Safe expanding excavation method for reconstruction and expansion of emergency parking strip of double-track tunnel in mountain ridge heavy hill area

Abstract

The invention relates to the technical field of tunnel reconstruction and extension, in particular to a mountain ridge heavy hill area double-track tunnel emergency parking strip reconstruction and extension safe expanding excavation method which comprises the following steps: refined modeling of a surrounding rock stress field and expanding excavation time sequence design, determining of an optimal expanding excavation sequence through three-dimensional geological detection and numerical simulation, and construction of an emergency parking strip. The method comprises the steps of mechanical arm precise dismantling guided by three-dimensional laser scanning, hydraulic-mechanical coupling graded unloading and step length optimization, dynamic unloading control, surrounding rock disturbance reduction, asymmetric supporting structure rapid forming, differential supporting scheme adoption according to stress distribution difference, whole-process digital management and control, and high-precision and high-efficiency construction of the asymmetric supporting structure. According to the mountain ridge heavy hill area double-track tunnel emergency parking strip reconstruction and extension safe expanding excavation method, a seepage field and stress field coupling equation is established, the influence of seepage water seepage on surrounding rock unloading is quantified, a grading unloading stage division standard is put forward, and the expanding excavation process is divided into three stages of initial unloading (the step length is smaller than or equal to 1.0 m), main unloading (the step length is 1.5-2.5 m) and final unloading (the step length is smaller than or equal to 0.5 m).

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel reconstruction and expansion construction, and in particular to a safe excavation method for reconstruction and expansion of an emergency parking strip of a double-track tunnel in a mountainous and hilly area. Background Art

[0002] As my country gradually enters a high-quality economic development trajectory, the demands for highway transportation have correspondingly increased. Earlier tunnels constructed in hilly and mountainous areas are no longer able to meet these demands, necessitating line expansion, redesign, and the addition of emergency lanes and cross tunnels. However, the expansion and renovation of tunnels in hilly and mountainous areas often present challenges, such as the need to construct excavations close to adjacent tunnels, disrupt the linings of adjacent tunnels, and determine the optimal excavation method for each location.

[0003] By reviewing relevant literature, we found that, for example, CN 118223912 A discloses a method for mountain tunnel expansion and reconstruction, which can handle different surrounding rock conditions and carry out safer mountain tunnel expansion and reconstruction. CN102852525A discloses a method for excavating an underground tunnel from a small single-track tunnel section to a large double-track tunnel section, which enhances soil stability during the expansion and construction process and improves construction efficiency. However, these in-situ, single-sided tunnel expansion methods have the following drawbacks and deficiencies: they are only applicable to single-sided expansion construction, and how to expand in different expansion situations is unclear, especially when facing safety risks that arise during the operation of existing tunnels, such as existing emergency parking lanes that do not meet current regulatory requirements and the inconsistent layout of emergency parking lanes in mountain tunnels.

[0004] The traditional tunnel parking strip expansion method has the following defects:

[0005] 1. The excavation sequence relies on experience, which can easily lead to the expansion of a large range of plastic zones in the surrounding rock;

[0006] 2. Insufficient mechanical demolition accuracy (error > 10mm), resulting in over-excavation and deterioration of the support structure stress;

[0007] 3. Before traditional parking strip excavation construction, the excavation area is usually pre-drained, and then the excavation construction is carried out. However, in actual construction, the effect of pre-drainage is unstable, and seepage water seepage occurs from time to time. In the actual construction process, the unloading step length is fixed, and the influence of the seepage field is not specifically considered. It cannot adapt to the dynamic response of the surrounding rock under the action of water-mechanical coupling.

[0008] 4. The symmetrical design of the support structure causes material waste and inhibits construction efficiency.

[0009] Therefore, there is an urgent need to provide a safe excavation method for the reconstruction and expansion of emergency parking strips in double-track tunnels in mountainous and hilly areas that is applicable to various expansion situations and has high construction safety.

[0010] To this end, we propose a safe excavation method for the reconstruction and expansion of emergency parking strips in double-track tunnels in mountainous and hilly areas. Summary of the Invention

[0011] One of the technical problems to be solved by this application is that the defects and shortcomings of the in-situ single-sided tunnel expansion construction method are: it is only applicable to the construction situation of single-sided expansion, and it is not clear how to expand in different expansion situations, especially when facing safety risks arising during the operation of existing tunnels, such as the existing emergency parking strips not meeting the current specifications and the uncoordinated layout of the emergency parking strips in mountain tunnels.

[0012] Excavation sequences for emergency stop lanes in existing mountain tunnels rely on experience, which can easily lead to widespread plastic zone expansion in the surrounding rock. Mechanical demolition accuracy is insufficient, with errors exceeding 10mm, leading to overexcavation and deterioration of support structure stresses. Fixed unloading steps, without specific consideration of the seepage field, fail to adapt to the dynamic response of the surrounding rock under hydraulic-mechanical coupling. The symmetrical design of the support structure results in material waste and inhibits construction efficiency.

[0013] In order to solve the problems existing in the above-mentioned in-situ single-sided expansion construction method of the tunnel, an embodiment of the present application provides a method for constructing an emergency parking strip for a double-track tunnel in a mountainous and hilly area according to the different locations to be expanded. When the existing tunnel main hole lacks an emergency parking strip, the existing main hole is expanded into a new emergency parking strip. When the emergency parking strip of the existing tunnel in the same driving direction does not meet the requirements of the current specifications, the existing emergency parking strip is expanded into a new emergency parking strip by a single-sided expansion method. When the emergency parking strip of the existing tunnel in the opposite driving direction does not meet the requirements of the current specifications, the existing emergency parking strip is expanded into a new emergency parking strip by a one-side expansion and the other-side backfilling method.

[0014] To address the aforementioned issues of relying on experience, inaccurate precision, fixed unloading step lengths, and symmetrical support structure design for the expansion of emergency stop strips in mountain tunnels, the present invention provides a safe excavation method for the expansion and reconstruction of emergency stop strips in double-track tunnels in hilly areas, including the following steps:

[0015] Step S1: Refined modeling of surrounding rock stress field and excavation sequence design, through 3D geological exploration and numerical simulation, to determine the optimal excavation sequence;

[0016] Step S2: The robotic arm guided by 3D laser scanning accurately dismantles the equipment, achieving millimeter-level contour control and reducing over-excavation and damage to the support structure.

[0017] Step S3: Hydro-mechanical coupling staged unloading and step size optimization. Through dynamic unloading control, the disturbance to the surrounding rock is reduced and the construction safety is improved;

[0018] Step S4: Rapid forming of asymmetric support structure. According to the stress distribution difference, a differential support scheme is adopted to improve the construction efficiency;

[0019] Step S5: Whole-process digital control. Realize the real-time monitoring and intelligent optimization of the construction process to improve the construction quality and safety.

[0020] In some embodiments, the refined modeling of the surrounding rock stress field and the excavation sequence design in step S1 include the following steps:

[0021] Step S101: Geological advanced detection. Adopt geological advanced detection - use 3D geological radar and borehole CT to conduct three-dimensional detection on the expansion section at a spacing of 5m to obtain the distribution of rock mass fissures, the occurrence state of groundwater and the data of the original stress field;

[0022] Real-time monitor the deformation of the existing structure through a distributed optical fiber sensing system and establish an initial database of mechanical parameters of the surrounding rock;

[0023] Step S102: Numerical simulation and time sequence optimization. Construct an asymmetric hydro-mechanical coupling model through the ABAQUS / FLAC3D platform and input the following parameters: elastic modulus of rock mass, Poisson's ratio dynamically corrected according to detection data, permeability coefficient tensor considering anisotropic seepage, and residual bearing capacity of the existing support structure;

[0024] Simulate the influence of different excavation sequences such as left-side priority, right-side priority, and alternate construction on the plastic zone of the surrounding rock, screen the time sequence scheme with the least disturbance, and output the priority list of the excavation section.

[0025] In some embodiments, the precise demolition of the robotic arm guided by 3D laser scanning in step S2 includes the following steps:

[0026] S201: High-precision 3D modeling. Adopt high-precision 3D modeling - use Trimble X7 laser scanner with an accuracy of 0.5mm within every 30-meter range to conduct a 360° scan of the existing structure, generate a textured BIM model, and identify the static measurement error of the area to be demolished ≤2mm;

[0027] Automatically label the weak points of the structure, such as cracks and hollow areas, through the point cloud-model comparison algorithm to generate a risk heat map;

[0028] S202: Robotic arm collaborative operation system. Deploy a hydraulic crushing robotic arm and integrate the following functional modules:

[0029] Visual guidance system: Based on binocular cameras and laser trackers for real-time positioning, dynamically compensate for the pose deviation of the robotic arm;

[0030] Adaptive crushing head: The adaptive crushing head is a hydraulic servo-controlled crushing head that adopts multi-stage pressure control technology. The adjustable pressure range is 0 - 50 MPa, and it automatically adjusts the impact energy according to the concrete strength.

[0031] In some embodiments, the specific steps of step S3 for hydro-mechanical coupling and step optimization for unloading are as follows:

[0032] S301: Construction of a multi-field coupling unloading model. Establish a hierarchical unloading equation considering seepage-stress coupling, input the mechanical parameters of the surrounding rock and groundwater seepage data, and simulate the influence of different unloading step lengths on the stress release of the surrounding rock.

[0033] S302: Dynamic unloading control process:

[0034] S3021: Initial unloading stage (construction step length ≤ 1.0 m): Adopt microseismic monitoring + fiber Bragg grating sensing to real-time feedback the expansion rate of the loosening zone of the surrounding rock. If it exceeds 5 mm / h, construction is suspended;

[0035] Simultaneously implement advanced small pipe grouting, namely cement-water glass double-fluid grout, to form a reinforced arch shell. The volume mixing ratio of the cement-water glass double-fluid grout is 1:0.5;

[0036] S3022: Main unloading stage (construction step length 1.5 - 2.5 m): Advance in a cycle of "excavation - support - monitoring". After each cycle ends, update the model parameters, measure the support reaction force through a pressure cell, and invert the stress release rate of the surrounding rock;

[0037] Use the TSP203 system to detect geological changes 20 m ahead;

[0038] S3023: Final unloading stage (construction step length ≤ 0.5 m): Activate the prestressed anchor cable compensation system (tensile force ≥ 200 kN) to control the convergence deformation within 3‰D (D is the tunnel diameter, unit: m).

[0039] In some embodiments, step S4 for rapid forming of the asymmetric support structure further includes the following steps:

[0040] Step S401: Design of differential support parameters. According to the stress field analysis results, adopt an asymmetric support scheme for the left and right sides of the excavation area:

[0041] High stress side: Arrange a double-layer steel mesh with a specification of steel bars with a diameter of 8 mm and a spacing of 150 mm × 150 mm, and arrange a grid steel frame with a specification of HW175 type and a spacing of 0.6 m;

[0042] Low-stress side: A single-layer steel mesh is arranged, with a specification of steel bars with a diameter of 6 mm and a spacing of 200 mm × 200 mm. Steel supports are arranged, with a specification of I20a and a spacing of 1.0 m.

[0043] Step S402: Intelligent shotcrete technology, using a robotic arm equipped with a slump self-adjusting system to achieve

[0044] Automatically adjust the spraying speed (3 - 8 m 3 / h) according to different rock surface roughness;

[0045] Add SiO2 and steel fibers (dosage of 1.5%) to improve the early strength. According to GB / T 50081-2019, the strength after 8 hours is not less than 10 MPa.

[0046] In some embodiments, the whole-process digital control of step S5 further includes the following steps:

[0047] S501: Build a digital twin platform, integrate BIM models, GIS geographic information, and IoT sensor data streams to construct a four-dimensional construction control platform. The function modules include:

[0048] Real-time stress cloud map display (update frequency ≤ 10 s);

[0049] Robotic arm operation trajectory simulation and collision warning;

[0050] Generation of automatic optimization suggestions for unloading step size;

[0051] S502: Risk intelligent warning, set multiple alarm thresholds,

[0052] When the vault settlement rate is greater than 3 mm / day, it is a yellow warning. When the vault settlement rate is greater than 5 mm / day, it is a red warning, and the stress compensation system is activated;

[0053] When the seepage water volume is greater than 5 L / min, it is a yellow warning. When the seepage water volume is greater than 10 L / min, it is a red warning, and the radial grouting robot is triggered to grout the water seepage area to reduce the seepage water volume.

[0054] In some embodiments, the crushing process of the hydraulic crushing robotic arm is as follows:

[0055] S2021: Define the virtual demolition boundary according to the BIM model, and the robotic arm performs millimeter-level contour cutting along the predetermined path;

[0056] S2022: Implement layered crushing of the core area, with a single-layer thickness ≤ 200 mm, and simultaneously start the dust removal system;

[0057] S2023: Real-time scan the demolished area, compare it with the design model, and trigger the error compensation mechanism.

[0058] In some embodiments, the operation process of the error compensation mechanism is as follows:

[0059] S2023.1: Real-time error detection. Use high-precision sensors (such as lidar, vision systems, encoders, etc.) to monitor the position and attitude of the end of the robotic arm in real time;

[0060] Compare the actual position with the target trajectory or target point, and calculate the error value.

[0061] S2023.2: Trigger condition judgment. Set an error threshold (such as 2 millimeters). When the detected error exceeds the threshold, the system automatically triggers the compensation mechanism;

[0062] S2023.3: Calculate the compensation amount. According to the detected error, use control algorithms (such as PID control, model predictive control, etc.) to calculate the angle or position change of the robotic arm joints that need to be adjusted;

[0063] Consider the dynamic characteristics of the robotic arm (such as inertia, friction) for accurate calculation.

[0064] S2023.4: Execute the adjustment. Apply the calculated compensation amount to the control system of the robotic arm to adjust the position and attitude of the robotic arm;

[0065] Ensure that the adjustment process is fast and accurate, and avoid introducing new errors or causing the robotic arm to be unstable;

[0066] S2023.5: Feedback and verification. After the adjustment is completed, detect the position and attitude of the robotic arm again to verify whether the error is within the allowable range;

[0067] If the error still exceeds the threshold, repeat the above steps until the error is within the allowable range;

[0068] S2023.6: Optimization and adaptation. According to historical data and actual operating conditions, optimize the control algorithm and compensation strategy to improve the response speed and accuracy of the system;

[0069] Combine multiple sensor data to enhance the reliability and accuracy of the system.

[0070] In some embodiments, the hierarchical unloading equation for seepage-stress coupling is where σ ij is the stress tensor, k is the permeability coefficient, is the hydraulic gradient, ε pl,kl is the plastic strain increment, and the specific derivation process is as follows:

[0071] According to the effective stress principle formula of Biot's theory, σ ij = σ′ ij + α·p·δij , where σ ij is the total stress, σ′ ij is the effective stress, α is the Biot coefficient, p is the pore water pressure, and δ ij is the unit tensor;

[0072] Taking the derivative of the Biot theory effective stress principle formula with respect to time gives

[0073] Introducing the elastoplastic constitutive relation, assuming under small deformation conditions, the relationship between effective stress and strain is σ′ ij = C ijkl (ε kl - ε pl,kl )(2), where C ijkl is the elastic stiffness tensor, ε kl is the total strain tensor, and ε pl,kl is the plastic strain tensor;

[0074] Combining the pore pressure change rate equation, the pore pressure change rate equation is where k is the permeability, reflecting the ability of the medium to conduct fluids, μ is the dynamic viscosity of the fluid, φ is the porosity, reflecting the proportion of void volume in the medium, S is the storage coefficient, reflecting the change in the fluid stock in a unit volume of rock under a unit pressure change, is the Laplace operator of the pore pressure, reflecting the non-uniformity of the pressure spatial distribution, ε v is the volume strain, and ε v = ε 11 + ε 22 + ε 33 ;

[0075] Combining the momentum conservation equation where ρ is the density of water and g is the acceleration due to gravity, taking the derivative of the momentum conservation equation with respect to time gives

[0076] Substituting (2) and (3) into (1) gives

[0077] Introducing the relationship between strain rate and displacement, the relationship between strain rate and displacement rate is where are the displacement rate components;

[0078] Combining the above equations gives where is the elastic response term, is the plastic dissipation term, is the seepage-strain coupling term, is the volume strain rate, describing the rate of swelling or compression of the surrounding rock;

[0079] In Equation (5), directly correlates the permeability k and directly quantifies the influence of the seepage field on stress changes where t0 is the initial time;

[0080] Define the stress release rate during the unloading process as where δσ is the stress release amount, the unloading step length δL and the stress release rate R σ The relationship is R σ = 1 - e -λ·δL , where λ is the stress release coefficient, taking 0.15, which is related to the stiffness of the tunnel surrounding rock and the initial stress state. The specific derivation process is as follows:

[0081] The influence of the stress release amount δσ corresponding to the unloading step length δL satisfies The equation shows that the stress release rate is proportional to the remaining unreleased stress (1 - R G );

[0082] (5) Deformation gives Solving gives -ln(1 - R σ ) = λ·δL + C. The initial condition is that when δL = 0, R σ = 0, getting C = 0, and getting R σ = 1 - e -λ·δL ;

[0083] The relationship between the unloading step length δL and the pore pressure change rate is as follows:

[0084] The excavation volume change corresponding to the unloading step length δL is δV = A·δL, where A is the excavation face area, and the volumetric strain change rate is where V0 = A·L0 is the volume under the action of the initial in-situ stress, L0 is the length under the action of the initial in-situ stress, δt = T is the single-step unloading time, and we get Substituting (6) into (3) gives where is the unloading driving term, and the unloading driving term directly quantifies the influence of the unloading rate on the pore pressure.

[0085] According to (3), ignoring the spatial diffusion term, that is we get Integrating gives the relationship between the pore pressure and time as

[0086] Considering the steady-state seepage condition, that is the pore pressure change rate equation is simplified to Assume one-dimensional seepage And introducing the boundary condition p(x = 0) = p0, we get the pore pressure gradient Considering the exponential variation of permeability with the unloading step, i.e., k = k0e β·δL , the equation obtained is where k0 is the initial permeability and β is the permeability stress sensitivity coefficient.

[0087] The relationship between the dynamic unloading control process and the unloading step is as follows:

[0088] In the initial unloading stage, according to the Taylor formula expansion and first-order approximation, the stress release rate R σ = 1 - e -λ·δL ≈λ·δL. At this time, the stress release rate approximately linearly increases, and it is necessary to strictly specify the unloading step δL to avoid mutations. The stress release rate at this time satisfies R σ ≤0.2, resulting in δL≤1.3m. Taking 1.0m as the safety threshold, i.e., δL≤1.0m;

[0089] In the main unloading stage, for the stress release rate R σ perform a first-order differential operation to obtain The stress release rate R σ decreases with the increase of the unloading step δL. At this time, the unloading needs to consider efficiency and safety. The optimization model at this time is where V L is the unloading rate, u max is the maximum displacement, is the pressure gradient. Solving jointly with (3) gives 1.5m≤δL≤2.5m;

[0090] In the final unloading stage, at this time R σ [[ID=۳۸]]= 1 - e -λ·δL ≈1. At this time, the unloading step for the release of the residual stress of the tunnel at the excavation area should be smaller than that in the initial unloading stage and the main unloading stage. The pore pressure change rate satisfies The unloading step δL≤0.5m is obtained.

[0091] In some embodiments, the operation process of the prestressed anchor cable compensation system includes the following steps:

[0092] S3023.1: Monitoring stage, install monitoring equipment, and arrange convergency meters, multi-point displacement meters, fiber optic sensors and other equipment in the tunnel to monitor the deformation of the surrounding rock and the support structure in real time;

[0093] Data acquisition, continuously acquire the tunnel convergence deformation data, including parameters such as lateral convergence, longitudinal displacement and surrounding rock pressure;

[0094] S3023.2: Trigger condition judgment, set a threshold value, and set the threshold value of the convergence deformation according to the design specifications (such as 3‰D);

[0095] Real-time comparison: Compare the monitored deformation data with the threshold value. When the deformation exceeds the set value, trigger the activation of the prestressed anchor cable compensation system;

[0096] S3023.3: Tension implementation, Select anchor cables: According to the calculation results, determine the positions and quantities of the prestressed anchor cables to be tensioned;

[0097] Tensioning process: Loading stage, Use tensioning equipment to apply prestress to the selected anchor cables, gradually increasing the tension to the calculated value;

[0098] Locking stage, After reaching the target tension, use locking devices to fix the anchor cables to ensure that the prestress is maintained;

[0099] Control accuracy, During the entire tensioning process, monitor the tension force and the elongation of the anchor cables in real time to ensure that the prestress is applied evenly and accurately;

[0100] S3023.4: Feedback and adjustment, Monitoring feedback, After the tensioning is completed, continue to monitor the convergence deformation of the tunnel and evaluate the effect of the prestress compensation;

[0101] Adjustment and optimization, If the monitoring results show that the deformation still does not reach the expected control target, the system will recalculate and make further tensioning adjustments until the convergence deformation meets the requirements;

[0102] S3023.5: Long-term maintenance, Continuous monitoring, After the compensation is completed, the system continuously monitors the deformation of the tunnel to ensure long-term stability.

[0103] Regular inspection, Regularly check the working status of the prestressed anchor cables, including the tension force, anchoring condition, etc., to prevent deformation out of control caused by anchoring failure;

[0104] S3023.6: Automation and intelligence, Automatic control, The system integrates an automatic control module to realize the full-automatic process of monitoring, calculation, tensioning, and feedback adjustment, reducing human intervention.

[0105] Intelligent algorithms, Adopt machine learning and intelligent algorithms to optimize the prestress application strategy according to historical data, improving the compensation efficiency and accuracy.

[0106] In some embodiments, the operation of the slump adaptive adjustment system includes the following steps:

[0107] S4021: Surrounding rock stress field analysis and working condition prediction, Stress monitoring, Install distributed optical fiber sensors to obtain the surrounding rock stress distribution (σ max ≤5MPa) in real time;

[0108] Unloading grading decision, Divide the unloading areas according to the stress gradient (Type A: high risk, Type B: medium risk, Type C: low risk), and predict the concrete performance requirements for each area;

[0109] S4022: Real-time detection of rock surface roughness and material state, IRI value scanning. A 3D laser scanner is mounted on the robotic arm to scan the rock surface before construction to generate an IRI heat map (accuracy ±0.1 mm / m).

[0110] Material monitoring, slump monitoring. A laser sensor is installed inside the pipeline to measure every 30 seconds (error ±5 mm).

[0111] Fiber dispersion detection, based on image recognition technology, to analyze the uniformity of steel fiber distribution (qualified rate ≥95%).

[0112] S4023: Generation of dynamic adjustment strategy, multi-objective optimization calculation. Input parameters: IRI value, surrounding rock stress level, current slump (S 实测 ), ambient temperature and humidity.

[0113] Output parameters: target slump (S 目标 ), spraying speed (V 喷射 ), accelerator dosage (C 速凝 ), fiber feeding frequency (F 纤维 ).

[0114] Optimization model: min(|S 实测 - S 目标 | + λ·|V 喷射 - V 理论 |), with the constraint that the 8-hour strength ≥10 MPa and the fiber dosage is 1.5% ±0.1%.

[0115] S4024: Issuing adjustment instructions. Slump adjustment: Control fluidity by increasing or decreasing mixing water (±3%) or high-range water reducer (±0.2%).

[0116] Spraying speed matching: For every 1 mm / m increase in the IRI value, the spraying speed is reduced by 1 m3 / h.

[0117] Fiber synchronous control: A variable-frequency vibrator adjusts the feeding rhythm of steel fibers to prevent agglomeration.

[0118] S4025: Precise execution and closed-loop verification. The wet spraying manipulator executes, and the robotic arm sprays according to the planned path, with the speed dynamically adjusted according to the IRI value (PID control accuracy ±0.2 m 3 / h);

[0119] SiO2 and steel fibers are accurately incorporated through a double-screw feeder (error ≤0.05%);

[0120] Real-time feedback and iteration, rapid strength detection: The penetration strength tester randomly checks the 8-hour strength every 30 minutes, and the data is transmitted back to the control system.

[0121] Adaptive correction: If the strength fails to meet the standard, trigger secondary adjustment (increase the accelerating agent by 0.1% or decrease the slump by 10 mm).

[0122] S4026: Stress-deformation collaborative management, post-unloading monitoring: After the staged unloading is completed, monitor the deformation of the concrete layer through strain gauges. If the deformation exceeds the limit (≥0.3 mm / m), start the supplementary spraying procedure.

[0123] Data precipitation: Store the construction parameters, adjustment records, and quality data in the database for iterative optimization of the machine learning model.

[0124] The present invention has at least the following beneficial effects:

[0125] 1. Based on the water-force coupling controlled excavation sequence, break the traditional symmetric construction mode. By establishing the coupling equation of the seepage field and the stress field, quantify the influence of seepage water penetration on the unloading of surrounding rock.

[0126] 2. Adopt a differential unloading step length design, propose the division standard for the staged unloading stage, and divide the excavation process into three stages: initial unloading (step length ≤ 1.0 m), main unloading (step length 1.5 - 2.5 m), and final unloading (step length ≤ 0.5 m).

[0127] 3. Adopt an asymmetric support structure design. According to the stress distribution difference (the stress value on the high-stress side ≥ 1.5 times that on the low-stress side), adopt differential support parameters. Brief Description of the Drawings

[0128] Figure 1 It is the process flow chart of the reconstruction and expansion of the emergency parking bay in the double-track tunnel in mountainous and hilly areas of the present invention;

[0129] Figure 2 It is the process flow chart of Step 1 of the present invention;

[0130] Figure 3 It is the process flow chart of Step 2 of the present invention;

[0131] Figure 4 It is the process flow chart of Step 3 of the present invention;

[0132] Figure 5 It is the process flow chart of Step 4 of the present invention;

[0133] Figure 6 It is the process flow chart of Step 5 of the present invention;

[0134] Figure 7 It is the relationship curve between the stress release rate and the unloading step length;

[0135] Figure 8 It is the relationship curve between the stress change rate and the unloading rate;

[0136] Figure 9 is the curve of pore pressure versus time

[0137] Figure 10 is the curve of pore pressure gradient versus unloading step length;

[0138] Figure 11 is the time history diagram of tunnel crown settlement and invert heave during the excavation of the tunnel parking bay in Example 1 of the present invention;

[0139] Figure 12 is the time history diagram of tunnel horizontal convergence during the excavation of the tunnel parking bay in Example 1 of the present invention;

[0140] Figure 13 is the time history diagram of tunnel crown settlement and invert heave during the excavation of the tunnel parking bay in Example 2 of the present invention;

[0141] Figure 14 is the time history diagram of tunnel horizontal convergence during the excavation of the tunnel parking bay in Example 2 of the present invention. Detailed implementation manners

[0142] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0143] Example 1: Please refer to Figure 1-6 , the present invention provides a technical solution: a safe excavation method for the reconstruction and expansion of an emergency parking bay in a double-track mountain tunnel in a mountainous and hilly area. According to different positions to be expanded, emergency parking bay construction is carried out on the double-track mountain tunnel in the mountainous and hilly area. When there is no emergency parking bay in the main tunnel of the existing tunnel, the existing main tunnel is excavated and expanded into a new emergency parking bay. When the emergency parking bay of the existing tunnel in the same driving direction does not meet the current specification requirements, the existing emergency parking bay is excavated and expanded into a new emergency parking bay by using the unilateral excavation method. When the emergency parking bay of the existing tunnel in the opposite driving direction does not meet the current specification requirements, the existing emergency parking bay is excavated and expanded into a new emergency parking bay by using the method of excavating one side and backfilling the other side.

[0144] It includes the following steps:

[0145] Step S1: Fine modeling of the surrounding rock stress field and design of the excavation sequence. Through three-dimensional geological exploration and numerical simulation, the optimal excavation sequence is determined;

[0146] Step S2: Precision demolition of the robotic arm guided by three-dimensional laser scanning to achieve millimeter-level contour control and reduce over-excavation and damage to the support structure;

[0147] Step S3: Hydro-mechanical coupling staged unloading and step size optimization. Through dynamic unloading control, the disturbance of surrounding rock is reduced and the construction safety is improved.

[0148] Step S4: Rapid forming of asymmetric support structure. According to the stress distribution difference, a differential support scheme is adopted to improve the construction efficiency.

[0149] Step S5: Whole-process digital control. By integrating BIM model, GIS geographic information and IoT sensor data, a digital twin platform is constructed to realize real-time stress cloud map monitoring (update frequency ≤ 10 seconds), robotic arm trajectory simulation and automatic optimization of unloading step size during the construction process, so as to improve the construction quality and safety.

[0150] Furthermore, the refined modeling of the surrounding rock stress field and the excavation sequence design in Step S1 include the following steps:

[0151] Step S101: Geological advanced detection. Geological advanced detection - using 3D geological radar and borehole CT, the expansion section is three-dimensionally detected at a spacing of 5m to obtain data on rock mass fissure distribution, groundwater occurrence state and original stress field.

[0152] The deformation of the existing structure is monitored in real time through a distributed optical fiber sensing system, and an initial surrounding rock mechanical parameter database is established.

[0153] Step S102: Numerical simulation and time sequence optimization. Through the ABAQUS / FLAC3D platform, an asymmetric hydro-mechanical coupling model is constructed and the following parameters are input: the elastic modulus of the rock mass, the Poisson's ratio is dynamically corrected according to the detection data, the permeability coefficient tensor considers anisotropic seepage, and the residual bearing capacity of the existing support structure.

[0154] Simulate the influence of different excavation sequences such as left-side priority, right-side priority, and alternate construction on the plastic zone of the surrounding rock, screen the time sequence scheme with the least disturbance, and output the priority list of the excavation section.

[0155] Furthermore, the precise demolition of the robotic arm guided by 3D laser scanning in Step S2 includes the following steps:

[0156] S201: High-precision 3D modeling. High-precision 3D modeling - using a Trimble X7 laser scanner with an accuracy of 0.5mm within a range of 30m, the existing structure is scanned 360° to generate a textured BIM model, and the static measurement error of the area to be demolished is identified ≤ 2mm.

[0157] Automatically label the weak points of the structure, such as cracks and hollow areas, through the point cloud-model comparison algorithm to generate a risk heat map.

[0158] S202: The robotic arm collaborative operation system deploys a hydraulic crushing robotic arm and integrates the following functional modules:

[0159] Vision guidance system: Based on binocular cameras and laser trackers for real-time positioning, dynamically compensating for the pose deviation of the robotic arm.

[0160] Adaptive crushing head: The adaptive crushing head is a hydraulic servo-controlled crushing head that uses multi-stage pressure control technology. The adjustable pressure range is 0 - 50 MPa, and it automatically adjusts the impact energy according to the concrete strength.

[0161] The crushing process of the hydraulic crushing robotic arm is as follows:

[0162] S2021: Define the virtual demolition boundary according to the BIM model, and the robotic arm performs millimeter-level contour cutting along the predetermined path.

[0163] S2022: Implement layered crushing on the core area, with a single-layer thickness ≤ 200 mm, and simultaneously start the dust removal system.

[0164] S2023: Real-time scan the demolished area, compare it with the design model, and trigger the error compensation mechanism.

[0165] Furthermore, the specific steps of step S3 for water-pressure coupling hierarchical unloading and step optimization are as follows:

[0166] S301: Construct a multi-field coupling unloading model, establish a hierarchical unloading equation considering seepage-stress coupling, input the mechanical parameters of the surrounding rock and groundwater seepage data, and simulate the influence of different unloading step lengths on the stress release of the surrounding rock.

[0167] S302: Dynamic unloading control process:

[0168] S3021: Initial unloading stage (construction step length ≤ 1.0 m): Use microseismic monitoring + fiber Bragg grating sensing to real-time feedback the expansion rate of the loosening zone of the surrounding rock. If it exceeds 5 mm / h, suspend construction;

[0169] Simultaneously implement advanced small pipe grouting, i.e., cement-water glass double-fluid grout, to form a reinforced arch shell.

[0170] S3022: Main unloading stage (construction step length 1.5 - 2.5 m): Advance in a cycle of "excavation - support - monitoring". After each cycle ends, update the model parameters, measure the support reaction force through pressure cells, and invert the stress release rate of the surrounding rock.

[0171] Use the TSP203 system to detect geological changes 20 m ahead.

[0172] S3023: Final unloading stage (construction step length ≤ 0.5 m): Activate the prestressed anchor cable compensation system (tensile force ≥ 200 kN), and control the convergence deformation within 3‰D (D is the tunnel diameter, unit: m).

[0173] Furthermore, step S4 for the rapid forming of the asymmetric support structure further includes the following steps:

[0174] Step S401: Design of differential support parameters. According to the stress field analysis results, adopt an asymmetric support scheme for the left and right sides of the excavation area:

[0175] High stress side: Arrange double-layer steel mesh with a specification of steel bars with a diameter of 8 mm and a spacing of 150 mm×150 mm, and arrange lattice girders with a specification of HW175 type and a spacing of 0.6 m;

[0176] Low stress side: Arrange single-layer steel mesh with a specification of steel bars with a diameter of 6 mm and a spacing of 200 mm×200 mm, and arrange section steel supports with a specification of I20a and a spacing of 1.0 m;

[0177] Step S402: Intelligent shotcrete technology. Adopt a wet shotcrete manipulator equipped with a slump self-adaptive adjustment system to automatically adjust the spraying speed (3 - 8 m 3 / h) according to different rock surface roughnesses;

[0178] Add SiO2 and steel fibers (dosage is 1.5%) to improve the early strength (strength at 8 h ≥ 10 MPa).

[0179] Furthermore, step five S5 for the whole-process digital control further includes the following steps:

[0180] S501: Build a digital twin platform, integrate BIM models, GIS geographic information, and IoT sensor data streams, and construct a four-dimensional construction control platform. The function modules include:

[0181] Real-time stress cloud map display (update frequency ≤ 10 s);

[0182] Robotic arm operation trajectory simulation and collision warning;

[0183] Generation of automatic optimization suggestions for unloading step length;

[0184] S502: Intelligent risk warning. Set multiple alarm thresholds.

[0185] When the vault settlement rate is greater than 3 mm / d, it is a yellow warning. When the vault settlement rate is greater than 5 mm / d, it is a red warning, and activate the stress compensation system;

[0186] When the water seepage volume is greater than 5 L / min, it is a yellow warning. When the water seepage volume is greater than 10 L / min, trigger the radial grouting robot.

[0187] Furthermore, the operation process of the error compensation mechanism is as follows:

[0188] S2023.1: Real-time error detection, using high-precision sensors (such as lidar, vision systems, encoders, etc.) to monitor the position and attitude of the end of the robotic arm in real time;

[0189] Compare the actual position with the target trajectory or target point, and calculate the error value;

[0190] S2023.2: Trigger condition judgment, set an error threshold (such as 2 mm), when the detected error exceeds the threshold, the system automatically triggers the compensation mechanism;

[0191] S2023.3: Calculate the compensation amount, according to the detected error, use control algorithms (such as PID control, model predictive control, etc.) to calculate the angular or positional changes of the robotic arm joints that need to be adjusted;

[0192] Consider the dynamic characteristics of the robotic arm (such as inertia, friction) for accurate calculation.

[0193] S2023.4: Execute the adjustment, apply the calculated compensation amount to the control system of the robotic arm, and adjust the position and attitude of the robotic arm;

[0194] Ensure that the adjustment process is fast and accurate, and avoid introducing new errors or causing the robotic arm to be unstable;

[0195] S2023.5: Feedback and verification, after the adjustment is completed, detect the position and attitude of the robotic arm again, and verify whether the error is within the allowable range;

[0196] If the error still exceeds the threshold, repeat the above steps until the error is within the allowable range;

[0197] S2023.6: Optimization and adaptation, according to historical data and actual operation conditions, optimize the control algorithm and compensation strategy to improve the response speed and accuracy of the system;

[0198] Combine multiple sensor data to enhance the reliability and accuracy of the system.

[0199] Furthermore, the hierarchical unloading equation for seepage-stress coupling is where σ ij is the stress tensor, k is the permeability coefficient, is the hydraulic gradient, ε pl,kl is the plastic strain increment, and the specific derivation process is as follows:

[0200] According to the effective stress principle formula of Biot theory, σ ij = σ′ ij + α·p·δ ij, where σ ij is the total stress, σ′ ij is the effective stress, α is the Biot coefficient, p is the pore water pressure, and δ ij is the unit tensor;

[0201] Taking the derivative of the Biot theory effective stress principle formula with respect to time gives

[0202] Introducing the elastoplastic constitutive relation, assuming small deformation conditions, the relationship between effective stress and strain is σ′ ij = C ijkl (ε kl - ε pl,kl )(2), where C ijkl is the elastic stiffness tensor, ε kl is the total strain tensor, and ε pl,kl is the plastic strain tensor;

[0203] Combining the pore pressure change rate equation, the pore pressure change rate equation is where k is the permeability, reflecting the ability of the medium to conduct fluids, μ is the fluid dynamic viscosity, φ is the porosity, reflecting the proportion of void volume in the medium, S is the storage coefficient, reflecting the change in fluid storage in a unit volume of rock under a unit pressure change, is the Laplace operator of the pore pressure, reflecting the non-uniformity of the pressure spatial distribution, ε v is the volume strain, and ε v = ε 11 + ε 22 + ε 33 ;

[0204] Combining with the momentum conservation equation where ρ is the density of water and g is the acceleration due to gravity, taking the derivative of the momentum conservation equation with respect to time gives

[0205] Substituting (2) and (3) into (1) gives

[0206] Introducing the relationship between strain rate and displacement, the relationship between strain rate and displacement rate is where are the displacement rate components;

[0207] Combining the above equations gives where is the elastic response term, is the plastic dissipation term, is the seepage-strain coupling term, is the volume strain rate, describing the rate of swelling or compression of the surrounding rock;

[0208] In equation (5), Directly relate the permeability k and Directly quantify the influence of the seepage field on stress changes where t0 is the initial time;

[0209] Define the stress release rate during the unloading process as where δσ is the stress release amount, the unloading step length δL and the stress release rate R σ The relationship is R σ = 1 - e -λ·δL , where λ is the stress release coefficient, taking 0.15, related to the stiffness of the tunnel surrounding rock and the initial stress state. The specific derivation process is as follows:

[0210] The influence of the stress release amount δσ corresponding to the unloading step length δL satisfies The equation shows that the stress release rate is proportional to the remaining unreleased stress (1 - R G );

[0211] (5) After deformation, Solve to get -ln(1 - R σ ) = λ·δL + C. The initial condition is that when δL = 0, R σ = 0, and C = 0 is obtained. Then R σ = 1 - e -λ·δL , which is applicable to the elastic stage of initial excavation. As Figure 7 shown, it can be seen from the figure that the stress release rate R σ has an approximately linear relationship with the unloading step length δL;

[0212] The relationship between the unloading step length δL and the pore pressure change rate is as follows:

[0213] The excavation volume change corresponding to the unloading step length δL is δV = A·δL, where A is the excavation face area, and the volumetric strain change rate is where V0 = A·L0 is the volume under the action of the initial in-situ stress, L0 is the length under the action of the initial in-situ stress, δt = T is the single-step unloading time, and we get Substitute (6) into (3) to get where is the unloading driving term, and the unloading driving term directly quantifies the influence of the unloading rate on the pore pressure. As Figure 8 shown, the horizontal axis is the unloading rate, the vertical axis is the stress change rate, and the pore pressure change rate has a negative linear relationship with the unloading rate, and the slope is From Figure 8 it can be seen that in the case of a high unloading rate, the pore pressure drops rapidly, the negative value increases, and in the case of a low unloading rate, the negative value decreases.

[0214] According to (3), ignoring the spatial diffusion term, that is we get Integrating gives the relationship between pore pressure and time as As Figure 9 shown is the curve of pore pressure varying with time. The horizontal axis is time T and the vertical axis is pore pressure p. It can be seen from Figure 9 that for a high unloading rate, that is, a steep curve slope, the pore pressure drops significantly in a short time; for a low unloading rate, that is, a gentle curve, the pore pressure drops slowly. Its physical meaning indicates that the pore pressure drops linearly with time, and the drop rate is controlled by the unloading step δL and the unloading time T.

[0215] Considering the steady-state seepage condition, that is the equation for the change rate of pore pressure is simplified to Assuming one-dimensional seepage and introducing the boundary condition p(x = 0) = p0, the pore pressure gradient is obtained as Considering the exponential variation of permeability with the unloading step, that is k = k0e β·δL , the equation obtained is where k0 is the initial permeability and β is the permeability stress sensitivity coefficient; as Figure 10 shown is the curve of the pore pressure gradient versus the unloading step. The horizontal axis is the unloading step and the vertical axis is the pore pressure gradient. For a small step (δL ≤ 1.0 m), the pore pressure gradient approximately linearly increases with the increase of the step; for a large step (δL ≥ 2 m), the permeability increases sharply due to crack expansion, resulting in a slowdown or even a decrease in the gradient growth.

[0216] The relationship between the dynamic unloading control process and the unloading step is as follows:

[0217] In the initial unloading stage, at this time, according to the Taylor formula expansion and first-order approximation, the stress release rate R σ = 1 - e -λ·δL ≈ λ·δL. At this time, the stress release rate approximately linearly increases, and it is necessary to strictly specify the unloading step δL to avoid sudden changes. The stress release rate at this time satisfies R σ ≤ 0.2, and it is obtained that δL ≤ 1.3 m. Taking 1.0 m as the safety threshold, that is, δL ≤ 1.0 m;

[0218] In the main unloading stage, for the stress release rate R σ performing a first-order differential operation gives The stress release rate R σ decreases with the increase of the unloading step δL. At this time, the unloading needs to consider both efficiency and safety. The optimization model at this time is where V L is the unloading rate, is the maximum displacement, where σ 初始 is the initial in-situ stress, E is the elastic modulus of the surrounding rock, and ν is the Poisson's ratio. is the pressure gradient. Solving jointly with (3) gives 1.5m ≤ δL ≤ 2.5m;

[0219] In the final unloading stage, at this time R σ = 1 - e -λ·δL ≈ 1. At this time, the unloading step length of the residual stress release of the tunnel at the excavation expansion is less than that in the initial unloading stage and the main unloading stage, and the pore pressure change rate satisfies The unloading step length δL ≤ 0.5m is obtained.

[0220] Furthermore, the operation process of the prestressed anchor cable compensation system includes the following steps:

[0221] S3023.1: Monitoring stage, install monitoring equipment, arrange convergence meters, multi-point displacement meters, fiber optic sensors and other equipment in the tunnel to monitor the deformation of the surrounding rock and the support structure in real time;

[0222] Data acquisition, continuously collect tunnel convergence deformation data, including parameters such as lateral convergence, longitudinal displacement and surrounding rock pressure;

[0223] S3023.2: Trigger condition judgment, set a threshold value. According to the design specifications, set the threshold value of the convergence deformation (such as 3‰D);

[0224] Real-time comparison, compare the monitored deformation data with the threshold value. When the deformation exceeds the set value, trigger the start of the prestressed anchor cable compensation system;

[0225] S3023.3: Tension implementation, select the anchor cable: According to the calculation results, determine the position and quantity of the prestressed anchor cables that need to be tensioned;

[0226] Tensioning process: Loading stage, use the tensioning equipment to apply prestress to the selected anchor cables, and gradually increase the tension to the calculated value;

[0227] Locking stage, after reaching the target tension, use the locking device to fix the anchor cables to ensure that the prestress is maintained;

[0228] Control accuracy, during the whole tensioning process, monitor the tension force and the elongation of the anchor cables in real time to ensure that the prestress is applied evenly and accurately;

[0229] S3023.4: Feedback and adjustment, monitoring feedback, after the tensioning is completed, continue to monitor the convergence deformation of the tunnel to evaluate the effect of the prestress compensation;

[0230] Adjustment and optimization, if the monitoring results show that the deformation still does not reach the expected control target, the system will recalculate and make further tensioning adjustments until the convergence deformation meets the requirements;

[0231] S3023.5: Long-term maintenance and continuous monitoring. After the compensation is completed, the system continuously monitors the deformation of the tunnel to ensure long-term stability.

[0232] Regular inspection. Regularly check the working status of the prestressed anchor cables, including tension, anchoring conditions, etc., to prevent out-of-control deformation caused by anchor failure.

[0233] S3023.6: Automation and intelligence. Automatic control. The system integrates an automatic control module to achieve a fully automatic process of monitoring, calculation, tensioning, and feedback adjustment, reducing human intervention.

[0234] Intelligent algorithms. Adopt machine learning and intelligent algorithms to optimize the prestress application strategy based on historical data, improving the compensation efficiency and accuracy.

[0235] Furthermore, the operation of the slump adaptive adjustment system includes the following steps:

[0236] S4021: Analysis of the surrounding rock stress field and prediction of working conditions. Stress monitoring. Install distributed optical fiber sensors to obtain the real-time distribution of the surrounding rock stress (σ max ≤5MPa);

[0237] Unloading classification decision. Divide the unloading areas according to the stress gradient (Category A: high risk, Category B: medium risk, Category C: low risk), and predict the concrete performance requirements for each area.

[0238] S4022: Real-time detection of the rock surface roughness and material state. IRI value scanning. The robotic arm is equipped with a 3D laser scanner to scan the rock surface before construction to generate an IRI thermal map (accuracy ±0.1mm / m);

[0239] Material monitoring. Slump monitoring. Install a laser sensor inside the pipeline to measure it every 30 seconds (error ±5mm);

[0240] Fiber dispersion detection. Based on image recognition technology, analyze the uniformity of the steel fiber distribution (qualified rate ≥95%);

[0241] S4023: Generation of dynamic adjustment strategies. Multi-objective optimization calculation. Input parameters: IRI value, surrounding rock stress level, current slump (S 实测 ), ambient temperature and humidity;

[0242] Output parameters: target slump (S 目标 ), spraying speed (V 喷射 ), accelerator dosage (C 速凝 ), fiber feeding frequency (F 纤维 );

[0243] Optimization model: min(|S 实测 -S 目标| + λ·|V 喷射 -V 理论 |), the constraint conditions are that the 8h strength ≥ 10MPa and the fiber content is 1.5% ± 0.1%.

[0244] S4024: Adjustment instruction issuance, slump adjustment: Control fluidity by increasing or decreasing mixing water (±3%) or high-range water reducer (±0.2%);

[0245] Spraying speed matching: For every 1mm / m increase in the IRI value, the spraying speed decreases by 1m3 / h;

[0246] Fiber synchronous control: The variable-frequency vibrator adjusts the feeding rhythm of steel fibers to prevent agglomeration;

[0247] S4025: Precise execution and closed-loop verification, executed by the wet shotcreting manipulator, the robotic arm sprays according to the planned path, and the speed is dynamically adjusted according to the IRI value (PID control accuracy ±0.2m 3 / h);

[0248] SiO2 and steel fibers are accurately incorporated through a double-screw feeder (error ≤ 0.05%);

[0249] Real-time feedback and iteration, rapid strength detection: The penetration strength tester randomly checks the 8h strength every 30 minutes, and the data is transmitted back to the control system;

[0250] Adaptive correction: If the strength does not meet the standard, trigger secondary adjustment (increase the accelerator by 0.1% or decrease the slump by 10mm);

[0251] S4026: Stress-deformation collaborative management, post-unloading monitoring: After the staged unloading is completed, monitor the deformation of the concrete layer through strain gauges. If the deformation exceeds the limit (≥ 0.3mm / m), start the supplementary spraying program;

[0252] Data precipitation: Store the construction parameters, adjustment records, and quality data in the database for iterative optimization of the machine learning model.

[0253] The following combines Figure 1-Figure 6 to specifically introduce this method:

[0254] According to different positions to be expanded, emergency parking bays are constructed for double-track tunnels in mountainous and hilly areas. When the main tunnel of the existing tunnel lacks an emergency parking bay, the existing main tunnel is excavated and expanded into a new emergency parking bay. When the emergency parking bay of the existing tunnel in the same driving direction does not meet the current code requirements, the existing emergency parking bay is excavated and expanded into a new emergency parking bay using the unilateral excavation method. When the emergency parking bay of the existing tunnel in the opposite driving direction does not meet the current code requirements, the existing emergency parking bay is excavated and expanded into a new emergency parking bay using the method of excavating one side and backfilling the other side.

[0255] The excavation process includes the following steps: First, it is the refined modeling of the surrounding rock stress field and the design of the excavation sequence. Through three-dimensional geological exploration and numerical simulation, the optimal excavation sequence is determined. Specifically, it is geological advanced exploration. Geological advanced exploration - using three-dimensional geological radar and borehole CT, a three-dimensional exploration of the expansion section is carried out at a spacing of 5m to obtain data on the distribution of rock mass fractures, the occurrence state of groundwater, and the original stress field; numerical simulation and sequence optimization. An asymmetric water-hydraulic coupling model is constructed through the ABAQUS / FLAC3D platform, and the following parameters are input: the elastic modulus of the rock mass, the Poisson's ratio is dynamically corrected according to the exploration data, the permeability coefficient tensor considers anisotropic seepage, and the residual bearing capacity of the existing support structure.

[0256] Simulate the influence of different excavation sequences such as left-side priority, right-side priority, and alternate construction on the plastic zone of the surrounding rock, screen the minimum disturbance sequence plan, and output the priority list of the excavation section.

[0257] Then, precise demolition by the robotic arm guided by three-dimensional laser scanning is carried out. Specifically: High-precision three-dimensional modeling. High-precision three-dimensional modeling - using the Trimble X7 laser scanner with an accuracy of 0.5mm within a range of 30 meters, a 360° scan of the existing structure is carried out to generate a textured BIM model, and the static measurement error of the area to be demolished is identified ≤2mm; automatically mark the weak points of the structure, such as cracks and hollow areas, through the point cloud-model comparison algorithm to generate a risk heat map; real-time monitor the deformation of the existing structure through the distributed optical fiber sensing system and establish an initial database of the mechanical parameters of the surrounding rock. Robotic arm collaborative operation system. Deploy a hydraulic crushing robotic arm and integrate the following functional modules: Visual guidance system: Real-time positioning based on binocular cameras and laser trackers to dynamically compensate for the pose deviation of the robotic arm; Adaptive crushing head: The adaptive crushing head is a hydraulic servo-controlled crushing head, using multi-stage pressure control technology, with an adjustable pressure range of 0-50MPa, and automatically adjusts the impact energy according to the concrete strength; The crushing process of the hydraulic crushing robotic arm is to demarcate the virtual demolition boundary according to the BIM model, and the robotic arm performs millimeter-level contour cutting along the predetermined path; Layered crushing is carried out on the core area with a single-layer thickness ≤200mm, and the dust removal system is started synchronously; The demolished area is scanned in real time and compared with the design model to trigger the error compensation mechanism.

[0258] The error compensation mechanism operates as follows: real-time error detection, using high-precision sensors (such as lidar, vision systems, encoders, etc.) to monitor the position and posture of the robotic arm's end in real time; comparing the actual position with the target trajectory or target point to calculate the error value; triggering condition judgment, setting an error threshold (such as 2 mm); when the detected error exceeds the threshold, the system automatically triggers the compensation mechanism; calculating the compensation amount, using a control algorithm (such as PID control, model predictive control, etc.) based on the detected error to calculate the angle or position change of the robotic arm joint that needs to be adjusted; taking into account the dynamic characteristics of the robotic arm (such as inertia and friction) for precise calculations. Execute adjustment, apply the calculated compensation amount to the control system of the robot arm, and adjust the position and posture of the robot arm; ensure that the adjustment process is fast and accurate to avoid introducing new errors or causing instability of the robot arm; feedback and verification, after the adjustment is completed, re-detect the position and posture of the robot arm to verify whether the error is within the allowable range; if the error still exceeds the threshold, repeat the above steps until the error is within the allowable range; optimization and adaptation, based on historical data and actual operation conditions, optimize the control algorithm and compensation strategy to improve the response speed and accuracy of the system; combine multiple sensor data to enhance the reliability and accuracy of the system.

[0259] Then, the hydraulic-mechanical coupled graded unloading and step size optimization are carried out. The specific steps are as follows:

[0260] A multi-field coupled unloading model is constructed, and a graded unloading equation considering seepage-stress coupling is established. The surrounding rock mechanical parameters and groundwater seepage data are input to simulate the influence of different unloading steps on the stress release of the surrounding rock. Dynamic unloading control process: Initial unloading stage (construction step length ≤ 1.0m): Use microseismic monitoring + fiber Bragg grating sensing to provide real-time feedback on the expansion rate of the surrounding rock loosening zone. If it exceeds 5mm / h, construction will be suspended; simultaneously implement advanced small-duct grouting, i.e. cement-water glass dual-liquid slurry, to form a reinforced arch shell; main unloading stage (construction step length 1.5-2.5m): proceed according to the "excavation-support-monitoring" cycle, update the model parameters after each cycle, measure the support reaction force through the pressure box, and invert the surrounding rock stress release rate; use the TSP203 system to detect geological changes 20m ahead; final unloading stage (construction step length ≤ 0.5m): start the prestressed anchor cable compensation system (tension force ≥ 200kN) and control the convergence deformation within 3‰D (D is the tunnel diameter in meters).

[0261] The operation process of the prestressed anchor cable compensation system is as follows: Monitoring stage: Install monitoring equipment, and arrange convergence meters, multi-point displacement meters, fiber optic sensors and other equipment in the tunnel to monitor the deformation of the surrounding rock and the support structure in real time; Data acquisition: Continuously collect tunnel convergence deformation data, including parameters such as lateral convergence, longitudinal displacement and surrounding rock pressure; Trigger condition judgment: Set thresholds. According to the design specifications, set the threshold of convergence deformation (such as 3‰D); Real-time comparison: Compare the monitored deformation data with the threshold. When the deformation exceeds the set value, trigger the start of the prestressed anchor cable compensation system; Tensioning implementation: Select anchor cables: According to the calculation results, determine the positions and quantities of the prestressed anchor cables to be tensioned; Tensioning process: Loading stage: Use tensioning equipment to apply prestress to the selected anchor cables and gradually increase the tension to the calculated value; Locking stage: After reaching the target tension, use a locking device to fix the anchor cables to ensure that the prestress is maintained; Control accuracy: During the entire tensioning process, monitor the tension force and the elongation of the anchor cables in real time to ensure that the prestress is applied evenly and accurately; Feedback and adjustment: Monitoring feedback: After the tensioning is completed, continue to monitor the convergence deformation of the tunnel to evaluate the effect of prestress compensation; Adjustment and optimization: If the monitoring results show that the deformation still does not reach the expected control target, the system will recalculate and perform further tensioning adjustments until the convergence deformation meets the requirements; Long-term maintenance: Continuous monitoring: After the compensation is completed, the system continuously monitors the deformation of the tunnel to ensure long-term stability. Regular inspection: Regularly check the working conditions of the prestressed anchor cables, including tension, anchoring conditions, etc., to prevent deformation out of control caused by anchoring failure; Automation and intelligence: Automatic control: The system integrates an automatic control module to realize the full-automatic process of monitoring, calculation, tensioning and feedback adjustment, reducing human intervention. Intelligent algorithms: Adopt machine learning and intelligent algorithms to optimize the prestress application strategy according to historical data, improving the compensation efficiency and accuracy.

[0262] Then perform the rapid forming of the asymmetric support structure, specifically the design of differential support parameters. According to the stress field analysis results, adopt an asymmetric support scheme for the left and right sides of the excavation area: High stress side: Arrange a double-layer steel mesh with a specification of Φ8@150×150, and arrange a lattice steel frame with a specification of HW175 type and a spacing of 0.6m; Low stress side: Arrange a single-layer steel mesh with a specification of Φ6@200×200, and arrange a steel section support with a specification of I20a and a spacing of 1.0m. Intelligent shotcrete technology: Use a wet shotcrete manipulator equipped with a slump self-adaptive adjustment system to automatically adjust the spraying speed (3 - 8m3 / h) according to different rock surface roughness; Incorporate SiO2 and steel fibers (dosage of 1.5%) to enhance the early strength (strength at 8h ≥ 10MPa).

[0263] The operation of the slump self-adaptive adjustment system is specifically as follows: Surrounding rock stress field analysis and working condition prediction, Stress monitoring: Install distributed fiber optic sensors to obtain the surrounding rock stress distribution (σ max≤5MPa); Unloading classification decision, dividing the unloading area according to the stress gradient (Class A: high risk, Class B: medium risk, Class C: low risk), predicting the concrete performance requirements of each area; Real-time detection of rock surface roughness and material state, IRI value scanning, a 3D laser scanner mounted on the robotic arm, scanning the rock surface before construction to generate an IRI thermal map (accuracy ±0.1mm / m); Material monitoring, slump monitoring, a laser sensor built into the pipeline, measuring once every 30 seconds (error ±5mm); Fiber dispersion detection, based on image recognition technology, analyzing the uniformity of steel fiber distribution (qualified rate ≥95%); Generating a dynamic adjustment strategy, multi-objective optimization calculation, input parameters: IRI value, surrounding rock stress level, current slump (S 实测 ), ambient temperature and humidity; Output parameters: target slump (S 目标 ), spraying speed (V 喷射 ), accelerator dosage (C 速凝 ), fiber feeding frequency (F 纤维 ); Optimization model: min(|S 实测 - S 目标 | + λ·|V 喷射 - V 理论 |), the constraint condition is that the 8-hour strength ≥10MPa, and the fiber dosage is 1.5% ±0.1%. Issuing adjustment instructions, slump adjustment: controlling fluidity by increasing or decreasing mixing water (±3%) or high-range water reducer (±0.2%); Spraying speed matching: for every 1mm / m increase in the IRI value, the spraying speed is reduced by 1m3 / h; Fiber synchronous control: a variable-frequency vibrator adjusts the steel fiber feeding rhythm to prevent agglomeration; Precise execution and closed-loop verification, executed by the wet spraying manipulator, the robotic arm sprays according to the planned path, and the speed is dynamically adjusted according to the IRI value (PID control accuracy ±0.2m 3 / h); SiO2 and steel fibers are accurately incorporated through a double-screw feeder (error ≤0.05%); Real-time feedback and iteration, rapid strength detection: the penetration strength tester randomly checks the 8-hour strength every 30 minutes, and the data is transmitted back to the control system; Adaptive correction: if the strength does not meet the standard, trigger secondary adjustment (increase the accelerator by 0.1% or reduce the slump by 10mm); Stress-deformation collaborative management, post-unloading monitoring: after hierarchical unloading is completed, monitor the deformation of the concrete layer through strain gauges. If the deformation exceeds the limit (≥0.3mm / m), start the supplementary spraying program; Data deposition: Store construction parameters, adjustment records and quality data in the database for iterative optimization of the machine learning model.

[0264] Finally, for the whole-process digital control, specifically, a digital twin platform is built to integrate BIM+GIS+IoT data streams, and a four-dimensional construction control platform is constructed. The function modules include: real-time stress cloud map display (update frequency ≤ 10s); robotic arm operation trajectory simulation and collision warning; automatic optimization suggestion generation for unloading step size; intelligent risk warning, with multiple alarm thresholds set. When the vault settlement rate is greater than 3mm / d, it is a yellow warning. When the vault settlement rate is greater than 5mm / d, it is a red warning, and the stress compensation system is activated. When the seepage water volume is greater than 5L / min, it is a yellow warning. When the seepage water volume is greater than 10L / min, the radial grouting robot is triggered.

[0265] The present invention has the following advantages: The excavation sequence control based on water-hydraulic coupling breaks the traditional symmetric construction mode. By establishing the coupling equation of seepage field and stress field, the influence of seepage water penetration on surrounding rock unloading is quantified. The differential unloading step size design is adopted, and the grading standard for unloading stage division is proposed. The excavation process is divided into three stages: initial unloading (step size ≤ 1.0m), main unloading (step size 1.5 - 2.5m), and final unloading (step size ≤ 0.5m). The asymmetric support structure design is adopted. According to the stress distribution difference (the stress value on the high-stress side ≥ 1.5 times that on the low-stress side), differential support parameters are adopted.

[0266] Taking the reconstruction and expansion project of a highway tunnel parking bay as an example, the road grade is highway, the designed driving speed is 100km / h, with eight lanes in both directions. The existing tunnel is 2902m long, the tunnel building limit has a net width of 9.75m and a net height of 5.0m, and a total of 3 emergency parking bays and vehicle cross passages are set. The tunnel constructed by this project is reconstructed to continue to use the original technical standard with a designed speed of 80km / h for the existing tunnel. Considering the connection with the main tunnel section, the driving lane and the inspection lane of the existing tunnel, as well as the requirements of the current specifications, the building limit of the newly added tunnel emergency parking bay is determined to have a net width of 13.5m and a net height of 5.0m.

[0267] Please refer to Figure 11 , Figure 11It is the time history diagram of the tunnel crown settlement and the invert heave during the excavation of the existing tunnel parking bay. The vertical coordinate is the displacement deformation with the unit of mm, and the horizontal coordinate is the time, i.e., the construction days, with the unit of day. A value greater than zero indicates that the tunnel deforms upward, and a value less than zero indicates that the tunnel deforms downward. It can be seen from the figure that in the initial stage of excavation disturbance, affected by unloading and stress redistribution, the settlement rate of the support structure is relatively fast, showing an upward curve. That is, before excavation, the crown settlement and invert heave of the tunnel are zero. As the process of removing the secondary lining and excavating the rock mass progresses, with unloading and stress redistribution, the crown settlement and invert heave of the tunnel gradually increase, and the corresponding time on the horizontal coordinate is 10 - 60 d. With the gradual stabilization of the support structure and the recovery of the surrounding rock's self-bearing capacity, the settlement rate gradually slows down, and the curve tends to be flat. That is, as the rock mass excavation is completed and the shotcrete spraying support process is carried out, the supporting force of the support structure comes into play and the stress of the surrounding rock is redistributed, and the displacement deformation rate and the displacement curve gradually become flat, presenting an approximately horizontal straight line, corresponding to the horizontal coordinate of 70 - 90 d. It can be seen from the figure that the cumulative deformation amount (the maximum deformation amount) of the tunnel crown settlement and invert heave during construction does not exceed 10 mm, meeting the requirements of the current specifications.

[0268] Please refer to Figure 12 , Figure 12 It is the time history diagram of the tunnel horizontal convergence during the excavation of the existing tunnel parking bay. The vertical coordinate is the horizontal convergence value of the tunnel with the unit of mm, and the horizontal coordinate is the time with the unit of day. A value greater than zero indicates that the tunnel sidewall expands outward, and a value less than zero indicates that the tunnel sidewall contracts inward. It can be seen from the figure that before excavation, the horizontal convergence of the tunnel is zero. As the process of removing the secondary lining and excavating the rock mass progresses, the rock mass on the tunnel sidewall undergoes unloading and stress redistribution, and the horizontal convergence value gradually increases, showing an approximately linear change trend, and the corresponding time on the horizontal coordinate is 10 - 60 d. With the completion of the shotcrete spraying support, the gradual stabilization of the support structure and the recovery of the surrounding rock's self-bearing capacity, the horizontal convergence curve gradually becomes flat, presenting an approximately horizontal straight line, corresponding to the horizontal coordinate of 70 - 90 d. It can be seen from the figure that the cumulative value of the horizontal convergence does not exceed 5 mm, meeting the requirements of the current specifications.

[0269] Compared with the conventional single-side excavation method, the comparison of indicators such as the maximum crown settlement, the maximum invert heave, the maximum horizontal convergence, the settlement rate, and the horizontal convergence rate of this method is shown in the following table:

[0270] Index Conventional single-side excavation This method Maximum crown settlement / mm -9.5 -8.2 Maximum invert heave / mm 5.4 3.9 Maximum horizontal convergence / mm -4.2 -2.7 Settlement rate / (mm / s) 1.5 1.0 Horizontal convergence rate / (mm / s) 1.2 1.0

[0271] It can be seen from the table that compared with the conventional single-side excavation method, the maximum crown settlement, the maximum invert heave, and the maximum horizontal convergence of this method are all smaller than those of the conventional single-side excavation method, and the settlement rate and the horizontal convergence rate controlled during construction of this method are better than those of the conventional single-side excavation method.

[0272] Example 2:

[0273] The following takes the reconstruction and expansion project of a parking bay in a highway tunnel as an example for specific description:

[0274] This tunnel adopts a separated two-way four-lane design. The total length of the left line is 15,231 m, and the total length of the right line is 15,173 m. The maximum buried depth of the tunnel body is about 629 m, belonging to an extra-long deep-buried tunnel. The design speed of the tunnel is 80 km / h. The clear height of the tunnel building limit is 5 m, the clear width of the building limit in the normal section is 10.25 m, the maximum excavation width is 13.6 m, the clear width of the building limit of the emergency parking bay is 13.25 m, and the maximum excavation width is 16.8 m.

[0275] Please refer to Figure 13 , Figure 13 for the time history curves of the crown settlement and floor heave of the tunnel during the excavation of the existing tunnel parking bay. The vertical coordinate is the displacement deformation, with the unit of mm, and the horizontal coordinate is the time, i.e., the construction days, with the unit of day. A value greater than zero indicates that the tunnel deforms upward, and a value less than zero indicates that the tunnel deforms downward. It can be seen from the figure that in the initial stage of excavation disturbance, affected by unloading and stress redistribution, the settlement rate of the support structure is relatively fast, showing an upward curve. That is, before excavation, the crown settlement and floor heave of the tunnel are zero. At 10 d, the crown settlement is -1.8 mm and the floor heave is 1 mm. As the process of removing the secondary lining and excavating the rock mass progresses, the crown settlement and floor heave of the tunnel gradually increase, and the corresponding time on the horizontal coordinate is 10 - 60 d. As the support structure gradually stabilizes and the self-bearing capacity of the surrounding rock recovers, the settlement rate gradually slows down, and the curve tends to be flat. That is, as the rock mass excavation is completed and the shotcrete support process is carried out, the displacement deformation rate and the displacement curve gradually become flat, presenting an approximately horizontal straight line, corresponding to the horizontal coordinate of 70 - 90 d. At 90 d, the crown settlement is -8.1 mm and the floor heave is 4.3 mm. It can be seen from the figure that the cumulative deformation amount (maximum deformation amount) of the crown settlement and floor heave of the tunnel during construction does not exceed 10 mm, meeting the requirements of the current specifications.

[0276] Please refer to Figure 14 , Figure 14It is the time history diagram of the horizontal convergence of the tunnel during the excavation of the existing tunnel parking belt. The vertical coordinate is the horizontal convergence value of the tunnel, with the unit of mm, and the horizontal coordinate is the time, with the unit of day. A value greater than zero indicates that the tunnel sidewall expands outward, and a value less than zero indicates that the tunnel sidewall contracts inward. It can be seen from the figure that before the excavation, the horizontal convergence of the tunnel is zero. At 10 days, the horizontal convergence is -0.8 mm. As the process of removing the secondary lining and excavating the rock mass progresses, the horizontal convergence value gradually increases, showing an approximately linear change trend, and the corresponding time on the horizontal coordinate is 10 - 60 days. As the shotcrete support is completed, the horizontal convergence curve gradually flattens and presents an approximately horizontal straight line, corresponding to the horizontal coordinate of 70 - 90 days. At 90 days, the horizontal convergence of the tunnel sidewall is -3.5 mm. It can be seen from the figure that the cumulative value of the horizontal convergence does not exceed 5 mm, meeting the requirements of the current specifications.

[0277] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0278] Although the embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention.

Claims

1. A safety excavation method for the reconstruction and expansion of an emergency parking bay in a double-track tunnel in a mountainous and hilly area, characterized in that: It includes the following steps: Step S1: Fine-grained modeling of the surrounding rock stress field and excavation sequence design. Through three-dimensional geological exploration and numerical simulation, determine the optimal excavation sequence; Step S2: Precision demolition by robotic arm guided by three-dimensional laser scanning, achieving millimeter-level contour control, reducing over-excavation and damage to the support structure; Step S3: Water-hydraulic coupling hierarchical unloading and step size optimization. Through dynamic unloading control, reduce the disturbance of the surrounding rock and improve construction safety; Step S4: Rapid prototyping of the asymmetric support structure. According to the stress distribution differences, adopt a differential support scheme to improve construction efficiency; Step S5: Whole-process digital management and control. By integrating BIM models, GIS geographic information, and IoT sensor data, build a digital twin platform to realize real-time stress cloud map monitoring (update frequency ≤ 10 seconds), robotic arm trajectory simulation, and automatic optimization of unloading step size during the construction process, so as to improve construction quality and safety.

2. A safe excavation method for the reconstruction and expansion of the emergency parking bay of a double-track tunnel in mountainous and hilly areas according to claim 1, characterized in that: The fine-grained modeling of the surrounding rock stress field and excavation sequence design in step S1 includes the following steps: Step S101: Geological advanced exploration. Adopt geological advanced exploration - use three-dimensional geological radar and borehole CT to conduct three-dimensional exploration of the expansion section at a spacing of 5 m to obtain data on rock mass fissure distribution, groundwater occurrence state, and original stress field; Real-time monitor the deformation of the existing structure through a distributed optical fiber sensing system, and establish an initial database of mechanical parameters of the surrounding rock; Step S102: Numerical simulation and time sequence optimization. Build an asymmetric water-hydraulic coupling model through the ABAQUS / FLAC3D platform and input the following parameters: elastic modulus of the rock mass, Poisson's ratio is dynamically corrected according to the detection data, permeability coefficient tensor considers anisotropic seepage, and residual bearing capacity of the existing support structure; Simulate the influence of different excavation sequences such as left-side priority, right-side priority, and alternating construction on the plastic zone of the surrounding rock, screen the time sequence scheme with the least disturbance, and output the priority list of the excavation section.

3. A safe excavation method for the reconstruction and expansion of the emergency parking bay of a double-track tunnel in mountainous and hilly areas according to claim 1, characterized in that: The precision demolition by robotic arm guided by three-dimensional laser scanning in step S2 includes the following steps: S201: High-precision three-dimensional modeling. Adopt high-precision three-dimensional modeling - use a Trimble X7 laser scanner with an accuracy of 0.5 mm within every 30 m range to scan the existing structure 360°, generate a textured BIM model, and identify static measurement errors in the area to be demolished ≤ 2 mm; Automatically mark the weak points of the structure, such as cracks and hollow areas, through the point cloud-model comparison algorithm, and generate a risk heat map; S202: Robotic arm collaborative operation system. Deploy a hydraulic crushing robotic arm and integrate the following functional modules: Visual guidance system: Based on binocular cameras and laser trackers for real-time positioning, dynamically compensate for the pose deviation of the robotic arm; Adaptive crushing head: The adaptive crushing head is a hydraulic servo-controlled crushing head that adopts multi-stage pressure control technology with an adjustable pressure range of 0 - 50 MPa, and automatically adjusts the impact energy according to the concrete strength.

4. A safety excavation method for the reconstruction and expansion of emergency parking bays in a double-track tunnel in mountainous and hilly areas according to claim 1, characterized in that: The specific steps of the water-hydraulic coupling staged unloading and unloading step optimization in step S3 are as follows: S301: Construct a multi-field coupling unloading model, establish a staged unloading equation considering seepage-stress coupling, input the mechanical parameters of the surrounding rock and groundwater seepage data, and simulate the influence of different unloading step lengths on the stress release of the surrounding rock; S302: Dynamic unloading control process: S3021: Initial unloading stage (construction step length ≤ 1.0 m): Use microseismic monitoring + fiber Bragg grating sensing to real-time feedback the expansion rate of the loosening zone of the surrounding rock. If it exceeds 5 mm / h, construction shall be suspended; Simultaneously implement advanced small pipe grouting, namely cement-sodium silicate double liquid grout, to form a reinforced arch shell. The volume mixing ratio of the cement-sodium silicate double liquid grout is 1:0.5; S3022: Main unloading stage (construction step length 1.5 - 2.5 m): Advance in a cycle of "excavation - support - monitoring". After each cycle ends, update the model parameters, measure the support reaction force through pressure cells, and invert the stress release rate of the surrounding rock; Use the TSP203 system to detect geological changes 20 m ahead; S3023: Final unloading stage (construction step length ≤ 0.5 m): Activate the prestressed anchor cable compensation system (tensile force ≥ 200 kN) to control the convergence deformation within 3‰D (D is the tunnel diameter, unit: m).

5. A safety excavation method for the reconstruction and expansion of emergency parking bays in a double-track tunnel in mountainous and hilly areas according to claim 1, characterized in that: The rapid forming of the asymmetric support structure in step S4 further includes the following steps: Step S401: Differential support parameter design. According to the stress field analysis results, adopt an asymmetric support scheme for the left and right sides of the excavation area: High stress side: Arrange a double-layer steel mesh with a specification of 8 mm diameter steel bars and a spacing of 150 mm × 150 mm, and arrange a grid steel frame with a specification of HW175 type and a spacing of 0.6 m; Low stress side: Arrange a single-layer steel mesh with a specification of 6 mm diameter steel bars and a spacing of 200 mm × 200 mm, and arrange a steel section support with a specification of I20a and a spacing of 1.0 m; Step S402: The intelligent shotcrete technology uses a robotic arm equipped with a slump self - adapting adjustment system to automatically adjust the spraying speed (3 - 8m 3 / h) according to different rock surface roughnesses; Incorporate SiO2 and steel fibers (dosage is 1.5%) to enhance the early strength. According to GB / T 50081-2019, the strength at 8 hours is not less than 10 MPa.

6. A safety excavation method for the reconstruction and expansion of emergency parking bays in a double-track tunnel in mountainous and hilly areas according to claim 1, characterized in that: The whole-process digital control in step S5 further includes the following steps: S501: Build a digital twin platform, integrate BIM models, GIS geographic information, and IoT sensor data streams, and construct a four-dimensional construction control platform. The function modules include: Real-time stress cloud map display (update frequency ≤ 10 s); Simulation of the operation trajectory of the robotic arm and collision warning; Generation of automatic optimization suggestions for unloading step lengths; S502: Intelligent risk warning, set multi-level alarm thresholds, When the vault settlement rate is greater than 3 mm / day, it is a yellow warning. When the vault settlement rate is greater than 5 mm / day, it is a red warning, and activate the stress compensation system; When the seepage water volume is greater than 5 liters per minute, it is a yellow warning. When the seepage water volume is greater than 10 liters per minute, it is a red warning, triggering the radial grouting robot to grout the water seepage area to reduce the seepage water volume.

7. A method for safely expanding the excavation of the emergency parking bay during the reconstruction and expansion of a double-track tunnel in mountainous and hilly areas according to claim 3, characterized in that: The crushing process of the hydraulic crushing robotic arm is as follows: S2021: Define the virtual demolition boundary according to the BIM model, and the robotic arm performs millimeter-level contour cutting along the predetermined path; S2022: Implement layered crushing on the core area, with a single-layer thickness ≤ 200 mm, and simultaneously start the dust removal system; S2023: Real-time scan the demolished area, compare it with the design model, and trigger the error compensation mechanism.

8. A method for safely expanding the excavation of the emergency parking bay during the reconstruction and expansion of a double-track tunnel in mountainous and hilly areas according to claim 7, characterized in that: The operation process of the error compensation mechanism is as follows: S2023.1: Real-time error detection, use high-precision sensors (such as lidar, vision system, encoder, etc.) to monitor the position and posture of the end of the robotic arm in real time; Compare the actual position with the target trajectory or target point, and calculate the error value. S2023.2: Trigger condition judgment, set an error threshold (such as 2 mm), when the detected error exceeds the threshold, the system automatically triggers the compensation mechanism; S2023.3: Calculate the compensation amount, according to the detected error, use control algorithms (such as PID control, model predictive control, etc.) to calculate the angle or position change of the robotic arm joints that need to be adjusted; Consider the dynamic characteristics of the robotic arm (such as inertia, friction) for accurate calculation. S2023.4: Execute the adjustment, apply the calculated compensation amount to the control system of the robotic arm, and adjust the position and posture of the robotic arm; Ensure that the adjustment process is fast and accurate, and avoid introducing new errors or causing the robotic arm to be unstable; S2023.5: Feedback and verification, after the adjustment is completed, detect the position and posture of the robotic arm again to verify whether the error is within the allowable range; If the error still exceeds the threshold, repeat the above steps until the error is within the allowable range; S2023.6: Optimization and adaptation, according to historical data and actual operation conditions, optimize the control algorithm and compensation strategy to improve the response speed and accuracy of the system; Combine multiple sensor data to enhance the reliability and accuracy of the system.

9. A method for safely expanding the excavation of the emergency parking bay during the reconstruction and expansion of a double-track tunnel in mountainous and hilly areas according to claim 4, characterized in that: The operation process of the prestressed anchor cable compensation system includes the following steps: S3023.1: Monitoring stage, install monitoring equipment, arrange convergence meters, multi-point displacement meters, fiber optic sensors and other equipment in the tunnel to monitor the deformation of the surrounding rock and support structure in real time; Data acquisition, continuously acquire tunnel convergence deformation data, including parameters such as lateral convergence, longitudinal displacement and surrounding rock pressure; S3023.2: Trigger condition judgment, set a threshold, according to the design specifications, set the threshold of convergence deformation (such as 3‰D); Real-time comparison: the monitored deformation data is compared with the threshold value. When the deformation exceeds the set value, the prestressed anchor cable compensation system is triggered to start; S3023.3: Tensioning Implementation, Anchor Cable Selection: Based on the calculation results, determine the location and number of prestressed anchor cables to be tensioned; Tensioning process: During the loading phase, prestressing equipment is used to apply prestress to the selected anchor cables, gradually increasing the tension to the calculated value; During the locking phase, after the target tension is reached, the anchor cable is fixed using a locking device to ensure that the prestress is maintained; Control precision: During the entire tensioning process, the tensioning force and the elongation of the anchor cable are monitored in real time to ensure uniform and accurate prestressing. S3023.4: Feedback and adjustment, monitoring feedback. After tensioning is completed, continue to monitor the convergence and deformation of the tunnel and evaluate the effectiveness of prestress compensation. Adjustment and optimization: If the monitoring results show that the deformation still does not meet the expected control target, the system will recalculate and make further tension adjustments until the convergence deformation meets the requirements; S3023.5: Long-term maintenance and continuous monitoring. After compensation is completed, the system continuously monitors the deformation of the tunnel to ensure long-term stability. Regular inspections should be carried out to check the working status of prestressed anchor cables, including tension and anchorage, to prevent uncontrolled deformation caused by anchorage failure; S3023.6: Automation and intelligence, automated control, system integration of automated control modules to achieve fully automatic processes of monitoring, calculation, tensioning and feedback adjustment, reducing human intervention. Intelligent algorithm uses machine learning and intelligent algorithms to optimize prestressing strategies based on historical data to improve compensation efficiency and accuracy.

10. The method for safely excavating and reconstructing an emergency stop lane in a double-track tunnel in a mountainous and hilly area according to claim 5, characterized in that: The operation of the slump adaptive adjustment system includes the following steps: S4021: Analysis of surrounding rock stress field and working condition prediction, stress monitoring, installation of distributed optical fiber sensors to obtain the surrounding rock stress distribution (σ max ≤ 5 MPa) in real time; Unloading classification decision-making: divide the unloading area according to the stress gradient (Class A: high risk, Class B: medium risk, Class C: low risk), and predict the concrete performance requirements of each area; S4022: Real-time detection of rock surface roughness and material status, IRI value scanning. The robotic arm is equipped with a 3D laser scanner to scan the rock surface before construction and generate an IRI thermal map (accuracy ±0.1mm / m); Material monitoring, slump monitoring, built-in laser sensor in the pipeline, measurement every 30 seconds (error ± 5mm); Fiber dispersion detection, based on image recognition technology, analyzes the uniformity of steel fiber distribution (qualified rate ≥ 95%); S4023: Dynamic adjustment strategy generation, multi-objective optimization calculation, input parameters: IRI value, surrounding rock stress level, current slump (S 实测 ), environmental temperature and humidity; Output parameters: target slump (S 目标 ), spraying speed (V 喷射 ), accelerator dosage (C 速凝 ), fiber feeding frequency (F 纤维 ); Optimized model: min(|S 实测 -S 目标 | + λ·|V 喷射 -V 理论 |), where the constraint is that the eight-hour strength is not less than 10 MPa and the fiber content is 1.5% ± 0.1%. S4024: Adjustment instructions are issued, slump adjustment: fluidity is controlled by increasing or decreasing mixing water (±3%) or high-efficiency water reducer (±0.2%); Jet velocity matching: for every 1 mm / m increase in the IRI value, the jet velocity decreases by 1 m 3 / h; Fiber synchronization control: frequency conversion vibrator adjusts the feeding rhythm of steel fiber to prevent agglomeration; S4025: Precise execution and closed-loop verification. The robotic arm executes, sprays along the planned path, and its speed is dynamically adjusted according to the IRI value (PID control accuracy ±0.2 m 3 / h); SiO2 and steel fiber are accurately added through a double-screw feeder (error ≤ 0.05%); Real-time feedback and iteration, rapid strength testing: The penetration strength meter checks the 8-hour strength every 30 minutes, and the data is sent back to the control system; Adaptive correction: If the strength does not meet the standard, trigger secondary adjustment (increase the accelerator by 0.1% or reduce the slump by 10mm); S4026: Stress-deformation collaborative management, post-unloading monitoring: After the staged unloading is completed, monitor the deformation of the concrete layer through strain gauges. If the deformation exceeds the limit (≥0.3 mm / m), start the supplementary spraying procedure; Data precipitation: Store the construction parameters, adjustment records, and quality data in the database for iterative optimization of the machine learning model.

Images