Evolution of the roof caving and fracture zones during mining of close range coal seams

Abstract

This study employs an integrated methodology combining slip line field theory, FLAC3D numerical simulation, and hydraulic injection testing to systematically investigate the evolutionary characteristics of roof two zones under repeated mining conditions in close range coal seams No.9 and 10 (average interburden thickness is 10.2 m). Theoretical modeling reveals a maximum failure depth of 6.26 m in the No.9 coal seam, with corresponding caving and fracture zone heights measuring 9.6 m and 33.4 m respectively. Numerical simulations demonstrate that subsequent extraction of the No.10 coal seam induces fracture zone expansion to 66 m vertical elevation, representing a 73% increase relative to single-layer mining conditions. Strong congruence was confirmed through MATLAB-based regression analysis. Hydraulic injection tests substantiate the spatial heterogeneity of two zones (caving zone and fracture zone) development, while mechanical analysis proposes a stress arch-dominated fracture propagation mechanism. These findings establish a mechanical framework for safe extraction of close range coal seam clusters, successfully informing optimization of combined bolt reinforcement and grouting curtain configurations in practical applications.

Introduction

With the continuous growth of global energy demand and the depletion of shallow coal resources, the exploitation of deep coal resources constitutes an imperative for ensuring energy security. As a critical national coal production base, China’s Jin-Shan mining area has experienced progressive increases in extraction depth in recent years, with average mining depths reaching 650–850 m; certain mines exceed depths of 1,000 m, exhibiting pronounced deep-mining characteristics. Under these conditions, coal resource distribution within the Jin-Shan mining area demonstrates a typical “proximity coal seam cluster” configuration, wherein coal seams with spacing less than 20 m comprise 35% of total reserves. The exploitation of these clusters is therefore crucial for mitigating regional energy supply-demand imbalances.

However, the mining of proximity coal seam clusters is significantly impacted by dynamic interlayer interactions due to narrow interlayer distances and strong structural correlations within the rock mass. The extraction of lower coal seams destabilizes residual coal pillars and overlying rock structures in upper seams. Concurrently, mined-out areas within upper seams alter the stress environment and fracture propagation pathways in underlying seams. This intensified “interlayer interaction” markedly exacerbates the coupling damage mechanisms within the overlying strata compared to single-seam mining, manifesting specifically as: expanded fracture zones in the overlying rock, complex stress redistribution pathways, and heightened propensity for water accumulation and fissure connectivity within goaf areas. These phenomena directly compromise roadway stability (resulting in deformation increases of 30%−50%) and undermine the effectiveness of roof water hazard prevention measures (with roof water-inrush accident incidence rates 2–3 times higher than in single-seam mining scenarios). Current research primarily concentrates on three critical dimensions:

Research progress on stress transfer mechanism: Through numerical simulation, Yang¹ demonstrated the stress arch effect in key bearing layers during multi-seam mining, showing stress transfer path control by interlayer rock mass stiffness. Based on similar material experiments, Cui² found that the downgoing mining sequence will lead to the occurrence of asymmetric stress concentration phenomenon in the roof of the lower coal seam. Fu³ studied the stress characteristics and control problems of coal rocks in the mining process of inclined coal seam under the condition of close range coal seam group. Feng⁴ analyzed the induction mechanism of strong pressure in the mining process of thick coal seam under the empty area. Zhang⁵ analyzed the deformation and stress transfer rules of the overlying rock strata in the process of coal seam mining. Teng⁶ analyzed the dynamic balance of coal column in close range coal seam group under repeated mining conditions. Zhang⁷ Zhang⁸ analyzed the specific influence of the coal column of the goaf on the mining surface pressure of the lower coal seam. Wei⁹ pointed out that the stress difference between the upper coal goaf and the coal pillar is the key factor leading to the significant increase in the shear stress of the lower coal roadway. However, extant research has predominantly focused on static stress field analysis while neglecting systematic characterization of dynamics during mining activities.

Rock mass failure mode recognition: Based on PFC numerical simulation technology, Xinlei¹⁰ studied the optimization of coal discharge parameters and top coal fall technology in deep buried hard coal seam. Peng¹¹ developed a fracture index model for close range coal mining using microseismic monitoring, finding the roof crushing angle rose 37% when layer spacing was below 15 m. Wang¹² confirmed via CT scans that lower coal seam roof fracture networks exhibit “butterfly-shaped” expansion due to upper seam mining disturbances. Xiong¹³ analyzed the influence of repeated mining on the strength characteristics of coal seam by experimental and simulation methods, and the mechanical model of coal seam instability is established. Du¹⁴ explored the dynamic evolution of the stress, energy field and plastic zone in deep coal seam. Yan¹⁵ investigated the disaster mechanism caused by the mining of the close range coal seam below the sandstone aquifer, especially the mutation of the overburden fracture and its catastrophic consequences. With the help of advanced ground penetrating radar (GPR) technology, Wang¹⁶ analyzed the stability of the residual pillar. Using the discrete element method (DEM) simulation analysis, Zang¹⁷ revealed the influencing factors of the stability of the residual coal pillar under close mining conditions. Chen¹⁸ studied the failure characteristics of surrounding rock and the expansion law of plastic area under the remaining coal column under the mining pressure. Lu¹⁹ studied the destruction mechanism of the upper coal floor in the process of close range coal seam mining. Current investigations exhibit a notable deficiency in elucidating the spatiotemporal evolution mechanisms inherent in stress system interactions under progressive extraction conditions.

Geological structure influence law: Yin²⁰ pointed out that in the deep close range coal seam group, the complex geological conditions and high intensity pressure. Tang²¹ developed a stress mutation criterion for faulted coal seams using fault activation theory, showing mining stress concentration coefficients up to 2.3 when fault dip angles exceed 45°. Kang²² found a negative exponential correlation between fracture height and coal seam spacing via regression analysis, but their model ignored interlayer rock mechanics parameters’ spatial heterogeneity. Cao²³ studied the evolution law of crack band and its height prediction method of roof plate. The existing research on destruction mode relies on a single observation method, and lacks a multi-dimensional verification system of theoretical calculation, numerical simulation and field detection.

These findings offer significant theoretical contributions to short-distance coal seam mining operations, yet critical scientific questions persist regarding two zones development patterns in close range coal seam mining:

(1) Establishing a comprehensive analytical framework for complex coupling damage mechanisms during close-rang coal seam mining, encompassing the characterization of static stress fields and dynamic mining-induced stress fields, and enabling multidimensional verification of fracture development spatiotemporal evolution, presents an imperative research need.

(2) Quantifying the spatial heterogeneity and dynamic evolution patterns of stress concentration phenomena within close range coal seams (interlayer spacing < 10 m) remains particularly challenging.

(3) The influence of upper coal seam floor failure on lower roof failure mechanisms, alongside the phased characteristics and governing factors dictating the development of overlying rock fracture networks under repeated mining disturbances, requires elucidation.

Addressing these knowledge gaps will facilitate the optimization of roadway stability maintenance strategies and roof water hazard prevention measures, thereby advancing safe and efficient mining technologies for close range coal seam clusters. This investigation focuses on the engineering scenario of close range coal seams (mean interburden 10.2 m) in Shanxi Coking Coal Huajin’s No.9 and No.10 workings. The coal seams displays nearly horizontal bedding. The immediate roof primarily comprises siltstone, whereas the main roof consists of alternating sandy mudstone and sandstone strata. The floor is characterized by interbedded fine sandstone, sandstone, and mudstone exhibiting significantly disparate mechanical properties. The mining methodology employs longwall top-coal caving along the strike direction using a descending sequence, inducing pronounced mining pressure in adjacent panels. Critical engineering challenges encompass fault-induced water inrush, repeated mining-induced stress, interconnected water-conducting fracture zones, and rockburst hazards. Resolution of these challenges is imperative to ensure safe extraction operations in coal seam No.9 & No.10. Employing slip-line field theory, we quantitatively determine floor damage depth and establish a model through FLAC3D numerical simulation. Hydraulic fracturing tests enable precise in-situ detection of two-zone vertical dimensions. The derived results establish a novel mechanical framework for safe extraction of close range coal seam clusters, particularly providing operational guidance for roadway stability maintenance and roof water hazard mitigation.

Floor damage range calculation

The breaking law of roof strata directly affects the mining pressure. Consequently, the extent of floor damage induced during upper coal seam extraction exerts a direct influence on pressure distribution characteristics during subsequent lower coal seam mining operations. The engineering geology significantly influences fracture development, as demonstrated in the following case study.

Project overview

The study case is located in Shanxi Coking Coal’s Huajin mining area. The mining area is located at the southern margin of the Hedong Coalfield in Shanxi Province, underlain by typical North China Carboniferous coal-bearing strata. The principal minable seams are Nos. 9 and 10, exhibiting near-horizontal bedding. The immediate roof predominantly comprises siltstone, while the main roof consists of interbedded sandy mudstone and sandstone. The floor strata feature an alternating sequence of fine sandstone, sandstone, and mudstone, resulting in significant mechanical property heterogeneity. Longwall mining is conducted along the strike direction utilizing a down-going mining sequence, which induces intense strata pressure within adjacent mining panels. Subject to the Ouyang gray water system, the static water level measures − 580 m, and the aquifer permeability coefficient is 0.48 m/d. The actual measured inflow rate at the working face is 20.3 m³/h, and there is a risk of a water-conducting fracture zone penetrating the area. The working face is developed along the XDF90 and XDF91 normal faults, which have a drop range of 0 ~ 9 m and an inclination of 70 ~ 75°, intersecting the working face’s advance direction at an angle of 22 ~ 30°. Three-dimensional seismic exploration indicates that the fault zone width ranges from 8 ~ 15 m, and the permeability of the fractured zone is 13 ~ 22 times that of the surrounding rock.

Brief of 9101 mining face: The mining layout of No.9 coal seam is shown in Fig. 1. The 9101 face is vertically aligned with the 10,101 face, exhibiting an interburden spacing of 10.20 m. The thickness varies between 0.65 m and 3.89 m, with a mean measured thickness of 3.57 m. The geological structure demonstrates a relatively flat-lying configuration, with seam dip angles measuring between 2° and 5°, exhibiting an average inclination of 3°. The immediate roof stratum of the working face predominantly comprises siltstone with a mean thickness of 4.83 m. Overlying this stratum, the main roof consists of interbedded sandstone and mudstone layers averaging 7.19 m in thickness, exhibiting low tensile strength and susceptibility to stratification and spalling phenomena. Notably, the floor formation of the 9101 working face structurally corresponds to the roof sequence of the adjacent 10,101 working face, composed of alternating layers of fine-grained sandstone, sandstone, and mudstone lithologies.

Brief of 10,101 mining face: The mining face extends 931 m in length with a vertical dimension of 135 m, exhibiting an average overburden depth of 852 m. The No. 10 coal seam constitutes the principal mining stratum within the 10,101 working face. This stratigraphic unit exhibits a thickness variation ranging from 3.75 m to 7.40 m, with a mean value of 5.10 m. The immediate roof stratum of the longwall face predominantly comprises fine-grained sandstone, exhibiting a mean thickness of 1.85 m, whereas the main roof consists of silty mudstone with an average stratigraphic thickness measuring 6.53 m.

The prevailing mining methodology predominantly employs down-dip sequential extraction techniques. However, recurrent mining-induced disturbances result in 60 ~ 80% escalation in surrounding rock failure rates compared with single-seam operations. Subjected to mining-induced disturbances at the working face, the basal strata undergo shear and compressive damage mechanisms. Empirical data demonstrate that the spatial superposition ratio of peak abutment pressure in adjacent longwall panels attains 82%, correlating with a 3 ~ 5 fold escalation in rockburst initiation probability. The vertical propagation extent of fractured strata exhibits a statistically significant inverse correlation with interburden thickness, with hydraulic conductivity coefficients of permeable fractures attaining values as high as 92% when interburden spacing is reduced below 15 m. Precise determination of the damage depth in the overlying coal seam proves crucial for establishing the roof structure configuration of the underlying coal seam.

Fig. 1.

Mining layout of No.9 coal seam.

Failure depth of the floor

The slip line field theory is applied to the analysis of the bottom plate failure of coal seams with a layer spacing around to 10 m. A mechanical model is established to characterize damage propagation in coal seam floor strata. As illustrated in Fig. 2, the slip line field model, grounded in the theory of limit equilibrium, delineates the base failure zone into three distinct regions:

• I.Active Zone: Positioned adjacent to the coal wall, this region experiences advancing abutment pressure. The principal stress orientation forms an angle of 45°-φ/2 with the coal wall, exhibiting radially distributed plastic slip lines.
• II.Transition Zone: This zone interconnects the active and passive regions, wherein the slip line radius of curvature varies as a function of the internal friction angle φ.
• III.Passive Zone: Situated proximate to the mined-out area, this region is subjected to unloading pressures. The maximum principal stress direction rotates to 45°+φ/2, generating logarithmic spiral slip trajectories.

Analytical results from adjacent goaf areas demonstrate that the calculated damage depth exhibits significant correlation with principal stress variations, indicating that stress field redistribution exerts a controlling influence on failure propagation mechanisms.

Fig. 2.

Sliding line field.

The equation for the floor damage is:

1

2

where j is the internal friction angle of the floor mass, °; q is the angle between r and r₀, °; X_a is the yield length of the coal pillar, m. The expression for the depth of floor damage is:

3

4

Take Formula (1) and (4) to substitute Formula (3):

5

By the dh/dq = 0, i.e.

6

reach

7

By substituting formulas (2) and (7) into formula (5), the maximum damage depth of the floor is:

8

By incorporating the plastic zone dynamic width correction coefficient (Eq. 9) and interlayer stress coupling parameter, the applicability of conventional models for predicting failure mechanisms in thin interlayer rock masses can be maintained. The width of the coal wall plastic zone X_a is:

9

where M is the mining height, m; K is stress concentration coefficient; g is the average volume force of the overlying rock, kN/m³; H is the buried depth, m; C is the bonding force of coal, MPa; and j₀ is the internal friction angle of coal, °. The selection of 52° for the j angle in slip-line field theory was validated through Mohr-Coulomb strength envelope fitting derived from triaxial test data. Field engineering applications have subsequently confirmed this methodological superiority. Boundary condition solutions were computed using parameter sets from Table 1. The stress concentration coefficient K = 3.02 was determined through real-time inversion analysis of microseismic monitoring data, corresponding to the measured peak value during working face advancement at 150 m.

Table 1.

Property of floor mass.

Property	M/m	K	g/(kN/m3)	H/m	C/MPa	j0/°	j/°
Value	3.57	3	25	860	2.5	36	52

Based on the geological characteristics of the target coal mine and sliding line field theory from plasticity mechanics, the maximum failure depth of the floor is calculated as 6.26 m.

Rock failure mode of close range coal seam mining

The structural configuration of coal seam roof strata exerts a direct influence on the pressure distribution characteristics within mining faces. The extraction process in upper coal seams disrupts the primordial stress equilibrium, resulting in progressive deformation and potential failure mechanisms within roof rock strata prior to subsequent coal seam excavation. Consequently, the kinematic patterns of roof strata during contiguous coal seam extraction diverge significantly from those observed in single-seam mining operations. Ensuring operational safety necessitates systematic investigation into overburden displacement characteristics during multi-seam extraction, coupled with predictive analysis of roof strata migration patterns and failure modes. This scientific approach forms the fundamental basis for establishing effective safety protocols in contiguous coal seam mining environments.

Establishment of the numerical model

FLAC3D was chosen for its suitability in analyzing overburden stress fields via continuum mechanics. FLAC3D adopts an explicit Lagrangian finite difference formulation within a continuum mechanics framework. Quasi-static solutions are achieved through the application of virtual mass damping, with the time step size automatically adjusted to satisfy numerical stability conditions. Computational stability is ensured by monitoring the convergence of displacements at designated monitoring points (default relative error tolerance < 1e-5). Model construction rigorously adheres to engineering geological conditions:

1) Geometric Boundaries: Dimensions of 300 m (length) × 80 m (width) × 100 m (height), as illustrated in Fig. 3, corresponding to a 1:1 scale representation of the actual mining area.

2) Displacement Constraints: Normal displacement constraints applied to lateral boundaries; complete displacement constraints applied to the bottom boundary.

3) Stress Boundaries: A vertical stress of 19.3 MPa applied at the upper boundary (equivalent to an overburden depth of 852 m, assuming a unit weight of 23 kN/m³).

4) Initial Conditions: A pre-equilibrated in-situ stress field is established using an elastic solver.

5) Material Parameters: Based on rock mass parameters categorized into six types according to GB/T 50266 − 2013 (refer to Table 2), the elastic modulus for coal is 0.42 GPa and for sandstone is 3.02 GPa.

Fig. 3.

Numerical model.

Table 2.

Geomechanical parameters of coal and rock mass obtained through laboratory testing.

Lithology	Bulk/GPa	Shear/GPa	Tension/MPa	Friction/°	Cohesion/MPa
Mud stone	0.78	0.54	0.99	46	0.54
Sandy mudstone	1.33	1.07	2.52	48	3.33
Siltite	1.23	0.65	1.98	40	1.18
Fine sandstone	1.92	0.89	2.69	52	1.35
Limestone	3.02	2.05	3.84	45	8.20
Coal	0.42	0.22	0.12	36	0.82

The No. 9 and No. 10 coal seams are extracted through sequential mining operations. A monitoring network comprising seven displacement observation lines was established above the No. 10 coal seam, with the No. 1 measurement line positioned 5 m above, while subsequent lines were arranged at 10 m vertical intervals.

Shift in overlying rock strata during no.9 coal seam mining

During extraction of coal seam No.9, deformation patterns in overlying strata are visualized through Fig. 4.

Fig. 4.

Strata displacement during extraction of No.9 coal seam.

At 120 m face advance, line 6 and line7 were not affected. At 150 m face advance, measuring line 2 records peak displacement of 1.82 m, with line 3 reaching 1.72 m. These values approximate coal seam thickness, confirming the 15 m strata above seam No.9 lie within the caving zone.

Lines 4–5 exhibit reduced displacements (1.37 m and 0.99 m respectively), yet demonstrate synchronous subsidence patterns with face progression. This dynamic response suggests the 35 m strata above the coal seam reside in the water-conducting fracture zone.

Line 6 shows minimal movement before 120 m advance, but significant displacement thereafter, indicating delayed failure of harder strata. Line 7 maintains marginal displacement (0.07 m maximum), though its gradual subsidence curve places both lines 6–7 within the curved subsidence zone.

Plastic zone development analysis reveals roof fracture evolution in Fig. 5.

Research demonstrates stepwise expansion of roof fracture height during extraction, attributed to layered strata mechanics. Mining-induced stress exceeds key layer tensile strength, triggering abrupt fracture propagation. Fracture height stabilizes at 38 m despite continued mining.

Fig. 5.

Fracture zone height changes in No.9 seam extraction.

Displacement monitoring data extracted from FLAC3D are processed using MATLAB(R2024a, https://www.mathworks.com). The two-band height evolution equation is fitted employing the Levenberg-Marquardt algorithm. Model reliability is validated through a residual normal distribution test. The result enables the derivation of a two-zone development height model:

10

where R² is degree of fitting.

Overlying rock layer movement during 10# coal seam mining

Following the completion of No.9 coal seam extraction, subsequent mining operations commenced on the No.10 coal seam. Figure 6 illustrates the characteristic patterns of strata movement and deformation observed in the overlying rock mass during No.10 coal seam exploitation.

Fig. 6.

Displacement characteristics of upper rock strata during advance of 10,101 working face.

Monitoring Line 1, positioned within the inter-seam rock stratum between No.9 and No.10 coal seams, demonstrates critical deformation patterns. Theoretical analysis presented in Chap. 2 confirms complete integration of No.10 seam’s immediate roof within the failure zone induced by No.9 seam floor deformation. The absence of significant structural discontinuities between these proximate coal seams facilitates full destruction of No.10 seam’s roof strata. The subsidence profile of Line 1 reveals maximum displacement reaching 6.21 m, approximating the cumulative mining thickness of both seams, thereby validating the theoretical framework established in Chap. 2.

Line 2 monitoring data indicates substantial prior displacement (5.03 m maximum) attributable to No.9 seam extraction, with actual rock mass disintegration evident. The measured subsidence magnitude remains below combined seam thickness due to fragment dilation effects, confirming that monitoring Lines 1–2 reside within the roof collapse zone.

Comparative analysis of Lines 3–7 demonstrates progressively diminishing displacement magnitudes: 4.51 m (Line 3), 3.68 m (Line 4), 2.43 m (Line 5), 1.66 m (Line 6), and 0.77 m (Line 7). These values, though proportionally small relative to total extraction thickness, exhibit progressive increase with face advancement. This displacement gradient suggests lateral extension of the fracture zone encompassing Lines 4–7.

Plastic zone evolution during No.10 seam extraction, as depicted in Fig. 7, demonstrates dynamic height variation corresponding to mining progression.

Fig. 7.

Fracture zone height changes in No.9 seam extraction.

The result reveals rapid fracture zone propagation during initial roof failure phases, stabilizing at approximately 66 m height regardless of continued face advance. This equilibrium state suggests establishment of competent stress arches within the overburden. Adjacent seam extraction induces cumulative rock displacement measuring 2.3 times that of single-layer mining, with shear displacement components increasing to 65% of total deformation. At this critical juncture, the overlying rock strata above the working face coalesce to form an asymmetric stress arch structure. This geological configuration exerts a substantial influence on advance support pressure, which exhibits a positive correlation with the fracture zone expansion rate. Through theoretical derivation grounded in key stratum theory, it was established that when L/H = 2.05, the fracture zone development height (H) and working face length (L) conform to the relationship H = 0.18 L + 28.3 (R² = 0.934), demonstrating a deviation margin of less than 5% relative to empirical measurements. Numerical simulation results were processed through MATLAB to derive empirical relationships for two zones development height during extraction, yielding the following formulation:

11

Measurement of the roof two zones

This study implemented hydraulic injection testing methodology to systematically investigate failure mechanisms in roof strata of close range coal seams. The experimental protocol was specifically designed to elucidate fracture network evolution and characterize crack propagation patterns.

Equipment and scheme

The drilling parameters were formulated in strict accordance with the hydrogeological exploration specifications stipulated in Detailed Rules for the Prevention and Control of Water in Coal Mines (GB/T 50218 − 2014, China) for close range coal seam group mining. Through systematic analysis of the stratigraphic dip angle and fracture development characteristics within the mining area, the borehole spacing and inclination angles were systematically optimized through orthogonal experimental methodology. The leakage of water injection is used to detect the development height of the water-conducting fracture zone. The device is a double-end water blocker, see Fig. 8. The double-end water plugging device is mainly composed of the inner plugging device and the outer measuring device. The inner occluder is a water pipe with sealed airbags at both ends. The off-hole meter is mainly composed of pressure gauge, flow meter and water injection valve. The principle of water water-conducting fracture zone detection is shown in Fig. 9.

Fig. 8.

Principle of fracture zone detection.

The double-ended water blocker system is engineered based on hydrodynamic principles and rock permeability assessment methodologies, specifically designed for determining the developmental elevation of hydraulic fracture zones in geological formations. The apparatus comprises two principal subsystems:

(1) Internal sealing assembly: Constructed with high-density polyethylene (HDPE) pipe (Φ 50 mm nominal diameter), incorporating pneumatic rubber bladders (pressure rating 1.5 MPa) at terminal positions. These bladders are pressurized through a dedicated pneumatic circuit, enabling formation of isolated test intervals within borehole environments.

(2) External monitoring instrumentation: Consists of a digital pressure transducer (0 ~ 5 MPa measurement range), electromagnetic flow sensor (± 0.5% FS accuracy class), and multi-port injection manifold with precision flow control.

The operational protocol follows these sequential procedures:

(1) Utilize XY-4 full-hydraulic drilling equipment to establish boreholes in compliance with GB/T 50,218 − 2014 geotechnical investigation specifications.

(2) Position the isolation apparatus at the target depth and pressurize terminal bladders to 0.8–1.2 MPa, creating a 2-meter sealed test interval.

(3) Conduct hydraulic injection at controlled rates (0.5 ~ 2 L/min) while continuously monitoring pressure-flow characteristic curves through integrated data acquisition systems.

(4) Identify fracture connectivity through abrupt pressure decay (≥ 15% baseline) accompanied by anomalous flow rate increases (> 20% setpoint). Each test horizon undergoes three replicate injection tests, with arithmetic mean values calculated upon achieving standard deviation < 5%. Fracture activation is confirmed when unit length leakage exceeds 0.6 L/(min·m) across three consecutive test cycles under constant pressure conditions.

The drilling field is established exterior to the 10,101 working face, positioned 20 m from the stop line. The configuration comprises three water injection monitoring boreholes (designated #1~#3) and one control borehole (#4), totaling four drilling operations. These monitoring boreholes are strategically located at the roof-wall intersection of the roadway. The borehole parameters are designed with the following specifications: initial casing diameter of F153 mm is maintained from 0 ~ 10 m depth, transitioning to F110 mm from 10 m depth to the borehole terminus. Specific parameters of probe drill hole is shown in Table 3. The design of drilling inclination 55 ~ 61° conforms to the relationship between the maximum development direction of fracture zone and the main stress direction of 75 ± 5°.

Table 3.

Parameters of drill hole.

No.	Dip/°	Intersection/°	Depth/m
#1	55	22	70
#2	44	38	80
#3	35	65	88
#4	61	30	73

See Fig. 9 for the layout profile of water injection detection hole.

Fig. 9.

Layout of detection holes.

Detect results of the two zones

The abnormal values are processed by wavelet noise reduction. The arithmetic mean value of three repeated tests is obtained.During the excavation of the No.9 coal seam, the 9101 working face implemented hydraulic injection methodology for monitoring the developmental characteristics of the two zones. To validate the numerical simulation findings presented in Chap. 3, this investigation incorporates experimental data from the feasibility study on water injection conducted at Duan Wang Coal Mine, with particular focus on the empirical analysis of two zones formation in the 9101 working face.

Field measurement results regarding the spatial dimensions of two zones are graphically represented in Fig. 10. The caving zone extend to 10.3 m, 9.2 m, and 9.4 m at three sampling points, yielding a mean height of 9.6 m. Corresponding water-conducting fracture zones exhibit height of 33.4 m, 34.6 m, and 32.1 m, with an average measurement of 33.4 m.

The congruence between these field measurements and computational simulation results substantiates the methodological validity.

Fig. 10.

Flow of detection borehole.

Through systematic analysis of field measurement data conducted in this study, the theoretical calculations demonstrate strong congruence with both numerical simulations and empirical field data. Comparative analysis with previous studies reveals critical insights into fracture zone development. In contrast to Cao²³, the collapse zone height measured in this study (9.6 m) aligns with their reported 10–12 m range. However, the fracture zone height (33.4 m) demonstrates an 18% reduction compared to their predictions, which may be attributed to variations in inter-seam geological spacing. Cross-referencing with the empirical model for predicting repeated mining-induced fracture zones (Model II) by Kang²², the observed fracture zone height of 66 m in the No.10 coal seam exceeds their formula-derived prediction by 9.3%. This discrepancy quantitatively validates the cumulative strata disturbance effects from successive extraction operations, as postulated in contemporary mining geomechanics theory.

Based on fracture zone height predictions, three principal engineering modifications have been implemented: (1) Enhanced roadway support systems employing 3.2 m anchor rods (increased from 2.4 m) with 40% denser reinforcement patterns; (2) Installation of a 3 m-thick grouted curtain barrier at fracture zone boundaries for hydrological control; (3) Activation of backfill mining technology when either predicted fracture zone height or interlayer spacing exceeds 2.5 times critical thresholds, effectively mitigating surface subsidence phenomena.

Conclusions

In close range coal seam mining, upper seam extraction disrupts lower seams’ roof strata, causing instability and safety risks. Previous studies lacked integrated experimental-field approaches. This study examines No.9 and No.10 seams to analyze roof failure, strata migration, and collapse/fracture zone heights:

(1) Slip-line field theory established a floor damage model, calculating No.9 seam’s maximum floor failure depth at 6.26 m. Simulations show 15 m collapse above No.9 seam with 38 m fractures. During No.9 + 10 coal seams mining, monitoring lines 1–2 lay within caving zone, lines 4–7 intersected fracture zone, with No.10’s fractures reaching 66 m. Hydraulic fracturing tests revealed average caving heights of 9.6 m, with fracture zones at 33.4 m. Field detection results align with simulation data.

(2) The synergistic evolution mechanism governing interlayer rock mass failure was elucidated, with analytical results demonstrating that the immediate roof stratum of the No.10 coal seam is entirely situated within the caving influence zone of the No.9 coal seam floor. This finding establishes a theoretical foundation for optimizing offset mining design in close range coal seam groups with interlayer distances < 10 m.

(3) The tripartite validation method combining FLAC3D, MATLAB, and water injection tests effectively addressed systematic deviations in traditional approaches. This approach particularly resolved the prevalent overestimation issue in two zones height prediction.

(4) The limitations of the study and recommendations for future research: The current model has not adequately account for the spatial coupling effect between active faults and principal stress orientation, potentially leading to underestimation of the amplification effect of fault activation on fracture zone height. Through the integration of fault activation theory with Discrete Element Method (DEM) simulation techniques, we propose to develop a comprehensive predictive model incorporating geological structures, mining disturbances, and hydrological responses.

Author contributions

Jianhua Li participated in the initial manuscript drafting; Wenyu Zhou was responsible for data analysis; Pengjiang Deng led the research design, theoretical model construction, and finalized manuscript revision; Jun Yan conducted FLAC3D numerical simulations, MATLAB data processing, and graph plotting; and Senbiao Chang implemented borehole parameter design, monitoring data collection, and two zones height calculation.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.