Cheng Zhu,Yong Yuan*,Wenmiao Wang,Zhongshun Chen,Shengzhi Wang,Huiwei Zhong

Key Laboratory of Deep Coal Resources Mining,Ministry of Education of China,China University of Mining and Technology,Xuzhou 221116,China

School of Mines,State Key Laboratory of Coal Resources & Safe Mining,China University of Mining and Technology,Xuzhou 221116,China

Keywords:Deep mining Large-section chamber‘‘Three shells”cooperative support Reasonable layout Surrounding rock control

ABSTRACT The‘‘three shells”cooperative support technology was proposed herein according to both the large deformation of the rock surrounding large-section chambers in deep mines and the precarious stability of the support structures therein.The development range of the plastic zone in the surrounding rock was controlled by a stress shell to reduce the difficulty of controlling the surrounding rock.Additionally,the residual strength of the rock mass in the plastic zone and the self-bearing capacity of the surrounding rock were improved by a reinforced load-bearing shell.Furthermore,a passive load-bearing shell could restore the triaxial stress state of the surrounding rock on the free surface,reduce the influence of the external environment on the surrounding rock,and reinforce the surrounding rock with the strength of the shell.Reasonable layouts of large-section chambers were determined by analyzing the control effect of the stress shell on the surrounding rock under three kinds of in situ stress fields.The orthogonal test method was applied to reveal the influences of different support parameters in the reinforced loadbearing shell and passive load-bearing shell on the surrounding rock stability.The surrounding rock control effect of the ‘‘three shells”collaborative support technology was analyzed through numerical simulation and field monitoring.The results show that the maximum displacement between the roof and floor of the coal preparation chamber in the Xinjulong coal mine was approximately 48 mm,and the maximum displacement between its two sides was approximately 65 mm,indicating that the technology proposed herein could meet the long-term control requirements of the surrounding rock stability for large-section chambers in deep mines.

1.Introduction

In recent years,underground mining,lifting,power supply,drainage,and other equipment have gradually evolved in the direction of large-scale use,intensive use,and intelligent technology.Accordingly,the section size of the underground chamber has increased,the axial length has become significantly larger than the section size,the layout has been centralized,etc.[1].At present,no unified standard for the division of roadway and chamber sections exists in the field of coal mining.Generally,the roadways and chambers are divided into small sections (3–10 m2),medium sections (10–50 m2),large sections (50–100 m2),and superlarge sections(≥100 m2)according to the section division recommendations of the International Tunneling Association.However,because of the joint influence of high ground stress,the complex tectonic stress field,and the rheological properties of deep rock masses[2],a large-section chamber in a deep mine faces the problems of strong deformation of the surrounding rock and the precarious stability of the support structures,which severely restrict safe and efficient mine production [3,4].

Aiming at the difficult problem of controlling the surrounding rock masses of large-section roadways and chambers in deep mines,many scholars have proposed a series of surrounding rock control countermeasures and support technologies [5].Hou [6]suggested that improving the stress state and mechanical properties of the surrounding rock,selecting a reasonable support form,increasing the support resistance,and optimizing the section shape and size are effective methods to realize the surrounding rock control of deep roadways.Kang et al.[7] put forward the concept of the ‘‘three-in-one”surrounding rock control technology,which combines bolt support,grouting modification,and pressure relief technology according to the strong rheology and large deformation of deep rock masses.Yuan et al.[8]established a system for rating the surrounding rock and proposed a theory of surrounding rock control in deep mines that considers the recovery and improvement of the stress state,strength improvement of the surrounding rock,fracture consolidation and damage repair,and stress transfer and expansion of load-bearing zones.Xie et al.[9]and Chang and Xie[10]divided the deformation of the surrounding rock into three stages;they suggested that strengthening during the relatively stable stage of deformation is conducive to maintaining the stability of the surrounding rock in deep mines.He and Zhang [11] analyzed the evolution of the deviatoric stress field in the surrounding rock of a deep roadway and discussed a method for controlling the stability of the surrounding rock under the influence of high horizontal stress.Wang et al.[12]postulated that the support concept should be changed from deformation control to stability control according to the characteristics of the ‘‘given deformation”of the surrounding rock in deep mines.Ju et al.[13]analyzed the influence of the excavation sequence on the stability of the rock surrounding large-section chambers in deep mines and proposed a full-section active support technology through a combination of bolting,anchor cables,lining,and reinforcing the chamber floor by inverted arch pouring.

Moreover,because the axial length of a large-section chamber is generally large,the influence of horizontal stress should be considered when arranging such a chamber [14].To date,many studies have been performed on the influence of horizontal stress on the surrounding rock stability of roadways;among the hypotheses from these studies,the maximum horizontal stress theory proposed by Gale and Blackwood [15] is the most widely used.Based on the maximum horizontal principal stress(MHPS) as the first main stress,this theory notes that (1) the surrounding rock is least affected by the principal stress when the axial direction of a roadway is parallel to the MHPS,which is most conducive to the maintenance of the roadway;(2) the surrounding rock is most affected by the principal stress when the axial direction of the roadway is perpendicular to the MHPS,which is most unfavorable for the stability of the roof and floor;and (3) the deformation and failure of the roof and floor are concentrated on one side of the roadway when the axial direction of the roadway intersects the MHPS obliquely.In addition,according to the theories of normal stress equilibrium and maximum shear stress,some scholars have proposed that the best layout of a roadway is not limited to the direction parallel to the MHPS and that the best layouts of roadways differ among various types of in situ stress fields [16].Other scholars proposed that (1) in a self-weight stress field,the stability of the surrounding rock is maximized when the axial direction of the roadway is parallel or perpendicular to the MHPS,and that (2)in a structural stress field,if the two-way lateral pressure coefficients are equal,the stability of the surrounding rock is maximized when the angle between the axial direction of the roadway and the MHPS is 0°,45°,or 90°.Hence,the roadway should be arranged according to the maximum horizontal stress theory when the two-way lateral pressure coefficients are not equal [17].

According to the above analysis,the current research on controlling the rock surrounding large-section roadways and chambers in deep mines is focused mainly on improving the stress state,strengthening the surrounding rock,providing coordinated deformation control,and establishing full-face collaborative support.Additionally,no conclusive agreement has been reached regarding the reasonable layouts of the roadways and chambers under different types of in situ stress fields.Nevertheless,a reasonable layout of a large-section chamber according to the MHPS can effectively decrease the plastic failure range of the surrounding rock.On this basis,establishing effective support according to the specific conditions of the surrounding rock is a universal method to realize the long-term stability control of the rock surrounding a large-section chamber in a deep mine.Therefore,it is necessary to study surrounding rock control technologies based on the reasonable layout of large-section chambers in deep mines.

In this paper,the‘‘three shells”cooperative support technology is first proposed based on the mechanisms for controlling the surrounding rock of the stress shell,reinforced load-bearing shell,and passive load-bearing shell.Subsequently,reasonable layouts of large-section chambers under different types of in situ stress fields are studied.Finally,taking the underground coal preparation chamber(CPC)in the Xinjulong coal mine as the engineering background,the influences of different support parameters on the control effect of the surrounding rock are revealed.A support scheme for the CPC is designed,and the effectiveness of the ‘‘three shells”cooperative support technology is verified by numerical simulation and field monitoring.

2.Engineering background

The Xinjulong coal mine,which is located in Heze city,Shandong province (Fig.1),has an approved production capacity of 6 Mt/a.The elevations of the first and second mining levels are -810 and -980 m,respectively,both of which are considered deep mining[18].The dense medium method is applied for underground coal preparation.The CPCs are planned to be arranged on the west side of the main roadway at a depth of -810 m.

The preparation chambers can be divided according to their uses and functions into a chamber for washing raw coal,a chamber for screening and crushing,a chamber for transferring products(CTP),a CPC,and a chamber for the clarification of slime water.The design section shape of each chamber is a three-centered arch.Ensuring the stability of the rock surrounding each chamber constitutes the premise for realizing high-efficiency coal preparation.The difficulty of controlling the surrounding rock is known to multiply with increasing section size.The net width and height of the CPC are 7.5 and 9 m,respectively,and the section size of the CPC is the largest among the chambers.Therefore,this paper takes the CPC as the research object to discuss the method for controlling the stability of the rock surrounding large-section chambers in deep mines.

Fig.1.Geographic location and mining map of the Xinjulong coal mine.

3.‘‘Three shells’’ cooperative support technology

3.1.Stress shell

Xie et al.[19] demonstrated that a macroscopic stress shell composed of high-stress bundles exists within the maximum principal stress field of the rock surrounding the roadway.The stress state of the rock mass outside this stress shell gradually changes to match that of the primary rock.The roadway and the rock mass inside the stress shell are in the stress relaxation area where the surrounding rock undergoes abrupt displacement and failure.Therefore,the stress distribution and deformation failure range in the rock surrounding the roadway are controlled by the shape and evolution of the stress shell.The large-section chambers are reasonably arranged according to the specific type of in situ stress field,in which the surrounding rock stability can be controlled to the maximum extent by the stress shell.However,the surrounding rock undergoes continuous deformation and failure when the selected support method and/or support parameters are unreasonable[20,21].As a result,the stress shell gradually moves to the far field,and its control effect on the surrounding rock weakens.Therefore,reasonable support measures should be taken to strengthen the surrounding rock to constrain its continuous deformation.Furthermore,such measures ensure that the stress shell is at a steady state,which always limits the deformation and failure range of the surrounding rock.

3.2.Reinforced load-bearing shell

The bolting support can form a compression arch in the surrounding rock by exerting a radial binding force,thereby improving the stress state of the surrounding rock and giving full play to the self-bearing capacity of the surrounding rock [13,22].The bolt-mesh-anchor support technology can be used to construct a stepped reinforced load-bearing shell in the surrounding rock composed of three parts i.e.the anchor load-bearing ring in the fracture zone,the anchor load-bearing ring in the plastic zone,and the anchor load-bearing ring in the elastic zone [23,24].If the rock mass in the fracture zone and that in the plastic zone are severely damaged and cannot be effectively reinforced by bolting support,grouting technology can be applied to improve the mechanical properties of the rock mass and improve the strength of the broken rock mass [25].

3.3.Passive load-bearing shell

The support structure outside the surrounding rock constructed by the technologies of shotcreting,arching,pipe sheds,U-steel supports,and concreting is called the passive load-bearing shell.The passive load-bearing shell depends on its own strength to reinforce the surrounding rock when the surrounding rock comes in close contact with the shell after deformation and failure[5].At present,the most widely used support technologies are shotcreting and Usteel supports.The shotcreting can support,fill,isolate,and transform the surrounding rock,restore the three-dimensional stress state of the surrounding rock,and form a closed protective layer.U-steel supports typically exhibit high resistance and contractibility,which are suitable in the context of surrounding rock with poor stability.

3.4.Principle and technological process of ‘‘three shells” cooperative support

By combining the surrounding rock control characteristics of the stress shell,the reinforced load-bearing shell,and the passive load-bearing shell,the ‘‘three shells”cooperative support technology is proposed.Its mechanical model is shown in Fig.2.The principle of the ‘‘three shells”cooperative support technology is as follows.

(1) The reasonable layout of a large-section chamber according to the specific type of in situ stress field can ensure that the stress shell is as close as possible to the large-section chamber,thereby decreasing the development range of the plastic zone (DRPZ) in the surrounding rock and reducing the difficulty of controlling the surrounding rock.

(2) The reinforced load-bearing shell and passive load-bearing shell can improve the stress condition of the surrounding rock,improve the self-bearing capacity of the surrounding rock,and realize the coordinated deformation and common bearing of the support structures and surrounding rock.The use of these two shells can prevent the stress shell from continuously moving to the far field due to the rheology of the deep rock mass.

(3) The long-term stability of the rock surrounding large-section chambers in deep mines is realized based on the collaborative control of the ‘‘three shells”within the surrounding rock.The key to this technology is to reasonably arrange the large-section chamber according to the type of in situ stress field and to support the surrounding rock effectively soon after the excavation of the large-section chamber.

The general technological process of the‘‘three shells”cooperative support technology is (1) the type of in situ stress field in the layout area of the large-section chamber is determined;(2)the reasonable layout of the large-section chamber is determined;(3)the large-section chamber is excavated;(4)the free surface of the surrounding rock is isolated by the first shotcreting process,and the thickness of the spraying layer is generally in the range of 40 to 60 mm;(5) the steel mesh is installed;(6) the U-steel supports are deployed;(7) rockbolts and anchor cables are installed;and(8) concrete is sprayed again to reach the design thickness.

4.Rational layouts of large-section chambers under different stress fields

4.1.Numerical simulation schemes

Based on the relationship between the MHPS σH,the minimum horizontal principal stress σh,and the vertical principal stress σv,the in situ stress field can be divided into three basic types,namely,σH-type (σH≥σh>σv),σHv-type (σH>σv>σh),and σv-type (σv>σH≥σh) [16,26].The in situ stress fields of shallow mines are mainly σH-type,and those of kilometer-deep mines are mainly σv-type.The in situ stress field is mainly σHv-type when the burial depth of the mine is intermediate.However,the specific type of in situ stress field where a mine is located still needs to be determined through field tests.

The reasonable layouts of large-section chambers under different types of in situ stress fields are studied by numerical simulation.The three types of in situ stress fields are divided into 6,9,and 6 groups of substress fields,as shown in Tables 1–3,respectively,by selecting different lateral pressure coefficients λH(σH/σv) and λh(σh/σv).The angle α between the axial direction of the large-section chamber and the MHPS takes values of 0°,15°,30°,45°,60°,75°,and 90° in each substress field.In addition,the optimal layout angle α0proposed in the literature[16]is considered in the σHvstress field.The value of α0is determined by Eq.(1).In total,151 numerical simulation schemes are designed.

Table 1 Classification of the substress field in the σH-type stress field.

Table 2 Classification of the substress field in the σHv-type stress field.

Table 3 Classification of the substress field in the σv-type stress field.

Fig.2.Mechanical model of the‘‘three shells”cooperative support technology(Note that σ1i is the value of the maximum principal stress at the boundary of plastic zone and σ1imax is the peak value of maximum principal stress).

According to each simulation scheme,numerical models are established by FLAC3Dsoftware (Fig.3).The shallow surrounding rock possesses known rheological properties after the deep largescale chamber is excavated.Therefore,the strain softening constitutive model is applied to the rock mass near the large-section chamber,and the Mohr-Coulomb constitutive model is adopted for the far-field rock mass [27,28].The mechanical parameters of the rock mass in the numerical model are consistent,as follows:the density ρ=2680 kg/m3,the bulk modulusK=5.6 GPa,the shear modulusG=3.2 GPa,the internal friction angle φ=33°,the cohesionC=2.5 MPa,and the tensile strength σt=1.5 MPa.The above parameters and gradient changes of the cohesion and internal friction angle in the strain softening constitutive model are determined by referring to the example in the FLAC3Dsoftware manual.A uniform load of 20 MPa is applied to the top of the model to simulate deep mining conditions with a burial depth of 800 m.The two-way lateral pressure coefficients are determined according to each simulation scheme.Normal displacement constraints are applied to the area and bottom of the model [29,30].The large-section chamber is excavated without support.The excavation is divided into five stages,each of which is 3 m.The numerical calculation automatically stops when the unbalanced force ratio is less than 1 × 10-5after each stage of excavation.

Fig.3.Numerical models established by the FLAC3D software (Note that q is the load exerted by overlying strata).

4.2.Evaluation of the surrounding rock stability

Zhu et al.[31] revealed the spatiotemporal evolution of the stress shell and its quantitative influence on the damage range of the surrounding rock.They indicated that the plastic zone is always surrounded by a stress shell and that the development of the plastic zone is consistent with the migration of this shell.The ratio of the maximum principal stress corresponding to the boundary of the plastic zone to the peak value of the maximum principal stress in the same radial direction (i.e.the boundary stress coefficient η) is 0.96.The value of η can provide a quantitative basis for determining the DRPZ in the surrounding rock.

The surrounding rock stability of the large-section chamber should be evaluated by simultaneously considering both the DRPZ and the maximum displacement [32,33].After the numerical calculation of each simulation scheme,the model information is input into Tecplot software by using the FISH language interface.Eight extraction lines are arranged in the middle of the large-section chamber (Fig.3a).The maximum displacement and the distribution curve of the maximum principal stress can be obtained by these extraction lines.On this basis,the DRPZ can be determined according to the boundary stress coefficient η.To quantitatively evaluate the surrounding rock stability of the chamber,the following terms are defined:P1,P3,P5,andP7are defined as the variation factors of the DRPZ in the roof,right rib,floor,and left rib,respectively.D1,D3,D5,andD7are defined as the variation factors of the maximum displacements of the roof,right rib,floor,and left rib,respectively.The DRPZ and the maximum displacement of the corresponding position are taken as reference values when α is 0°in the σv6stress field.Then,P1is the ratio of the DRPZ in the roof of each simulation scheme to the corresponding reference value.D1is the ratio of the maximum displacement of the roof to the corresponding reference value.The calculation methods ofP3,P5,P7,D3,D5,andD7are analogous.As shown in Eq.(2),the comprehensive evaluation coefficientSof the surrounding rock stability is obtained by multiplying the eight variation factors of each simulation scheme.Note that the smaller the value ofSis,the better the surrounding rock stability is.

4.3.Reasonable layout of the large-section chamber under the σH-type stress field

Fig.4 shows radar maps of the maximum displacement of the rock surrounding the large-section chamber and the DRPZ in the σH-type stress field.As shown in Fig.4,changes in the angle α have little effect on the DRPZ and the maximum displacement when λHis equal to λh(in the substress fields σH1,σH3,and σH6).This result shows that α is not the main factor affecting the deformation and failure of the surrounding rock at this time.The DRPZ and the maximum displacement are in direct proportion to the lateral pressure coefficient if λHis equal to λh.This result shows that the surrounding rock stability decreases with an increasing lateral pressure coefficient in the bidirectional isobaric stress field.The maximum displacement and the DRPZ in the roof and floor are all in direct proportion to α when λHis not equal to λh(in the substress fields σH2,σH4and σH5).When α is 0°,the DRPZ and the maximum displacement are close to those in the bidirectional isobaric stress field with the lateral pressure coefficient λh;when α is 90°,they are close to those in the bidirectional isobaric stress field with the lateral pressure coefficient λH.When λHis not equal to λhand λhremains constant,the DRPZ in the rib decreases in the range of 0° ≤α ≤45°with increasing λH,while the DRPZs in other areas increase.And when λHis not equal to λhand λHremains constant,the DRPZs in the roof and floor do not change significantly with increasing λh,while that in the rib increases significantly and is proportional to α.

Fig.4.Maximum displacement of the surrounding rock and the development range of its internal plastic zone in the σH-type stress field.

According to the previous definition of the comprehensive evaluation coefficientS,theSvalues of the 42 simulation schemes in the σH-type stress field are obtained.The calculation results are shown in Table 4.The followings are worth noting.

(1) When λHis equal to λh,Sis inversely proportional to α in the range of 0°≤α ≤30°,whileSis proportional to α in the range of 30°<α ≤90°.Therefore,the surrounding rock stability of the large-section chamber is maximized when α is 30°.

(2) In addition,λHis the main factor affecting the stability of the surrounding rock,which generally decreases with increasing λH.

(3)Sis proportional to λhwhen λHand α remain constant.

(4)Sis proportional to α when λHis not equal to λh.

These results show that the layout of a large-section chamber should follow the maximum horizontal stress theory under the conditions described herein.

4.4.Reasonable layout of the large-section chamber in the σHv-type stress field

Fig.5 shows the maximum displacement of the rock surrounding the large-section chamber and the DRPZ under the σHv-type stress field.It can be seen that the maximum displacement is proportional to α.In addition,λHis still the main factor affecting the stability of the surrounding rock.The maximum displacement and the DRPZ increase significantly with increasing λH.When λHis constant,the DRPZ decreases gradually with increasing λh.And when λhis constant,the DRPZ is highly discrete in the range of 0° ≤α ≤45° with increasing λH,especially the DRPZ in the rib.However,the influence of λHon the DRPZ decreases in the range of 45° <α ≤90°.

TheSvalues of the simulation schemes under the σHv-type stress field are listed in Table 5.Table 5 shows that the larger λHand the larger the difference between λHand λhare,the greater the averageSvalue is,indicating worsening stability of the rock surrounding the large-section chamber.The value of α0calculated by Eq.(1) is not necessarily the optimal arrangement angle for alarge-section chamber.In each substress field,Scan take a minimum value in the range of 0° ≤α ≤15°.Therefore,a reasonable layout of the large-section chamber is achieved when the angle between the axial direction and the MHPS is 0° to 15° under the σHv-type stress field.

Table 4 Comprehensive evaluation coefficients of the surrounding rock stability in the σH-type stress field.

Table 5 Comprehensive evaluation coefficients of the surrounding rock stability in the σHv-type stress field.

Fig.5.Maximum displacement of the surrounding rock and the development range of its internal plastic zone in the σHv-type stress field.

4.5.Reasonable layout of the large-section chamber under the σv-type stress field

The maximum displacement of the rock surrounding the largesection chamber and the DRPZ under the σv-type stress field are shown in Fig.6.As can be seen in Fig.6,a change in the angle α has little effect on the DRPZ and the maximum displacement when λHis equal to λh.Moreover,the maximum displacements of the roof and floor show a pattern of first decreasing and then increasing with an increasing lateral pressure coefficient in the bidirectional isobaric stress field.The DRPZs in the roof and in the floor are consistent with this interpretation.The DRPZ in the rib exhibits no obvious change with an increasing lateral pressure coefficient.However,the maximum displacement of the rib increases gradually.The maximum displacement is proportional to the angle α if λHis not equal to λh.When α is 0°,the DRPZ and the maximum displacement are close to those in the bidirectional isobaric stress field with the lateral pressure coefficient λh;when α is 90°,they are close to those in the bidirectional isobaric stress field with the lateral pressure coefficient λH.When λHis not equal to λhand λhremains constant,the DRPZ and the maximum displacement increase with increasing λH.In addition,the discreteness of the maximum displacement of the rib is enhanced with a change in the angle α.When λHis not equal to λhand λHremains constant,with increasing λh,the discreteness of the DRPZ and the maximum displacement decrease gradually with a change in the angle α.

Table 6 shows theSvalue of each simulation scheme under the σv-type stress field.The followings are worth noting.

(1) If λHis equal to λh,theSvalue is the minimum when α is equal to 45°,which shows that the stability of the rock surrounding the large-section chamber is maximized.

(2) The larger λHand the larger the difference between λHand λhare,the larger theSvalue with increasing angle α,indicating worsening stability of the surrounding rock.

(3)Sis proportional to α when λHis not equal to λh.

The result shows that the layout of a large-section chamber should follow the maximum horizontal stress theory under the conditions described herein.

4.6.Layout plan for coal preparation chambers

The field test shows that the in situ stress of the Xinjulong coal mine is a σHv-type stress field[31].According to the above analysis,for the optimal layout of CPCs,the angle between the axial direction and the MHPS is between 0° and 15°.The section sizes of the chamber for screening and crushing and the CPC are the largest among the chambers,and they are the main functional chambers to realize underground coal preparation.Therefore,the layout of these two chambers should be given priority.The final layout plan of the CPCs is shown in Fig.7.The chamber for washing raw coal,the chamber for screening and crushing,and the CPC are arranged in parallel with each other.The angle between their axial direction and the MHPS is 13°.The angle between the axial direction of the CTP and the MHPS is 77°,and the angle between the axial direction of the chamber for the clarification of slime water and the MHPS is 87°.

5.Analysis of the surrounding rock control effect with different support parameters

5.1.Establishment of the numerical model

The CPC is designed to be arranged in siltstone with a thickness of 19.87 m in the roof of seam #3.The physical and mechanical parameters of the measured coal measure strata shown in Table 7 were obtained by laboratory testing.Because the strain softening constitutive model is adopted in the rock mass near the CPC in the numerical model,the residual cohesion,residual internal friction angle and other parameters need to be determined in advance.Strain softening was observed in the rock samples during the uniaxial compression tests [34–36].That is,when the compressive stress applied to the rock mass exceeds its peak strength,the strength of the rock mass decreases with increasing strain [37].Therefore,as shown in Fig.8,the relevant softening parameters of each rock stratum can be determined through mechanical tests and numerical simulations of uniaxial compression [38,39].

Table 6 Comprehensive evaluation coefficients of the surrounding rock stability in the σv-type stress field.

A numerical model is established according to the actual geological conditions of the Xinjulong coal mine (Fig.9a).A uniform load of 19.25 MPa is applied on the top of the model to simulate the weight of the overlying strata with a thickness of 770 m.According to the ground stress test results,the two-way lateral pressure coefficients are 1.4 and 0.83 [31].The boundary conditions are the same as those above.First,the CPC is excavated in stages without support.The excavation is divided into five stages,each of which is 14 m.The evolution of the deformation and failure of the surrounding rock is analyzed to provide guidance for the support design of the CPC.

Fig.6.Maximum displacement of the surrounding rock and the development range of its internal plastic zone in the σv-type stress field.

Fig.7.Layout plan of CPCs.

5.2.Damage characteristics of the surrounding rock

The failure characteristics of the surrounding rock during the excavation of the CPC are shown in Fig.9b.As illustrated in Fig.9b,the stress shell gradually moves to the far field,and the peak value of the maximum principal stress in the shell gradually decreases.Based on the consistency between the development of the plastic zone and the migration of the stress shell,the DRPZ is gradually expanded.The DRPZ is the largest in the floor,the second largest in the roof,and the smallest in the two ribs.The plastic failure in the surrounding rock is mainly shear failure,but the application of rockbolts and anchor cables can improve the shear strength of the surrounding rock [25].During the excavation of the CPC,the surrounding rock should be effectively supported early enough to enhance the self-bearing capacity of the rock mass.This process also prevents the stress shell from continuously moving to the far field and reduces the DRPZ.

5.3.Analysis of the key factors in supporting a large-section chamber

The‘‘three shells”coordinated support technology is adopted to control the stability of the rock surrounding the CPC.The reinforced load-bearing shell is constructed by the bolt-mesh-anchor support method,and the passive load-bearing shell is constructed by shotcreting.Because many support parameters are involved,determining which parameter has the greatest influence on the support effect and which parameters are the main factors affecting the stability of the surrounding rock are the primary issues to beconsidered before supporting the CPC.To solve these problems,the orthogonal test method is used to design various simulation schemes with multiple support parameters and multiple value levels.On the premise of greatly reducing the number of tests,the control effect of the surrounding rock with different support parameter combinations can be comprehensively analyzed.Furthermore,the primary and secondary factors affecting the support can be determined by the range method.The support parameters considered include the diameterx1,lengthy1,pretensionz1,and spacing and row spacingy2of the rockbolts,the diameterx2,lengthy3,pretensionz2,and spacing and row spacingy4of the anchor cables,and the compressive strengthu1and thicknessx3of the shotcrete.As shown in Table 8,each support parameter has three value levels.Therefore,an L27(3^13)-type orthogonal table is selected,i.e.27 simulation schemes are designed (see Table 9 for details).

Table 7 Physical and mechanical parameters of the rock mass.

Fig.8.Stress–strain curves and failure modes in the laboratory tests and numerical simulations of the siltstone uniaxial compressive strength tests.

The cable element available in FLAC3Dsoftware is used to simulate the rockbolts and anchor cables,and the geogrid element is used to simulate the steel mesh.A node–node connection is established between cable and geogrid elements to realize the combined support of the bolt-mesh-anchor structure (Fig.10).Shotcreting is simulated by reassigning parameters to blocks.The elastic constitutive model is adopted for the blocks of the spray layer to give full play to its passive load-bearing effect [40].The mechanical parameters of each support material according to the field test and relevant research data are shown in Table 10 [41].

The 27 simulation schemes are numerically simulated.The DRPZ and maximum displacement of the corresponding position are taken as reference values after the CPC in scheme 1 is arranged.According to the above assumptions and Eq.(2),theSvalue of each simulation scheme can be calculated,as shown in Table 11.

According to the range method,the sum of the indicator values isRiwhen the value level of a support parameter isi(1 ≤i≤3).Riis divided by the number of occurrences of this value level in the column,and thenriis obtained [42].The range is the difference between the maximum and the minimum ofr1,r2,andr3in the same column.The greater the range is,the greater the influence of this factor on the support effect is.The range analysis results of the 27 simulation schemes are calculated and listed in Table 12.The range for the thickness of the shotcrete is the largest,which indicates that it has the greatest influence on the stability of the rock surrounding the CPC.The degrees of influence for the other support parameters on the surrounding rock stability decrease in the following order:y2,u1,x2,y1,x1,y4,z1,y3,andz2.

6.Results and discussion

6.1.Optimization of support parameters

The rigidity of shotcrete is proportional to its thickness.Shotcrete cannot adapt to the deformation of the surrounding rock when its rigidity is large.The thickness of shotcrete in actual construction generally ranges from 50 to 150 mm.On this basis,here,the shotcrete thickness is considered to be 150 mm.According to the above analysis,the influence of the strength of the shotcrete on the surrounding rock stability is relatively small.To reduce the economic cost,the strength grade of the shotcrete is determined to be C20.

The influence of the spacing and row spacing of the rockbolts on the surrounding rock stability is second only to that of the shotcrete thickness.Therefore,the authors optimize the spacing and row spacing of the rockbolts and anchor cables.Five support schemes are developed,as shown in Table 13.In the numerical model,the dimensions of the rockbolt are Φ20 × 2500 mm,and the pretension is 110 kN.The dimensions of the anchor cables are Φ15.2 × 8500 mm,and the pretension is 160 kN.The shape of the reinforced load-bearing shell in the rock surrounding the CPC after the application of each support scheme is shown in Fig.11.The followings should be noted.

Table 8 Value ranges of the support parameters.

Table 9 Orthogonal test schemes.

Table 10 Mechanical parameters of the support material.

Table 11 Results of the orthogonal test.

Table 12 Range analysis of the comprehensive evaluation coefficient of surrounding rock stability.

Table 13 Spacing and row spacing of rockbolts and anchor cables in different support schemes.

(1) The compressive stress in the surrounding rock increases gradually with the decreasing spacing and row spacing of the rockbolts.

(2) When schemes 1 and 2 are adopted,a complete reinforced load-bearing shell cannot be formed in the surrounding rock.

(3) When schemes 3,4,and 5 are adopted,the radial binding forces provided by the rockbolts and anchor cables can be superposed upon each other,thus forming a continuously distributed anchor load-bearing zone in the surrounding rock

(4) TheSvalues of the support schemes 1 to 5 are 1.03,0.95,0.93,0.74,and 0.57,respectively.

The results show that the stability of the surrounding rock is significantly improved when the spacing and row spacing of the rockbolts are less than 1 m.However,when schemes 4 and 5 are adopted,practical problems (namely,a complex construction process,high construction cost,and slow excavation speed) are encountered due to the small spacing and row spacing of therockbolts and anchor cables.In addition,the support effect of the surrounding rock can be improved by adjusting the other support parameters.Thus,scheme 3 is applied to the actual construction.

Based on the above analysis,the support scheme for the CPC is finally determined.

(1) Screw steel resin rockbolts with dimensions of Φ22 × 2500 mm are adopted;the spacing and row spacing are 1 m×1 m,the pretention is 130 kN,and the dimensions of the anchor trays are 200 mm × 200 mm × 10 mm.

(2) The dimensions of the anchor cables are Φ22 × 8300 mm,the spacing and row spacing are 2 m × 3 m,the pretention is 180 kN,and the dimensions of the anchor trays are 300 mm × 300 mm × 16 mm.

(3) The steel mesh is composed of 6-mm diameter steel bars with dimensions of 1200/1000 mm × 2000 mm.

(4) The shotcrete thickness is 150 mm,and the strength grade is C20.

6.2.Control of the surrounding rock with different support schemes

The distributions of the maximum principal stress and the DRPZ in the surrounding rock are compared and analyzed when the CPC adopts bolt-mesh-anchor support technology,shotcreting,and the‘‘three shells”cooperative support.The CPC is excavated in stages without support as a reference.Fig.12a illustrates that after supporting the CPC,the two ribs of the stress shell prominently contract.The contraction in the two ribs of the stress shell is the largest when the ‘‘three shells”cooperative support is applied.Only when the ‘‘three shells”cooperative support technology is applied does the top of the stress shell prominently contract.The DRPZ is the largest when adopting the bolt-mesh-anchor support,the second largest when shotcreting is used,and the smallest when applying the ‘‘three shells”cooperative support.

Fig.11.Influence of spacing and row spacing on the distribution of compressive stress in the surrounding rock.

Fig.12.Surrounding rock control of different support schemes.

Extraction lines are arranged in the middle of the CPC to obtain the distribution curves of the displacement and maximum principal stress(Fig.12b and c).It can be observed that the displacement of the surrounding rock is the largest when adopting the boltmesh-anchor support,the second largest when shotcreting is used,and the smallest when applying the‘‘three shells”cooperative support.The changes in displacement are consistent with those in the DRPZ.After supporting the CPC,the peak value of the maximum principal stress in the ribs and floor increases and moves to the free face.The peak value of the maximum principal stress in the roof remains constant when bolt-mesh-anchor supports and shotcreting are applied.However,the peak value of the maximum principal stress moves prominently to the free face when the ‘‘three shells”cooperative support is utilized.After supporting the CPC,the maximum principal stress in the rock mass of the plastic zone is significantly increased,indicating that the strength and integrity of the rock mass are significantly improved and that the self-bearing capacity is enhanced.The control effects of the bolt-mesh-anchor supports and shotcrete on the roof stability are essentially the same.However,the control effects of the bolt-mesh-anchor supports on the stability of the ribs and floor are weaker than those of the shotcrete.

Clearly,the control of the surrounding rock stability is the best when the ‘‘three shells”cooperative support is applied.This approach can give full play to the cooperative load-bearing of active and passive support structures and prevent the stress shell from moving to the far field,thereby controlling the stress shell of the surrounding rock in a steady state.

6.3.Engineering application

The surface displacement of the surrounding rock is monitored by cross measurement.The monitoring data for the deformation of the rock surrounding the CPC and the CTP are shown in Fig.13.The monitoring results indicate that the deformation of the surrounding rock enters a stable period 40–50 days after the CPC is excavated.The convergence of the two ribs is approximately 65 mm,and that of the roof and floor is approximately 48 mm.The overall control effect of the surrounding rock is good,meeting the normal use requirements for the CPC.The support scheme and parameters of the CTP are consistent with those of the CPC,and the section size of the former is relatively small.However,the final convergence of the surrounding rock and the time of entering the stable period for the CTP are significantly higher than those for the CPC.This contrast shows that the limiting effect of the stress shell on the deformation and failure of the surrounding rock is not fully exerted when the large-section chamber is not reasonably arranged.In this case,ideal control of the surrounding rock cannot be achieved even if the surrounding rock is supported with high strength.

The rock integrity in the ribs and roof of the CPC in the range of 0 to 10 m can be obtained by using the borehole electronic peep technique.The rock in the roof is relatively intact when the distance from the free surface is more than 4 m,and the rock in the ribs is relatively intact when the distance from the free surface is more than 5 m.The results of the borehole electronic peep and numerical simulation are essentially consistent.These data show that the proposed ‘‘three shells”cooperative support technology can control the stability of the rock surrounding large-section chambers in deep mines.

7.Conclusions

This paper introduces the principle,key points,and technological process of the proposed ‘‘three shells”cooperative support technology.Reasonable layouts of large-section chambers under three kinds of in situ stress fields are determined according to FLAC3Dnumerical simulations.The key factors influencing the support effect of the surrounding rock are revealed.A support scheme for the CPC is designed,and the effectiveness of the ‘‘three shells”collaborative support technology is further verified through numerical simulation and field monitoring.The main conclusions of this study are as follows.

(1) Under the σH-type stress field,for the optimal layout of a large-section chamber,the angle between its axial direction and the MHPS is 30°when λHis equal to λh,whereas the layout should follow the maximum horizontal stress theory when λHis not equal to λh.Under the σHv-type stress field,for the optimal layout of a large-section chamber,the angle between its axial direction and the MHPS ranges from 0° to 15°.Under the σv-type stress field,for the optimal layout of a large-section chamber,the angle between its axial direction and the MHPS is 45° when λHis equal to λh,and the layout should still follow the maximum horizontal stress theory when λHis not equal to λh.

(2) The influences of the support parameters on the surrounding rock stability decrease in the following order:the thickness of the shotcrete,spacing and row spacing of the rockbolts,shotcrete strength,anchor cable diameter,rockbolt length,rockbolt diameter,spacing and row spacing of the anchor cables,rockbolt pretension,anchor cable length,and anchor cable pretension.

(3) The distributions of the maximum principal stress,the DRPZ,and the displacement in the surrounding rock are compared and analyzed when the CPC adopts the bolt-mesh-anchor support method,the shotcreting approach,and the ‘‘three shells”cooperative support.The results show that the‘‘three shells”cooperative support technology can give full play to the cooperative load-bearing of active and passive support structures and realize control of the stress shell in the surrounding rock at a steady state.

Fig.13.Control effect and deformation curve of the rock surrounding the CPC.

(4) The maximum displacement between the roof and floor of the CPC in the Xinjulong coal mine is approximately 48 mm,and the maximum displacement between the two ribs is approximately 65 mm,which tends to be stable in the later stage of monitoring.The results show that the ‘‘three shells”cooperative support technology proposed in this study can meet the long-term control requirements of the surrounding rock stability for large-section chambers in deep mines.

Acknowledgements

This work was supported by the Fundamental Research Funds for the Central Universities (No.2019XKQYMS61).