Modelling the Small Throw Fault Effect on the Stability of a Mining Roadway and Its Verification by In Situ Investigation

Transcription

1 energies Article Modelling Small Throw Fault Effect on Stability a Mining Roadway and Its Verification by In Situ Investigation Małkowski Piotr 1, * ID, Ostrowski Łukasz 1 ID and Bachanek Piotr 2 1 Faculty Mining and Geoengineering, AGH University Science and Technology, Kraków, Poland; lukost@agh.edu.pl 2 Tytan sp. z O.O. Mining Company, Mysłowice, Poland; piotrbachanek@o2.pl * Correspondence: malkgeom@agh.edu.pl; Tel.: Received: 7 September 2017; Accepted: 30 November 2017; Published: 7 December 2017 Abstract: The small throw fault zones cause serious problems for mining engineers. The knowledge about range fractured zone around roadway and about roadway s contour deformations helps a lot with right support design or its reinforcement. The paper presents results numerical analysis effect a small throw fault zone on convergence mining roadway and extent fracturing induced around roadway. The computations were performed on a dozen physical models featuring various parameters rock mass and support for purpose to select settings that reflects most suitably behavior tectonically disturbed and undisturbed rocks around roadway. Finally, results calculations were verified by comparing m with in situ convergence measurements carried out in maingate D-2 in Borynia-Ziówka-Jastrzębie coal mine. Based on results measurements it may be concluded that rock mass displacements around a roadway section within a fault zone during a year were four times in average greater than in section tectonically unaffected. The results numerical calculations show that extent yielding zone in ro reaches two times throw fault, in floor 3 times throw, and horizontally approx. 1.5 to 1.8 times width modelled fault zone. Only a few elasto-plastic models or models with joints between rock beds can be recommended for predicting performance a roadway which is within a fault zone. It is possible, using se models, to design roadway support sufficient load bearing capacity at tectonically disturbed section. Keywords: fault; roadway stability; roadway convergence; numerical calculations; in situ measurements 1. Introduction The diverse geological structure Upper Silesian Coal Basin and its mining history result in frequent changes rock mass stability conditions along mining galleries [1]. One instability factors is previously undetected small throw fault zones (a series small dislocations along roadway) within which roadway is prone to convergence, deformation support, and a collapse in worst circumstances. The effect faults or fault zones on underground workings is recognized, qualitatively well-known and researched [2 9]. However, quantitative evaluations this effect are scarce and, in particular, stand-up time roadway. The next possible step following earlier reported research [10], would be an attempt at numerical modelling effect fault on mining roadway to enable its quantitative evaluation, i.e., extent fracturing zones and characteristics Energies 2017, 10, 2082; doi: /en

2 Energies 2017, 10, rock mass deformation. This information is indispensable for proper support design or a decision on additional reinforcement such as rock bolts. In this paper is presented a numerical analysis an effect a small throw fault zone on magnitude convergence mining roadway, and on extent fracturing zone forming around roadway due to convergence. The computation was performed on a dozen rock mass models various parameters rock and support, to determine model best representing behavior tectonically disturbed rock around opening. The results were compared with sections roadway where faults were not present. Although re are some works which employed numerical modelling [2,5 9] authors didn t study alternative way modelling, which could produce significantly different results. Instead, results numerical analyses were ultimately verified by convergence measurements purposely carried out in maingate D-2 in Borynia-Ziówka-Jastrzębie coal mine. Two convergence stations were installed outside and in fault zone, and measurements carried out over 15 months. There are few works which compare results modelling with underground observations [2,7,9]. 2. Weakening Rock in a Fault Zone In areas that underwent strong tectonics mining roadways commonly intersect with minor faults. A weakness zone is associated with each fault and its extent depends on cohesion rock, intensity rock mass tectonic deformation, dip angle and throw magnitude fault. In article [11] it can be found that weakness zone is 1 m wide, while Kidybiński states [3] that around 2 to 7 m throw faults, zone significant weakening is 1 m, yet deterioration reaches up to 5 m on both sides fault plane. Similar observations are brought up by Nieć [4] who reports more than twold reduction uniaxial compressive strength (UCS) sandstone within 4 m zone along Kłodnicki Fault 5 m throw, and within 5 m zone along Bytomski Fault 12 m throw. Shen et al. [5] demonstrates from numerical models that extent rock weakening along fault (distance from fault plane) is numerically two to four times fault s throw. Su et al. [6] suggests, that largest dislocations related to roadways sections being inside fault zone are six meters. Yao et al. [9], in course research, determined minimum fault throw as low as 2 m that is significant for stability a mining roadway. If fault throw is larger than 2 m roadway should be designed and supported considering fault effect. Yao et al. [9] states also that in a case throw smaller than 2 m roadway can be designed and supported as if for tectonically unaffected areas i.e., disregarding fault effect; this is because changes stress around opening are negligible and rock mass is in an undisturbed condition. However, researchers point out necessity additional supports roadways within fault zones e.g., cable bolts, rock grouting [7,8], or reinforced yokes clamps steel support frames to withstand large loads, possibly a dynamic character [12]. Still, it needs to be reminded that most researches were carried out in geological conditions entirely different to those known in Upper Silesian Coal Basin. Polish standards mine roadway support design [13,14], based on works by Zorychta, provide an empirical relationship between extent fault zone, fault throw (h u ), and dip angle fault plane (Equation (1)): L u = 2.5 h u sin β u (1) Therefore, considering typical dip 60 and 5 m throw affected fault zone becomes 6.45 m, while for 4 m throw it is 4.1 m. The extent zone varies significantly toger with shallow fault dipping. However, or researches indicate large differences in assessment fault related extent rock deterioration around roadway.

3 Energies 2017, 10, Research in Mine 3.1. Geological Conditions in Area The analysis fault s impact on stability opening was carried out for maingate D-2 for longwall in Borynia-Ziówka-Jastrzębie mine in coal seam 412. The mine is located in south Poland, 15 km away from border with Czech Republic. The coal seam belongs to Dolnorudzkie beds m thick series Carboniferous strata. In terms its lithology Dolnorudzkie beds comprise claystones (30% uniaxial compressive strength (UCS) = MPa), mudstones (14% UCS = MPa, conglomerates (1% UCS = MPa), sandstones (49% UCS = MPa) and coals (6% UCS = MPa). Usually 5 10 coal seams from this series beds are subject to exploitation. The studied opening belongs to block D mine, which is outlined by position high throw faults (35 m and 50 m), and protection pillar for Main Conveyer Trunk. Fault density coefficient, defined as a ratio between total length faults and area encompassing faults, amounts to m/ha. It is a typical geological situation at stone coal mines in Upper Silesian Coal Basin. Except for major faults which are mainly borders mining fields, re are many small throw faults that intersect driven headings. Maingate D-2 is located at an average depth 1000 m. The sidewall roadway is in its full height formed in a coal bed which is 4.1 m thick (locally interlayered with m claystone) and dips at several degrees (varied 2 22 /E, NE, NEE sometimes SEE). The ro is within an overlying 1.5 m thick bed coaly shale, n 8.5 m sandstone, while floor exposes top 8.3 m thick bed claystone, interlayered with 0.5 m coal. The or two coal seams lie next to seam 412 are: seam 411/3 in a vertical distance m above, and seam 413/ m below. The position strata around roadway was drawn based on technical documentation [15] The Maingate D-2 Support The roadway was entirely supported with ŁPC Bor 12/4/V32 type steel yielding support. This is a steel yielding support, consisting four arched elements, jointed with two clamps each, made V shape section steel type S480W, which is characterized by heightened mechanical properties (Table 1). The maximum width frame (at floor) is 6.5 m, and height m. The arches were spaced at 0.6 m. Maingate D-2 was completed in December Spacing V-Shape Height Table 1. Parameters steel yielding support. V-Shape Cross Section Area V-Shape Moment Inertia Steel Young Modulus Steel Poisson Ratio Steel Yield Limit Steel Tensile Strength [m] [m] [m 2 ] [m 4 ] [GPa] [-] [MPa] [MPa] The In Situ Measurement Methodology In this paper are presented convergence measurement results from two monitoring stations (Figure 1): at chainage ch. 930 m where re were no faults found or mining related effects, and at chainage ch. 757 m where roadway intersected a series faults a relatively small throw ranging from 0.3 to 3 m over 12 m section (Figure 1). Convergence measurements were carried out until July 2016 i.e., until closing longwall excavation started showing up in results.

4 Energies 2017, 10, Energies 2017, 10, D-2 unaffected by faults. The results underground measurements became verification furr numerical calculation studies. Energies 2017, 10, D-2 unaffected by faults. The results underground measurements became verification furr numerical calculation studies. (a) 0.3 ~3.0 ~ Maingate D-2 700m 750m 800m m m 950m 1000m convergence station (a) (b) 0.3 ~3.0 ~ Figure 1. Faults position and 0.2 convergence stations in 2.2 Maingate D-2; (a) The Maingate map Maingate D-2 D-2 Figure 1. Faults position and convergence stations in Maingate D-2; (a) The map Maingate D-2 (b) The convergence stations layout. (b) The convergence stations layout. 700m 750m 800m m m 950m 1000m 1.9 The convergence maingate convergence D-2 was station investigated by a series in situ measurements to (b) determine: change height opening H, change width W, and change floor heave u fl (FigureFigure 2). The 1. Faults values position wereand calculated convergence with reference stations in Maingate to design D-2; (a) dimensions The map Maingate opening. D-2 (b) The convergence stations layout. Figure 2. Scheme convergence measurement station and measurement methodology in D-2 Maingate. Figure Figure 2. Scheme 2. convergence measurement station station and and measurement measurement methodology methodology in D-2 in D-2 Maingate.

5 Figure 4. Maingate D-2 convergence station at chainage 757 m (ΔW roadway width change, Energies 2017, 10, The aim measurement was to show difference rock mass behavior when roadway is located within zone faults with small throws comparing to section Maingate D-2 unaffected by faults. The results underground measurements became verification furr Energies 2017, numerical 10, 2082 calculation studies Results Convergence Measurements 3.4. Results Convergence Measurements Analysis convergence monitoring shows that in 19 months since completion Analysis convergence monitoring shows that in 19 months since completion roadway D-2, at ch. 930 station, height opening decreased 38.5 cm ( H Figure 3) roadway D-2, at ch. 930 m station, height opening decreased 38.5 cm (ΔH Figure 3) and measured 3.84 m, which is 91% initial (nominal) dimension. Apparently, heave and measured 3.84 m, which is 91% initial (nominal) dimension. Apparently, heave floor was cause change opening height. The width roadway within same floor was cause change opening height. The width roadway within same period decreased 20 cm ( W) and measured 5.9 i.e., 96.7% initial (nominal) width 6.10 m. period decreased 20 cm (ΔW) and measured 5.9 m i.e., 96.7% The cross-sectional area roadway decreased by initial (nominal) width 6.10 m. i.e., 13% initial figure [10]. The The cross-sectional area roadway decreased by 2.9 m 2 i.e., 13% initial figure [10]. The convergence development versus time is presented on Figure 3. convergence development versus time is presented on Figure 3. Figure 3. Maingate D-2 convergence station at chainage 930 m (ΔW roadway width change, Figure 3. Maingate D-2 convergence station at chainage 930 m ( W roadway width change, ΔH roadway height change, urf ro displacement, ufl floor heaving). H roadway height change, u rf ro displacement, u fl floor heaving). In contrast, convergence roadway at station ch. 757 m located within fault zone shows much Inhigher contrast, magnitudes. convergence The height roadway opening at station within ch months m located decreased within1.7 fault m zone mainly shows due much to higher floor heave. magnitudes. Horizontal, The height inward movement opening within sidewalls 19 months amounted decreased to m, m mainly i.e., 12% due to initial floor width heave. Horizontal, roadway inward (Figure movement 4). Based on measured sidewallsconvergence, amounted to 0.73reduction m, i.e., 12% cross-sectional initial widtharea could roadway be calculated (Figureas 4). 46% Based on initial measured value [10]. convergence, To maintain reduction maingate D-2 cross-sectional operational area floor could had be to calculated be brushed asthree 46% times: initial by 80, value 60, and [10]. 70 Tocm maintain which is included maingate D-2 operational record and convergence floor had toresults be brushed corrected threeaccordingly times: by 80, (Figure 60, and 4). 70 cm which is included in record and convergence results corrected accordingly (Figure 4). It may be noted that within Maingate D-2 section beyond tectonic deformation, apparently stress condition stabilized approx. after 11 months since completion roadway (Figure 3). Such a behavior is typical for Carboniferous rock mass and is proved by or underground research in Poland [16]. In case roadway section in fault zone, small deformation rock mass around opening continued even after 600 days (approx. 20 months) (Figure 4). It can be also concluded that change height and width opening within fault zone was respectively 4.5 and 3.5 greater comparing to section tectonically unaffected. Generally, at both monitoring stations floor heave contributed mainly to vertical convergence, while downward movement ro recorded was from several to over a dozen centimeters. The contribution floor heave to total displacement is due to fact that re is no

6 Figure 3. Maingate D-2 convergence station at chainage 930 m (ΔW roadway width change, ΔH roadway height change, urf ro displacement, ufl floor heaving). A numerical study was carried out using Phase 2 computer program that utilizes finite elements analysis (FEA) method. The analysis was carried out as for a plane state strain. The In contrast, convergence roadway at station ch. 757 m located within fault zone shows Energies much higher 2017, 10, magnitudes The height opening within 19 months decreased 1.7 m mainly 6 due 21 to floor heave. Horizontal, inward movement sidewalls amounted to 0.73 m, i.e., 12% initial width roadway (Figure 4). Based on measured convergence, reduction cross-sectional support in floor area and could its be strength calculated is relatively as 46% low, so initial it is value only [10]. place To where maintain high maingate stresses D-2 can operational be released. The floor floor heave had to is be caused brushed by large three horizontal times: by stresses 80, 60, and [17], 70 that cm arises which within is included fault in zone. The record summary and convergence in situ measurements results corrected are presented accordingly in Table (Figure 2. 4). Energies 2017, 10, It may be noted that within Maingate D-2 section beyond tectonic deformation, apparently stress condition stabilized approx. after 11 months since completion roadway (Figure 3). Such a behavior is typical for Carboniferous rock mass and is proved by or underground research in Poland [16]. In case roadway section in fault zone, small deformation rock mass around opening continued even after 600 days (approx. 20 months) (Figure 4). It can be also concluded that change height and width opening within fault zone was respectively 4.5 and 3.5 greater comparing to section tectonically unaffected. Generally, at both monitoring stations floor heave contributed mainly to vertical convergence, while downward movement ro recorded was from several to over a dozen centimeters. The contribution floor heave to total displacement is due to fact that re is no support Figure 4. Maingate D-2 convergence station chainage 757 (ΔW roadway width change, Figure in 4. Maingate floor and D-2 its convergence station strength is relatively at low, chainage so it 757 is m only ( W roadway place where width high change, stresses can be ΔH roadway height change, urf ro displacement, ufl floor heaving). H roadway released. The height floor change, heave uis caused rf ro displacement, by large horizontal u fl floor stresses heaving). [17], that arises within fault zone. The summary in situ measurements are presented in Table 2. Table 2. Results convergence measurements in Maingate D-2 [10]. Table 2. Results convergence measurements in Maingate D-2 [10]. Measured Measured Results Results Change Height Change Mointoring Change Width Change Height Floor Heave Roadway Station Roadway Roadway Floor Heave Roadway Station Roadway Roadway Cross-Section Chainage Cross-Section Chainage ΔW W H u W ΔW [%] ΔH H ΔH [%] fl min u fl max u fl av A ufl min ufl max ufl av ΔA [cm] [cm] [cm] [cm] [cm] [m 2 A [%] ΔA ] [cm] [%] [cm] [%] [cm] [cm] [cm] [m 2 ] [%] m m Figure 5 graphically depicts changes cross sectional areas at at both bothmonitoring stations. It It demonstrates that that in in tectonically undisturbed section as aswell as in fault affected section floor heave contributed mainly to to vertical convergence roadway D-2. D-2. This This kind kind behavior behavior rock rock mass mass around around openings openings is regarded is regarded typical typical for Carboniferous for Carboniferous formation formation [18], particularly [18], particularly where where underlying underlying rock is rock closely is closely stratified stratified and aand low strength a low strength [2,19]. [2,19]. Figure 5. Maingate D-2 convergence: (a) station at ch. 930 m; (b) station at ch. 757 m. 4. Numerical Evaluation Stability Opening 4.1. Assumptions Models

7 Energies 2017, 10, Numerical Evaluation Stability Opening 4.1. Assumptions Models A numerical study was carried out using Phase 2 computer program that utilizes finite elements analysis (FEA) method. The analysis was carried out as for a plane state strain. The position geological layers and ir mechanical parameters, juxtaposed with parameters corresponding with Hoek-Brown failure criterion are presented in Table 3. Mechanical parameters rock types were based on laboratory testing carried out by authors, and on technical documentation roadway [15]. The position strata on model were inferred from geological log an exploratory core holes in a close vicinity both monitoring stations in maingate D-2. Table 3. Rock mass parameters in numerical model. Stratum Rock Type Thickness h [m] Specific Weight γ [kn/m 3 ] UCS R c [MPa] Young Modulus (Lab. Test) Young Modulus (Rock Mass) Poisson Ratio Hoek Brown Failure Criteria E i [GPa] E m [GPa] ν [-] m s s Fault Breccia Ro Sandstone Coal bed 411/ Mudstone Sandstone Coaly shale Sidewall Coal bed Floor Claystone Coal Claystone Coal bed Mudstone Claystone The opening location at an average depth 1000 m and related to it geostatic stress condition were assumed. Discretization model was in form three node triangle mesh each triangle constituting a single element. In vicinity opening mesh element density has been increased. The furr change mesh element density didn t affect results numerical calculations. The condition model excluded horizontal movement nodes on vertical boundaries and vertical movement nodes on horizontal boundaries. The roadway support was a steel frame arches m high and 6.5 m span at foot, made V shape section steel type S480W with yield limit 480 MPa and tensile strength 650 MPa (standard ŁPC Bor 12/4/V32 type support ). The arches were spaced at 0.6 m. For model purpose, support was represented as Reinforced Concrete and by selection Reinforcement option only. This option allows to consider, in contrary to Standard Beam, spacing individual supports Section Roadway Unaffected by Fault A model rock mass for section roadway tectonically unaffected was outlined within m plane (Figure 6). The total number nodes and elements for model amounted to 6785 and 13,122. The computation was carried out on 13 different numerical models to produce a series strain-stress characteristics in surrounding roadway, as well as extent rock failure zones (Table 4). Nine models represented rock mass elasto-plastic properties, strengned or weakened, or perfect elasto-plastic properties, while four models assumed elastic behavior rock. In elasto-plastic model no. 3, and in elastic model no. 12, Young modulus rock strata was empirically estimated assuming that Geological Strength Index is equal to Rock Mass

8 Energies 2017, 10, Rating GSI = RMR [20], whereas for or models it was determined through laboratory testing. The joint type interbed contact has been used in three models no. 7, 8 and 9 (Table 4). In model Energies 2017, 10, no. 7 contact between all beds was assumed as Phase 2 default setting for normal and shear stiffness, i.e., k n = 100,000 MPa/m and k s = 50,000 MPa/m ( case 1). In model no. 8 contact internal friction 33, while for contacts between gangue rocks tensile strength 0.2 MPa, type joint was applied to contact between coal, sandstone and shale where k n = 220,000 MPa/m cohesion 2 MPa, and angle internal friction 35. For both situations, residual values and k s = 110,000 MPa/m, while for or gangue rocks k n = 880,000 MPa/m and k s = 440,000 MPa/m tensile strength and cohesion were assumed to be zero. Furrmore, for both situations, a reduction (case 2). angle internal friction to 28 and 30 was specified as suggested by Jie et al. [2]. Figure Figure Model Model plane plane roadway roadway beyond beyond fault fault effect. effect. Table 4. Physical models rock mass section beyond fault zone. Table 4. Physical models rock mass section beyond fault zone. Post-Failure Support Contact Physical Model Young No. Physical Post-FailureStrength Support Support Young Frame Slide Contact between No. (Isotropic) Model Modulus Strength E Support Frame Slide Modulus E Change Possible Strata (Isotropic) Change Possible Strata Yes, steel material 1 1 Elasto-plastic Laboratory Laboratory - Yes, - steel yielding support No No material yielding support 2 Elasto-plastic Laboratory - Yes, steel yielding support Yes; 0.5% material Yes, steel material 2 3 Elasto-plastic Rock mass Laboratory - Yes, - steel yielding support No Yes; 0.5% material yielding support 4 Elasto-plastic Laboratory - No - material Yes, steel material 3 5 Elasto-plastic Laboratory Rock mass +20% Yes, - steel yielding support No No material yielding support 6 Elasto-plastic Laboratory 20% Yes, steel yielding support No material material 4 Elasto-plastic Laboratory - No - 7 Elasto-plastic Laboratory - Yes, steel yielding support No joint type case 1 8 Elasto-plastic Laboratory - Yes, steel yielding support Yes, steel No joint typematerial case 2 5 Elasto-plastic Laboratory +20% No 9 Elasto-plastic Laboratory - Yes, steel yielding yielding support support No joint type case 3 10 Elastic Laboratory - Yes, steel yielding support Yes, steel No material material 6 Elasto-plastic Laboratory 20% No 11 Elastic Laboratory - Yes, steel yielding yielding support support Yes; 0.5% material Yes, steel joint type 7 12 Elasto-plastic Elastic Rock mass Laboratory - Yes, - steel yielding support No No material yielding support case 1 13 Elastic Laboratory - No - material Yes, steel joint type 8 Elasto-plastic Laboratory - No yielding support case 2 In case model no. 9, joint type contacts were given Yes, steel ability slide due to yield joint effect type 9 Elasto-plastic Laboratory - No along contact plane and governed by Coulomb-Mohr criterion yielding support (case 3). In this model, for contacts case3 coal-gangue rocks tensile strength was assumed 0.2 MPa, cohesion Yes, steel 2 MPa, and angle material internal 10 Elastic Laboratory - No friction 33, while for contacts between gangue rocks yielding tensilesupport strength 0.2 MPa, cohesion 2 MPa, and Yes, steel material 11 angle Elastic internal friction Laboratory 35. For both - situations, residual values Yes; tensile 0.5% strength and yielding support 12 Elastic Rock mass - Yes, steel yielding support 13 Elastic Laboratory - No - No material material

9 Energies 2017, 10, Energies 2017, 10, cohesion were assumed to be zero. Furrmore, for both situations, a reduction angle internal friction Table to5 28presents and 30 a was compilation specified as suggested results bycalculated Jie et al. [2]. total displacements along contour Table opening. 5 presents Displacement a compilation ro results Δurf and calculated floor Δufl totalare displacements juxtaposed along with contour values vertical opening. ΔH and Displacement horizontal ΔW convergence. ro u rf andthe floor u final results fl are juxtaposed underground with measurements values vertical are presented H and horizontal in Table W 5 as convergence. well. The coefficient The final results conformity underground between measurements measured areand presented modelled in displacement Table 5 as well. is also Theprovided coefficient as an conformity average between all four abovementioned measured andvalues. modelled displacement is also provided as an average all four abovementioned values. Table 5. Physical models rock mass section beyond fault zone. Table 5. Physical models rock mass section beyond fault zone. Model Δurf [cm] Δufl [cm] ΔH [cm] ΔW [cm] Conformity Model 1 u rf [cm] 4 u fl [cm] H [cm] 20 W [cm] 96.90% Conformity % % % 43.78% % 38.22% % 31.29% % 73.52% % 80.70% % 89.13% % % % % % % 9.86% % 52.97% % 29.11% Measured Measured values values Bold conformity up up to to 20% 20% between value from a a model and and measurement. Analysis obtained results shows that in all elastic models coefficient conformity between modelled and measured values values appears appears significantly lower lower than inthan elasto-plastic in elasto-plastic models. models. Generated Generated displacements displacements are generally are generally very low very (Figure low 7), (Figure and even 7), and decreasing even decreasing Young modulus Young modulus and increasing and increasing deformability deformability rocks didrocks not did change not change output values output towards values towards closer tocloser to deformations deformations measured measured in maingate maingate D-2. D-2. Figure Figure Elastic Elastic model model no. no. 10 map 10 map total total displacements. displacements. For this reason, elastic model rock mass was decided as not applicable for estimation For this reason, elastic model rock mass was decided as not applicable for estimation stress-strain condition around workings in Carboniferous formation. The results furr confirmed stress-strain condition around workings in Carboniferous formation. The results furr confirmed that elastic models can be used for homogenous or quasi-homogenous rock mass while good that elastic models can be used for homogenous or quasi-homogenous rock mass while good results for stratified rock mass can be achieved from elasto-plastic models [18]. results for stratified rock mass can be achieved from elasto-plastic models [18]. Models, in which steel yielding support was assumed with sliding, showed larger inward movement crown and sidewalls working (Table 4) in comparison with or models. It showed that models with yielding support (frames with sliding ability) are not suitable for this purpose.

10 Energies 2017, 10, Models, in which steel yielding support was assumed with sliding, showed larger inward movement Energies 2017, 10, 2082 crown and sidewalls working (Table 4) in comparison with or models It showed that models with yielding support (frames with sliding ability) are not suitable for this purpose. An interesting conclusion was suggested by analysis models where Young modulus for Anrock interesting mass was conclusion empirically wasestimated. suggested Despite by analysis evident difference models where between Young modulus behavior for a rock rocksample mass was and empirically rock mass it estimated. is commonly Despite recommended evidenta difference reduction between mechanical behavior parameters a rock for sample rock mass and[20 22], rock mass researches it is commonly prove such recommended a reduction a reduction for sedimentary mechanical rock, that parameters are typically for rock low mass anyway, [20 22], is not researches well substantiated. prove suchin a reduction case for maingate sedimentary D-2, both rock, models that arewhere typically estimated low anyway, Young is not modulus well substantiated. for a rock mass In was case applied, maingate D-2, results both were models largely where discrepant estimated Young from modulus in situ for ameasurements. rock mass wasthe applied, elastic model results(no. were 12) largely generated discrepant results from even lower in situ than measurements. for model The based elastic on model laboratory (no. derived 12) generated Young results modulus even and lower on yielding than forsupport. model The based elasto-plastic on laboratory model derived (model Young no. 3) modulus showed results and oncertainly yieldingtoo support. high, The exceeding elasto-plastic twice model underground (model no. measurements 3) showed results on average, certainlythat too high, can be exceeding expressed twice as coefficient underground conformity measurements 38% between on average, predicted that canand be expressed measured as values. coefficient The chart conformity total displacements 38% betweenand predicted mechanism and measured failure is values. presented The on chart Figure total 8. The displacements zone failure and is mechanism formed in failure floor is presented opening ondue Figure to tension 8. The zone and shearing. failure is Its formed extent in is 5.6 floor m. In sidewalls opening due failure to tension is caused and shearing. by Its stress extent and isreaches 5.6 m. In 3.2 m deep. sidewalls In crown, failure is caused yielded by elements shear stress zone and appears reaches in 3.2entire m deep. 1.5 Inm thick crown, layer yielded coaly shale. elements However, zone appears such an inextensive entirefailure 1.5 m thick zone layer does not coaly exist shale. in reality. However, such an extensive failure zone does not exist in reality. Figure 8. Model elasto-plastic with with interpreted Young Young modulus map total displacement total displacement and failure and mechanism failure mechanism around around opening. opening. Almost Almost entire entire conformity conformity between between generated generated and measured and measured values was values achieved was forachieved elasto-plastic for model elasto-plastic no. 1 (coefficient model no. 1 (coefficient conformity 96.9% Table conformity 96.9% Table 4). The chart 4). The total chart displacement total displacement generated from generated this model from this is presented model is on presented Figure 9. on Movement Figure 9. Movement floor appears floor nearly appears uniform nearly on uniform full width on full width opening and reaches opening approx. and reaches 34 cm, which approx. is consistent 34 cm, which with in is situ consistent measurements with in in situ mine. measurements The chart in yielded mine. elements The chart around yielded elements opening is around shown on opening Figure 10. is shown It is worth on Figure pointing 10. out It is that worth it is pointing almost out identical that it with is almost one identical generated with from one model generated no. 3 (compare from model with no. Figure 3 (compare 7) in which with Figure Young s 7) in modulus which Young s rock mass modulus was estimated rock empirically. mass was The estimated downward empirically. extent The yield downward zone appears, extent reaching yield 5.8 zone m from appears, floor, reaching whereas 5.8 m in from crown floor, it spreads whereas over in overlying crown it layer spreads coaly over shale overlying and shows layer a clear coaly shale and on shows above a clear layer strong on sandstone. above layer Towards strong sidewalls sandstone. yield Towards zone is 3 sidewalls m deep. Therefore, yield zone it can is be 3 m concluded deep. Therefore, that change it can be concluded Young s modulus, that change which indicates Young s modulus, ratio between which indicates stress and strain, ratio strongly between affects stress total and strain, displacements. strongly However, affects total it does displacements. not significantly However, affect it does extent not significantly yield zone affect around extent opening. yield zone around opening.

11 Energies 2017, 10, Energies 2017, 10, Figure 9. Model no. 1 elasto-plastic map total displacements. Figure 10. Model no. 1 elasto plastic yielded rock mass elements. Satisfactory Satisfactory results results were were obtained obtained from from elasto-plastic elasto-plastic models models no. no. 7 and 7 and 8 that 8 that utilized utilized joint type joint contact type contact between between layers. layers. It can It be can noted be noted that in that models in models no. 7 and no. 8 7 and selected 8 selected parameters parameters normal and normal tangential and tangential stiffness stiffness joint type contact joint type did contact not significantly did not significantly alter convergence alter comparing convergence to model comparing no. 1. to It model appears no.1. different It appears in different case in model case no. 9 where model no. 9 joint where type contacts joint type were contacts given were ability given slide ability due to yield slide effect due to along yield effect contact along plane. contact However, plane. introduction However, introduction se settings to se settings model no. to 9 resulted model no. in 9 resulted least conformity in least between conformity generated between and generated measured and convergence measured (i.e., convergence approx. 38% (i.e., against approx. approx. 38% against 88% for approx. models 88% no. for 7 models and 8). no. 7 and 8). The The most most satisfactory satisfactory results results were were obtained obtained from from model model no. no. 6 with with 20% 20% post-failure post-failure strength strength weakening weakening rock rock mass. mass. Considering Considering that that roadway roadway needs needs to to be be maintained maintained and and used used over over a period period time, time, model model no. no. 6 may may prove prove to to be be most most one one suitable suitable for for predicting predicting behavior behavior opening opening under under given given conditions. conditions. It It has has been been demonstrated demonstrated that that even even after after days days from from completion completion opening opening deformation deformation rock rock mass mass may may continue. continue. The The chart chart total total displacement displacement is is shown shown on on Figure Figure 11, 11, and and chart chart stress stress distribution distribution around around opening opening on on Figure Figure Due Due to to weakening post-failure parameters about 20% yield zone below floor increases from 5.8 to 6.6 m comparing to equivalent model yet without weakening rock mass (compare to Figure

12 Energies 2017, 10, Energies Energies 2017, 2017, 10, 10, weakening post-failure parameters about 20% yield zone below floor increases from 5.8 to 6.6 m comparing to equivalent model yet without weakening rock mass (compare to Figure 7), 7), 7), whereas decreasing post-failure parameters does not not affect zone yield above ro whereas decreasing post-failure parameters does not affect zone yield above ro and and beyond sidewalls. Directly at at contour opening destruction rock is is due due to to and beyond sidewalls. Directly at contour opening destruction rock is due tension, to tension, where where depth depth tension tension zone zone is is is from from approx. approx cm cm cm in in in crown crown and and and sidewalls sidewalls to to to approx. approx m 1.5 below m below floor. floor. Figure Figure Figure Model Model no. 6 6 elasto-plastic elasto-plastic with with with 20% post-failure 20% 20% post-failure post-failure strength weakening strength strength weakening weakening rock mass map rock rock mass map mass map total displacements total total displacements displacements around around opening. opening. opening. Figure Figure Figure Model Model no. no. 6 elasto-plastic elasto-plastic with with with 20% post-failure 20% 20% post-failure post-failure strength weakening strength strength weakening weakening rock mass map rock rock mass map mass map stress distribution stress stress distribution distribution around around around opening. opening. opening Section Roadway within Fault Zone For For section intersecting fault zone dimensions rock mass model were: height m, m, width m (Figure 13). 13). The The total number nodes and and elements for for model in in this this case amounted to to 7042 and and 13,803, but but mesh density and and size size elements next to to roadway were same, so so discretization didn t influence results calculations. The The fault is is represented in in model as as a zone lower mechanical parameters. The The extent zone is is

13 Energies 2017, 10, Section Roadway within Fault Zone For section intersecting fault zone dimensions rock mass model were: height 60 m, width 100 m (Figure 13). The total number nodes and elements for model in this case amounted to 7042 and 13,803, but mesh density and size elements next to roadway were same, Energies 2017, 10, so discretization didn t influence results calculations. The fault is represented in modelled as aas zone a 30 cm lower thick mechanical band tectonic parameters. breccia, The as observed extent in zone Maingate is assumed D-2, and to befeaturing 5 m on both: ca. 10 times footwall lower and mechanical hangingparameters wall side, based than on intact experience rocks in discussed this area. in Chapter 2 [2,3,5,9]. For all models, The it2-d is assumed geological also model that depicts compressive dislocation strength UCS strata within consistently tectonically with disturbed 3 m throw zone is 50% fault, smaller its slip compared direction to and magnitude. unaffected The section. fault The plane fault dip plane is assumed is modelled to be as65 a 30after cm thick geological band documentation tectonic breccia, as observed area (Figure in 13). Maingate The change D-2, and featuring fault angle ca. parallel 10 times to lower roadway mechanical axis parameters hasn t been than considered intact rocks in 2-D inmodel. this area. Figure 13. Model planes roadway intersecting fault zone. When modelling fault effect on stability roadway authors utilized The 2-D geological model depicts dislocation strata consistently with 3 m throw experience from modelling section beyond fault zone. For this purpose, eight numerical fault, its slip direction and magnitude. The fault plane dip is assumed to be 65 after geological models were elaborated excluding elastic models. In all models, elasto-plastic rock mass was documentation area (Figure 13). The change fault angle parallel to roadway axis hasn t assumed. A minor leakage was found during site visits from fault fracture in maingate D-2. been considered in 2-D model. The leakage causes corrosion many ro arches support (Figure 14). When modelling fault effect on stability roadway authors utilized experience from modelling section beyond fault zone. For this purpose, eight numerical models were elaborated excluding elastic models. In all models, elasto-plastic rock mass was assumed. A minor leakage was found during site visits from fault fracture in maingate D-2. The leakage causes corrosion many ro arches support (Figure 14). On that section, observed movements on ro arches joints and ro-sidewall arches clamps were not bigger than several millimeters. Based on se observations it was specified, that for stiffened support structure, half models did not feature ability clamp movement, whereas in second half this option is enabled and is set as 0.5% support cross section contour, i.e., movement approx. 7 cm along joint (models no. 1, 4, 6 and 8). Figure 14. View on ro roadway at monitoring station at chainage 757 m. On that section, observed movements on ro arches joints and ro-sidewall arches clamps were not bigger than several millimeters. Based on se observations it was specified, that for

14 When modelling fault effect on stability roadway authors utilized experience from modelling section beyond fault zone. For this purpose, eight numerical models were elaborated excluding elastic models. In all models, elasto-plastic rock mass was Energies assumed. 2017, A 10, minor 2082 leakage was found during site visits from fault fracture in maingate 14D The leakage causes corrosion many ro arches support (Figure 14). Figure 14. View on ro roadway at monitoring station at chainage 757 m. Figure 14. View on ro roadway at monitoring station at chainage 757 m. On that section, observed movements on ro arches joints and ro-sidewall arches clamps were The not features bigger than all several physical millimeters. models designed Based on to analyze se observations effect it was fault specified, on deformation that for stiffened roadway support are compiled structure, in Table half 6. Normal models stiffness did k not n and feature shear stiffness ability k s are clamp provided movement, in cases where whereas contact in types second played half a this role option in model. is enabled In this and case is set (models as 0.5% no. 5 8) support stiffness cross considered section contour, inter-bed i.e., movement contacts inside approx. damage 7 cm along zone is joint assumed (models dozen no. hundred 1, 4, 6 and times 8). lower than for intact rock beds as in models studied for roadway section beyond fault zone. It was assumed that normal to shear stiffness ratio is equal 3:1 that is suggested as optimal [23], refore normal and shear stiffness for models no. 5 and 6 amount to k n = 150 MPa/m, k s = 50 MPa/m (case 4) and for models no. 7 and 8 amount to k n = 300 MPa/m, k s = 100 MPa/m (case 5). Parameters contact for all models are given in Table 6. The applied higher values contact stiffness are typical for Carboniferous sandstone [23], which is present in ro roadway being analyzed. Table 6. Physical models for roadway section inside fault zone. Model No. Isotropic Model Young Modulus E Post-Failure Strength Change Support Support Frame Slide Possible Contact Beds on Fault Plane 1 Elasto-plastic Laboratory - Yes, steel yielding support Yes; 0.5% material 2 Elasto-plastic Laboratory - Yes, steel yielding support No material 3 Elasto-plastic Laboratory 20% Yes, steel yielding support No material 4 Elasto-plastic Laboratory 20% Yes, steel yielding support Yes; 0.5% material 5 Elasto-plastic Laboratory 20% Yes, steel yielding support No joint type case 4 6 Elasto-plastic Laboratory 20% Yes, steel yielding support Yes; 0.5% joint type case 4 7 Elasto-plastic Laboratory 20% Yes, steel yielding support No joint type case 5 8 Elasto-plastic Laboratory 20% Yes, steel yielding support Yes; 0.5% joint type case 5

15 Energies 2017, 10, The comparison results generated from models and in situ measurements in mine is presented in Table 7. Generally, after exclusion several models on first stage investigation, generated results are largely consistent with in situ measurements despite that obtained horizontal convergence values were too high up to 70% (model no. 4). In case vertical deformations and upheaval floor modelled and factual values differed not more than 26%, and most ten difference was not bigger than 10%. All results from models no. 1 and 2 (except movement ro in model no. 1) show only a dozen or so percent difference to measurements on monitoring stations (Table 7). Settings elasto-plastic model with 20% post-failure strength weakening for section affected by fault (models no. 3 and 4) Energies 2017, 10, produced values most consistent with in situ measurements at monitoring stations. Exceptionally good conformity is shown in model no. 3 in which support is assumed without movement along joints. Table 7. Total deformations roadway generated by models in comparison with in situ measurements. Table 7. Total deformations roadway generated by models in comparison with in situ measurements. ΔH ΔW Model Δufl max Δufl Δurf Average Conformity Δurf + Δufl Left Right L + R Model u fl H W Average u fl u [cm] [-] rf Max u rf + u fl 63 Left 42 Right 105 L + R Conformity 75.42% [cm] % [-] % % % % % 72.60% % 76.45% % 81.25% % 78.46% Measured % % - values Measured values Bold conformity up to 20% between value from a model and measurement. Bold conformity up to 20% between value from a model and measurement. Charts total displacements opening contour generated from model no. 3 and model no. 4 are Charts presented total on displacements Figures 15 and 16 opening respectively. contourthe generated charts from show model qualitatively no. 3 andvery model similar no. 4 deformation are presentedaround on Figures maingate 15 and 16 D-2 respectively. and characteristic The chartsasymmetrical show qualitatively movement very similar deformation contour in relation around to maingate fault D-2 plane. and characteristic Quantitatively, asymmetrical deformation movement rock contour mass is in better relation reproduced to fault by model plane. no. Quantitatively, 3. deformation rock mass is better reproduced by model no. 3. Figure 15. Model no. 3 Maingate D-2 in fault zone map total displacement.

16 Energies 2017, 10, Energies 2017, 10, Figure 16. Model no. 4 Maingate D-2 in fault zone map total displacement. Based Basedon on analysis analysis obtained results it it can can be be concluded concluded that that in in models models set set with with ability ability slip slipon onjoints (clamps) calculated movements ro ro are are characterized by by highest highest conformity with in in situ situ measurement, measurement, while while calculated calculated change change width opening width opening lowest. On lowest. or hand, On in or models hand, set with in stiffmodels support set structure, with stiff deformations support structure, ro deformations were underestimated ro bywere up tounderestimated 50%, yet most accurate by up to as50%, to yet change most accurate width as to opening. change The width reason behind opening. se observations The reason behind may bese different observations than assumed may be (geostatic) different distribution than assumed (geostatic) stress around distribution roadway. However, stress around geostatic roadway. stress was However, not investigated geostatic prior tostress modelling. was not investigated Furrprior investigation to modelling. included aspect contact character between fault and rock beds [2,24,25]. ForFurr that purpose, investigation models included no. 5, 6, aspect 7 and 8 contact character type jointbetween was selected. fault These and rock models beds [2,24,25]. achievedfor that highest purpose, conformity in coefficient models no. between 5, 6, 7 and generated 8 contact and measured type joint situ was values; selected. approx. These models 10% higher achieved than models highest without conformity jointcoefficient type contacts. between However, generated models and twold measured higher in situ stiffness values; approx. parameters 10% between higher than layers (models no. without 7 and joint 8) were type characterized contacts. However, by muchmodels better, than twold modelshigher no. 5 stiffness and 6, fit parameters generated between andlayers in situ(models measuredno. deformation 7 and 8) were characterized floor (Table 7). by much better, than models Figures 5 and 17 6, fit and 18 depict generated distribution and in situ measured displacements deformation around opening floor (Table generated 7). from models Figures no. 717 and and A significant depict difference distribution (yet small displacements in terms absolute around value around opening generated 8 cm) canfrom be models noted in no. displacement 7 and 8. A significant hangingdifference wall (yet fault small that in is bigger terms in model absolute no. value around 8, in which 8 support cm) can be was noted set with in displacement option arch sliding hanging (Figure wall 17). Qualitatively, fault that is deformation bigger in model around no. maingate 8, which D-2 in support both models was does set with not differ, option except arch for sliding above (Figure detail. 17). However, Qualitatively, in terms magnitude, deformation it needs around to maingate be pointed D-2 out in that both in models model does no. not 8, differ, increase except deformation above in detail. rohowever, 8 cm resulted in terms in a magnitude, decrease it needs floorto upheaval be pointed by 9out cm. that in model no. 8, increase deformation in ro 8 cm Byresulted comparing a decrease stress condition floor around upheaval analyzed by 9 cm. maingate D-2 within and beyond fault section it can be noted that destruction zone in floor is 2.5 times deeper in fault affected section than beyond it (Figures 9, 19 and 20). In case sidewalls destruction is 3.5 times deeper in average, while in ro this zone is up to 6 to 7 times larger.

17 Energies 2017, 10, 2082 Energies 2017, 10, Energies 2017, 10, Figure 17. Model no. 7 Maingate D-2in in fault zone joint type rock mass and rigid arch Figure support 17. Model map no. total 7 Maingate displacement. D-2 in fault zone joint type rock mass and rigid arch support map total displacement. Figure 18. Model no. 8 Maingate D-2 in fault zone joint type rock mass and yielding arch support Figure 18. Model map no. total 8 Maingate displacement. D-2 in fault zone joint type rock mass and yielding arch Figure 18. Model no. 8 Maingate D-2 in fault zone joint type rock mass and yielding arch support map total displacement. support map total displacement. By comparing stress condition around analyzed maingate D-2 within and beyond fault By section comparing it can be noted stress that condition destruction around zone analyzed in maingate floor is 2.5 D-2 times within deeper and in beyond fault affected fault In section applied section it can numerical than be beyond noted that models, it (Figures destruction destruction 9, 19 and 20). zone zone In in mainly case floor coincides is 2.5 sidewalls times with deeper area destruction in strength fault is 3.5 affected parameters times section reduced deeper than to in average, beyond reflect while it fracturing (Figures in 9, caused ro 19 this and by zone 20). In fault. is up to case Introduction 6 to 7 times sidewalls joint larger. type destruction contact into is 3.5 times model In deeper (Figure applied in 19) numerical average, affectswhile extent models, in ro destruction this zone zone only zone is up in mainly to 6 ro to coincides 7 times larger. roadway i.e., reduces its with area strength downward parameters In applied movement reduced numerical approx. to reflect fracturing models, by 10% comparing caused destruction to model by zone fault. mainly without Introduction coincides joint contact joint with (Figure type area 20). contact strength into model parameters (Figure reduced 19) affects to reflect extent fracturing zone caused only by in fault. ro Introduction roadway i.e., joint reduces type contact its downward into movement model (Figure approx. 19) affects by 10% comparing extent to model zone only without in joint ro contact roadway (Figure 20). i.e., reduces its downward movement approx. by 10% comparing to model without joint contact (Figure 20).

18 Energies 2017, 10, Energies 2017, 10, Figure 19. Model no. 7 Fault zone, joint type contact, rigid support map yielded rock mass elements around roadway. Figure 20. Model no. 3 Fault zone, yielding support map yielded rock mass elements around roadway 5. Discussion In this paper are presented results convergence measurements mining roadway in which alignment is within and outside fault zone. The measurements were carried out over a period a dozen months to show dynamics and magnitude changes deformation, and with purpose analyzing impact tectonically affected zone on stability opening. Based on on numerical numerical analysis analysis it possible it is possible to achieve to very achieve highvery agreement high between agreement convergence between values convergence generated values by generated models by and measured models inand situ. measured Based on experience, in situ. Based on difference experience, between modelled difference and between factual modelled values isand around factual 20%, values hence is acceptable around 20%, andhence valid, acceptable from rock mechanics and valid, point from rock view, mechanics for prediction point view, deformations prediction mining deformations roadways. Although mining it roadways. seems that Although 3D modelling it seems small that 3D fault modelling zones should small give fault more zones accurate should results, give it ismore limited accurate by unknown results, it spatial is limited orientation by unknown faultspatial planesorientation and ir true length. fault For planes thisand reason, ir true fault length. is modeled For this asreason, a straight inclined fault is plane modeled [9]. as a straight inclined plane [9]. It should be underlined that modelling rock mass with fault zones is extremely difficult and entails necessity or additional assumptions to model concerning this zone. There are

19 Energies 2017, 10, It should be underlined that modelling rock mass with fault zones is extremely difficult and entails necessity or additional assumptions to model concerning this zone. There are some phenomena on contact between rocks and fault plane which are known qualitatively but not quantitatively. The first problem is how much mechanical parameters rocks within fault zone should be reduced due to se phenomena. It is reasonable to relate reduction to fault throw, so for set small throw zones, as a rule thumb, we decided to reduce initial values by 50%, but for high throw fault and very small rock pieces breccia probably more. Introduction joint type contact between rock beds may lead to a better fit numerical output and factual deformation values. However, due to difficulties in estimation mechanical parameters contacts, final effect may be opposite, what is demonstrated by numerical calculations. Yet in case fault zones around mining roadways this type physical models may still be best methodology option, even though it is a 2D modelling [2]. It is interesting that for example Jie et al. [2] uses also model with joints, described separately for successive rock beds, but assumes very high stiffness, close to intact rocks contact (several GPa/m). The model built by Yao team [9] had normal stiffness on fault plane 1.5 GPa/m and shear stiffness 0.5 GPa/m. It seems to be too high, but both authors didn t explain this assumption. Undoubtedly, or numerical analyses and underground research from various countries would be very valuable for calibration and verification fault zone influence on mining openings. Also, a long-lasting in situ measurements carried out next to fault zone in openings would be a source meaningful information. An inclusion in model initial stress condition in rock mass would be advantageous for estimation deformation around opening, too. Yet in case reported in this paper this information was not available. The last problem is influence rheological and frictional slip properties on fault mechanics. However, for short-time existing openings as e.g., gates it can be considered as marginal importance. In our opinion, such studies should focus on resolving mining problems. A large damage zone forming around a roadway imposes very high capacity rock bolt system or even use reinforced concrete and rock mass injections a common occurrence in mines e.g., in Poland and China. The examination various physical numerical models a small throw fault zones and ir effect on mining roadway brought a new light on problem and opened prospects for furr studies. 6. Conclusions The results study can be summarized as follows: 1. The modelling fault zones should be preceded by initial analysis various physical models tectonically unaffected section, n a selection models that better reflect rock mass condition in investigated area. 2. In case roadway section within a fault zone extent yielding rock in sidewalls is 3.5 times in average greater than in section tectonically unaffected. In instance an investigated fault zone, which intersected roadway at 65, extent rock yielding in ro reached two times throw fault, in floor 3 times throw, and horizontally approx. 1.5 to 1.8 times width modelled fault zone. Noticeably, extent yielded rock mass in fault section is asymmetrical and this asymmetry changes toger with fault throw and its inclination. 3. All numerical models deformations around mining roadway should be verified by in situ measurements in purpose to determine applicability physical model to existing geological and mining conditions and for calibration models. 4. The numerical analysis stability maingate D-2 pointed selection adequate physical models and opened furr prospects for modelling mining roadways in faulted areas. However, cases mining roadways driven within complicated set small throw faults various positions and dips should be studied only with help 3D stware packages.

20 Energies 2017, 10, Author Contributions: Małkowski Piotr conceived and designed experiments; Ostrowski Łukasz and Bachanek Piotr preformed numerical and mining research; Małkowski Piotr and Ostrowski Łukasz analayzed numerical and mining results; Bachanek Piotr analayzed numerical results; Małkowski Piotr wrote paper. Conflicts Interest: The authors declare no conflict interest. References 1. Bukowska, M. The exploitation depth and bump hazard in mines Upper Silesian Coal Basin. In Deep Mining Challenges, International Mining Forum 2009; CRC Press: London, UK, 2009; pp [CrossRef] 2. Gao, J.; Zhao, W.; Lu, Y.; Zhai, J.; Yang, G. Control Mechanism and Countermeasures for Stability Roadway Surrounding Rock in Fault Fracture Zone. China Univ. Min. Technol. 2014, 19, Kidybiński, A. Podstawy Geotechniki Kopalnianej; Wydawnictwo Śl ask: Katowice, Poland, (In Polish) 4. Nieć, M. Geologia Kopalniana; Wydawnictwo Geologiczne: Warszawa, Poland, (In Polish) 5. Shen, H.C.; Cheng, Y.F.; Wang, J.Y. Finite element study on effects faults on ground stress field. Pet. Geol. Oilfield Dev. Daqing 2007, 4, Su, Y.; Zhang, M.; Zhang, Z. The Influence Large Fault on Tunnels in Underground Mines. Electron. J. Geotech. Eng. 2017, 22, Wang, H.; Jiang, Y.; Xue, S.; Mao, L.; Lin, Z.; Deng, D.; Zhang, D. Influence fault slip on mining-induced pressure and optimization roadway support design in fault-influenced zone. J. Rock Mech. Geotech. Eng. 2016, 8, [CrossRef] 8. Yan, S.; Bai, J.; Li, W.; Chen, J.; Li, L. Deformation mechanism and stability control roadway along a fault subjected to mining. Int. J. Min. Sci. Technol. 2012, 22, [CrossRef] 9. Yao, Q.; Li, X.; Pan, F.; Wang, T.; Wang, G. Deformation and Failure Mechanism Roadway Sensitive to Stress Disturbance and Its Zonal Support Technology. Shock Vib. 2016, 2016, [CrossRef] 10. Małkowski, P.; Ostrowski, Ł.; Bachanek, P. The impact low throw fault on stability roadways in a hard coal mine. Stud. Geotech. Mech. 2017, 39, [CrossRef] 11. Xi, X.; Guo, Q.; Liu, T.; Yan, Z. Comprehensive Monitoring and Stability Assessment Roadway with Water Gushing in a Fault Zone. Electron. J. Geotech. Eng. 2015, 20, Brodny, J. Analysis operation new construction frictional joint with resistance wedge. Arch. Min. Sci. 2012, 57, Central Mining Institute. Simplified Principles Standing Support Designing for Headings Driven in Hard Coal Mines; Series Instructions No. 15; Central Mining Institute: Katowice, Poland, (In Polish) 14. Silesian Technical University. The Principles Standing Support Designing for Headings Driven in Hard Coal Mines, 2nd ed.; Corrected; Institute for Exploitation Deposits, Silesian TU: Gliwice, Poland, (In Polish) 15. Borynia,-Ziówka-Jastrzębie Hard Coal Mine. Technical Documentation Maingate D-2, Unpublished work (In Polish) 16. Małkowski, P.; Niedbalski, Z.; Majcherczyk, T. Investigations Hard Coal Mine Roadways Stability in Stratified Rock. In Proceedings 8th International Symposium on Ground Support in Mining and Underground Construction: Ground Support 2016, Lulea, Sweden, September Yu, W.; Wang, W.; Chen, X.; Du, S. Field investigations high stress st surrounding rocks and deformation control. Int. J. Rock Mech. Min. Sci. 2015, 7, [CrossRef] 18. Prusek, S. Empirical-statistical model gateroads deformation. Arch. Min. Sci. 2010, 55, Małkowski, P. The impact physical model selection and rock mass stratification on results numerical calculations state rock mass deformation around roadways. Tunn. Undergr. Space Technol. 2015, 50, [CrossRef] 20. Hoek, E.; Diederichs, M.S. Empirical estimation rock mass modulus. Int. J. Rock Mech. Min. Sci. 2006, 43, [CrossRef] 21. Bieniawski, Z.T. Determining rock mass deformability: Experience from cases histories. Int. J. Rock Mech. Min. Sci. 1978, 15, [CrossRef]

21 Energies 2017, 10, Dinc, O.S.; Sonmez, H.; Tunusluoglu, C.; Kasapoglu, K.E. A new general empirical approach for prediction rock mass strength st to hard rock masses. Int. J. Rock Mech. Min. Sci. 2011, 48, [CrossRef] 23. Małkowski, P. Behavior joint in sandstones during shear test. Acta Geodyn. Geomater. 2015, 12, Pilecki, Z. Modelowanie numeryczne pola naprężenia w górotworze naruszonym wielopokładowa eksploatacja węgla kamiennego w warunkach silnego zagrożenia sejsmicznego. Zeszyty Naukowe Instytutu Gospodarki Surowcami Mineralnymi i Energią PAN 2011, 80, (In Polish) 25. Prusek, S.; Walentek, A. Ocena zmian zachodzacych w górotworze w bezpośrednim otoczeniu ściany. Wiadomości Górnicze 2015, 12, (In Polish) 2017 by authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under terms and conditions Creative Commons Attribution (CC BY) license (