Research on large deformation control technology of highly weathered carbonaceous slate tunnel

: In order to further explore the large deforma - tion control technology of high ground stress soft rock tunnel, this study takes the highly weathered carbonac - eous shale section of Muzhailing tunnel as the research background,thelargedeformationcontrole ﬀ ectofscheme1 “ three - step method ” and scheme 2 “ pilot tunnel expansion method ” are compared and analysed by ﬁ eld test and numerical simulation. The results show that the numerical simulation results are consistent with the ﬁ eld test, and the deformation trend of the tunnel is the same. The maximum deformation of the tunnel occurs at the middle step. From the perspective of tunnel deformation, control e ﬀ ect is reduced by about 10% compared with scheme 1, and the deformation of both schemes does not exceed the reserved deformation ( 400 mm ) . From the perspective of construc - tion e ﬃ ciency, the construction e ﬃ ciency of scheme 2 is 23.07% lower than that of scheme 1. Taking into account the deformation control e ﬀ ect and construction e ﬃ ciency, it is recommended that the three - step method should be adoptedintheconstructionofthissection,andtheresearch results can provide a reference for the construction of the carbonaceous slate section of the tunnel.


Introduction
Disasters caused by tunnel deformations have become increasingly common in recent decades.Typically, once large deformations occur, they can last from weeks to months, and deformations in soft rock tunnels can even last for a year [1].In Japan, the Enayama II tunnel experienced uncontrolled deformation after construction and the tunnel continued to deform for several months after excavation [2].The Muzhailing tunnel has experienced significant deformation since its construction.The cumulative settlement of the tunnel reached 1,700 mm and the cumulative horizontal convergence reached 1,080 mm, seriously affecting the structural safety and efficiency of the tunnel [3].They cause damage to the support structure, invade the section limits and, if not managed properly, may lead to collapse.This results in fatalities, construction delays, and increased costs.Many engineering examples show that large deformations are the key problem in tunnel construction, so further research on large deformations is needed [4,5].
The International Society of Rock Mechanics defines large deformation as compressional deformation, which usually occurs around the excavation face in underground spaces, and emphasises that large deformation in tunnels has a temporal effect [6,7].Jiang et al. also consider large deformation as a type of destructive deformation with time effect in the surrounding rock of tunnels and underground works, but they point out that large deformation is a type of plastic damage different from rock bursting and surrounding rock collapse [8].Some researchers start with the actual tunnel project [9,10].Among them, Meng et al. summarised three factors of large deformation as plastic flow of soft rock, shear sliding of wedges, and bending of thin layer of soft rock, respectively [10].Bian et al. concluded that two causes of large deformation in the Huangjiazhai tunnel were the plastic flow caused by tunnel excavation under high geostress and low rock strength and the hydrated mechanical coupling process between shales and water [11].
The conditions under which large deformations occur have also been studied to some extent by experts.Anagnostou argues that the main factors influencing large deformations are the strength of the rock and the thickness of the overlying soil layer, while other factors have less influence [12].Zhang et al. argue that the main factors for tunnel deformation are joint inclination, burial depth, water absorption, and softening of the surrounding rock [13].
To control the large deformation of the tunnel, many experts have also proposed some control measures.Ortlepp and Stacey studied the performance of tunnel support structures.They showed that under high ground stress and deep burial conditions, the support structure should absorb energy during yielding [14].Gasc-Barbier et al. proposed the energy support theory, which suggests that the support structure can balance the energy released from the surrounding rock and the energy absorbed by the support structure [15].Liu et al. proposed an analytical method for the seismic stability evaluation of tunnels based on the deformation reinforcement theory [16].Zhang et al. proposed the short bench construction method to control the plastic zone of the surrounding rock and the deformation of the surrounding rock [17].Mu et al. found that the stability of curved tunnels was better than that of straight tunnels, and the rock deformation rate showed a strong time effect from large to small [18].Wang et al. developed a mechanical test system for composite arches in large section tunnels and compared the acceptance mechanisms of conventional I-beam composite arches and restrained concrete [19].
Therefore, in order to further explore the large deformation control technology of highly weathered carbonaceous shale tunnel, this study, based on the project of highly weathered carbonaceous shale section of Muzhailing tunnel, conducts a comparative study on the deformation of the tunnel caused by different excavation methods by using two methods of field test and numerical simulation.The research results can accumulate experience for controlling large deformation of tunnel with high ground stress soft rock.

Engineering background
The Muzhailing tunnel on the Lanzhou-Chongqing Railway is approximately 19 km long, which is a super-long tunnel with a maximum buried depth of 715 m.The geological condition of the tunnel is very complex.There are 11 faults in the tunnel body which cross 3 anticlines and 2 synclines.In the tunnel, there are six trackless inclined shafts and two trackless inclined shafts, which are the key control projects of the Lanzhou-Chongqing Railway.The distribution of the tunnel inclined shafts is shown in Figure 1.
The Dazhangou inclined shaft crosses the Dazhangou anticline.The surrounding rocks of this section are mainly weakly weathered thin to medium-thick carbonaceous slate and carbonaceous slate intercalated with sandstone.The joints and fractures are developed and the stability of the surrounding rock is poor.The tunnel is located in the high ground pressure zone.The maximum and minimum horizontal principal stresses measured by the hydraulic fracturing method are 17 and 9 MPa, respectively, and the lateral pressure coefficient is 0.61-2.26.
3 Large deformation disaster and control scheme

Large deformation disaster
During the construction of the tunnel, five inclined shafts are subject to varying degrees of large deformation.Shotcrete spalling occurs in the deformation zone and the steel frame is twisted (Figure 2).Most of the

Control scheme
In the construction process of soft rock tunnel, the selection of excavation method is very important to control the large deformation of soft rock tunnel.The selection of excavation method is the premise of controlling the large deformation of soft rock tunnel.Improper selection of excavation method can easily lead to large deformation of tunnel and even cause accidents.According to the geological characteristics of the tunnel high stress carbonaceous slate, combined with the tunnel buried depth, span, equipment configuration, and other characteristics, the tunnel excavation method selection scheme is determined.
Because the full section method is not suitable for the class IV-V soft rock stratum, it is not conducive to the deformation control of the surrounding rock and the tunnel during the construction of the tunnel in high geostress soft rock.If the two-step method is used, the tunnel benches are high and the requirements for the working procedure are high, which further increases the difficulty of the arch advance support, and at the same time it is difficult to achieve selfstability during the construction, which may lead to tunnel collapse.Therefore, the three-step method and the pilot tunnel expansion method are selected in this trial to compare and select the better deformation control (Table 1).
The section DYK188 + 045 ∼ DYK188 + 075 of the right line of the main tunnel is selected as the test section of the three-step method.The length of each bench of the three-step method is 4 m.The inverted arch is constructed after the lower bench is 15 m.The heights of the upper, middle, and lower benches are 3.2, 3, and 4 m, respectively.The vertical section is shown in Figure 3.The section DYK187 + 996 ∼ DYK188 + 034 of the right line of the main tunnel is selected as the test section for the extended excavation method.The overall construction sequence of the pilot tunnel expansion method is as follows: first, the small pilot tunnel is excavated and supported, then the support is removed; second, the main tunnel is expanded and excavated; finally, the main tunnel is supported.The pilot tunnel is 32 m long, 4.5 m wide, and 4.5 m high.The excavation method of the pilot tunnel is the full section method and the threestep method is adopted for the expansion of the main tunnel.The bench length and height of the pilot tunnel expansion method are the same as the three-step method, and the section diagram is shown in Figure 4.
When the tunnel is constructed in the highly weathered carbonaceous slate section, it is necessary to determine appropriate support parameters.The support parameters for the two excavation methods are given in Table 2.
4 Effect analysis of control scheme   analysis, heat conduction analysis, fatigue analysis and coupling analysis.Especially, in solving non-linear problems, ABAQUS has great advantages.
In the large deformation analysis of tunnels, ABAQUS can simulate the tunnel excavation by killing element, and support measures such as bolts and steel frames can also be realised by beam element.Thus, ABAQUS can be used to study the large deformation of the tunnel.

Model and parameters
This study takes the highly weathered carbonaceous slate section of the Dazhangou inclined shaft of the Muzhailing tunnel as the research background, a calculation model is established by using numerical simulation software.The width of the model is seven times the tunnel width, about 80 m in width, 80 m in height, and 32 m in depth.The bottom of the model is fully constrained, with horizontal and longitudinal constraints around the model.For the calculation, the applied horizontal in situ stress is 17 MPa, the Mohr-Coulomb criterion is used for the soil, the hexahedron element is used for the soil element, the beam element is used for the bolt simulation, and the linear elastic model is used for the lining element.The calculation mode is shown in Figure 5.The pilot tunnel is supported by C25 mesh shotcrete of 16 cm; the main tunnel support parameters are the same as scheme 1 Intensity class in Code for design of concrete structures [20].
According to the geological survey data and field data, the rock mass around the tunnel is Grade IV, the tunnel lining material is C30 concrete, and the steel arch is H175 steel.By consulting literature and specifications, various parameters of the materials can be determined.The material parameters used in the model are given in Table 3.

Monitoring arrangement
The monitoring section of the numerical simulation is set at 8 m longitudinal position of the model, and the threestep method is adopted for excavation in both working conditions.Therefore, monitoring points for horizontal convergence of upper bench, middle bench, and lower bench are set for monitoring sections, and monitoring points for vault settlement are set.The monitoring layout is shown in Figure 6.

Scheme 1: three-step method
The displacement results of the numerical simulation monitoring points in scheme 1 are extracted, and the displacement time history curve of the numerical simulation of the three-step method is drawn, as shown in Figure 7.
It can be seen from Figure 7 that the displacement and deformation rates of each monitoring point in scheme 1 are larger in the early stage of construction, and is obviously smaller after the tunnel lining is arched.After the construction of the inverted arch, the tunnel deformation is basically stable.The numerical simulation results of scheme 1 show that the settlement of the arch crown is about 41.66 mm, the final deformation of the upper bench convergence is 107.95 mm, the final deformation of the middle bench convergence is 136.28 mm, and the final deformation of the lower bench convergence is 101.51 mm.According to the field test results, the maximum deformation occurs at the middle bench, and the total displacement deformation does not exceed the reserved deformation.

Scheme 2: pilot tunnel expansion method
The displacement results of the numerical simulation monitoring points in scheme 2 are extracted, and the displacement time history curve of the numerical simulation of the pilot tunnel expansion method is drawn, as shown in Figure 8.It can be seen from Figure 8 that the displacement and deformation rates of the monitoring points in scheme 2 are slightly lower than those in condition 1 at the early stage of construction, indicating that the pilot tunnel can relieve the stress.After inverting, the convergence deformation of the tunnel is basically controlled and the deformation growth is slow.The numerical simulation results of scheme 2 show that the settlement of the arch crown is about 44.30 mm, the final deformation of the upper bench convergence is 100.80 mm, the final deformation of the middle bench convergence is 130.05 mm, and the final deformation of the lower bench convergence is 97.89 mm.The maximum deformation occurs at the middle bench, and the final deformation of the middle bench is basically the same, and the total deformation displacement is much smaller than the reserved deformation.
Judging from the numerical simulation results, the final deformation of scheme 2 is slightly smaller than that of scheme 1, indicating that the pilot tunnel has the effect of stress relief.In scheme 2, the maximum deformation of upper bench is reduced by 6.6%, the maximum deformation of middle bench is reduced by 4.6%, and the maximum deformation of lower bench is reduced by 3.6% compared with scheme 1.The reduction is smaller and the reason may be that the stress release phase is shorter and the stress is not fully released.

Field test analysis 4.2.1 Monitoring arrangement
Scheme 1 selects section DYK188 + 056 as monitoring section, and scheme 2 selects section DYK188 + 016 as monitoring section, monitoring arrangement is the same as numerical simulation.

Scheme 1: three-step method
The displacement measurement data in the construction process of scheme 1 are extracted, and the displacement time history curve of the three-step method is drawn, as shown in Figure 9.It can be seen from Figure 9 that all the deformations have obvious convergence when the three-step method is applied.The settlement of the arch crown reaches about 60 mm, the final deformation of the upper bench convergence is 114.73 mm, the final deformation of the middle bench convergence is 145.94 mm, and the final deformation of the lower bench convergence is 122.80 mm.The deformation of the middle and lower benches of the section is large, and the deformation of the middle bench is the largest.Overall, the deformation of the tunnel is fast in the initial stage of construction.After the invert excavation, the deformation rate decreases, and the deformation tends to be stable, and the deformation does not exceed the reserved deformation.Therefore, the threestep method can be used as the excavation method for the construction of highly weathered carbonaceous slate tunnel.

Scheme 2: pilot tunnel expansion method
The displacement measurement data of the construction process of working scheme 2 are extracted, and the displacement time history curve of the pilot tunnel expansion method is drawn, as shown in Figure 10.
It can be seen from Figure 10 that when the pilot tunnel expansion method is adopted, all the deformations have obvious convergence phenomenon, among which the settlement of arch crown reaches about 30 mm, the final deformation of upper bench convergence is 51.31 mm, the final deformation of middle bench convergence is 129.74 mm, and the final deformation of lower bench convergence is 124.01 mm.In scheme 2, the maximum deformation of upper bench is reduced by 55.28% and the maximum deformation of middle bench is reduced by 11.10% compared with scheme 1.The deformation of the upper bench in Scheme 2 is significantly less than in Scheme 1 because the pilot tunnel expansion allows earlier stress relief and does not cause large disturbances to the upper bench soil.In the early stage of tunnel construction, the deformation rate of the tunnel decreases compared with scheme 1, which indicates that the stress release effect of the pilot tunnel is better.At about 16 days, the deformation of the tunnel increased sharply, because the inverted arch construction was not applied in time and the support did not form a ring, causing the deformation to increase.After the construction of the inverted arch, the overall deformation of the tunnel tends to be stable, and the convergence deformation does not break through the reserved deformation.Therefore, the pilot tunnel expansion method can be used as the excavation method for the construction of highly weathered carbonaceous slate tunnel.

Comparative analysis
The field test data of the two excavation methods are extracted and compared with the numerical simulation results.The results are given in Table 4.
As can be seen from Table 4, (1) The maximum deformation error between numerical simulation of three-step method and field test is 6.62%.The maximum deformation of numerical simulation of pilot tunnel expansion method is basically the same as that of field test, and the maximum deformation position is middle bench, which shows the accuracy of the numerical simulation.(2) The tunnel is 12.2 m high, so it is impossible to use the double bench method or the full section method.Scheme 1 is three-step method construction, the maximum deformation position is at the middle bench step, which may be caused by large horizontal geostress.The maximum deformation of field test and numerical simulation is smaller than the reserved deformation.The tunnel deformation can be effectively controlled by using the three-step method.At the same time, the construction efficiency is 1.3 m/day and the efficiency is high.(3) Compared with the excavation method of scheme 2, the excavation and support of pilot tunnel are added in scheme 1, which makes the stress release step of surrounding rock increase before excavation.The maximum deformation of field test and numerical simulation is slightly smaller than the maximum deformation of scheme 1, which is much less than the reserved deformation.The maximum deformation is located at the middle bench.The large deformation of the tunnel can also be effectively controlled by using the pilot tunnel expansion method, but the construction efficiency is greatly reduced to 1 m/day.

Conclusion
(1) Due to the large height and width of the highly weathered carbonaceous slate tunnel, the bench height should not be too large during excavation, so the full section and double bench method cannot be used.The horizontal ground stress of this tunnel section is large, so the side wall is the key part of large deformation disaster.The side wall deforms to the inside and the vault subsides downward.(2) From the perspective of controlling the large deformation of the tunnel, the maximum deformation of the field test and numerical simulation of scheme 1 is 145.95 and 136.28 mm, while that of scheme 2 is 129.74 and 130.05 mm, which is not much less than that of condition 1.The deformation of the two schemes is much less than the reserved deformation, so the three-step method and the pilot tunnel expansion method can effectively solve the problem of large deformation of highly weathered carbonaceous slate tunnel.(3) From the perspective of construction efficiency, the construction efficiency of scheme 1 is 1.3 m/day.Compared with condition 1, the construction efficiency of scheme 2 is reduced by 23.07% and the construction efficiency of scheme 1 is much higher than that of scheme 2.
(4) According to the comprehensive control effect and construction efficiency of the large deformation of the tunnel, the three-step method is the best method to control the large deformation of the tunnel.

Figure 1 :
Figure 1: Distribution of inclined shafts in tunnels.
deformation sections are controlled by arching and other measures.The deformation section of the Dazhangou inclined shaft reaches more than 80%, which is the most severe deformation zone in the whole tunnel.The deformation time of the inclined shaft section is long and the deformation range is wide.The maximum deformation is more than 100 cm and the maximum deformation rate is 80 mm/day, most of the deformation sections are controlled by arching.The deformation of the Daping inclined shaft section is characterised by abrupt change and large initial deformation rate.The maximum daily deformation is 1.9 m.The shotcrete layer in the deformation section cracks and falls off, as well as circumferential and longitudinal cracks, the inverted arch heave is also severe, about 1 m.

4. 1
Numerical simulation analysis4.1.1Simulation methodAs the most widely used simulation software, ABAQUS can be used in various fields such as displacement

Figure 3 :
Figure 3: Vertical section of three-step method.

Figure 4 :
Figure 4: Section diagram of pilot tunnel expansion method.

Figure 7 :
Figure 7: Numerical simulation of displacement time history curve by three-step method.

Figure 8 :
Figure 8: Numerical simulation of displacement time history curve of pilot tunnel expansion method.

Figure 9 :
Figure 9: Displacement time history curve of three-step method.

Figure 10 :
Figure 10: Displacement time history curve of pilot tunnel expansion method.

Table 1 :
Scheme grouping Large deformation control techniques for tunnel  3

Table 2 :
Support parameters 1Full ring H175 steel + grid steel frame, steel frame spacing 0.6 m; the system bolt is 4.5 m long in the tunnel arch and 6 m long at the side wall of the tunnel, the bolt spacing is 1.2 m × 1 m; C30 1 early high-strength concrete is sprayed and the thickness is 33 cm; the thickness of the reserved deformation is 40 cm 2

Table 3 :
Material parameters

Table 4 :
Results of comparative analysis