1 Introduction

With the deep mining of coal resources has entered the deep strata in China (Zhang et al. 2019; Liu et al. 2020; Xie et al. 2019; Cao et al. 2020; Guo et al. 2024), the "three high" complex harsh environment of the surrounding rock of the deep mine roadway has gradually become obvious, and the surrounding rock has significant characteristics such as low strength and weak crushing (Hou. 2017a, b; Hou. 2017a, b). After excavation of the roadway, the surrounding rock is prone to exhibit nonlinear and unsteady rheological characteristics under harsh environment for a long time (Huang et al. 2020b, a). There are significant differences in the occurrence state of deep soft rock roadway, resulting in different causes affecting the instability of surrounding rock (Xia et al. 2023). Traditional support methods are difficult to effectively ensure the long-term stability of surrounding rock, and need to be repaired several times (Zhan et al. 2022), which seriously affects the safe and efficient production of mine. Therefore, it is of great significance to study the creep deformation and failure characteristics of deep mine soft rock roadway and put forward the creep control method and engineering practice of deep mine soft rock roadway.

In the field of deep soft rock roadway surrounding rock creep deformation characteristics and control techniques, numerous studies have been conducted by scholars from both domestic and international backgrounds. In terms of creep testing research, (Yang et al. 1999; Fujii et al. 2011; Maranini. 1999; Zhang et al. 2012; Liu et al. 2016) have carried out uniaxial and triaxial compression creep experiments on siltstone, salt rock, weathered sandstone and soft rock such as coal. (Wang et al. 2018a, b; Wang et al. 2019) simulated the stress environment and stress change path of deep coal bodies using the method of constant axial pressure and step-by-step unloading of the surrounding pressure, obtaining the creep characteristics of coal bodies. (Yan et al. 2023; Ru et al. 2023; Hu et al. 2023) carried out triaxial creep tests on anthracite and weakly cemented soft rock under different confining pressures, respectively. The conditions required for the initial creep, stable creep and accelerated creep stages of the rock mass are analyzed. It is obtained that the time for the rock mass to enter the stable creep stage is related to the initial confining pressure, and the greater the confining pressure on the rock mass, the faster the time to enter the stable stage. Liu et al. (2023) considered the influence of clay mineral composition and content on the rheological properties of soft rocks. Liu et al. (2017) revealed the creep damage mechanism caused by confining pressure from a microscopic perspective. With the increase of confining pressure, joint fracture and joint fracture development, grain boundary fracture, trans-granular fracture and grain boundary fracture co-exist at the fracture fracture of coal and rock mass. In terms of creep numerical simulation research, many scholars use ABAQUS (Yu et al. 2024), FLAC3D (Wang 2022; Gao, et al. 2022), PFC (Hu et al. 2019), UDEC (Wang et al. 2022) and other software to analyze the creep characteristics of rock under different load conditions. Zhang et al. (2015) used PFC discrete element software to describe the creep of rock using the Burgers constitutive model, which effectively captures different stages of rock creep at the micro-particle contact level. Xu et al. (2018) used the built-in Burgers numerical model in FLAC3D to describe the creep characteristics of mudstone and coal under triaxial compression conditions, comparing it with the Burgers theoretical model to demonstrate the feasibility of the creep numerical model under triaxial compression. As for the control techniques for surrounding rock in complex deep roadway conditions, they mainly fall into five categories: surface support, anchoring, grouting modification, unloading, and combined control (Kang et al. 2021; Kang et al. 2022a, b; Kang et al. 2021; Zuo et al. 2023). WEN et al. (2023) established a model of soft rock roadway floor heave and analyzed the control mechanism, proposing a new optimized support method. Wang et al. (2023a, b) proposed a combined support scheme based on high elongation, high constant resistance, and pre-stressed negative Poisson's ratio anchor rods, in conjunction with anchor cables, metal mesh, grouted anchor rods, and shotcrete, and validated the effectiveness of the scheme using FLAC3D numerical simulation. Chen et al. (2021) proposed a combined support control technology of 'anchor cable shed + deep and shallow grouting + floor unloading + local reinforcement' by analyzing the influence of different support structures on roadway stability. Wang et al. (2018a, b) used a finite element model to simulate the support effect of the 'U-shaped steel support + surrounding rock anchoring grouting' combined support technology and verified the reliability of the support technology.

Many experts and scholars have made fruitful achievements in the control of surrounding rock in deep mine roadways. However, few scholars have started from the creep deformation characteristics of soft rock under triaxial stress, using numerical models to verify constitutive models suitable for the creep characteristics of the soft rock roadways, and proposing corresponding control methods. Due to the complexity of the actual environment in roadways, the creep characteristics of surrounding rock in soft rock roadways vary, leading to limitations in various control techniques. Therefore, further research is needed on the creep characteristics and control techniques of surrounding rock in deep mine soft rock roadways. In view of this, this paper takes the 61–71 track uphill in the fourth mining area of a mine in Suzhou, Anhui Province as the research background. By using a combination of field surveys, laboratory experiments, numerical simulations, and industrial-scale tests, the paper analyzes the deformation and failure characteristics of surrounding rock in soft rock roadways, conducts triaxial stress tests on mudstone creep characteristics, and combines FLAC3D numerical simulations to elucidate the creep deformation and failure laws of surrounding rock in soft rock roadways. Based on this, a combined support and control strategy of 'anchor net cable shotcrete + floor and two corners in floor bolt-grouting + deep and shallow hole grouting + secondary reinforcement support' is proposed. The feasibility of the control scheme is verified through numerical simulation and on-site mine pressure monitoring data. The research results provide a certain reference basis for the control technology of surrounding rock creep in deep mine soft rock roadways.

2 Engineering background

2.1 Engineering geological overview

The track leading to the mountain in the fourth mining area of a certain mine in Suzhou, Anhui, is located between the 61coal seam and the 71coal seam. The roadway is buried at a depth of 600 m, with a designed total length of 886 m. The cross-section is characterized by a straight wall with a semi-circular arch shape, with dimensions of 4.2 m × 3.5 m, and a net cross-sectional area of 12.8 m2. The track leading to the mountain is relatively distant from the stop lines of 922 and 3214 working faces. and the nearby mining activities have minimal impact on the stability of the track leading to the mountain. The surrounding rock of the track leading to the mountain is primarily composed of mudstone, characterized by soft and fragmented rock with poor overall integrity, exhibiting the characteristics of soft rock. Under high stress conditions, the surrounding rock of the roadway will exhibit creep deformation characteristics, exerting long-term deformation pressure on the support system of the surrounding rock, exacerbating the degree of roadway deformation and deterioration. This falls under the category of typical high-stress soft rock roadway. The original support plan for the track leading to the mountain involved anchor net spraying combined with U-shaped shed support. The anchor rods used were Φ 20 mm × 2400 mm, with a total of 7 Φ 17.8 mm × 5500 m anchor cables arranged across the entire cross-section. The U-shaped shed utilized the U29 type, with a shed spacing of 0.5 m. The original support section of the roadway is shown in Fig. 1, and the lithological columnar diagram is shown in Fig. 2.

Fig. 1
figure 1

Original support section of the roadway

Full size image
Fig. 2
figure 2

Histogram of rock layers

Full size image

2.2 Characteristics of deformation and damage of the surrounding rock of the roadway

Through the on-site tracking research and engineering feedback on the 61–71 track uphill of the four mining areas, the overall serious deformation of the roadway of the original support program and the failure of the support structure are following four aspects:

Shown in Fig. 3, which sums up the deformation and damage characteristics of the peripheral rock of the roadway as the.

  1. (1)

    Overall large deformation of the surrounding rock of the roadway. The roof of the roadway sank and the floor swelled, and the section shrinkage was large, which seriously affects the production and transport of the mine. The weak parts such as the floor was easy to be damaged first, and the deformation of the sidewall was smaller than the deformation of the roof and floor as a whole. The deformation of the sidewall was smaller than the deformation of the roof and the floor as a whole, the maximum deformation of the sidewall was 1123 mm, and the average deformation was 607 mm, the maximum deformation of the roof and floor was 1650 mm, and the average deformation is 847 mm.

  2. (2)

    Bottom corner damage and gang expansion deformation. The arch of the roadway sinks as a whole, and the gang was strongly extruded towards the roadway, and the bottom corner damage and gang expansion damage were most significant.

  3. (3)

    Serious failure of support structure. With the continuous deformation of the surrounding rock, the spray layer slurry skin fell off, metal net was torn, and the anchor rods were loose or even twisted and pulled off.

  4. (4)

    The deformation of surrounding rock lasts a long time. The roof and floor of the surrounding rock of the roadway were mainly composed of soft rocks such as mudstone and argillous sandstone, and the strength of the surrounding rock was relatively low. Shortly after excavation, the surrounding rock continues to deform under the action of high ground stress, and the deformation rate of the surrounding rock can reach 20 mm/d at the initial stage. The continuous deformation rate of surrounding rock decreased to only 2.0 mm/d in the later period, which led to roadway excavation and repair, and the repair effect was poor for many times.

Fig. 3
figure 3

Deformation and damage of track uphil

Full size image

3 Creep characteristics of mudstone under triaxial stresses

3.1 Characteristics of deformation and damage of the surrounding rock of the roadway

3.1.1 Experimental scheme

In order to obtain the creep characteristics of mudstone under triaxial stress conditions, taking the mudstone of the roof and floor of the track uphill roadway in the fourth mining area as the experimental research object, the TYJ-1500 M rock mechanics test system was used to carry out the mudstone creep test under triaxial stress conditions.

The lithology of the roof and floor of the 61–71 track uphill roadway in the fourth mining area is mainly dominated by mudstone, and the compressive strength of mudstone is 20 MPa. The typical mudstone samples of this roadway are processed and polished into ISRM standard cylindrical samples to satisfy the 'Standard of Engineering Rock Body Test' (GBT 50266-2013), and HC-U81 ultrasonic detector was used to select the rock samples with no obvious damage and deviation of longitudinal wave velocity less than 10%.

In laboratory experiments, different confining pressures are often applied to simulate the deep in-situ stress state (Wang et al. 2023a, b). Therefore, combined with the field engineering background, triaxial creep tests of mudstone with confining pressures of 5 MPa, 10 MPa, 15 MPa and 20 MPa are carried out to simulate roadways with different buried depths. The test is divided into four groups: A, B, C and D, with 3 specimens in each group, numbered 1 ~ 3, a total of 12 specimens. The prepared rock sample is wrapped with a heat shrinkable casing, and the hot air gun shrinks to ensure the sealing effect of the rock sample and prevents the hydraulic oil from infiltrating into the rock sample. After the rock sample is placed in a triaxial pressure chamber, the rock sample is adjusted to the center position. The test equipment is TYJ-1500 M rock mechanics test system, as shown in Fig. 4.

Fig. 4
figure 4

TYJ-1500M rock mechanics test system

Full size image

As shown in Fig. 5, The experimental was loaded according to the conventional triaxial path, the loading path and loading experimental scheme. Creep test to 0.01 MPa/s rate loading peripheral pressure to the set value to remain unchanged, and then use graded loading to apply axial pressure, at this time the rock samples can be viewed as affected by the different size of the difference in the principal stress load, each group of tests the first level of the difference in the principal stress of 10 MPa, the gradient of the difference in the principal stress is set to 5Mpa, the gradient of the loading level is set to 7, and then with a rate of 0.01 MPa/s The rate of loading to the set value, each level of principal stress difference to maintain 10 h, to be rock samples loaded to the set time, the next level of load is applied in turn, when loading to the last level of load does not occur in the accelerated rheological damage, then continue to load the specimen at a rate of 0.01 MPa/s, until the rock samples occur creep damage.

Fig. 5
figure 5

Schematic diagram of the loading path

Full size image

3.1.2 Analysis of experimental results

Under the four initial confining pressures of 5 MPa, 10 MPa, 15 MPa and 20 MPa, the mudstone was subjected to gradient axial compression to simulate triaxial creep test. The test results are shown in Table 1, and the typical test results are shown in Figs. 6 and 7. From Table 1 and Fig. 6, it can be seen that the axial strain is positive, the compressive strain, the lateral strain is negative, and the tensile strain. The specimens in the different principal stress difference loading process all appeared obvious transient strain increment and creep strain increment; axial strain are dominated by transient strain increment, lateral strain are dominated by creep strain increment. When the peripheral pressure of the specimen is 20 MPa, the destructive strength, maximum axial strain and maximum lateral strain of the specimen are 1.81, 1.13 and 1.61 times of the peripheral pressure of 10 MPa, respectively; with the increase of the peripheral pressure, the destructive strength of the specimen, the maximum axial creep strain increment and the maximum lateral creep strain increment are all increasing gradually.

Table 1 Creep test results under true triaxial stresses
Full size table
Fig. 6
figure 6

Triaxial creep test curve of mudstone under different confining pressures

Full size image
Fig. 7
figure 7

Characteristic curve of creep of mudstone under different circumferential pressure

Full size image

Under each type of pressure, a set of smooth creep curves without obvious mutation were selected to analyse the triaxial creep characteristics of mudstone, due to the existence of historical cumulative damage effect of the rock samples in the loading process, in order to analyse the creep characteristics under the influence of different peripheral pressures, therefore, based on the improved Boltzmann's principle of superposition (Wang et al. 2008), the creep curves under the graded loading conditions were converted into the creep curves of the rock samples loaded by the loading conditions of the different values of the difference between the principal stresses, as shown in Fig. 7. The specimen undergoes three stages of transient elastic deformation, decelerating creep, and isochronous creep during each level of loading, and when the last level of loading is applied, the specimen undergoes unsteady state creep.

From Fig. 7a–d, when the peripheral pressure is kept constant, the deformation of the specimen increases with the increase in the difference of principal stresses, and when the last level of loading is applied, the creep strain and instantaneous strain increase sharply, and the specimen undergoes creep damage. When the axial pressure is kept constant, the deformation of the specimen under each level of loading gradually decreases with the increase of the peripheral pressure. The breaking strength at the last level of loading increases and the time of destruction of the specimen grows. When the peripheral pressure is 10 MPa, the specimen at the same time there is shear and cleavage composite damage, this is due to the peripheral pressure on the specimen constraint capacity is small, the specimen along the axial direction of the tensile cracks and damage, and the macroscopic rupture surface is relatively rough; when the peripheral pressure is 20 MPa, the peripheral pressure on the specimen constraint capacity increases, the specimen gradually manifested in the macroscopic shear damage. It can be seen that under high peripheral pressure conditions, the specimen will be transformed from brittle to ductile, resulting in creep, and with the increase of the peripheral pressure, the bearing capacity of the specimen is increased, and the creep damage strength is subsequently increased.

From Fig. 8, it can be seen that when the confining pressure is the same, with the increase of the difference of axial principal stress, the proportion of creep increment to the total axial strain under single-stage load also increases. When the confining pressure is 5 MPa, the maximum value can reach 63.1%; when the principal stress difference is the same, with the increase of confining pressure, the proportion of axial deformation under each level of load gradually decreases. When the load is close to the ultimate strength of the sample, the proportion of deformation increment increases rapidly until the sample is destroyed. As the confining pressure increases, the principal stress difference required for the failure of the specimen increases, and the time for the creep failure of the specimen increases.

Fig. 8
figure 8

The proportion of creep deformation increment to axial deformation under each load level

Full size image

3.2 Numerical simulation of creep characteristics of roadway

3.2.1 Creep modelling and parameter identification

Based on the engineering geological conditions of the track uphill in the fourth mining area, the roadway has been repaired for many times due to the serious creep deformation, resulting in the roadway size increasing from 4.2 m × 3.5 m to 4.8 m × 4 m. Based on this, we established a FLAC3D numerical model according to the actual occurrence conditions of the roadway. The numerical model has a length × width × height of 90 m × 60 m × 90 m, a straight wall semi-circular arch roadway section, and a width × height of 4.8 m × 4.0 m. The influence of creep on the deformation and failure of surrounding rock in soft rock roadway is further analyzed. Because the working face layout is far away from the track uphill roadway, the numerical model does not consider the influence of coal seam mining on the roadway disturbance. The Burgers model is used to study the influence of the creep characteristics of mudstone on the deformation of the roadway. The Mohr–Coulomb model is used in the first and third layers to restore the lithology of the surrounding rock of the roadway as much as possible (Ding et al. 2021).

The displacement boundary conditions are limited around and at the bottom of the model, cable unit is used for anchor rods, shell unit is used for concrete grouting, gravity load of 22 MPa is applied to the top of the overlying rock layer, and six monitoring points are arranged around the model from the shallow part to the deep part, and the locations of the numerical model and the monitoring points are shown in Fig. 9. The parameters of the numerical model are shown in Table 2. Among them, the four parameters of Maxwell shear modulus, Kelvin shear modulus, Maxwell viscosity and Kelvin viscosity of mudstone are obtained from the constitutive model of Burgers model and the creep curve of mudstone when the confining pressure is 20 MPa by Matlab software. The tensile strength, bulk modulus, cohesion and internal friction angle of sandstone are calculated by laboratory test and the parameters of various sandstones in the numerical model of reference (Chang et al. 2023).

Fig. 9
figure 9

Numerical simulation and location of monitoring points

Full size image
Table 2 Setting of each parameter of the numerical model
Full size table

In order to verify the feasibility of Burgers model, the creep curves of mudstone under different confining pressures in the laboratory were calibrated. A numerical calculation model of Φ 50 × 100 mm is established, and the size is consistent with the cylindrical standard sample. According to the requirements of the conventional triaxial compression rheological test of rock, the vertical displacement is restrained at the bottom of the model, the axial pressure P1 is applied at the top of the model, and the confining pressure P2 is applied at the side of the model. Three displacement monitoring points are arranged at the top of the model to monitor the vertical displacement of the model during the triaxial compression creep process, and the average is taken as the vertical displacement of the sample. Therefore, the axial strain in the triaxial creep numerical model can be obtained. The numerical model is shown in Fig. 9. The values of P1 and P2 are consistent with those in Table 2, and the triaxial creep tests under different confining pressures are simulated. The Burgers model is selected as the creep model in the numerical calculation of the model. The model parameters are selected from the mudstone Burgers model parameters in Table 1.

The test results and numerical simulation results are shown in Fig. 10. It can be seen from Fig. 10 that the numerical calculation results of Burgers model are basically consistent with the experimental results in the creep characteristic curves of mudstone under different confining pressures, which proves the reliability of creep numerical calculation (Fig. 11).

Fig. 10
figure 10

Schematic diagram of the numerical calculation model of triaxial creep

Full size image
Fig. 11
figure 11

Comparison of experiment and numerical simulation

Full size image

Based on the ground stress distribution characteristics of the 61–71 track uphill in the four mining areas, the creep characteristic curves of the selected specimens at a peripheral pressure of 20 MPa were used to calibrate several built-in rheological models in the FLAC3D software. The calibrated results found that the Burgers model can more accurately represent the long-term creep properties of mudstone, in which the fitting effect of the Burgers model under different stress conditions is shown in Fig. 12.

Fig. 12
figure 12

Effect of Burgers' model fitting curve

Full size image

As can be seen from Fig. 12, when the stress level is 10 ~ 35 MPa, the mudstone undergoes steady state creep, and when the stress level is 40 MPa, the mudstone undergoes unsteady state creep, when the peripheral pressure is kept unchanged, with the increase of the difference of the principal stresses, the instantaneous strain and instantaneous strain rate both decrease gradually, the creep strain and creep strain rate both grow gradually, the time of peripheral rock damage grows, the time of failure is shortened remarkably, this is in agreement with the characteristic curves of the mudstone creep under different peripheral pressures of Fig. 7, and it verifies the feasibility of the Burgers' model for simulation of the creep characteristics of the mudstone.

3.2.2 Numerical simulation results

The surrounding rock of soft rock roadway in deep mines redistributes the ground stress after excavation, and is prone to show nonlinear and nonstationary rheological properties under the influence of harsh environment. In order to study the creep damage mechanism of the surrounding rock of soft rock tunnel and further analyse the causes of creep damage, numerical simulation analysis was carried out on the track uphill tunnel under unsupported condition. After 60d of roadway excavation, the distribution characteristics of stress, strain and deformation of the surrounding rock in the unsupported condition of the 61–71 track uphill in the four mining areas are shown in Fig. 13.

Fig. 13
figure 13

Distribution characteristics of stress, strain and deformation of surrounding rock under unsupported conditions in the roadway

Full size image

The redistribution of the stress state of the surrounding rock after the excavation of the soft rock roadway leads to the susceptibility of the roadway gang and the roof and floor to stress concentration. As can be seen from Fig. 13, the horizontal stress concentration at the bottom corner of the bottom plate is greater than that at the gang (Fig. 13b). The deformation amount of the roadway gang is 778 mm (Fig. 13a), and the deformation amount of the roof and floor is 1403 mm (Fig. 13b), the deformation amount of the roadway gang is obviously smaller than the deformation amount of the roof and floor, and the deformation amount of the roadway is basically the same as that of the on-site monitoring results. The horizontal stress concentration area at the two bottom corners and the roof and floor of the roadway is large, in which the maximum tensile strain increment in the horizontal direction at the gang corner is 0.166 and the maximum compressive strain increment at the floor is 0.178 (Fig. 13d). Under the influence of vertical stress, there are stress concentration phenomena in the bottom plate and gang part of the roadway, with the maximum compressive strain increment at the bottom corner being 0.211, the maximum tensile strain increment at the floor being 0.136 (Fig. 13e), and the maximum shear strain increment being 0.398, and the horizontal stress increment and vertical stress increment are obviously smaller than the shear strain increment.

From the numerical simulation results, it can be seen that the tunnel peripheral rock has been in a high-stress environment for a long time, and the excavation leads to stress concentration in the roadway, so that the floor, bottom corner and the gang part of the tensile strain, compressive strain and shear strain, which exacerbates the degree of creep deformation and damage to the roadway, and the support bearing structure is not enough to resist the deformation of the surrounding rock to produce a significant plastic state, resulting in the instability of the roadway peripheral rock damage.

However, the four mining area 61–71 track uphill only on the roadway roof and the sidewall for support, the roadway floor, gang angle and the roof and no local reinforcement measures, the bottom corner of the stress concentration caused by the destruction of the gradual extension of the sidewall, and ultimately the entire section of the roadway tensile strain, shear strain are larger, the peripheral rock deformation and damage from the shallow part of the crushing gradually shifted to the deep part of the shallow anchors are not supported in the stable layer of rock, so that the support structure of the bearing structure of the lack of point of gravity, weakened the support structure of the bearing ability, and further aggravated by the soft rock roadway peripheral rock creep and deformation.

4 Soft rock roadway creep control technology

According to the characteristics and causes of creep damage of soft rock roadway in deep mines, numerical simulation is used to analyse the control effect of different support methods, put forward the method of controlling the surrounding rock of soft rock roadway in deep mines and carry out industrial experiments to monitor the effect of roadway support.

4.1 Basic control ideas and new support program

Based on the reasons of creep damage to the surrounding rock of the soft rock roadway in deep mines, the joint support method of 'anchor net cable shotcrete + floor and two corners in floor bolt-grouting + deep and shallow hole grouting + secondary reinforcement support' is proposed based on the concept of coordinated support in all sections and the combination of rigid and flexible design, and the program is as follows:

Anchor net cable shotcrete means that anchor rods, anchor ropes, anchor nets and spraying slurry are used for joint support. The roof and sidewall adopt Φ 20 × 3000 mm left-handed threaded anchor rods without longitudinal tendons as the basic support for the surrounding rock, and the specifications of the roof anchor ropes are Φ 21.8 × 6300 ~ 10300 mm, with the spacing between rows of 800 × 800 mm, and the surface of the roadway is paved with disc round steel metal mesh, with the width of the mesh being 300 × 1000 mm, and the thickness of the shotcrete is not less than 100 mm.

The anchor cable Φ 20 × 4000 mm grouting bolt with Φ 21.8 × 6300 ~ 10300 mm is used in the 'floor and bottom corner bolting and grouting'. The row spacing is 1600 × 2400 mm, and the grouting pressure is not less than 6 MPa. The weak parts of the floor and bottom corner are strengthened and supported.

Deep and shallow hole grouting that first use Φ 20 × 4000 mm grouting anchors, grouting pressure is not less than 3 MPa, Φ 20 × 6000 mm grouting anchors, grouting pressure is not less than 6 MPa, grouting anchors are arranged in five alternating rows, the spacing between rows of 1600 × 2400 mm, to ensure that all the shallow peripheral rock fissures are filled by the slurry. Then use Φ 21.8 × 6300 ~ 10300 mm grouting anchors to strengthen the grouting support of deep surrounding rock, with the spacing between rows of 1600 × 2400 mm and the grouting pressure of not less than 6 ~ 8 MPa, to improve the filling effect of deep surrounding rock fissures, and to strengthen the bearing capacity of surrounding rock.

Secondary reinforcement support that the roof and sidewall are used disc round steel metal mesh with Φ 20 × 3000 mm left-rotating non-longitudinal bar threaded anchor rods, the spacing between rows of 800 × 800 mm, after the installation of the anchor rods and sprayed concrete, the thickness of the concrete should not be less than 100 mm, to carry out the secondary reinforcing support. The current support scheme of the roadway is shown in Fig. 14.

Fig. 14
figure 14

The current support scheme of the roadway

Full size image

4.2 Analysis of the simulation effect of the new support method

Based on the joint support control countermeasures for the surrounding rock of the soft rock tunnel in deep mines, FLAC3D numerical software was used to simulate the creep control effect of the soft rock tunnel in different support forms within 180d, respectively. The simulation schemes of different support forms are shown in Table 3, in which the support grouting effect is simulated by increasing the parameters such as cohesion and internal friction angle in the grouting area (Huang, et al. 2020b, a). The simplified schematic diagram of numerical simulation of different support forms is shown in Fig. 15, and the support effect is shown in Figs. 16, 17, 18, 19, 20 respectively.

Table 3 Simulation scenarios for different forms of support
Full size table
Fig. 15
figure 15

Simplified schematic diagram of simulation of different support forms

Full size image
Fig. 16
figure 16

Anchor and net shotcrete support

Full size image
Fig. 17
figure 17

Floor and two corners in floor bolt-grouting

Full size image
Fig. 18
figure 18

Deep and shallow hole grouting

Full size image
Fig. 19
figure 19

Secondary reinforcement support

Full size image
Fig. 20
figure 20

Combined support

Full size image

As can be seen from Figs. 16, 17, 18, 19, 20, Anchor and net shotcrete support reduced the deformation of the top, bottom and gang sections of the roadway by 45.25%, 33.50% and 55.78%, respectively, compared with the unsupported condition (Fig. 16a,b), and the deformation of the roadway is significantly reduced, which is due to the fact that the gang part of the anchors is used in conjunction with the roof anchors and anchors, which is directly anchored to the deeper rock body, and improves the range of anchoring with the rock body and avoids the unstable rock layers at shallow depths. Floor and two corners in floor bolt-grouting, compared with the unsupported conditions of the roof, floor, and sidewall deformation were reduced by 46.13%, 37.48%, and 60.15%, respectively (Fig. 17a,b), strengthened support for the foot of the arch, the anchor cable of the foot of the arch increases the anchorage range, bottom plate grouting support improves the mechanical properties of the bottom plate, improves the load-bearing capacity of the surrounding rock, so that the shear deformation of the foot of the arch decreases, and then the deformation of the vault is effectively controlled (Du et al. 2020). Compared with the unsupported condition, the deformation of the top plate, bottom plate, and gang section were reduced by 48.38%, 38.47%, and 62.98%, respectively, with the deep and shallow hole reinjection support (Fig. 18a, b). With the use of deep and shallow hole grouting in the roof, floor and sidewall of the roadway, the slurry gradually fills up from the shallow surrounding rock fissures to the deep surrounding rock fissures with the grouting pressure, and gels with the rock body to form a monolithic structure, which effectively improves the cohesion, angle of internal friction, and other mechanical properties of the surrounding rock, greatly enhances the coupling effect of the supporting body and the surrounding rock, and restricts the deformation of the surrounding rock in the roadway (Hou et al. 2021). The supplementary support with anchors reduced the deformation of the roof, floor and sidewall by 41.25%, 31.30%, and 52.96%, respectively, compared with the unsupported condition (Fig. 19a, b). Compared with the unsupported condition, the deformation of the roof, floor and side of the combined support is reduced by 84.25%, 80.76% and 93.57% respectively (Fig. 20a, b).

From Figs. Figure 16c, 17, 18, 19, 20c, it can be seen that the maximum shear strain increment of roadway surrounding rock under five different support forms is significantly reduced, which is reduced by 45.66%, 47.48%, 49.90%, 41.88% and 88.92%, respectively compared with that of no support condition.

From the numerical simulation results, it can be seen that after the excavation of the soft rock roadway, the initial reinforcement of the roadway peripheral rock is carried out by using the Anchor and net shotcrete support which improves the peripheral rock bearing capacity and reduces the deformation of the roadway peripheral rock; through the grouting support of the roadway, the roadway rock fissures can be glued into an integral whole, which can effectively repair the solidified ruptured peripheral rock, improve the peripheral rock mechanical properties, and control the deformation of sidewall and the roof; by adopting the anchoring support of the roadway bottom plate. On the one hand, it can effectively enhance the anchoring range of the arch foot, further enhance the shear resistance of the base plate, and prevent the base plate from shear slip phenomenon; on the other hand, grouting support for the roadway base plate can effectively improve the strength of the base plate, control the tensile damage of the base plate, and reduce the amount of the bottom bulge; the numerical simulation results of combined support show that the deformation of surrounding rock of roadway is significantly reduced after combined support, which further ensures the long-term stability of roadway.

4.3 Roadway support effectiveness monitoring

In order to further verify the feasibility of the combined support method of 'anchor net cable spray + bottom plate and bottom angle anchor injection + deep and shallow hole compound injection + secondary composite support', the 61–71 track uphill in the fourth mining area is constructed according to the above scheme, and the specific construction process of the partial support structure of the roadway is shown in Fig. 21, in which Fig. 21a is anchor cable construction; Fig. 21b is grouting anchor construction; Fig. 21c is the construction of grouting anchor cable; Fig. 21d shows the shotcrete construction, and two measuring stations are arranged on the track uphill. The 'cross measuring point method' is used to monitor the surface displacement of the roadway for 6 months. In the first 30d, the roadway completes the anchor network cable and the first shotcrete, and strengthens the support for the bottom plate and bottom corner; in 30 ~ 60 d, installs the grouting anchor and implements the deep and shallow hole grouting; in 60 ~ 90 d, installs the grouting anchor cable to grout the deep peripheral rock, and in 90 ~ 120 d, and installs the anchor and the shotcrete again to carry out the second composite support for the roadway.

Fig. 21
figure 21

Construction process of support structure

Full size image

The control effect of the surrounding rock of the roadway on the track is shown in Fig. 22a. After the completion of the roadway construction, the section is complete, the surrounding rock has no obvious deformation, and no phenomena such as bolt fracture and metal mesh tearing are found. It can be seen from Fig. 22c and d that the overall deformation of roadway surrounding rock is small. The average deformation of roof and floor of No.1 station and No.2 station is 111.9 mm, and the average deformation of two sides is 62.5 mm, which is 13.2% and 10.3% of the deformation under the original support scheme respectively. The growth curve of the surface displacement of the roadway in the first 90 days showed a fluctuating trend, because the roadway was during the installation of the support structure, the bearing capacity of the surrounding rock was weak. The curve gradually changed from fluctuation to gentle within 90 ~ 120 days, and the deformation rate of the surrounding rock of the roadway was small. The reason is that the support of the first 90 days has played a better control role in the surrounding rock of the roadway. In the 90d ~ 120d, the curve gradually changes from fluctuation to gentle, and the deformation rate of roadway surrounding rock is small. The reason is that the support of the first 90d has a good control effect on the surrounding rock of the roadway. The deformation of surrounding rock tends to be stable during 120 ~ 180d of roadway maintenance, and the deformation rate of surrounding rock is less than 0.2 mm/d. As shown in Fig. 22b, the grouting reinforcement can fill the fractured and swollen rock body or fissure damaged rock body well, and further enhance the overall stability of surrounding rock. It shows that the joint support control scheme of 'anchor and net shotcrete + floor and two corners in floor bolt-grouting + deep and shallow hole grouting + secondary reinforcement support' has improved the support strength and stiffness of the overall bearing system of the roadway peripheral rock, achieved effective control of the soft rock roadway of the deep shaft, and met the needs of safe production in the mine.

Fig. 22
figure 22

Effect of perimeter rock support on the upper hill of the track

Full size image

5 Conclusion

  1. (1)

    The creep process of the specimen under the three axial stress conditions all showed obvious transient strain increment and creep strain increment, and the axial strain was dominated by transient strain, and the lateral strain was dominated by creep strain. When the peripheral pressure is kept constant, the deformation of the specimen increases with the increase of the difference of the principal stress; when the axial pressure is kept constant, it gradually decreases with the increase of the peripheral pressure, the destructive strength of the last level of loading increases, the destructive time of the specimen grows, the load carrying capacity of the specimen improves, and the creep destructive strength increases with it.

  2. (2)

    It clarifies the main reasons of creep damage of soft rock roadway in deep mines. Under the condition of high-stress and complex geological structure, the stress concentration level at the roof and floor of the roadway and the two bottom corners of the roadway leads to creep damage of the roadway; the low strength of the surrounding rock, weak and fragmented, poor integrity, failure of the supporting structure, irrational supporting method and other factors lead to the low bearing capacity of the surrounding rock, which further exacerbates the creep damage of the roadway.

  3. (3)

    The joint support scheme of 'anchor and net shotcrete + floor and two corners in floor bolt-grouting + deep and shallow hole grouting + secondary reinforcement support' is proposed, and the effect of joint support on the control of tunnel peripheral rock is clarified. Anchor net and cable spray strengthens the strength of peripheral rock and closes the surface of the roadway, the anchor injection of the base plate and the bottom corner improves the deformation resistance of the base plate and the strength of peripheral rock, and the shallow and shallow holes reinjection fill in the shallow peripheral rock fractures to the deeper peripheral rock fissures, which is glued into the whole body of the rock to improve the stability of peripheral rock as a whole, and reinforcement of the support for the weak parts of the tunnel is made up, which is beneficial to the realization of the long-term stability of the peripheral rock in the roadway.

  4. (4)

    The monitoring results of the peripheral rock of the roadway show that the overall deformation of the peripheral rock of the soft rock roadway under the joint support program is small, and the average deformation of the roof and floor and the sidewall is 111.9 mm and 62.5 mm respectively, which is 13.2% and 10.3% of the deformation under the original support program, so the support effect is significantly improved, and it achieves the effective control of the soft rock roadway of the deep mine.