al., 2024; Huang et al., 2023). This method, taking into account the characteristics of CBM reservoirs, such as gas adsorption and desorption, gas flow processes, and various influencing factors, enabled the simulation and prediction of the dynamic production of CBM wells under different conditions (Xu et al., 2024; Wei et al., 2022). Another method, production decline analysis, operated on the premise that the production of CBM wells decreased with time. This method was suitable for wells that had reached a stable production stage and exhibited a clear declining trend in production. By fitting the production decline curve, future production levels could be forecasted for a specific period (Yehia et al., 2023). The Analytical method involved predicting the production of coalbed methane wells by establishing an analytical model of gas flow in coalbeds and utilizing mathematical formulas. This method was characterized by rapid prediction speed, low accuracy, and a limited application range, typically suitable for simple geological conditions and stable gas flow scenarios (Jia et al., 2024). The Empirical method (Ya et al., 2021; Shi et al., 2021) relied on empirical formula derived from statistical analysis of extensive production data, which was used to predict the production of coalbed methane wells. These formulas took into consideration key parameters such as coal seam thickness, permeability, gas content, and gas pressure at the well bottom hole (Zhou F. et al., 2014). While simple and feasible, the empirical method had a restricted scope of application and might be affected by the geological conditions. The analogy method (Stopa and Mikołajczak, 2018; Zhang et al., 2023), on the other hand, relied on empirical data and expert judgment to predict the productivity of a target coalbed methane reservoir by comparing it with similar reservoirs in terms of geological conditions and development history. This method was suitable when detailed geological and production data are lacking. The accuracy of analogies hinged on the similarity between the objects of comparison and the expertise of the analyst. Advancements in artificial intelligence and machine learning technologies had also been integrated into the prediction of coalbed methane development (Du et al., 2023a; Du et al., 2023b). By training machine learning models, the relationships between geological conditions, development patterns, production data, and productivity could be learned and used to forecast the production of coalbed methane wells or blocks (Zhao et al., 2023). This method excelled in handling complex and nonlinear relationships but necessitates substantial data support.

In practical applications, the selection of appropriate prediction methods should be based on the specific conditions of coalbed methane reservoirs and data availability. Furthermore, verifying and comparing various methods was essential to enhance the accuracy and reliability of prediction outcomes. In contrast to ground wells, underground boreholes were shorter and horizontal, which were constructed and located in the drilling sites. Even if they played an equally important role in gas extraction with surface wells from the view of gas production, research on production prediction and fundamental extraction characteristics for underground boreholes remains inadequate. This gap is primarily due to their initial aim being gas disaster reduction rather than efficient utilization, leading to a lack of systematic analysis of production history and key controlling factors. The dual-carbon strategy has brought new significance to gas extraction from underground boreholes, aiming not only to prevent gas disasters but also to ensure the efficient utilization of gas with higher concentration. This necessitated effective extraction management and a focus on analyzing the fundamental underground extraction unit and the gas extraction characteristic curve of the borehole drilling field. Unfortunately, the production history of underground boreholes was often overlooked or roughly estimated. The lack of detailed analysis and research on the production history of underground boreholes hindered the realization of effective control and management, as it failed to provide a scientific guidance for gas extraction.

Despite extensive research on coalbed methane (CBM) drainage, previous studies have primarily focused on laboratory-scale permeability evolution or numerical simulations of single processes such as desorption or seepage. However, the dynamic coupling between seepage, diffusion, and operational control parameters in field-scale underground boreholes remains poorly understood. The key scientific problems addressed in this study therefore include: (1) the unresolved dynamic characteristics and stage evolution of gas production from underground boreholes; (2) the unclear conversion mechanism between gas seepage and diffusion during long-term extraction; and (3) the quantitative impacts of operational parameters—particularly vacuum pressure and borehole length—on extraction efficiency and methane concentration. To bridge these gaps, this study performs a systematic rate-transient analysis (RTA) of extensive historical extraction data from the No. 3 coal seam of the Wangpo Coal Mine. Unlike previous models that rely solely on assumed flow regimes, our approach integrates field data interpretation with flow-regime diagnostics to capture stage-wise production behavior and identify transitions between flow mechanisms. The work provides quantitative insight into how operational control modifies gas transport processes, establishing a methodological framework that connects laboratory-scale diffusion–seepage theory with real mining conditions. The conclusions are of direct practical value for optimizing CBM drainage design, enhancing gas disaster prevention, and supporting safe face replacement in underground operations.

 Filed situation Coal seam situation Geological conditions of the coal seam

Wangpo coal mine is situated in the Qinshui coalfield in the southeastern region of Shanxi Province, an area of significant interest for methane drainage through underground boreholes, as shown in Figure 1. The primary coal seam is the No.3 coal seam, originating from the Permian period and known for its abundant methane content. Its thickness is stable, which varies from 4.20 m to 6.70 m in the minefield with an average value of 5.76 m. The variation coefficient of coal thickness is 17.6%, and the recoverability index is 1. The structure is generally simple and stable with strip and laminar structure that can be mined in the whole area. The coal seams within the Qinshui Basin are notable for their high gas content but exhibit strong heterogeneous permeabilities ranging from 0.1 to 10 mD (Zhou F. et al., 2014). The No. 3 seam primarily consists of anthracite, characterized by bright and mirror coals, which constitute approximately 75% of the coal reserves and are found in thicker seams that are either stratified or lenticular. Dark coal and silky charcoal are less prevalent. The microscopic characteristics are outlined in Table 1. The solidity coefficient varies from 0.32 to 1.49, with values below 0.5 observed when coal samples are obtained from the thin layer of tectonic coals. For a more comprehensive understanding of coal geology, please refer to the relevant literature (Liu et al., 2022; Su et al., 2005).

The location of Wangpo coal mine.

Map of China highlighting provinces with major cities and geographical landmarks. An inset in the top-right shows the location of Wangpo coal mine near Taiyuan. A blue arrow indicates its position relative to the rest of China.

Microscopic characteristics of No.3 coal seam.

Microscopic composition Vitrinite Inertinite Minerals
Content (%) 50.4∼66.0 19.2∼32.8 11.6∼18.2
Character Mainly composed of unstructured homogeneous vitrinite, followed by glial vitrinite and matrix vitrinite Dominated by oxidized filamentous bodies distributed in a fragmented or lens like manner Mainly composed of clay minerals, filled in cell cavities or distributed as lens shaped aggregates
Coal pore structure

The permeability of coal and gas production are influenced by the intricate pore fractures within coal formations, primarily determined by the characteristics of natural fractures (Liu et al., 2024; Zhao et al., 2024; Zhang, 2024). The pore structure of coal samples was analyzed at various scales through a combination of low-pressure CO2 adsorption, low-temperature N2 adsorption, and high-pressure mercury intrusion porosimeter (MIP) experiments. The coal samples with the diameter of 0.18–0.25 mm and 1–3 mm were used in the adsorption test and MIP, respectively. These testing procedures were rigorously conducted in compliance with the standards GB/T 21650.1–2008, GB/T 21650.2–2008, and GB/T 21650.3–2011.

The pore volume, specific surface area, and pore size distribution were obtained, as shown in Table 2. During testing, variations in relative pressure and temperature were observed between the CO2 and N2 adsorption methods. Consequently, there are discrepancies in the calculated pore volume and surface area using the DFT and HK theories based on the data obtained from both methods. Typically, the CO2 adsorption method is more sensitive to detecting smaller pores. Analysis of Tables 3, 4 reveals that the pore size and volume derived from the adsorption process are greater than those from the desorption process according to the BJH and D-H theories. Conversely, the surface area calculated from the adsorption process is smaller than that from the desorption process. Permeability, a crucial parameter for characterizing gas migration in coal seams, was determined to be 73569.5608 mDarcy from the MIP test of the NO.3 coal seam. The permeability is influenced by the pore and fracture structure.

Pore structure parameters of the coal sample (CO2).

DFT HK DR DA
V T (cm3/g) (≤1.048 nm) A P (m2/g) (>1.048 nm) A TP (m2/g) (>0.367 nm) V MP (cm3/g) W MP (nm) A MS (m2/g) C M (cm3/g) A MS (m2/g) V LM (cm3/g)
0.03746 75.53 201.535 0.049 0.6614 280.0525 61.314 273.614 0.1169

(), different determined method; DFT, density functional theory; HK, Horvath-Kawazoe theory; DR, Dubinin-Radushkevich theory; DA, Dubinin-Astakhov theory; V T , total volume; A P , area in pores; A TP , total area in pores; V MP , maximum pore volume; W MP , median pore width (nm); A MS , micropore surface area; C M , monolayer capacity; V LM , limiting micropore volume.

Pore structure parameters of the coal sample (N2).

DFT HK Surface area (m2/g)
V P (×10−3 cm3/g)(<1.269 nm) V T (×10−3 cm3/g)(≤343.337 nm) A TP (m2/g)(>1.269 nm) V MP (×10−3 cm3/g) W MP (nm) BET Langmuir ATPM ATPE
0.25 1.39 0.049 0.365 0.6271 0.6984 0.9646 0.646 0.052
Parameters between 1.7000 nm and 300.0000 nm diameter
Pore size (nm) Pore volume (×10−3 cm3/g) Surface area (m2/g)
BJH AD 42.0117 1.254 0.1194
BJH DE 10.5024 0.420 0.1601

V P , volume in pores; A TPM , t-Plot micropore area; A TPE , t-Plot external surface area; AD, adsorption; DE, desorption; BJH, Barrett–Joyner–Halenda theory.

Pore structure parameters of the coal sample (Hg).

V T (cm3/g) A TP (m2/g) D AP (nm) P (mDarcy) F CF Porosity (%) Tortuosity
0.2574 7.262 141.81 73569.5608 0.143 26.9603 3.9864

A TP , total pore area; D AP , average pore diameter; F CF, conductivity formation factor; P, permeability.

As the average pore radius of coal decreases, the impacts of gas slippage and Knudsen diffusion become evident. The interplay between gas slippage and Knudsen diffusion can counteract the reduction in permeability, resulting in a complex pattern of permeability changes that do not simply decrease (Liu et al., 2015; Li et al., 2015). Nevertheless, an increase in the quantity of large pores enhances the connectivity of coal pores, expanding the permeability volume within the coal sample. This enhancement is advantageous for the flow capacity of coalbed methane (Li H. et al., 2024).

 Gas distribution in coal seam

The gas content in the coal seam was determined using the desorption method, revealing values ranging from 6.11 to 12.80 m3/t. The methane concentration exceeded 80%. The gas pressure within the coal seam ranged from 0.40 to 0.65 MPa. Further details on the industrial analysis and adsorption characterization of the coal sample can be found in Table 5.

The industrial analysis and adsorption characteristics of No.3 Coal Seam.

Langmuir parameters Apparent density (g/mL) Porosity (%) Moisture (%) Ash (%)
Volume (mL/g) Pressure (MPa)
38.35 0.66 1.47 2.65 1.16 14.04
Drilling sites

The boreholes are usually used to control methane flow from the fractured zone and are drilled from the roadway to a depth that situates them within the coal seam, as shown in Figure 2. The coal mine employed the regional extraction through directional drilling along the layer using the progressive method. Directional drilling had demonstrated greater efficacy compared to conventional drilling techniques due to the ability to create significantly longer boreholes. This approach enabled gas extraction not only from the working face but also from the coal strip that would be excavated along the tunnels. It could anticipate subsequent roadway excavation. The method involved connecting and pumping during construction.

The borehole layout diagram for the working face 3301.

Diagram showing borehole locations and trajectories along 3301 and 3219 transport roadways near a water sump and return laneway. Multiple paths diverge in pink and red, labeled with measurements and identifiers.

Figure 2 show the boreholes of the drill site 1#. These boreholes were drilled from the transport roadway of working face 3301 to the transport roadway of working face 3219. It covered the working face 3301 and its return air trough. The drilling was driven by nitrogen gas without the effect of moisture on the extraction process (Pan et al., 2014; Liu et al., 2015). The construction of the drilling site 1# began on May 29, and was completed on October 28 at same year. The completion time of each borehole is shown in Figure 3. There are 21 main boreholes, and 25 branch boreholes in total, and the drilling length is 11092 m. All the boreholes are collected into one pipeline with the inner diameter of 75 mm. Notably, the drill bit fell in the D6 borehole which was treated as a waste hole.

The completion time of each borehole.

Timeline graph showing dates from May 19 to November 15 with numbered markers above and below, indicating events labeled one through twenty-one. Events are marked on June 8, June 28, July 18, August 7, August 27, September 16, October 6, and October 26 with varying vertical positions. Test device

To monitor production history of the underground boreholes, the measurement instrument typed YDC5 for gas drainage pipeline was applied to manually record various data including the total gas flow, methane percentage, pipeline vacuum, temperature, and ambient pressure, as shown in Figure 4. To ensure data reliability during manual collection, the following measures were implemented: (1) Regular calibration of the YDC5 device before each measurement session; (2) Standardized training for field personnel following fixed protocols; (3) Triple measurements for critical parameters with averaging to reduce variability; (4) Cross-validation with operational logs to identify anomalies. This device employs internal circulation sampling technology of gas and the laser inspection technology to ensure accurate. The key operational parameters are detailed in Table 6.

Determination diagram of methane concentration and pressure inside pipelines. (a) Determination of methane concentration. (b) Measurement of pressure inside pipelines.

Two diagrams depicting instruments measuring pipeline conditions. (a) Shows a device determining methane concentration with a probing rod inserted into a pipeline, indicating gas flow. (b) Shows a pressure measurement setup with a probe inside a pipeline, labeled pressure tap, also indicating gas flow.

Functional parameters of measurement device.

Parameters Measurement range Resolution Error
Differential pressure (Pa) 0∼5000 0.01 ≤±25
Absolute pressure (kPa) −100∼100 0.1 ≤±1
Temperature (°C) −10∼50 0.1 ≤±1
Methane percentage (%) 0.00∼1.00 0.01 ≤±0.06
1.01∼9.99 0.01 ≤±6
10.0∼100 0.1 ≤±6

This device is capable of acquiring the gas flow parameters within the pipeline, which includes gas mixing and gas purity. The vacuum pressure of extraction refers to the pressure difference between the extraction pipeline and the underground roadway environment. A higher vacuum value indicates a more effective pumping process. Therefore, both the ambient pressure and the quality of the pump play a crucial role in gas flow and production. The data collected are presented in Table 7.

Some data of gas extraction (D1).

Measurement time Extraction time (d) Vacuum pressures (kPa) Methane percentage (%) Mixed gas rate (m3/min) Methane rate (m3/min)
2022/6/13 4 12.7 6.36 1.537 0.0977532
2022/6/17 7 14.15 19.34 0.26 0.050284
2022/6/20 10 14.28 6.72 0.378 0.0254016
2022/6/23 13 9.6 12.27 0.23 0.028221
2022/6/26 16 10.2 11.08 0.254 0.0281432
2022/6/29 17 10.2 11.08 0.254 0.0281432
2022/6/30 19 12.3 7.95 0.245 0.0194775
2022/7/2 20 12.5 11.9 0.231 0.027489
2022/7/3 21 12.7 7.4 0.227 0.016798
2022/7/4 22 10.8 2.7 0.29 0.00783
2022/7/5 23 12.5 6.2 0.31 0.01922
2022/7/6 24 12.5 2.8 0.08 0.00224
2022/7/7 25 12.7 10.2 0.24 0.02448
2022/7/8 26 16.3 12 0.23 0.0276
…
Characteristics of gas extraction in underground boreholes Production data analysis

The drilling process at a drilling site is a time-consuming procedure that occurs simultaneously with gas extraction. While manual recording introduces potential human error, our methodology implemented systematic controls including timed interval verification and duplicate data entry checks. Future studies would benefit from automated monitoring systems to achieve higher frequency data collection with reduced human intervention. Various techniques were implemented in the data processing stage to enhance the assessment of the gas extraction operations.

It is assumed that the acquisition time interval is an integer multiple of 24 hours.

The state of gas extraction between the current and previous moments is represented by the data available at the present time. To calculate the net amount of gas extraction, the following formula can be used:

Q G n − Q G n − 1 = Q T n − Q T n − 1 C n where, Q G n denotes the net amount of gas extraction in the nth time periods, Q T n denotes the total amount of pipeline extraction, C n is the methane concentration of pipeline gas.

3. For a more precise comparison, it is recommended to average the borehole extraction scenario over a 1-day period.

 Gas extraction in drilling sites Evaluation of vacuum pressure and methane concentrations

A rudimentary assessment might conclude that pressure and concentration data showed the extraction condition. Figure 5 illustrates the vacuum pressures and methane concentrations in the main pipeline during the drainage periods at the drilling site since its inception, which spanned over 600 days. However, not every borehole can sustain effective operation for such an extended period. Figure 5 depicts both parameters exhibiting irregular fluctuations over time. Initially, the concentration percentage fluctuates from 4.8% to a peak of 62.3%, with an average of 37.1%. This site encompasses simultaneous drilling construction and gas extraction processes, leading to a reciprocal influence between the two activities. Moreover, a small quantity of adsorbed methane initially infiltrates the borehole through fractures. As gas extraction progresses, this is attributed to the rapid depletion of a substantial amount of in-situ free methane, causing a swift rise in the methane concentration gradient between the fracture and the matrix. Consequently, more adsorbed methane is desorbed, diffused into the fracture, and subsequently enters the borehole (Liu et al., 2018).

The history for methane concentrations and vacuum pressure at the drilling site.

Line graph showing methane concentration (black line) and vacuum pressure inside a pipeline (red line) from May 30, 2022, to March 10, 2024. Methane concentration ranges from 4.8% to 62.3%, with an average of 37.1%. Vacuum pressure ranges from 1.4 to 38.8 kilopascals, averaging 19.5 kilopascals. Significant changes noted after drills completed on October 7, 2022.

The methane concentration curve exhibits a gradual decline after reaching its peak value, followed by significant fluctuations over a period. These fluctuations are expected to stabilize once drilling is completed. As the total quantity of adsorbed methane in the coal matrix remains constant, a portion of the adsorbed methane migrates towards the borehole. Consequently, the methane concentration gradient between the fracture and the matrix diminishes gradually, leading to a reduction in the methane diffusion capacity within the matrix and a decrease in the diffusion quantity (Kang et al., 2022).

As depicted in the figure, the vacuum pressure ranges from 1.4 kPa to 38.8 kPa, with an average of 19.5 kPa. During the latter period, the pressure consistently exceeds the average value. This discrepancy may be attributed to the fact that the average is calculated based on varying data sampling frequencies. The reduced frequency of data acquisition in the later period leads to a lower overall average. A similar issue is observed with the concentration data. Nevertheless, it is evident that the extraction process in the drilling field transitions into a relatively stable phase following the completion of all drilling activities.

The vacuum pressure conditions partially reflect the bottom-hole pressure and influence the gas pressure differential. Stable daily gas production denotes the average daily output of gas from a specific well during its stable production phase.

 Evaluation of total gas production rates

Since the main production of the boreholes is the gas composed of two main components (air and methane), total gas production and methane percentage, from which pure methane production can be calculated, were selected as performance parameters in this study. Figure 6 shows gas production history of monitored gas boreholes. It mainly includes accumulation, production rate, and average daily yield of methane. The latter two ones have the same unit but different calculation methods. The methane production rate is calculated based on total gas production and methane percentage, and average daily yield is to average the cumulative amount to the day that has been drawn. The methane production rate has a more sensitivity than the average amount. The change curve of the average daily amount with time is gentler, which will reduce the impact of data noise on the overall future trend development (Alwin Blasingame et al., 1989). The four-stage curve (slow growth, steep growth, stable increase, and accumulative rate decay) delineates the evolution of gas migration dynamics: Stage I is dominated by drilling disturbances and sealing issues, where air ingress dilutes methane; Stage II reflects free gas flow from fractures under pressure gradients; Stage III indicates a balance between desorption, diffusion, and seepage, crucial for long-term production stability; and Stage IV signals resource depletion, emphasizing the need for timely interventions like new drillings or pressure adjustments. This characterization aids in predicting production declines and optimizing extraction scheduling based on mechanistic insights. Although the curves are different as the calculation method of each parameter is different, the general trends in observed gas productions show that the ventholes initially produce at higher gas rates and then enter a decline period with increasing time.

Methane production at the drilling site.

Graph showing accumulated volume and methane production rate over 600 days. Accumulated volume (black line) rises steadily, and methane production rate (red line) fluctuates with periods marked I to IV. Key phases and average yield are indicated.

As the primary output of the boreholes is gas, mainly composed of air and methane, total gas production and the methane fraction, allowing for the calculation of pure methane yield, were chosen as key performance indicators in this research. Figure 6 displays the historical gas production data for the monitored gas wells, primarily presenting methane accumulation, production rate, and average daily yield. The latter two parameters, though measured in the same units, are derived using distinct calculation methods: the methane production rate is determined from the total gas output and the methane fraction, while the average daily yield is calculated by dividing the cumulative methane production by the number of production days. The methane production rate exhibits greater sensitivity compared to the average yield, with a less variable trend over time, thereby mitigating the influence of data variability on the projection of future trends (Alwin Blasingame et al., 1989). Despite differences in calculation methods, the observed gas production trends indicate that wells initially exhibit higher gas production rates, followed by a decline phase as time progresses.

As illustrated in Figure 6, the gas production curves can be categorized into four distinct stages. (I) Slow growth stage: The number of boreholes contributes to the cumulative extraction volume, which increases slowly due to the time-consuming nature of drilling construction. Additionally, the gas production rate initially decreases over time because several boreholes, with poor completion quality, allow mine air to be drawn into the extraction area, thereby diluting the methane. (II) Steep growth stage: The cumulative extraction volume increases significantly, and the methane production rate exhibits significant fluctuations. The production rate is primarily influenced by methane concentration, which stabilizes above the average value after the completion of drilling field construction. At this point, the basic coverage and influence area of the drilling field have been established, with gas mainly emanating from the coal body around the borehole wall, creating a high concentration gas supply. During the initial to second stage, the average daily methane production rate generally increases with extraction time, reaching its maximum value. (III) Stable increase stage: The accumulation exhibits a stable increasing trend over an extended period due to a stable gas production rate. This stage is strongly controlled by sorption time and the relative permeabilities to gas (Salmachi and Yarmohammadtooski, 2015). (IV) Accumulative rate decay stage: After 550 days of extraction, the methane production rate experiences a precipitous decline, while the daily average yield shows a gradual decrease.

Throughout the entire process, gas flow within the coalbed transitions from an unstable to a steady state. This transformation is attributed to the complex structure of natural coal, simplified as a dual poroelastic medium containing fractures and pores (Sun et al., 2018). During coalbed methane extraction, adsorbed methane within the matrix first diffuses into the fractures and then, under the pressure differential between the fracture system and the extraction borehole, transitions into free methane, migrating to the borehole through seepage. Consequently, gas within the coal matrix exists in both adsorbed and free states, influencing the migration patterns of methane (Sciencedirect, 2024a). From a flow dynamics perspective, coal seam gas flow is predominantly seepage-driven in the early stages, shifting to diffusion in the later stages (Si et al., 2019). The roles of seepage and diffusion evolve over time and across spatial gradients (Liu et al., 2018), offering new insights for gas extraction regulation.

 Production character diagnosis analysis

In the analysis and diagnosis of productivity data, it is assumed that all boreholes at a drilling site contribute equivalently to a single representative borehole. In contrast to groundwater and oil flow, the gas flow must consider the compressibility of gas, with the driving force being the square of the pressure drop, as shown in Equation 2. Q ∝ P i 2 − P w f 2 where Q is the gas flow rate, Pi is the initial reservoir pressure; Pwf is the flowing bottomhole pressure.

We note that the most important diagnostic tool is ultimately the reservoir model - whether it is provided as a type curve, it will be adjusted based on the data. Comparing production data with consistent reservoir models is a basic task for diagnosing production data regardless of the type of reservoir model. The material balance time tc, a critical parameter in reservoir engineering, is defined as the ratio of the current cumulative production Np to the daily production rate q. An equivalent relationship can be established between production at a constant rate and production at a variable rate (Clarkson, 2013; Clarkson, 2021). The integral function of the production rate and its corresponding derivative are presented in Equations 3, 4 (Doublet et al., 1994), respectively. q △ p i = 1 t c ∫ 0 t c q △ p d t c q △ p i d = t c d d t c q △ p i △ p = P i 2 − P w f 2 where, △p denotes the pressure drop, MPa2; P i and P wf denote initial reservoir pressure and flowing bottomhole pressure, respectively. [] i denote the integral; [] id denote the integral derivative.

Figure 7 presents a schematic representation of the integral-derivative function of the data (q/p) with respect to mass balance time and extraction time. It is important to note that these data functions are derived from the exact same reservoir model and are plotted in this format due to historical preference (Jiang et al., 2022; Ruiz Maraggi et al., 2023). During the initial phase of methane extraction, prior to the completion of the drilling site, the rate integral-derivative functions exhibit a rapid decline, indicative of the significant impact of drilling construction at this stage. Subsequently, after 150 days, once the drilling site is completed, the function curves gradually ascend, reaching a peak when the drawdown time approaches 550 days, after which they begin to decline.

Methane production rate changes with time at the drilling site. (a) Rate integral derivative vs. time (b) Rate integral derivative vs. mass balance time.

Two graphs compare rate integral-derivative. (a) Plots it against time in days, showing an increase after an initial flat phase marked by "Drilling." (b) Plots it against mass balance time, showing a similar trend with a rise after an initial decline. Both graphs feature black data points and a red trend line.

In reality, the quality, quantity, or accuracy of production data seldom matches theoretical expectations. The rate and pressure history is meticulously compiled from daily records, yet the pressure data can be infrequent, imprecise, or even absent (Ilk et al., 2010). Figure 7 reveals discrepancies with the typical attenuation trends of natural gas productivity observed in unfractured wells (Yehia et al., 2023; Kamari et al., 2017). A comparison of the two graphs in Figure 7 shows similar curve shapes, but with inconsistent temporal characteristics. Particularly in the later stages of extraction, the mass balance time significantly exceeds the actual extraction time due to the reduction in extraction flow.

Figure 7 illustrates the function [(q/p)] in the format commonly attributed to Blasingame format (OnePetro, 2024), which does not fully recognize the distinct characteristic features that could aid in understanding the relationship between flow rate and time for underground boreholes. To delve deeper into this relationship, Figure 8 displays the logarithm of the (q/p) integral derivative, which is a different linear partition of the logarithm of the material balance time. The slope of this partition remains nearly constant over time, offering positive implications for gas production forecasting. This constant slope may suggest a fractal characteristic of flow rate and time, potentially arising from the fractal structure of coal pores (Zhang, 2024). The complex, multi-scale pore-fracture network in coal often exhibits fractal geometry. Gas flow through such a fractal system can deviate from the classical Darcy flow behavior, manifesting as a power-law decline in production rate. The constant slope m observed in Figure 8 could thus be a direct reflection of the fractal dimension of the coal’s pore structure, which controls the accessibility and pathways for gas migration from the matrix to the borehole.

The log-log relationship between mass balance time and flow rate.

Scatter plot showing the rate integral-derivative-log against mass balance time N/q-log in days. Data points form descending diagonal clusters, separated by red trend lines, indicating a pattern of decline. Changes in methane concentration during extraction

The methane concentration serves as a critical performance parameter in evaluating borehole production, often dictating the extent of gas utilization extracted from underground borehole (Pan et al., 2014; Lou et al., 2024). Figure 9 illustrates the variation in methane concentration across twenty boreholes throughout their production lifespans. Typically, boreholes initiate with a high methane concentration, which subsequently decreases over time. The divergent concentration profiles—such as sudden spikes (e.g., D1’s increase from 9.84% to 48.9% on day 146) indicating fracture connectivity, stable linear declines (e.g., D7) demonstrating effective sealing, and premature drops (e.g., D2 and D17) revealing leakage issues—underscore the impact of geological heterogeneity and borehole quality on gas migration. These trends necessitate real-time monitoring and adaptive management, such as dynamic pressure adjustments based on concentration data, to maximize methane recovery and operational efficiency. The observed initial low and subsequent high methane concentrations during the observation period may be attributed to factors such as drilling construction, exemplified by boreholes D1 and D5.

Methane concentration and accumulated volume of each borehole (DX is the borehole number in the graph) (a–t).

Twenty graphs display concentration and accumulated volume over time. Each graph, labeled from (a) to (t), shows fluctuating lines indicating variations in concentration and accumulation trends. Axes are labeled with time and concentration percentage, and each graph includes a legend.

For enhanced analysis and comparative purposes, the data for methane concentration and accumulated volume have been categorized into four groups based on drainage time: (1) 0–650 days; (2) 0–600 days; (3) 0–550 days; (4) 0–500 days; and (5) 0–250 days Figure 9 further reveals that each borehole exhibits a distinct life cycle, with variations in the methane concentration extracted. This variability can be ascribed to three main factors: firstly, the differing initiation times of the boreholes themselves, and secondly, the presence of fractures and perforations among the boreholes due to coal creep (Danesh et al., 2015), and thirdly, air leakage of surrounding coal around boreholes (Wang et al., 2022). The inter-borehole stringing, which is prone to occur during extraction, signifies the end of the drilling life. This phenomenon can be readily identified through vacuum pressure measurements.

Figure 9 can categorize boreholes into two distinct groups based on their notable methane concentration behaviors, which are worthy of discussion. The majority of boreholes have an extraction duration exceeding 500 days, while a few, such as D2 and D17, have a lifespan of less than 200 days and gradually become non-functional, as depicted in Figure 9, where they exhibit a significant deviation from the initial group. Our initial focus is on boreholes that demonstrate a favorable extraction effect. For instance, Figure 9a illustrates the classical behavior of borehole D1 during and after the completion of the drilling site. A sudden shift in methane concentration is observed, which can even alter the cumulative methane production curve. The average methane concentration for D1 is 7.9%, with a slow initial increase in cumulative methane production, as shown in Figure 10a. On the 146th day of extraction, the methane concentration experiences a dramatic increase from 9.84% to 48.9%, stabilizing around 50% thereafter, thereby accelerating the growth of methane production. Such mutations are not unique to this borehole; similar phenomena, including both abrupt increases and decreases, are observed in other boreholes.

The detailed information for methane concentration and accumulated volume of D1 amd D7 (a), (b).

Two graphs labeled (a) and (b) showing concentration and accumulated volume over time. Both graphs feature concentration in black and accumulated volume in dashed purple. Graph (a) indicates an average concentration of 7.9% and 49.9%, while graph (b) shows 53.1%. Both axes are labeled with time and percentage or volume. Red lines highlight average values.

Furthermore, uneven acquisition time intervals can also lead to abnormal changes in methane production, as illustrated in the A-A section of Figure 10a. This occurs because high concentration data collected at long intervals result in a sharp change in methane production. Table 8 indicates that changes in methane concentration lead to a significant alteration in methane production, excluding boreholes where the gas concentration has sharply decreased to zero. The sharp change in concentration is attributed to the connection between the borehole and the fracture development area due to the prolonged deformation of the coal surrounding the borehole. The second observation pertains to the behavior of boreholes, particularly D7, as shown in Figure 10b. During extraction, the cumulative methane production increases linearly with time. Even with significant fluctuations in gas concentration, the overall trend remains stable. This stable characteristic enhances its utility in the study of productivity prediction, akin to the integral parameter of the extraction rate (Li et al., 2023; Fang et al., 2020).

Concentration mutation in the process of borehole extraction.

Boreholes Started time Mutation time Extraction days Concentration change (%)
D1# 2022/6/13 2022/11/6 146 9.84–48.9
D2# 2022/6/15 2022/7/29 44 1.89–33
D5# 2022/6/20 2022/12/5 168 1.7–17.9
2022/12/13 176 16.4–65.7
D8# 2022/10/28 2022/11/15 18 62.6–18.9
D13# 2022/8/12 2022/9/15 34 25.6–69.2
D14# 2022/8/20 2023/2/19 183 1.7–52.8
D21# 2022/10/29 2022/11/26 29 66.3–19.7

In the subsequent analysis, since the sub-boreholes inherit the trajectory angle of the main borehole, we treat the main borehole and its branches as a single entity, as shown in Table 9. The inclination angle varies from 4° to 8.5°, while the azimuth angle ranges from 87° to 149°. It is noted that some boreholes possess multiple branches, whereas others do not. Examination of Figure 9 and Table 9 reveals that alterations in the geological orientation of the borehole exert minimal influence on the extraction concentration within a confined area. Observations from Figures 2, 9 indicate that this concentration is correlated with the extent of the borehole’s control area. Furthermore, field measurements demonstrate that a borehole with a favorable extraction effect often has adjacent boreholes that are susceptible to stringing, resulting in suboptimal extraction performance. This suggests that the density of boreholes at the drilling site may be excessive for effective gas extraction.

The length, inclination, and azimuth of boreholes.

Boreholes Angle (°) Drilling length (m)
Inclination Azimuth Depth Length Summing
D1 7 128 0–267 267 267
D2 7 118 0–312 312 312
D3 7 115 0–252 252 393
D3-1 189–279 90
D3-2 261–312 51
D4 6 110 0–315 315 315
D5 5 101 0–312 312 330
D5-1 78–96 18
D6 7 104 0–210 210 210
D7 7 98 0–330 330 849
D7-1 105–333 228
D7-2 42–333 291
D8 7 94 0–330 330 330
D9 7 90 0–327 327 327
D10 7 87 0–333 333 819
D10-1 66–201 135
D10-2 126–333 207
D10-3 159–303 144
D11 8.4 126 0–288 288 561
D11-1 39–312 273
D12 8.4 132 0–330 330 816
D12-1 69–267 198
D12-2 42–330 288
D13 7 134 0–324 324 600
D13-1 33–309 276
D14 8.5 138 0–213 213 714
D14-1 135–342 207
D14-2 42–336 294
D15 8 141 0–320 320 868
D15-1 84–336 252
D15-2 48–344 296
D16 4 138 0–336 336 752
D16-1 140–317 176
D16-2 28–252 224
D16-3 156–172 16
D17 7 146 0–336 336 604
D17-1 136–148 12
D17-2 56–312 256
D18 7 149 0–320 320 756
D18-1 60–320 260
D18-2 88–264 176
D19 6 105 0–332 332 564
D19-1 80–312 232
D20 5.6 102 0–320 320 348
D20-1 104–132 28
D21 5 96.3 0–312 312 312

x-n is the nth branch hole of the main borehole x.

 Effect of vacuum pressure

The application of extraction vacuum pressure is pivotal in establishing a vacuum environment within the extraction pipeline system. This vacuum stimulates the migration of methane from the surrounding coal into the boreholes, thereby enhancing extraction efficiency. It is widely accepted that a greater vacuum pressure is more conducive to extraction during the initial stages (Dong et al., 2017). The flow of gas within fractures creates a pressure differential between the coal matrix and the fracture, facilitating gas extraction. However, as extraction time increases, the effect of suction pressure on methane migration diminishes, potentially leading to significant air leakage (Yi et al., 2021). A well-managed vacuum pressure ensures an optimal flow velocity of gas within the coal seam, enabling rapid and substantial gas transfer into the pipeline system. Thus, managing vacuum pressure from underground boreholes in the drilling field is of paramount importance.

Monitoring the changes in vacuum pressure during the borehole extraction process is crucial for timely identification of the borehole’s working condition. To evaluate the impact of suction pressure on CBM extraction, vacuum pressures from twenty boreholes were recorded over multiple collections, as depicted in Figure 11. Each borehole’s pressure fluctuates with the progression of gas extraction, exhibiting varying trends in maximum, minimum, and average values. For instance, the maximum pressure recorded ranges from 25.5 kPa to 35.8 kPa, with an average of 33.5 kPa. For the average pressure across the 20 boreholes, the minimum is 2.8 kPa, the maximum is 33.5 kPa, and the average value is 19.5 kPa, which coincides with the pipeline pressure as illustrated in Figure 5. This average value essentially reflects the influence of the vacuum pressure exerted by each borehole in the drilling field.

Vacuum pressure changes of each borehole.

Chart depicting pressure in kilopascals across various drills numbered one to twenty-one. It includes maximum, minimum, and average pressure values shown by black squares, purple circles, and blue triangles respectively. Key values marked are 33.5, 35.8, 25.5, 13.6, 19.5, 24.0, 2.8, 0, and 8.6. Red dashed lines indicate thresholds at approximately ten and thirty-five kilopascals.

The magnitude of drilling and extraction pressures significantly influences extraction efficiency and gas concentration. It is observed that lower extraction pressures allow for a greater influx of methane and air into the borehole per unit time (Liu et al., 2018). To scrutinize the impact of suction pressure on the volume of extraction, Figure 12 presents an analysis where the average vacuum pressure is plotted on the horizontal axis, and the daily absolute and mixed gas volumes during the extraction process are plotted on the vertical axis. According to Figure 12, the vacuum pressure fluctuates between 13 kPa and 25 kPa, indicating a relatively narrow range. The mixed gas volume ranges from 313.0 m3/d to 873.4 m3/d, with the majority of boreholes clustering near the average value of 467.6 m3/d. The absolute gas volume varies from 60.3 m3/d to 322.8 m3/d, with an average volume of 159.4 m3/d. From a distribution standpoint, the data for mixed volume is more tightly concentrated, whereas the data for absolute volume is more dispersed.

Gas extraction amount under different vacuum pressure.

Scatter plot comparing absolute and mixed volumes in cubic meters per day against pressure in kilopascals. Black squares represent absolute volumes, while purple circles represent mixed volumes. Key data points are labeled with numerical values, such as 873.4, 467.6, and 313.0. A dashed purple line indicates a trend across the graph.

As vacuum pressure increases, the mixed volume remains largely unchanged, while the absolute volume exhibits a slight overall increase. Notably, the maxima for both mixed and absolute volumes occur at the lower pressure range, below 16 kPa. Within this narrow range, vacuum pressure appears to have no significant effect on either the mixed or absolute volumes. However, adjusting vacuum pressure from 25 kPa to 13 kPa is of considerable importance for overall pipeline regulation and energy conservation. Thus, it can be inferred that within a certain vacuum pressure range, vacuum pressure adjustment does not significantly affect field extraction. There exists a threshold for the minimum vacuum pressure that is effective.

Figure 13 illustrates the distribution of methane concentration in relation to increasing vacuum pressure. The distribution patterns for the boreholes are based on effective extraction time, presenting a similarity to Figure 9. Notably, boreholes exhibit varying concentration distributions, with frequencies in the high vacuum pressure range generally exceeding those in the low vacuum pressure range, as depicted in Figure 13. This suggests that high vacuum pressure correlates with higher methane concentrations, particularly when the borehole sealing is effective.

Methane concentration distribution under different vacuum pressure (a–t).

Multiple scatter plots labeled (a) to (t) are shown, each with vacuum pressure (kPa) on the x-axis and cavitation (%) on the y-axis. Various data points are plotted in black, blue, red, green, and purple, with red lines highlighting trends in some plots.

The increase of vacuum pressure augments the gas pressure gradient, which in turn increases the gas flow rate and permeability. Furthermore, under vacuum pressure conditions, the methane desorption rate and the diffusion coefficient DD escalate, diminishing the mass transfer resistance within fractures. The vacuum pressure environment surrounding coal particles modifies the desorption kinetic parameters, facilitating the desorption and diffusion of methane within the coal matrix (Du et al., 2018). The distribution characteristics evident in Figure 13 can be categorized into three distinct patterns: (1) a step distribution, where methane concentration and distribution probability surge sharply when vacuum pressure exceeds 17.5 kPa, as seen in Figures 13a,c,k,n,o; (2) a linear strip pattern, with concentration data linearly distributed between two profiles, as observed in Figures 13i,j,s,t; and (3) a diffused type, characterized by uniform distribution of concentration data across vacuum pressures, as represented by the remaining figures. These concentration data distributions reflect the quality of borehole gas extraction to a certain extent.

Figure 13 illustrates the relationship between vacuum pressure and methane concentration across multiple boreholes. A clear threshold phenomenon is observed: once the vacuum pressure exceeds approximately 17.5 kPa, further increases yield no significant enhancement in methane concentration. This plateau indicates that beyond a critical negative pressure, gas desorption and seepage rates become diffusion-limited, suggesting that the coal matrix and fracture network, rather than suction strength, control the overall gas transport. Consequently, excessive vacuum pressure not only fails to improve extraction efficiency but may reduce economic viability due to increased energy consumption and potential borehole instability.

In contrast, boreholes with shorter service lives (<250 days), as shown in Figures 13p–t, exhibit lower frequencies of maximum vacuum pressure, with most values remaining below 25 kPa. This pattern reflects suboptimal sealing or early borehole degradation, which leads to gas leakage and reduced effective vacuum within the borehole. The lower frequency of high-pressure operation thus serves as an indirect indicator of borehole quality and sealing integrity. When combined with pump suction pressure data, this parameter can quantitatively assess both technical performance and economic returns of drilling operations. Overall, Figure 13 demonstrates that optimal extraction lies within a moderate vacuum-pressure range, where the system maintains stable gas flow without unnecessary energy expenditure or leakage losses.

 Effect of borehole length

Figure 14 illustrates the variation in pipeline pressure and mixed volume in relation to drill length, which is the cumulative extent of a main borehole and its branches. This length ranges from 250 m to 900 m. The corresponding pipeline pressure fluctuates between 13 kPa and 24 kPa, exhibiting a distribution of peak bands for maximum and minimum values. Meanwhile, the mixed volume ranges from 300 m3/d to 900 m3/d, with a concentration around 400 m3/d. Both parameters exhibit random variation and demonstrate minimal correlation with drill length.

Pressure and mixed volume with the borehole length.

Scatter plot showing borehole length in meters on the x-axis compared to pressure in kilopascals on the left y-axis and mixed volume in cubic meters per day on the right y-axis. Pressure is represented by black squares and mixed volume by purple circles. Data points are dispersed across borehole lengths from 300 to 900 meters.

The elongation of borehole length increases the control area over the coalbed, facilitating a greater flow of methane through the borehole. Consequently, the cumulative absolute volume of gas extracted has a positive relationship with borehole length, as depicted in Figure 15. However, this increase in length also results in a reduction of the cumulative absolute volume per unit length, potentially diminishing the extraction efficiency of the borehole. This necessitates a re-evaluation of the optimal borehole length to achieve the best economic benefits from extraction.

Accumulative absolute amount and amount per unit length along the borehole.

Scatter plot showing accumulative absolute amount and amount per unit length against borehole length in meters. Black squares represent total accumulative amounts and blue triangles indicate amounts per unit length, with values on respective y-axes ranging from zero to three hundred fifty and zero to one. Data points vary across borehole lengths from two hundred to nine hundred meters.

Furthermore, a reduction in the control area can shorten the effective extraction time, thereby saving considerable time for the transition to mine production. The area encompassed by the borehole length is 40,913 m2. Given a coal seam thickness of 4.5 m, the volume of coal under control is estimated at 265,000 tons. The raw coal in this area has a gas content of 9.92 m3/t. The 3219 transport trough is scheduled for excavation in August 2023, with the anticipated gas extraction outcomes detailed in Table 10.

Prediction of pre-extraction effect in drilling field.

Advance progress (m) Extraction effect (m3/min) Time (d) Estimated extraction volume (×104m3) Residual gas content (m3/t)
11092 2 270 93 6.41
Implications on gas extraction

During the gas extraction process via underground boreholes, vacuum pressures within the pipelines and boreholes draw free gas from the fracture system. As the free gas in the coalbed diminishes, adsorbed gas within the coal matrix desorbs and diffuses into the fractures, eventually penetrating the boreholes (ScienceDirect, 2024b). Initially, extraction relies on free gas, transitioning to a dependence on adsorbed gas. Gas migration can be categorized into three stages (Liu et al., 2018): (1) the seepage-dominated stage; (2) the transformation-dominated stage; and (3) the diffusion-dominated stage. The most evident outcome is the skewed distribution of daily gas production rates over time. The significance of desorption and diffusion in controlling gas production has been acknowledged and is applied in suction pressure regulation. For instance, a time-based suction pressure regulation method enhances methane extraction efficiency and concentration compared to constant pressure extraction. Different extraction strategies are essential, contingent upon the stage of gas migration in the coalbed. In the initial extraction stage, a high negative extraction pressure can rapidly extract gas from the seam, increasing concentration. As extraction progresses and gas concentration declines, reducing the vacuum pressure can minimize air leakage, thereby enhancing extraction concentration and extending extraction time (Dong et al., 2017).

The suction pressure significantly influences pumping efficiency and methane concentration. As depicted in Figure 13, increasing the extraction suction pressure within a certain range can improve methane concentration by augmenting the gas flow rate in the pipeline. Excessive suction pressure can increase resistance in the borehole and pipeline and cause air leakage, reducing extraction efficiency and methane concentration (Pan et al., 2014). Conversely, if suction pressure is too low, it fails to provide adequate extraction power. Therefore, selecting a reasonable suction pressure and maintenance duration is crucial, based on coal seam conditions, equipment, and other factors. A sharp increase in methane concentration across all underground boreholes is observed when the suction pressure exceeds 17.5 kPa. Higher suction pressures are employed in the early stages to initiate methane flow and overcome resistance. In later stages, when gas concentration is low, a lower suction pressure can improve extraction efficiency and methane concentration. Additionally, emphasizing the quality of borehole sealing in the actual extraction process is vital. Effectively sealing air leakage channels in tunnels into boreholes is key to addressing low-concentration, small-flow engineering issues. A proper seal can substantially increase methane concentration and gas production duration (Lou et al., 2024; Xia et al., 2014; Lou et al., 2025; Longinos et al., 2023; Longinos et al., 2025a; Longinos et al., 2025b).

As borehole length increases, so does the inner area of the coalbed exposed, enabling the capture and discharge of more gas by the extraction system, thereby increasing gas extraction volume. Longer boreholes facilitate the formation of stable flow channels in the coal seam, reduce localized flow resistance, and enhance borehole gas flow rates, which are less susceptible to geological condition changes, allowing for greater gas release. This contributes to the stable operation of the gas extraction system and improves the reliability of the extraction effect. While an increase in borehole length typically results in higher efficiency and total gas extraction, it also entails greater investment in drilling equipment, tools, and manpower, leading to higher initial costs. However, the long-term benefits of longer boreholes, such as higher gas extraction volumes and more stable extraction effects, can reduce methane concentration risks and improve mine safety production levels. Determining a reasonable borehole length is essential for optimizing gas extraction effects, considering factors like coal seam geological conditions, gas accumulation states, and extraction system parameters.

In practical development and application, extraction engineering design should be tailored to the drilling field, considering not only borehole spacing and permeability but also geographical conditions to enhance extraction effects, particularly the orientation and direction of fractures. Improving drilling speed is crucial for overall drilling field construction and gas extraction. By altering the original pipeline’s resistance distribution pressure, the system now relies on concentration ratios to adjust pressure for effective gas mining. On this foundation, developing an intelligent pressure regulation system could provide an algorithm for the system to automatically adjust suction pressure. Such a system would not only enable online monitoring of methane extraction but also effectively improve methane utilization. The ultimate goal is to develop an intelligent pressure regulation system capable of automatically adjusting suction pressure, shifting the pressure distribution caused by pipeline resistance to that dictated by methane concentration ratios for effective extraction. The integral variable of gas production rate is a better predictor of productivity than the differential variable. The daily flow rate is more favorable for judging gas production trends and explaining dynamic attenuation. However, the average daily gas production amount is more appropriate as it is primarily stable and less susceptible to data noise.

 Conclusion

Based on the field data of underground borehole extraction from the No. 3 coal seam of the Wangpo Coal Mine, the evolution of gas flow rate, vacuum pressure, methane concentration, and cumulative extraction was systematically analyzed to reveal the dynamic characteristics and control mechanisms of coalbed methane (CBM) production. The main conclusions are as follows:

The gas-production rate exhibits a rapid-rise and gradual-decay pattern, reflecting the combined contribution of free gas release and adsorbed-gas desorption within the coal matrix. This behavior confirms that production dynamics are governed by both the pore–fracture structure and the staged pressure decline during extraction.

The rate-integral and derivative analysis shows a distinct linear segment in log–log space, indicating a shift from early-time transient flow to late-time boundary-dominated flow. This feature provides a quantitative basis for forecasting gas-production performance and identifying the control area of each borehole.

The cumulative gas-production curve can be divided into slow-growth, rapid-increase, stable-growth, and decline stages. The steep-growth period corresponds to the drainage of free gas, whereas the subsequent stable stage reflects the diffusion and percolation of adsorbed gas toward the borehole. The rate-decay stage marks the boundary of effective control for each borehole.

Field statistics reveal a threshold vacuum pressure of ∼17.5 kPa, above which methane concentration no longer increases significantly. Excessive suction may even lower efficiency through air leakage and diffusion limitation. Thus, the extraction system should adapt vacuum pressure dynamically according to gas concentration and extraction time, while maintaining strict borehole sealing management to prevent leakage losses.

Extended extraction duration stabilizes gas supply but increases leakage risk due to evolving fractures. Optimal operation should therefore balance extraction time, borehole layout, and permeability evolution, supported by continuous field monitoring. These findings provide a quantitative and practical framework for enhancing CBM drainage design and ensuring the safety and efficiency of underground gas-extraction systems.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

 Author contributions

BW: Visualization, Funding acquisition, Writing – review and editing, Writing – original draft, Conceptualization, Methodology. XZ: Supervision, Methodology, Writing – original draft, Investigation, Funding acquisition, Formal Analysis, Writing – review and editing. QM: Formal Analysis, Methodology, Resources, Investigation, Writing – review and editing. ZW: Writing – review and editing, Investigation, Methodology, Visualization. GW: Funding acquisition, Project administration, Formal Analysis, Writing – review and editing, Investigation, Supervision. QL: Formal Analysis, Methodology, Investigation, Writing – review and editing. JL: Methodology, Investigation, Writing – review and editing, Formal Analysis. HY: Methodology, Investigation, Writing – review and editing, Formal Analysis. JC: Writing – review and editing, Investigation, Methodology, Formal Analysis. LD: Methodology, Investigation, Writing – review and editing, Formal Analysis. XX: Investigation, Methodology, Writing – review and editing, Formal Analysis. MZ: Investigation, Writing – review and editing, Formal Analysis, Methodology.

 Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

 Generative AI statement

The authors declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

 Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Edited by: Jinlong Liu, Zhejiang University, China

Reviewed by: Ruomiao Yang, Zhejiang University, China

Haowei Yao, Zhengzhou University of Light Industry, China

Tianfang Xie, Purdue University, United States

Sotirios Longinos, University of Calgary, Canada

 References Alwin Blasingame T. Blasingame T. A. Johnston J. (1989). Type-curve analysis using the pressure integral method. 10.2118/18799-ms Chen F. Wu C. Huo Z. Zhao Q. (2024). Productivity characteristics and controlling factors of coalbed methane vertical well in a coal block of qinshui basin. Energy Explor. and Exploitation 42 (3), 856–877. 10.1177/01445987231219766 Chu P. Xie H. Gao M. Li C. Shang D. Liu Q. (2024). Influence of desorption hysteresis effects on coalbed methane migration and production based on dual-porosity medium model incorporating hysteresis pressure. Comput. Geotechnics 165, 105893. 10.1016/j.compgeo.2023.105893 Clarkson C. (2013). Production data analysis of unconventional gas Wells: Review of theory and best practices. Int. J. Coal Geol. 109, 101–146. 10.1016/j.coal.2013.01.002 Clarkson C. (2021). Introduction to the concept of RTA, RTA methods, and the RTA workflow, 1–106. Danesh N. Chen Z. Aminossadati S. Kizil M. Pan Z. Connell L. D. (2015). Creep, a neglected phenomenon in coal permeability evolution and coalbed methane production. 10.2118/176980-ms Dong J. Cheng Y. Jin K. Zhang H. Liu Q. Jiang J. (2017). Effects of diffusion and suction negative pressure on coalbed methane extraction and a new measure to increase the methane utilization rate. Fuel 197, 70–81. 10.1016/j.fuel.2017.02.006 Doublet L. Doublet L. E. Pande P. K. Blasingame T. A. (1994). Decline curve analysis using type Curves--Analysis of oil well production data using material balance time: application to field cases. 10.2118/28688-ms Du Y. Chen X. Li L. Wang P. (2018). Characteristics of methane desorption and diffusion in coal within a negative pressure environment. Fuel 217, 111–121. 10.1016/j.fuel.2017.12.087 Du S. Wang J. Wang M. Yang J. Zhang C. Zhao Y. (2023a). A systematic data-driven approach for production forecasting of coalbed methane incorporating deep learning and ensemble learning adapted to complex production patterns. Energy 263, 126121. 10.1016/j.energy.2022.126121 Du S. Wang M. Yang J. Zhao Y. Wang J. Yue M. (2023b). An enhanced prediction framework for coalbed methane production incorporating deep learning and transfer learning. Energy 282, 128877. 10.1016/j.energy.2023.128877 Fang B. Hu J. Xu J. Zhang Y. (2020). A semi-analytical model for horizontal-well productivity in shale gas reservoirs: coupling of multi-scale seepage and matrix shrinkage. J. Petroleum Sci. Eng. 195, 107869. 10.1016/j.petrol.2020.107869 Flores R. M. (1998). Coalbed methane: from hazard to resource. Int. J. Coal Geol. 35 (1), 3–26. 10.1016/s0166-5162(97)00043-8 Guo H. Yu Y. Wang Y. Wang K. Yuan L. Xu C. (2024). Experimental study on the desorption law and diffusion kinetic characteristics of gas in raw coal and tectonic coal. Energy 289, 129924. 10.1016/j.energy.2023.129924 Guo X. Liu Y. Zhang L. Deng C. Song L. (2024). Regulatory mechanism of microscopic pore structure on anisotropy of gas multimodal seepage in original coals. Energy 300, 131611. 10.1016/j.energy.2024.131611 Huang Q. Shen J. Zhang B. Zhao G. Cheng M. Cai Y. (2023). Real-time monitoring of coalbed methane production network following liquid CO2 injection in a low-efficiency well network: response to gas and water production characteristics. Energy, 285. Ilk D. Anderson D. M. Stotts G. W. J. Mattar L. Blasingame T. A. (2010). “Production-data analysis—challenges, pitfalls, diagnostics”. SPE Reservoir Eval. Eng. 13 (3), 538–552. 10.2118/102048-pa Ji P. Lin H. Li S. Kong X. Wang X. Zhang J. (2024). Technical system and prospects for precise methane extraction in the entire life cycle of coal mining under the goal of carbon peak and carbon neutrality. Geoenergy Sci. Eng. 238, 212855. 10.1016/j.geoen.2024.212855 Jia Y. Shi J. Zhang L. Li W. He Y. Li Y. (2024). An analytical method for timely predicting of coal seam pressure during gas production for undersaturated coalbed methane reservoirs. Processes 12 (4), 777. 10.3390/pr12040777 Jiang T. Sun H. Wang H. Xiao X. (2022). Reserves estimation for geopressured gas reservoirs. Elsevier. Kamari A. Mohammadi A. H. Lee M. Bahadori A. (2017). Decline curve based models for predicting natural gas well performance. Petroleum 3 (2), 242–248. 10.1016/j.petlm.2016.06.006 Kang J. Elsworth D. Fu X. Liang S. Chen H. (2022). Influence of water on elastic deformation of coal and its control on permeability in coalbed methane production. J. Petroleum Sci. Eng. 208, 109603. 10.1016/j.petrol.2021.109603 Li J. Liu D. Lu S. Yao Y. Xue H. (2015). Evaluation and modeling of the CO2 permeability variation by coupling effective pore size evolution in anthracite coal. Energy and Fuels 29, 717–723. 10.1021/ef502674y Li J. Zhou X. Gayubov A. Shamil S. (2023). Study on production performance characteristics of horizontal Wells in low permeability and tight oil reservoirs. Energy 284, 129286. 10.1016/j.energy.2023.129286 Li C. Sun L. Zhao Z. Zhang J. Li Y. Meng Y. (2024). A method for evaluating coalbed methane reservoir productivity considering drilling fluid damage. Energies 17 (7), 1686. 10.3390/en17071686 Li H. Lv X. Lu J. Liu M. Yang W. Hong Y. (2024). Effect of cyclic thermal stimulation on the pore structure and fluid space of coal and inspiration for coalbed methane production. Energy 289, 129994. 10.1016/j.energy.2023.129994 Lin H. Ji P. Kong X. Li S. Long H. Xiao T. (2023). Experimental study on the influence of gas pressure on CH4 adsorption-desorption-seepage and deformation characteristics of coal in the whole process under triaxial stress. Fuel 333, 126513. 10.1016/j.fuel.2022.126513 Liu Q. Cheng Y. Zhou H. Guo P. An F. Chen H. (2015). A mathematical model of coupled gas flow and coal deformation with gas diffusion and klinkenberg effects. Rock Mech. Rock Eng. 48, 1163–1180. 10.1007/s00603-014-0594-9 Liu Z. Cheng Y. Dong J. Jiang J. Wang L. Li W. (2018). Master role conversion between diffusion and seepage on coalbed methane production: implications for adjusting suction pressure on extraction borehole. Fuel 223, 373–384. 10.1016/j.fuel.2018.03.047 Liu D. Jia Q. Cai Y. Gao C. Qiu F. Zhao Z. (2022). A new insight into coalbed methane occurrence and accumulation in the qinshui basin, China. Gondwana Res. 111, 280–297. 10.1016/j.gr.2022.08.011 Liu Z. Han J. Yang H. Lv J. Dong S. (2024). A new model for coal gas seepage based on fracture-pore fractal structure characteristics. Int. J. Rock Mech. Min. Sci. 173, 105626. 10.1016/j.ijrmms.2023.105626 Longinos S. N. Dillinger A. Wang L. Hazlett R. (2023). Uniaxial compressive strength (UCS) and SEM study of liquid nitrogen for waterless hydraulic fracturing in coalbed methane reservoirs of Karaganda basin in Kazakhstan. Gas Sci. Eng. 115, 204998. 10.1016/j.jgsce.2023.204998 Longinos S. N. Begaliyev D. James M. Konysbekov A. (2025a). Experimental and Numerical Investigation of Cryogenic Fracturing in Coal Rocks Using Compression and Dog-Bone Testing Methods: longinos et al. Rock Mech. Rock Eng., 1–18. Longinos S. N. Begaliyev D. Asif M. Tuleugaliyev M. (2025b). Cryogenic fracturing of coal with LN2 treatment: a sustainable approach for enhancing coalbed methane extraction in water-scarce regions. Gas Sci. Eng. 142, 205686. 10.1016/j.jgsce.2025.205686 Lou Z. Wang K. Kang M. Zhao W. Wei G. Yue J. (2024). Plugging methods for underground gas extraction boreholes in coal seams: a review of processes, challenges and strategies. Gas Sci. Eng. 122, 205225. 10.1016/j.jgsce.2024.205225 Lou Z. Wang K. Yao H. Zhao W. Qin H. Wu Z. (2025). A novel dynamic filling material for plugging fractures around underground gas extraction boreholes: experimental and engineering performances. Energy 314, 134202. 10.1016/j.energy.2024.134202 OnePetro (2024). “Application of blasingame type curves to a multi-well gas-condensate reservoir: field case study,” in SPE Trinidad and Tobago section energy resources conference. Available online at: https://onepetro.org/SPETTCE/proceedings-abstract/18TTCE/3-18TTCE/D031S026R003/215468. Pan R. Cheng Y. Yuan L. Yu M. Dong J. (2014). Effect of bedding structural diversity of coal on permeability evolution and gas disasters control with coal mining. Nat. Hazards 73, 531–546. 10.1007/s11069-014-1086-7 Paper (2024). The new policy on methane emission control forces the utilization of low concentration gas to speed up. [EB/OL]. Available online at: http://paper.people.com.cn/zgnyb/html/2023-11/20/content_26028646.htm. Ruiz Maraggi L. M. Lake L. W. Walsh M. P. (2023). Limitations of rate normalization and material balance time in rate-transient analysis of unconventional reservoirs. Geoenergy Sci. Eng. 227, 211844. 10.1016/j.geoen.2023.211844 Salmachi A. Yarmohammadtooski Z. (2015). Production data analysis of coalbed methane wells to estimate the time required to reach to peak of gas production. Int. J. Coal Geol. 141, 33–41. 10.1016/j.coal.2015.02.006 Sciencedirect (2024a). Modeling of gas migration in a dual-porosity coal seam around a borehole: the effects of three types of driving forces in coal matrix - sciencedirect[eb/ol]. Available online at: https://www.sciencedirect.com/science/article/abs/pii/S0360544222030675. ScienceDirect (2024b). Gas diffusion behavior of coal and its impact on production from coalbed methane reservoirs - sciencedirect[eb/ol]. Available online at: https://www.sciencedirect.com/science/article/abs/pii/S0166516211000619. Shi J. Jia Y. Wu J. Xu F. Sun Z. Liu C. (2021). Dynamic performance prediction of coalbed methane wells under the control of bottom-hole pressure and casing pressure. J. Of Petroleum Sci. And Eng., 196. Si L. Li Z. Yang Y. Gao R. (2019). The stage evolution characteristics of gas transport during mine gas extraction: its application in borehole layout for improving gas production. Fuel 241, 164–175. 10.1016/j.fuel.2018.12.038 Stopa J. Mikołajczak E. (2018). Empirical modeling of two-phase CBM production using analogy to nature. J. Petroleum Sci. Eng. 171, 1487–1495. 10.1016/j.petrol.2018.07.018 Su X. Lin X. Liu S. Zhao M. Song Y. (2005). Geology of coalbed methane reservoirs in the southeast qinshui basin of China. Int. J. Coal Geol. 62 (4), 197–210. 10.1016/j.coal.2005.01.002 Sun Z. Shi J. Zhang T. Wu K. Feng D. Sun F. (2018). A fully-coupled semi-analytical model for effective gas/water phase permeability during coal-bed methane production. Fuel 223, 44–52. 10.1016/j.fuel.2018.03.012 Upadhyay R. Kagdi I. Parida B. (2024). A novel alternative method of compositional simulation: a semi-analytical approach for predicting and evaluating reservoir performance of a multicomponent coalbed methane (CBM) well. Acs Omega 9 (5), 5637–5660. 10.1021/acsomega.3c08030 38343921 Wang Z. Sun Y. Li Z. Wang Y. You Z. (2022). Multiphysics responses of coal seam gas extraction with borehole sealed by active support sealing method and its applications. J. Nat. Gas Sci. Eng. 100, 104466. 10.1016/j.jngse.2022.104466 Wang L. Li J. Han Y. Liu M. Song F. Tian W. (2023). A new coal seam seepage parameter prediction method based on gas pressure recovery curve for adjacent aquifer environment. Fuel 353, 129240. 10.1016/j.fuel.2023.129240 Wang C. Zhao H. Liu Z. Wang T. Chen G. (2024). A fully coupled gas–water–solids mathematical model for vertical well drainage of coalbed methane. Energies 17 (6), 1497. 10.3390/en17061497 Wei M. Liu C. Liu Y. Liu J. Elsworth D. Tivane O. A. (2022). Long-term effect of desorption-induced matrix shrinkage on the evolution of coal permeability during coalbed methane production. J. Petroleum Sci. Eng. 208, 109378. 10.1016/j.petrol.2021.109378 Wei J. Su E. Xu G. Yang Y. Han S. Chen X. (2024). Comparative analysis of permeability rebound and recovery of tectonic and intact coal: implications for coalbed methane recovery in tectonic coal reservoirs. Energy 301, 131702. 10.1016/j.energy.2024.131702 Xia T. Zhou F. Liu J. Hu S. Liu Y. (2014). A fully coupled coal deformation and compositional flow model for the control of the pre-mining coal seam gas extraction. Int. J. Rock Mech. Min. Sci. 72, 138–148. 10.1016/j.ijrmms.2014.08.012 Xu F. Zhang W. Li Z. Zhang L. Zhang J. Hou Q. (2023a). Coalbed methane reservoir description and enhanced recovery technologies in baode block, ordos basin. Nat. Gas. Ind. 43, 96–112. Xu F. Hou W. Xiong X. Xu B. Wu P. Wang H. (2023b). The status and development strategy of coalbed methane industry in China. Petroleum Explor. Dev. 50 (4), 765–783. 10.1016/s1876-3804(23)60427-6 Xu H. Qin Y. Yang D. Wang G. Huang Q. Wu F. (2024). Quantification of gas transport behavior during coalbed methane extraction in A coal seam considering a dual-porosity/single-permeability model. Nat. Resour. Res. 33 (1), 321–345. 10.1007/s11053-023-10291-4 Yan X. Zhang S. Tang S. Li Z. Guan W. Zhang Q. (2021). A prediction model for pressure propagation and production boundary during coalbed methane development. Energy and Fuels 35 (2), 1219–1233. 10.1021/acs.energyfuels.0c03354 Ye Z. Chen D. Wang J. (2014). Evaluation of the non-darcy effect in coalbed methane production. FUEL 121, 1–10. 10.1016/j.fuel.2013.12.019 Yehia T. Abdelhafiz M. M. Hegazy G. Elnekhaily S. A. Mahmoud O. (2023). A comprehensive review of deterministic decline curve analysis for oil and gas reservoirs. Geoenergy Sci. Eng. 226, 211775. 10.1016/j.geoen.2023.211775 Yi M. Wang L. Hao C. Liu Q. Wang Z. (2021). Method for designing the optimal sealing depth in methane drainage boreholes to realize efficient drainage. Int. J. Coal Sci. and Technol. 8, 1400–1410. 10.1007/s40789-021-00448-y Yu H. Li Z. Bai Y. Wang Y. Hu H. (2023). Research on gas injection to increase coalbed methane production based on thermo-hydro-mechanical coupling. Fuel 354, 129294. 10.1016/j.fuel.2023.129294 Zhang X. (2024). Fractal character of coal nanopore and effect of deviation corrected, coal rank, and gas adsorption. Microporous Mesoporous Mater. 367, 112972. 10.1016/j.micromeso.2023.112972 Zhang X. Zhu Z. Wen G. Lang L. Wang M. (2021). Study on gas desorption and diffusion kinetic behavior in coal matrix using a modified shrinking core model. J. Petroleum Sci. Eng. 204, 108701. 10.1016/j.petrol.2021.108701 Zhang X. Chen B. zhuang Z. Z. Feng Q. Fan B. (2023). New methods of coalbed methane production analysis based on the generalized gamma distribution and field applications. Appl. Energy 350, 121729. 10.1016/j.apenergy.2023.121729 Zhao D. Liu J. Pan J. (2018). Study on gas seepage from coal seams in the distance between boreholes for gas extraction. J. Loss Prev. Process Industries 54, 266–272. 10.1016/j.jlp.2018.04.013 Zhao Z. Chen Y. Zhang Y. Mei G. Luo J. Yan H. (2023). A deep learning model for predicting the production of coalbed methane considering time, space, and geological features. Comput. and Geosciences, 173. Zhao Z. Liu P. Li Q. Nie B. Zhao K. Zhao Y. (2024). Evaluating coal pore structure and gas sorption-diffusion behavior alteration induced by ultrasound stimulation using sorbing tests and matrix diffusion modeling. Geoenergy Sci. And Eng. 234, 212642. 10.1016/j.geoen.2024.212642 Zhou H. Yang Q. Cheng Y. Ge C. Chen J. (2014). Methane drainage and utilization in coal mines with strong coal and gas outburst dangers: a case study in luling mine, China. J. Nat. Gas Sci. Eng. 20, 357–365. 10.1016/j.jngse.2014.07.023 Zhou F. Yao G. Wang J. (2014). Effects of variogram characteristics of coal permeability on CBM production: a case study in southeast qinshui basin, China. Energy Explor. and Exploitation 32 (2), 263–279. 10.1260/0144-5987.32.2.263