Numerical modelling techniques for studying longwall geotechnical problems under realistic geological structures

Nội dung text: Numerical modelling techniques for studying longwall geotechnical problems under realistic geological structures

1. Journal of Mining and Earth Sciences Vol. 62, Issue 3 (2021) 87 - 96 87 Numerical modelling techniques for studying longwall geotechnical problems under realistic geological structures Dung Tien Le *, Tung Manh Bui Faculty of Mining, Hanoi University of Mining and Geology ARTICLE INFO ABSTRACT Article history: th Longwall - associated geotechnical problems have been extensively Received 12 Feb 2021 studied by using numerical modelling methods. However, proper Accepted 23rd April 2021 representation of its geological structures remains a challenging task. Available online 30th June 2021 This paper presents a systematic understanding of numerical modelling Keywords: techniques for studying longwall coal mining with geological structures. Discrete Fracture The modelling techniques derived from conventional and advanced continuum and discontinuum methods were reviewed in detail with Grain - Based Model (GBM), emphasiz on their mechanic's formulation and applications. This study Longwall, suggests that the successful selection of a proper modelling technique Network(DFN), should be based on the physical principle of longwall problem, texture and Trigon logic. shape of materials, and mechanics formulation of the numerical program used for modelling. The paper’s conclusions assist numerical modellers in quickly and properly selecting modelling technique for investigating a site - specific longwall problem. Copyright © 2021 Hanoi University of Mining and Geology. All rights reserved. stability, face/rib spalling, pillar yielding, roof 1. Introduction fall/caving, and floor heave are the objects of Longwall is one of the most productive many investigations. Depending on the research underground coal mining methods, and has been methods used, those studies shed different lights widely applied in many coal industries. The on the mechanisms driving the problems. For method’s application has been extended to example, according to Galvin (2016), empirical various geo - mining conditions due to the methods (which are based on experiment, field significant improvement in face equipment (i.e., data and observations) may produce an actual shield support) and deeper insight into rock mass response of rock mass around the longwall face. A behaviour around the mined - out area. Longwall clear understanding of the underlying physical geotechnical problems such as entry/roadway phenomenon and database for the development is a key for the method's success, as seen in Yu et al. ___ (2015; 2017). Due to the high cost in setting *Corresponding author experiment and the long time for implementation E - mail: letiendung@humg.edu.vn at longwall panel, empirical methods should be DOI: 10.46326/JMES.2021.62(3).10
2. 88 Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 carefully designed before their application. site - specific longwall problem. Analytical methods, on the other hand, may reveal fundamental physical principles of longwall 2. Continuum modelling techniques behaviours, as can be seen in Wang and Wang Finite Difference Methods (FDM) are widely (2019) and Le (2021). However, the methods used continuum techniques for studying longwall require significant time and effort for solving coal mining. By using the methods, a longwall equations when the problem involves many domain is discretised into many sub - domains detailed structures that typically occur in coal whose connectivity between them remains measure rocks. On laboratory scale, physical unchanged during the modelling (Kelly et al., modelling methods can obtain similar results to 2002). The methods - with representative field experiment at a more reasonable cost, as programs such as FLAC and FLAC3D - have seen in Sui et al. (2015) and Zhang et al. (2019). interface elements or slide lines that allow the The methods still require significant time and modelling of discontinuum domain such as effort for constructing models compared to the longwall coal seam/rock strata to some extent project time frame. (Itasca Consulting Group, 2019). For example, Bai Compared to the above methods, numerical et al. (2016a) used FLAC2D to study coal wall modelling methods, as based on mathematical micro - responses caused by longwall retreat. equations founded on analytically derived Although different geometrical configurations of formulae/algorithms (Galvin, 2016), are geological structures exist in the stratigraphic particularly suitable for studying longwall column of the area, it seems that the structures are problems. In particular, the methods can implicitly represented in the model (Figure 1). investigate detailed micro - mechanisms of large - This representation techniques can also be seen in scale longwall problems within a reasonable time. Li et al. (2018) and Zhang et al. (2019), in which Compared to physical modelling and analytical FLAC3D was used to investigate rockburst and methods, numerical methods are more efficient in roof fracture caused by longwall top coal caving incorporating many geological structures. respectively. A longwall panel containing sub - Nevertheless, compared to empirical methods, horizontal parting planes was explicitly modelled proper modelling of geological structures is more by FLAC3D using interface elements in Behera et challenging due to the complex geometrical al. (2020). The modelling technique did not distribution of structures in the field. While the consider other typical structures such as joints derivation of rock material properties at within coal seam (Figure 2). To realistically laboratory scale is reliable through different represent both parting/bedding planes and joints testing techniques, the derivation at field scale for within thick coal seam, Wang et al. (2020) used large - scale modelling such as longwall is limited FLAC3D coupled with a Discrete Fracture due to available apparatus and inhomogeneity of Network (DFN) for the modelling (Figure 3). This geological structures. Although some numerical modelling technique accordingly captured well modelling techniques have been developed for failure mechanisms caused by longwall mining. the realistic representation of longwall geological However, due to the continuum nature of the structures, they can be very different in the program, the technique could not consider mechanic’s formulation and may be misused if realistic caving development in LTCC, which is a insufficiently understood. key phenomenon to the success of this mining This paper presents a systematic method. Further details on DFN are described in understanding of numerical modelling techniques Section 3.2. for studying longwall geotechnical problems under realistic geological structures. Different 3. Discontinuum modelling techniques techniques derived from continuum and discontinuum codes are carefully reviewed. The Distinct Element Methods (DEM) are widely paper’s conclusions and suggestion assist used discontinuum techniques for studying numerical modellers in quickly and properly longwall mining. By using the methods, a longwall selecting modelling technique for investigating a domain is divided into an assemblage of discrete
3. Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 89 Figure 1. FLAC2D model of longwall retreat (Bai et al., 2016a). Figure 2. FLAC3D model of longwall with parting planes (Behera et al., 2020).
4. 90 Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 Figure 3. Longwall model with DFN - FDM (Wang et al., 2020). a) pre - existing fracture orientations; (b) pre - existing fracture lengths; (c) DFN realization; (d) FDM model; (e) FDM - DFN model; (f) fracture influenced zones; and (g) geometrical size of model. blocks formed by connected fractures (Le et al., of structures are defined by statistical geometrical 2017). The methods - with representative parameters such as dip angle, trace length, gap programs such as UDEC, 3DEC, PFC2D, PFC3D - length, spacing, and spatial location. This are most suitable for moderately fractured rock modelling technique has been widely used to mass as in longwall coal mining. The use of DEM study coal caving/spalling under typical in representing geological structures in longwall structures such as horizontal bedding planes and model can be classified into conventional, vertical joints (Wang et al., 2016; Kong et al., 2019; Discrete Fracture Network (DFN), and Grain - Le and Bui, 2020). Note that in these studies, the Based Model (GBM) technique as follows. structures are persistent throughout the strata. At a smaller longwall - scale problem (i.e., longwall 3.1. Conventional technique entry), the conventional joint generator was also A conventional joint generator built - in DEM used to model non - persistent vertical joint sets, based programs (e.g., UDEC) is used to create sets which can be considered as DFNs, as seen in of geological structures in a longwall model. A set Abousleiman et al. (2020) (Figure 4).
5. Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 91 because the field measurement of fracture characteristics can only be implemented at limited space (e.g., outcrop, underground exposure, drill core), the reliability of the modelled DFN is highly dependent on the number, availability, and quality of field measurement. Application of DFN in mining and civil geomechanics along with challenges in building conceptual model, collecting field data, and simplifying, calibrating, and validating DFN are described in detail in Lorig et al. (2015). For stratified rock mass as in underground Figure 4. Longwall entry model with DFN coal mine, DFN has been used mostly in the (Abousleiman et al., 2020). “Synthetic Rock Mass - SRM” technique (Pierce et al., 2007) for studying the effect of scale and anisotropy on coal mechanical properties 3.2. DFN technique (Deisman et al., 2010; Scholtès et al., 2011; Gao et The basic concept, generation, advantage, al., 2014b; Wang et al., 2019a; Wang et al., 2019b; and disadvantage of DFN technique can be found Karimi Sharif et al., 2019). In such studies, the in many papers and textbooks. According to Jing dimension of the SRM is commonly about several (2003), Elmo and Stead (2010), Mas Ivars et al. meters, which is much less than a typical longwall (2011), and Hadjigeorgiou (2012), DFN was dimension. The use of DFN in the 3D longwall developed since the 1980s to study fluid flow and problem remains, however, very limited. This is transport process within jointed rock masses because when the problem scale increases, the through a network of connected fractures. A DFN detail level of the DFN model significantly is generated based on the geometry of the fracture increases that requires excessive solution time. system and the persistence of individual fractures. The detail level may decrease in large - scale 2D The fracture system geometry is established longwall problems, as for longwall entry stability based on the stochastic distribution of (Abousleiman et al., 2020; Zhu et al., 2020) characteristics such as density, orientation, size, (Figure 5). It should be noted that in the aperture, and persistence. One major advantage mentioned entry study, the DFN was created by of this technique is the realistic simulation of using a built - in conventional joint generator, fracture distribution at the field. However, while in the roof study, a third - party DFN Figure 5. Longwall gob - size entry model with DFN - DEM (Zhu et al., 2020).
6. 92 Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 generator named GEOFRAC (Ivanova et al., 2014) Cundall, 1987; Trueman et al., 1996) or contact was used. Other DFN generators can also be listed, stiffness (Nicksiar, 2013). Since 2010, a few such as FracMan (Golder Associates, 2020) used standard calibration processes were established in Wang et al. (2019b), or MoFrac for the derivation of a unique set of contact micro ( 2020) used in - properties for matching on - site rock mass Farahmand et al. (2018) and Wang and Cai response (Kazerani and Zhao, 2010; Kazerani, (2019). 2013; Gao and Stead, 2014). Due to the standard process in calibration and advantage in modelling 3.3. GBM technique failure development, GBM has been widely used GBM reflects the use of polygonal structures in UDEC for studying longwall stability such as (e.g, grain mesh, Voronoi, and Trigon roadway (Coggan et al., 2012; Gao et al., 2014c; tessellations) to model grain - shaped material Bai et al., 2016b), roof strata (Gao et al., 2014a; Li mostly in DEM based numerical programs et al., 2016), surface subsidence (Zhang et al., (Damjanac et al., 2007; Potyondy, 2010; Lan et al., 2017), coal face (Yao et al., 2017), and coal pillar 2010; Gao and Stead, 2014). GBM was first used in (Wu et al., 2019) (Figures 6 - 7). The use of GBM, UDEC by Lorig and Cundall (1987) to overcome however, must take into consideration the the intrinsic limitation of UDEC that this program disadvantages and uncertainties of the technique. cannot explicitly model intact block failure. GBM For example, according to Gao and Stead (2014), generation is done by subdividing an intact block micro - properties of contacts created by Voronoi into smaller polygonal blocks through Voronoi logic seem to be overestimated after calibration logic or triangular blocks through Trigon logic in and the solution time increases significantly. arbitrary sizes (Itasca Consulting Group, 2019). Mayer and Stead (2017) and Zhang and Wong The contacts (or boundaries) between polygons (2018) emphasise that Voronoi logic reduces are not representative of natural fractures but kinematic freedom of polygonal blocks that play potential paths for failure development in the facilitates tensile failure. Meanwhile, Trigon logic rock mass. The micromechanical properties of increases the kinematic freedom of triangular contacts and polygons must be calibrated to blocks that facilitates shear failure. The two GBM match the modelled rock mass’s behavior against logics may misrepresent realistic failure reality. The calibration process can be simple by, mechanisms if they do not sufficiently represent for example, adjusting contact strength (Lorig and the texture and shape of minerals constituting rock mass. Figure 6. Longwall roof stability model with Trigon logic (Gao et al., 2014a).
7. Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 93 Figure 7. Longwall surface subsidence model with Voronoi logic (Zhang et al., 2017). Note that due to the 3D nature of geological 4. Conclusions structure distribution, the 2D modelling must be This paper presents a systematic implemented with care for retaining the understanding of numerical modelling techniques representativeness of the problem’s structure. At for studying longwall geotechnical problems the same time, GBM logics are suited for the under realistic geological structures. The representation of explicit failure and caving modelling techniques derived from conventional development along well - calibrated fictitious and advanced continuum and discontinuum fractures. This study suggests that the successful methods were reviewed in detail with emphasis selection of a proper modelling technique should on their mechanics formulation and applications. be based on the physical principles of longwall For FDM - based techniques, the current study problems, textures and shapes of materials confirms that they well represent the micro - constituting problems, and mechanics mechanics of rock mass failure rather than formulation of the numerical program used for explicit caving caused by unjointed or moderately modelling. bedded coal seam/roof strata in longwall. The Author contributions continuum formulation of the programs, on the one hand, limits the complete and large Dung Tien Le reviewed and wrote the detachment of elements in the domain. On the introduction and discussion; Tung Manh Bui other hand, it may break down the calculation collected documents and wrote the conclusions. process when many discontinuities (e.g., geological structures) are incorporated. For DEM Acknowledgements - based techniques, the study finds a wide use This research is funded by Hanoi University from longwall entry stability to coal wall spalling, of Mining and Geology, Vietnam. roof caving, and roof fracture. Conventional joint generator built in the programs is found to be References limited to modelling deterministic fractures which are persistent. DFN logics are seen to be Abousleiman, R., Walton, G. & Sinha, S. (2020). suited for explicit representation of stochastic Understanding roof deformation mechanics fractures as in practice, but their use in longwall and parametric sensitivities of coal mine problems is currently limited and mainly in 2D. entries using the discrete element method.
8. 94 Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 International Journal of Mining Science and Galvin, J. M. (2016). Ground engineering - Technology, 30, 123 - 129. Principles and practices for underground coal mining, Cham, Springer International Bai, Q. - S., Tu, S. - H., Chen, M. & Zhang, C. (2016a). Publishing. Numerical modeling of coal wall spall in a longwall face. International Journal of Rock Gao, F., Stead, D. & Coggan, J. 2014a. Evaluation of Mechanics and Mining Sciences, 88, 242 - 253. coal longwall caving characteristics using an innovative UDEC Trigon approach. Computers Bai, Q. - S., Tu, S. - H., Zhang, C. & Zhu, D. (2016b). and Geotechnics, 55, 448 - 460. Discrete element modeling of progressive failure in a wide coal roadway from water - Gao, F., Stead, D. & Kang, H. (2014b). Numerical rich roofs. International Journal of Coal investigation of the scale effect and anisotropy Geology, 167, 215 - 229. in the strength and deformability of coal. International Journal of Coal Geology, 136, 25 - Behera, B., Yadav, A., Singh, G. S. P. & Sharma, S. K. 37. (2020). Numerical Modeling Study of the Geo - mechanical Response of Strata in Longwall Gao, F., Stead, D. & Kang, H. (2014c). Simulation of Operations with Particular Reference to Indian roof shear failure in coal mine roadways using Geo - mining Conditions. Rock Mechanics and an innovative UDEC Trigon approach. Rock Engineering, 53, 1827 - 1856. Computers and Geotechnics, 61, 33 - 41. Coggan, J., Gao, F., Stead, D. & Elmo, D. (2012). Gao, F. Q. & Stead, D. (2014.) The application of a Numerical modelling of the effects of weak modified Voronoi logic to brittle fracture immediate roof lithology on coal mine modelling at the laboratory and field scale. roadway stability. International Journal of Coal International Journal of Rock Mechanics and Geology, 90–91, 100 - 109. Mining Sciences, 68, 1 - 14. Damjanac, B., Board, M., Lin, M., Kicker, D. & Leem, Golder Associates (2020). FracMan. Seattle: J. (2007). Mechanical degradation of Washington. emplacement drifts at Yucca Mountain - A Hadjigeorgiou, J. (20120. Where do the data come modeling case study: Part II: Lithophysal rock. from? Transactions of the Institutions of Mining International Journal of Rock Mechanics and and Metallurgy, Section A: Mining Technology, Mining Sciences, 44, 368 - 399. 121, 236 - 247. Deisman, N., Mas Ivars, D., Darcel, C. & (2020). MoFrac Chalaturnyk, R. J. (2010). Empirical and [Online]. Available: numerical approaches for geomechanical [Accessed 21 June characterization of coal seam reservoirs. 2020]. International Journal of Coal Geology, 82, 204 - 212. Itasca Consulting Group (2019). UDEC – Universal Distinct Element Code, Ver. 7.0. Minneapolis: Elmo, D. & Stead, D. (2010). An Integrated Itasca. Numerical Modelling–Discrete Fracture Network Approach Applied to the Ivanova, V. M., Sousa, R., Murrihy, B. & Einstein, H. Characterisation of Rock Mass Strength of H. (2014). Mathematical algorithm Naturally Fractured Pillars. Rock Mechanics development and parametric studies with the and Rock Engineering, 43, 3 - 19. GEOFRAC three - dimensional stochastic model of natural rock fracture systems. Farahmand, K., Vazaios, I., Diederichs, M. S. & Computers & Geosciences, 67, 100 - 109. Vlachopoulos, N. (2018). Investigating the scale - dependency of the geometrical and Jing, L. (2003). A review of techniques, advances mechanical properties of a moderately jointed and outstanding issues in numerical modelling rock using a synthetic rock mass (SRM) for rock mechanics and rock engineering. approach. Computers and Geotechnics, 95, 162 International Journal of Rock Mechanics and - 179. Mining Sciences, 40, 283 - 353.
9. Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 95 Karimi Sharif, L., Elmo, D. & Stead, D. (2019). Location of Critical Rock Block Fracture on Improving DFN - geomechanical model Crack Evolution in a Gob - side Filling Wall. integration using a novel automated approach. Rock Mechanics and Rock Engineering, 49, Computers and Geotechnics, 105, 228 - 248. 1041 - 1058. Kazerani, T. (2013). A discontinuum - based Li, Z. - L., He, X. - Q., Dou, L. - M. & Song, D. - Z. model to simulate compressive and tensile (2018). Comparison of rockburst occurrence failure in sedimentary rock. Journal of Rock during extraction of thick coal seams using top Mechanics and Geotechnical Engineering, 5, - coal caving versus slicing mining methods. 378 - 388. Canadian Geotechnical Journal, 55, 1433 - 1450. Kazerani, T. & Zhao, J. (2010). Micromechanical parameters in bonded particle method for Lorig, L. J. & Cundall, P. A.( 1987) Modeling of modelling of brittle material failure. reinforced concrete using the distinct element International Journal for Numerical and method. In: SHA, S. P. & SWARTZ, S. E., eds. Analytical Methods in Geomechanics, 34, 1877 - International Conference on Fracture of 1895. Concrete and Rock, Houston, Texas. 276 - 287. Kelly, M., Luo, X. & Craig, S. (2002). Integrating Lorig, L. J., Darcel, C., Damjanac, B., Pierce, M. & tools for longwall geomechanics assessment. Billaux, D. (2015). Application of discrete International Journal of Rock Mechanics and fracture networks in mining and civil Mining Sciences, 39, 661 - 676. geomechanics. Mining Technology, 124, 239 - 254. Kong, D., Liu, Y. & Zheng, S. (2019). Sensitivity analysis of influencing factors and control Mas Ivars, D., Pierce, M. E., Darcel, C., Reyes - technology for coalface failure. Arabian Journal Montes, J., Potyondy, D. O., Paul Young, R. & of Geosciences, 12, 550. Cundall, P. A. (2011). The synthetic rock mass approach for jointed rock mass modelling. Lan, H., Martin, C. D. & Hu, B. (2010). Effect of International Journal of Rock Mechanics and heterogeneity of brittle rock on Mining Sciences, 48, 219 - 244. micromechanical extensile behavior during compression loading. Journal of Geophysical Mayer, J. M. & Stead, D. (2017). Exploration into Research, 115, B01202. the causes of uncertainty in UDEC Grain Boundary Models. Computers and Geotechnics, Le, T. D. (2021). Analytical Study on the Stability 82, 110 - 123. of Longwall Top Coal Caving Face. In: BUI, X. N., LEE, C. & DREBENSTEDT, C. (eds.) Proceedings Nicksiar, M. (2013). Effective parameters on crack of the International Conference on Innovations initiation stress in low porosity rocks. PhD for Sustainable and Responsible Mining (ISRM Thesis, University of Alberta. 2020). Cham: Springer International Pierce, M., Mas Ivars, D., Potyondy, D. & Cundall, P. Publishing. A. A Synthetic Rock Mass Model for Jointed Le, T. D. & Bui, X. N. (2020). Effect of Key Rock. In: EBERHARDT, STEAD & MORRISON, Parameters on Top Coal First Caving and Roof eds. Rock Mechanics: Meeting Society's First Weighting in Longwall Top Coal Caving: A Challenges and Demands, 2007. London: Case Study. International Journal of Taylor & Francis Group, 341 - 349. Geomechanics, 20, 04020037. Potyondy, D. 2010. A grain - based model for rock: Le, T. D., Mitra, R., Oh, J. & Hebblewhite, B.( 2017). Approaching the true microstructure. In: LI, C. A review of cavability evaluation in longwall C., GRØNENG, G., OLSSON, R. & ENGEN, S., eds. top coal caving. International Journal of Mining Rock Mechanics in the Nordic Countries, 2010. Science and Technology, 27, 907 - 915. Kongsberg, Norway: Norwegian Group for Rock Mechanics, 225–234. Li, X., Ju, M., Yao, Q., Zhou, J. & Chong, Z. (2016). Numerical Investigation of the Effect of the Scholtès, L., Donzé, F. - V. & Khanal, M. 2011. Scale
10. 96 Dung Tien Le, Tung Manh Bui/Journal of Mining and Earth Sciences 62(3), 87 - 96 effects on strength of geomaterials, case study: Wu, W. - D., Bai, J. - B., Wang, X. - Y., Yan, S. & Wu, S. Coal. Journal of the Mechanics and Physics of - X. 2019. Numerical Study of Failure Solids, 59, 1131 - 1146. Mechanisms and Control Techniques for a Gob - Side Yield Pillar in the Sijiazhuang Coal Mine, Sui, W., Hang, Y., Ma, L., Wu, Z., Zhou, Y., Long, G. & China. Rock Mechanics and Rock Engineering, Wei, L. 2015. Interactions of overburden 52, 1231 - 1245. failure zones due to multiple - seam mining using longwall caving. Bulletin of Engineering Yao, Q., Li, X., Sun, B., Ju, M., Chen, T., Zhou, J., Liang, Geology and the Environment, 74, 1019 - 1035. S. & Qu, Q. 2017. Numerical investigation of the effects of coal seam dip angle on coal wall Trueman, R., Coulthard, M. A. & Poulsen, B. A. stability. International Journal of Rock Numerical methods for estimating the stability Mechanics and Mining Sciences, 100, 298 - 309. of unsupported spans for highwall mining. In: AUBERTIN, M., HASSANI, F. & MITRI, H., eds. Yu, B., Zhao, J., Kuang, T. & Meng, X. (2015). In situ 2nd North American Rock Mechanics investigations into overburden failures of a Symposium, 1996/1/1/ 1996 Montreal. super - thick coal seam for longwall top coal Rotterdam: A. A. Balkema. caving. International Journal of Rock Mechanics and Mining Sciences, 78, 155 - 162. Wang, C., Zhang, C., Li, Y. & Zheng, C. 2019a. Numerical investigation of the mechanical Yu, B., Zhao, J. & Xiao, H. (2017). Case study on properties of coal masses with T - junctions overburden fracturing during longwall top cleat networks under uniaxial compression. coal caving using microseismic monitoring. International Journal of Coal Geology, 202, 128 Rock Mechanics and Rock Engineering, 50, 507 - 146. - 511. Wang, J. & Wang, Z. 2019. Systematic principles of Zhang, C., Mitra, R., Oh, J., Canbulat, I. & surrounding rock control in longwall mining Hebblewhite, B. (2017). Numerical analysis on within thick coal seams. International Journal mining - induced fracture development of Mining Science and Technology, 29, 65 - 71. around river valleys. International Journal of Mining, Reclamation and Environment, 32, 463 Wang, J., Wang, Z. & Li, Y. 2020. Longwall Top Coal - 485. Caving Mechanisms in the Fractured Thick Coal Seam. International Journal of Zhang, X., Gong, P., Wang, K., Li, J. & Jiang, Y. Geomechanics, 20, 06020017. (2019). Characteristic and Mechanism of Roof Fracture Ahead of the Face in an LTCC Panel Wang, J., Yang, S. & Kong, D. 2016. Failure When Passing an Abandoned Roadway: A Case mechanism and control technology of longwall Study from the Shenghua Coal Mine, China. coalface in large - cutting - height mining Rock Mechanics and Rock Engineering, 52, method. International Journal of Mining 2775 - 2788. Science and Technology, 26, 111 - 118. Zhang, Y. & Wong, L. N. Y. (2018). A review of Wang, X. & Cai, M. (2019). A DFN–DEM Multi - numerical techniques approaching scale Modeling Approach for Simulating microstructures of crystalline rocks. Tunnel Excavation Response in Jointed Rock Computers and Geosciences, 115, 167 - 187. Masses. Rock Mechanics and Rock Engineering, 53, 1053–1077. Zhu, G. L., Sousa, R. L., He, M. C., Zhou, P. & Yang, J. (2020). Stability Analysis of a Non - pillar - Wang, X., Kang, H. & Gao, F. 2019b. Numerical Mining Approach Using a Combination of investigation on the shear behavior of jointed Discrete Fracture Network and Discrete - coal mass. Computers and Geotechnics, 106, Element Method Modeling. Rock Mechanics 274 - 285. and Rock Engineering, 53, 269 - 289.