Journal of Mining Science, Vol. 46, No. 2, 2010
MINERAL MINING TECHNOLOGY COAL EXTRACTION FROM THICK FLAT AND STEEP BEDS
V. I. Klishin and S. V. Klishin
UDC 622.273:274+539.3
The paper offers an underground geotechnology for thick coal beds based on the controllable force-feed extraction of pre-broken coal with using a re-designed powered roof support. Based on the discrete element method, the mathematical model has been developed for the numerical simulation of gravitation movement of granular materials. The two-dimensional problem about sublevel coal caving in thick beds is studied, and the effect exerted by the order of opening the roof support outlets on the coal extraction parameters is illustrated. Powered roof support, underground coal extraction, gravitation movement, numerical model, discrete element method INTRODUCTION
Extraction of thick coal increasingly involves powered roof supports to cut coal in layers adjacent to roofs, and in interlayers. Such technologies involve coal destruction with the help of the strata pressure. This fact allowed the powered supports to be functionally enriched with extraction control for coal above and behind the support. In Russian, Chinese, Kazakh, French and Czech mines, such supports have been used in coal cutting and extraction onto a front conveyor located in front of the support, closer to the working face, or onto a rare conveyor adjacent to the rare part of the back-end of the support [1 – 5]. Conventional labor-intensive slicing methods will certainly be replaced with the technologies including such supports as a basic element advantageous for: greatly cut down development work, investment and maintenance cost, energy intensity, coal combustion risk, as well as higher applicability to complex geology conditions and extractability of coal left in pillars. As a result, efficiency and safety of coal extraction grow while workload and concentration of mining increase. With all the advantages, the technology has implementation difficulties primarily associated with coal extraction ratio, mechanization of coal haulage as well as safety and efficiency of a production face. Coal lost in a caved area may self-ignite. Besides, cut coal mixes with broken dirt roof rocks and final coal ash content increases. So, it is necessary to analyze the process of coal extraction through outlets in the powered roof support and to design a new coal cutting and extraction complex. COAL EXTRACTION TECHNOLOGIES
There are two versions of the technology for cutting top coal in the roofs of mine workings: 1) coal feed to a front (face) conveyor with mining complexes KTU or KNKM (Russia), or VHP-731 (Hungary) illustrated in Fig. 1a and 2) coal feed onto a rear conveyor with mining complexes OKPV-70 and KM81V (Russia), or ZFC (China) shown in Fig. 1b. Institute of Mining, Siberian Branch, Russian Academy of Sciences, E-mail:
[email protected],
[email protected], Novosibirsk, Russia. Translated from Fiziko-Tekhnicheskie Problemy Razrabotki Poleznykh Iskopaemykh, No. 2, pp. 69-81, March-April, 2010. Original article submitted February 18, 2010. 1062-7391/10/4602-0149 ©2010 Springer Science + Business Media, Inc. 149
Fig. 1. Coal extraction methods: a — upper feed to a front (face) conveyor; b — lower feed to a rare conveyor
Technology version 1 features location of the support outlet close to the face. This allows for a small span of the support unit but provides no proper preparation of coal for cave-in as there is a short distance between the beam and the outlet. As a consequence, coal is to be pre-loosened, its extraction entails high dust generation and raises operational risk. Technology version 2 (coal feed onto a rear conveyor) creates favorable conditions for deformation caving of the roof-adjacent coal layer. However, the support units should be long in this case, while the rare conveyor complicates the structure of the support and of the load-transfer device placed at the longwall juncture with a belt entry, which impedes its maintenance. Common drawbacks of the both technological versions are that coal flow through the outlet is limited and, as a consequence, coal losses and ash content get higher, especially when coal fragments are coarse. With introducing coal extraction lengthwise the entire longwall, through outlets of all units of the support at the same time, the coal – rock contact will come down uniformly, which will allow for the extraction controllability throughout the longwall. The throughout extraction makes it possible to cut coal in thick and complex structure strata, including strata in tectonically faulted rock masses, with variable thickness and dip angles, it also becomes possible to extract coal from previously left pillars, with largely reduced development works and no money needed for purchase of auxiliary high-price equipment, etc. In terms of the Kuznetsk Coal Basin, above 142 Mt coal reserves are extractable with the throughout technology. Based on the research results achieved at the Institute of Mining, Siberian Branch, Russian Academy of Sciences [6 – 8], it has been offered to control coal flow in the course of its force-feed extraction by using a plunger feeder to deliver coal onto a front conveyor and a new design of the powered roof support (Fig. 2). The adjustable and proportioned coal flow above the roof support ensures higher coal production and lower dilution. In addition, the load of the front conveyor is under control, too, and no auxiliary conveyor is involved, which simplifies the geotechnology, especially at the longwall junctures with other workings. The lipped feeder lies between hydraulic legs, under the roof top (outlet); its wedge-shape wavy surface ensures minimum resistance to movement towards the front conveyor and maximum friction and cohesion in movement towards the outlet. A hydraulic jack of the feeder is attached to the rigid platform (rod) and to lips (housing). There is an unloading platform to transfer coal from the feeder to the conveyor. The roof top and the base form the straight-line four bar linkage (Chebyshev linkage). The outlet screen is made as telescope jointed plates, has a hydraulic jack, and is hinge-supported in the roof top. The support unit can additionally include facilities to holdup a face breast, to ensure the unit stability, to support the front conveyor, to ensure effective backup when the support is advanced, to cover lateral clearances, and others. 150
Fig. 2. Powered roof support for the controlled coal feed to the front push conveyor, general view
This thick coal geotechnology and the new design powered roof support with the controlled coal extraction to the front conveyor include the known technological advantages and exclude their drawbacks. Thus, we face new prospects in the development of a science-driven thick coal longwall technology. The application of the described support to coal extraction from thick and steep strata in the Prokopievsk-Kiselevsk Coalfield was limited due to the faulted state and occurrence difficulty of coal. This problem solution included creation of a “support – entry” complex advanced strike-line a coal bed in the course of undercutting with a sublevel drift. This complex has many design variations, e.g., KPP (movable sublevel complex, Keselevsk Machine Works after Chernykh), KPO (sublevel cutting and caving complex, SIBGORMASH JSC, Novosibirsk), APV (sublevel stoping equipment, KuzNIUI), which proved to be efficient [9 – 11]. The recent years’ application of the technology took place at “Kazimierz-Juliusz” coal mine in Poland, by GEOTECH Co., under the name “shrinkage method of extraction”. In this method, entries are driven on the floor of a panel, and the mining equipment, specifically designed in Slovakia, is installed at the ends of the entries. The mining equipment includes two-unit support to maintain the entry and protect the push conveyor between the support legs. This mining equipment enjoyed a success in a 20 m thick coal bed with a dip of 45° at the “KazimierzJuliusz” coal mine [12], which proved the sublevel stoping effect in thick and steep coal beds. However, in terms of the Prokopievsk-Kiselevsk Coalfield, the technology is confined by the absence of blastingfree loosening methods for entry pillars coal. Researchers of the Institute of Mining, Siberian Branch, Russian Academy of Sciences have proposed and validated the following blast-free treatment of the roof-adjacent and interlayer coal: vibroseismic treatment with using a vibroseis source installed in a subdrift; or a directional hydraulic fracturing through holes drilled from subdrifts. As shown in Fig. 3a the offered geotechnology suggests cutting strike-line a coal bed, lengthwise a mining block, with sublevels 1 and haulage drift 2 that are connected through cut-through 3. Between the sublevel drifts, there are strike-line balancing subdrifts 4 for weakening the coal pillar between the sublevel drifts. A balancing subdrift 4 can be connected to a sublevel drift 3 with ventilation holes 5 for airing a dead-end face (Fig. 3b). The most complicated operation in this technology is the safe and effective coal extraction from the weakened pillar to the sublevel drift. To implement this technology and execute powered and controlled coal extraction from the broken inter-level pillar on the sublevel drift, a new sublevel stoping complex has been designed. The sublevel stoping support complex (KPV1) includes two hydraulic-leg support units 6, loader 7 and gate conveyor 8 (Fig. 3c). Loader 7 is placed between the support units furnished with side shields meant for portioning coal flow. The complex is supplied by a high-head plant arranged in a niche in the sublevel drift. 151
Fig. 3. Sublevel stoping method: a — flowsheet; b — coal pre-weakening by vibroseismic treatment; c — coal pre-weakening by directional hydraulic fracturing
The vibroseismic weakening of coal is realized with the use of vibration source 9 installed in subdrift 4 (Fig. 3b), or by the directional hydraulic fracturing through fractures 10 from blastholes 11 drilled from sublevel drift 1 or from subdrift 4 (Fig. 3c). The powered support units have outlets to control the force-feed coal extraction with the help of the adjustable feeders and loader. The design was based on the early designed powered support with controllable feeders. The complex KPV1 in Fig. 4 includes two shielding-supporting units 1, their shields have outlets 2 and reversible shields 3, and feeders 4. The structure of loaders PSP-26 is arranged on the drift floor and is connected with support units 1 via feeder hydraulic cylinder 5. The complex is advanced by alternatively displacing the support units 1 and the loader structure by the feed hydraulic cylinders 5. The motion span depends on the roof state in the sublevel drift and the coal – rock contact in the tail part of the unit. After the units are displaced and abutted, coal from the pre-weakened pillar is extracted through successively opened outlets onto the loader and, then, to a drift conveyor. After coal is extracted from the roof and rock starts flowing through the outlets, their screens are closed. An advantage of this complex is controllability of the coal extraction flow by means of successive opening of the outlets in the support units, which ensures completeness of coal extraction from a mined layer.
Fig. 4. Sublevel stoping suppport complex KPV1 152
For the analysis of the controlled coal extraction in thick and steep coal beds as well as for the more detailed description of the process, based on laboratory research [6], numerical modeling of the gravitation extraction and the assessment of the stress-strain state in the pre-broken coal pillar are discussed below. MATHEMATICAL MODELING OF COAL EXTRACTION
The coal and rock can be presented as separate elements in the discrete element method (DEM) first developed for the analysis of the dense medium behavior [13]. We briefly introduce the basic procedures of the discrete element method as it is sufficiently described in the literature. In the mechanics of granular media, DEM implies digitization of a study domain into elements (particles), as well as assigning them mechanical properties, geometrical parameters and initial conditions for the Cauchy problem. Based on these conditions, a system of equations is generated for description of movement of the particles, considering their collision dynamics. A rock as a granular material is usually considered as a dry noncoherent medium, its individual particles interact only when contact one another. Generally, collision of separate particles can be described with laws of the contact mechanics, based on the continuum approach [14] and realized numerically by using the finite element method. This way involves very high computational effort, and now we have a number of simplified models where particles of a deforming medium keep their shapes and do not break [15 – 17]. Together with classical collision models, the rock behavior analysis can use models with cohesion [18] and models based on the theory of mobile cellular automations [19]. Usually a particle is assumed to be spherical in a 3D space or cylindrical (disk) in plane problems. If a body has a more complex shape, it is often approximated by a combination of spheres (or disks) that are then joined in clusters [20]. The interaction of such clusters can be described by analyzing the interaction of their spherical components. MATHEMATICAL MODEL
In a general case, motion of a spherical particle with radius Ri and mass mi consists of translation and rotation, and is described with the second principle of dynamics: mi
d 2 xi = dt 2 Ii
N
∑ (F
n ij
+ Fijt ) + mi g,
(1)
j =1, j ≠ i
d 2ϕi = dt 2
N
∑ Tij ,
(2)
j =1, j ≠ i
where x i is position vector; I i is inertia moment; ϕi is angular displacement of an i-th particle; N is total quantity of particles. The right-hand side of (1) includes gravity and contact forces between the i-th and j-th particles. Their contact force interaction is described by repulsive force Fijn and friction force Fijt , which are conditioned by normal overlap δ n and tangential overlap δ t (Fig. 5a, b); vector
Tij in (2) is a moment of force Fijt relative to the center of the i-th particle. Current collision models can describe both elastic and inelastic interaction of particles. We apply a linear visco-elastic model (Fig. 5c): the contact forces between the i-th and j-th particles are expressed as follows: 153
Fijn = −k nδ n n ij − η n v ijn ,
(3)
Fijt = −k t δ t t ij − η t v tij ,
(4)
where v ijn and v tij are relative collision velocities of the particles. If the absolute value of vector Fijt is not higher than the limit of the contact friction force, i.e. condition that Fijt > µ Fijn is valid, where µ is dry friction coefficient, then the particles slide over one another and the tangential repulsive force is calculated with the pre-set µ value: Fijt = − µ | Fijn | t ij .
(5)
In Eqs. (3) and (4), kn and kt are elastic coefficients in normal n ij and tangential t ij directions at the contact point; they depend on the size of particles and physical properties of material of the particles; η n and ηt are normal and tangential viscosity, respectively. The parameter η n depends on the pre-defined coefficient of velocity recovery [15]. The interaction between the particles and a boundary in the form of a set of straight intervals is described with the same constitutive relations as above with the only difference that the boundary’s rounded radius at the contact point and mass are assumed as infinite. The system of second-order differential equations (1), (2) with provision for (3) – (5) for every i-th particle in time t + ∆t ( ∆t is integration step) can be solved with using different methods [21]. The present authors chose the Verlet algorithm [22] as one of the simplest and most commonly used approaches to integration with high accuracy at a large ∆t. By the Verlet algorithm, position of a particle at t + ∆t is defined by its two previous positions:
x i (t + ∆t ) = 2x i (t ) − x i (t − ∆t ) + a i (t )∆t 2 , where ai (t ) is particle acceleration obtained from inserting the calculated xi (t ) to the right-hand side of Eq. (1). Velocity of particles is found as: v i (t + ∆t ) =
x i (t + ∆t ) − x i (t − ∆t ) . 2∆t
Such relations are valid for calculating angular displacement ϕ i (t + ∆t ) and angular velocity ωi (t + ∆t ) in (2).
Fig. 5. Mathematical model: a — contact conditions and geometry; b — contact forces; c — rheological model of visco-elastic collision of two particles 154
TABLE Characteristic
Young’s modulus Е, GPa Poisson’s ratio ν Density ρ , kg/m3 Friction coefficient µ
Coal
Rock
5 0.25 1400 0.4
20 0.3 2500 0.4
NUMERICAL EXPERIMENT
A computer program launched to assist in solving application tasks about rock stress-strain state enables 2D numerical modeling of movement of granular materials with various physical properties at different boundary conditions [23, 24]. Let a coal bed be 8 m thick, a sublevel be 12 m high and a bed dip be 60º (Fig. 6). The outlets I – III are 1 m wide and spaced at 0.5 m (Fig. 6). The granular material (coal and rock) is a combination of particles with their radii in a range from 0.05 to 0.1 m. The Table above gives the material characteristics. Weight of a particle is calculated by the assigned radius and density values. First, the material package was realized with the random pushpool algorithm [25]; then, the particles were colored depending on their position. After the material settled in a time, it acquired an equilibrium state equal to the state of a pre-stress rock mass. Simulation of gravitation flow was implemented by removal of the boundary intervals that closed the outlets, and the material started moving. The flow stages in Fig. 6 show the case when the outlets were opened at the same time. The darkcolor and light-color stripes illustrate the material motion. Kinematically, this picture resembles the laboratory experiment. It is seen that the span between the outlets maintains a wide flow and does not allow the flow to be split into three individual flows. The earlier in situ observations and laboratory research showed that the material flow will be continuous, with no outlets chocked if an average fragment size is approximately 1/5 of the outlet width [6]. Actually, this condition is achieved by the direct weakening and, also, due to destruction of large fragments while moving. However, even in this case, it is difficult to maintain the flow continuity.
Fig. 6. Throughout coal extraction in a thick steep coal bed: I, II, III — outlets; L — width of the analyzed domain; h — height of the analyzed domain 155
Fig. 7. Granular material passage velocity: a — free flow through the outlets; b — free flow is limited by the feeder
In the discrete element model, the particles collide without failure but their size distribution satisfies the above condition on the average. The numerical experiments showed that coal extraction involves loosening and compaction cycles in the flow zone above the outlets, and the flow density changes in jumps in the course of the extraction process. This phenomenon results from that the flow particles in the bottom of the flow periodically chock the outlets and the material is “hung up” over the outlets; then, the top flow particles break by gravity the blockade, and the material continue its motion with cave-ins inside the deformation area. The plots in Fig. 7 illustrate the described above phenomenon in terms of the flow passage velocity through the outlets in terms of amount of particles, N, passing through the outlet per time unit. The free flow is shown in Fig. 7a, and Fig. 7b plot is for the case when the particle velocity is limited, for instance, by the support feeder. This motion pattern is typical for the case when three outlets are open, it remains as well when two neighbor outlets are open (I and II, or II and III), and only the width of the flow becomes smaller in the latter case. The pattern of the movement is kinematically different when the outlets I and III are open: the flow regularly chocks the outlet, the rock is hung up and domes arise over them (Fig. 8). Figure 9 exemplifies numerical simulation of coal extraction onto the front conveyor of the support in a thick flat coal bed in the pre-weakened state and at the moment of closing the outlet when rock comes to it. The stress state of rocks above the mined-out space and in the extraction zone is shown in Fig. 10.
Fig. 8. Outlet status: a — force-feed lines when two opposite outlets are open; b — hung up rock domes over the two opposite outlets open; c — loose flow with three outlets open; L — width of the analyzed domain; h — height of the analyzed domain 156
Fig. 9. Coal extraction onto the front conveyor in a thick flat bed: a — initial state; b — kinematics of the extraction
Fig. 10. Coal extraction onto the front conveyor: a — force-feed lines; b — vertical stresses (Pa)
CONCLUSION
The underground geotechnologies for cutting roof coal in thick flat and steep beds should be centered at the idea of controlling the pre-broken rock movement via the forced-feed extraction onto a conveyor placed in front of a powered roof support. The validated design of such support includes feeders arranged in-between the support units, that control the coal extraction process lengthwise entire longwall. The controllable coal extraction on the front conveyor is advantageous for: (1) the roof coal is caved under the action of rock pressure, which reduces energy intensity of mining; (2) coal bed is cut through its entire thickness, which contributes to higher concentration of mining, lower operational losses and weaker spontaneous fire risks; (3) the developmental work and its cost are reduced by a factor of 1.5 to 2; (4) expenses for stoping equipment and coal haulage are decreased, as well; (5) the cost of coal is cut down as the production rate per longwall increases sharply while personnel number is to be increased just a little. The developed mathematical model for gravitation motion of granular materials is based on the discrete element method that seems promising for solving problems about weakened coal mass. The DEM has no drawbacks specific of the continuum models, which show themselves when a discontinuity appears in the material, or as a result of the discreteness of its internal structure. 157
The present paper describes the mathematical model of the gravitation motion of coal and rocks in the sublevel caving technology and solves the problem about coal extraction. The authors have shown that the extracted coal flow and the stress-strain state of the medium near the outlets of the support undergo an impulsive impact of alternating modes “weakening – compaction”. The control over the extraction flow ensures its stabilized velocity when passing the outlets. The study was conducted with financial support from the Russian Academy of Sciences, Basic Research Program ONZ-3.3. REFERENCES
1. A. G. Salamatin, Underground Development of Thick Flat Coal Beds [in Russian], Nedra, Moscow (1997). 2. I. A. Shundulidi, A. S. Markov, S. I. Kalinin, et al., Selecting Parameters of Geotechnology for Cutting Roof and Interlayer Coal in Thick Beds [in Russian], Kemerovo (1999). 3. L. N. Gapanovich, P. F. Savchenko, and V. A. Bernatsky, “Development of powered roof support and coal extraction geotechnology,” Ugol, No. 11 (1986). 4. A. S. Saginov and S. S. Zhetesov, Advanced Mining Technology for Thick and Flat Coal [in Russian], “Kazakhstan”, Alma-Ata (1981). 5. A. S. Saginov and S. S. Zhetesov, Two-Face Cutting of Thick and Flat Coal [in Russian], Nauka, Alma-Ata (1982). 6. V. I. Klishin, Yu. S. Fokin, D. I. Kokoulin, et al., Powered Roof Supports and the Controlled Coal Extraction in Thick Beds [in Russian], Nauka, Novosibirsk (2007). 7. V. I. Klishin, V. N. Vlasov, and B. Kubanychbek, Powered Roof Support for the Force-Feed Extraction of Roof-Adjacent Coal [in Russian], GIAB, Moscow (2003). 8. V. I. Klishin, Yu. S. Fokin, and D. I. Kokoulin, “Joined coal and gas extraction from thick methanesaturated coal beds,” in: International Science & Practice Conference “Kazakhstan Mining Sciences: Outcome and Outlook” [in Russian], Vol. 68, Part 1, Almaty (2004). 9. L. P. Tomashevsky, V. P. Levochko, P. A. Borovikov, et al., “Development and validation of sublevel caving method and parameters for the extractive powered complex “support – drift”,” in: Advanced Geotechnology for Steep Coal Beds of the Kuznetsk Coal Basin. Collection of Scientific Papers [in Russian], No. 25, KuzNIUI, Prokopievsk (1974). 10. L. P. Tomashevsky, Current Technology for Thick, Steep and Faulted Coal Beds in the Kuznetsk Coal Basin and Its Advance Ability. Review [in Russian], TSNIEIugol, Moscow (1978). 11. S. N. Dmitriev, S. I. Zapreev, L. S. Sen’ko, et al., Basics for Coal Extraction with Flexible Supports [in Russian], Nedra, Moscow (1967). 12. S. Gajos, “Experience and practical aspects of utilizing a shrinkage metod of extraction at “KazimierzJuliusz” coal mine in Sosnowiec,” in: International Mining Forum for New Technologies in Underground Mining, Safety in Mines, Cracow-Szczyrk-Wieliczka, Poland (2004). 13. P. A. Cundall and O. D. L. Strack, “A discrete numerical model for granular assemblies,” Geotechnique, 29 (1979). 14. K. L. Johnson, Contact Mechanics, Cambridge University Press, Cambridge (1989). 15. G. Kuwabara and K. Kono, “Restitution coefficient in collision between two spheres,” Japanese Journal of Applied Physics, 26 (1987). 16. H. Kruggel-Emden, “Review and extension of normal force models of the discrete element method,” Powder Technology, 171 (2007). 17. J. Schafer, S. Dippel, and D. E. Wolf, “Force schemes in simulations of granular materials,” Journal de Physique, 6 (1996). 18. E. P. Rusin, S. B. Stazhevsky, and G. N. Khan, “Geomechanical aspects of the genesis of exo- and endokarst,” Journal of Mining Science, No. 2 (2007). 158
19. S. G. Psakh’e, A. Yu. Smolin, S. Yu. Korostelev, et al., “Method of mobile cellular automations and its application to different scale modeling,” in: Mechanics — from Discrete to Continuous [in Russian], SO RAN, Novosibirsk (2008). 20. G. G. W. Mustoe, M. Miyata, and M. Nakagawa, “Discrete element methods for mechanical analysis of systems of general shaped bodies,” in: Proceedings of the 5th International Conference on Computational Structures Technology, Leuven, Belgium (2000). 21. H. Kruggel-Emden, M. Sturm, S. Wirtz, et al., “Selection of an appropriate time integration scheme for the discrete element method (DEM),” Computers and Chemical Engineering, 32 (2008). 22. M. P. Allen and D. J. Tildesley, Computer Simulation of Liquids, Clarendon Press, Oxford (1987). 23. V. I. Klishin and S. V. Klishin, “DEM analysis of coal drawing from high coals in sublevel caving systems,” in: Extended Abstracts of ECCOMAS Thematic Conference “Multibody Dynamics 2009”, Warsaw (2009). 24. S. V. Klishin, “Discrete element analysis of the gravitation flow of a granular material in the narrowing channels,” in: Proceedings of the Conference “Geodynamics and Stress State of the Earth’s Interior” [in Russian], IGD SO RAN, Novosibirsk (2008). 25. K. Han, Y. T. Feng, and D. R. J. Owen, “Sphere packing with a geometric based compression algorithm,” Powder Technology, 155 (2005).
159
