Positioning construction method for tunnel reserved rock ballast channel

Abstract

The invention discloses a positioning construction method for a tunnel reserved rock ballast channel, and belongs to the technical field of tunnel engineering construction. According to the method, before a main tunnel is excavated, a rock ballast channel three-dimensional digital reference is established based on a design model; during construction, a total station, a laser orientation instrument and an inertial navigation system are fused for multi-source cooperative measurement, and the spatial position of a channel is obtained in real time; comparing measured data with a design model through a digital twin platform, dynamically generating a deviation correction instruction, and linking tunneling or supporting equipment to automatically adjust operation parameters; and meanwhile, the channel supporting opportunity and strength are adaptively optimized in combination with surrounding rock deformation monitoring data. According to the method, closed-loop management of measurement, analysis, regulation and verification is achieved, accumulative errors are effectively restrained, the positioning precision and the construction efficiency are improved, and the method is suitable for long and large tunnel engineering with various section forms such as a round section form, a rectangular section form and a horseshoe section form.

Description

Technical Field

[0001] This invention belongs to the field of tunnel construction positioning technology, and specifically relates to a positioning construction method for reserved rockfill channels in tunnels. Background Technology

[0002] In the construction of long tunnels, reserved rockfill channels are crucial temporary structures for ensuring the efficient removal of excavated soil during tunneling. Their positioning accuracy and forming quality directly affect the main tunnel excavation progress, the safety of transport equipment, and the reliability of subsequent sealing construction. Traditional rockfill channel construction often uses manual layout combined with simple guiding devices for positioning, which has significant technical drawbacks: First, relying solely on total stations or laser pointers is easily affected by interference in the dusty, poorly lit tunnel environment, leading to drift of the positioning reference and difficulty in controlling accumulated errors; second, positioning operations are severely disconnected from tunneling and support construction processes, lacking a real-time coordination mechanism, and deviations are often only discovered after formation, resulting in high correction costs; third, facing complex geological conditions such as soft surrounding rock and well-developed joints, existing methods lack the ability to dynamically perceive and respond to surrounding rock deformation, leading to delayed adjustments to support parameters and easily causing local collapses or cross-sectional encroachment of the channel.

[0003] Although some projects have attempted to introduce technologies such as 3D laser scanning or inertial navigation to assist positioning, these efforts have largely remained at the data acquisition level, failing to establish a closed-loop control system encompassing "perception-decision-execution-verification." There is a lack of effective mapping between positioning data and construction equipment commands, and adjustment strategies rely too heavily on human experience, resulting in low levels of automation and intelligence. This makes it difficult to meet the high-precision, high-efficiency, and high-safety construction requirements of long-distance, multi-section (circular, rectangular, horseshoe-shaped) tunnels with rockfill passages. Therefore, a new positioning and construction method that deeply integrates multi-source sensing, intelligent modeling, and collaborative control is urgently needed to address the core bottlenecks of existing technologies in terms of accuracy, efficiency, and adaptability. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a positioning construction method for reserved stone ballast channels in tunnels. The present invention effectively eliminates the error drift of single measurement methods in dusty and low-light environments by using multi-source collaborative measurement and real-time comparison with digital twin models, realizes high-precision dynamic control of the spatial position of stone ballast channels, avoids cross-sectional encroachment or deviation, and ensures reliable connection of the subsequent sealing structure.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for positioning and constructing a pre-reserved rockfill channel in a tunnel, comprising the following steps: S1: Deploy a multimodal sensor array to collect three-dimensional geometric data, surrounding rock physical parameters, and environmental perception data of the entire tunnel area. After spatiotemporal alignment and noise suppression processing, generate a high-precision positioning reference dataset and extract theoretical positioning parameters of the tunnel axis. Simultaneously complete the debugging of construction equipment and calibration of the multimodal sensor array to clarify the collaborative reference for positioning and construction. S2: Based on the high-precision positioning reference dataset, construct a tunnel-channel coupled geometric model, and combine the tunnel cross-section type with the construction process parameters to generate a comprehensive construction plan that includes the excavation boundary, support points and construction path; calculate the global deviation parameters by comparing the measured data with the preset theoretical benchmark in the tunnel-channel coupled geometric model. S3: Start the positioning-construction collaborative operation. The construction equipment performs segmented excavation and support along the positioning axis indicated by the theoretical positioning parameters of the channel axis. Simultaneously, the multimodal sensor array is used to monitor the construction trajectory and surrounding rock deformation in real time. If the positioning deviation or surrounding rock deformation exceeds the preset threshold, the dynamic adjustment mechanism is triggered. S4: Perform differentiated adjustments based on the type of deviation: If it is positioning drift, deviation compensation is achieved by adjusting the attitude parameters of the construction equipment and correcting the theoretical positioning parameters of the channel axis; if it is surrounding rock deformation, deformation development is controlled by optimizing support parameters and slowing down the construction pace, and the high-precision positioning reference dataset is updated simultaneously. S5: After the segmented construction is completed, the constructed channel is reconstructed in three dimensions using laser point cloud registration technology. The reconstruction is then compared with the theoretical positioning parameters of the channel axis to detect the axis offset and cross-sectional flatness quality indicators. If the indicators do not meet the standards, the positioning and construction data are backtracked, the parameters are optimized and adjusted, and the construction is restarted. S6: After the completion of the entire section of construction, carry out final acceptance, integrate positioning data, construction records and quality inspection results to generate technical archives, and establish a positioning benchmark for channel operation and maintenance to provide coordinate reference for subsequent stone debris removal and maintenance.

[0006] As a preferred embodiment, S1 includes: S1.1: Multiple sets of multimodal sensing units are deployed along the tunnel axis. Each set of multimodal sensing units integrates a lidar for acquiring spatial geometric information, an inertial navigation module for acquiring the motion attitude of the equipment, a geological sensor for detecting the physical state of the surrounding rock, and an environmental sensor for sensing environmental conditions. The lidar, the inertial navigation module, the geological sensor, and the environmental sensor synchronously collect three-dimensional point cloud data of the tunnel planning area, real-time attitude data of the construction equipment, data on the density or stress state of the surrounding rock, and environmental parameters such as temperature, humidity, and dust concentration. S1.2: A time synchronization algorithm is used to align the timestamps of sensor data with different sampling frequencies, and the global coordinates of each multimodal sensor unit in the tunnel space are obtained by combining an external absolute positioning system. A unified spatiotemporal reference framework is established to fuse multi-source heterogeneous data into a structured multi-source fusion dataset. S1.3: Analyze the various noise characteristics in the multi-source fusion dataset. For the range jump noise and attitude drift noise output by the inertial navigation module in the lidar point cloud, a combined filtering algorithm consisting of statistical filtering and Kalman filtering is used for multiple rounds of iterative processing to effectively suppress random and systematic noise and output a clean dataset with high signal-to-noise ratio. S1.4: Extract key geometric and geological features reflecting the channel design intent from the clean dataset, including the three-dimensional theoretical coordinate sequence of the channel's central axis, cross-sectional design profile dimensions, surrounding rock stability level or disturbance sensitivity index, to form a structured positioning benchmark dataset; S1.5: Conduct joint no-load and load testing of the tunneling machine and hydraulic support trolley construction equipment, calibrate the mapping relationship between the construction equipment's travel trajectory, the cutter excavation boundary and the positioning reference dataset, and ensure that the construction equipment's actions and positioning commands are consistent in space to meet the construction coordination accuracy requirements.

[0007] As a preferred embodiment, S2 includes: S2.1: Based on the aforementioned positioning benchmark dataset, original tunnel design drawings, and geological survey data, a tunnel-channel coupled geometric model is constructed in the digital modeling platform, which includes geological layering structure, surrounding rock mechanical parameters, channel geometric constraints, and construction technology limitations. The tunnel-channel coupled geometric model clearly expresses the spatial nesting relationship between the channel and the main tunnel. S2.2: Automatically adapt differentiated construction strategies based on tunnel cross-sectional geometry: For circular cross-sections, optimize the excavation angle and support density distribution of the tunnel in the bottom arch area; for rectangular cross-sections, formulate the construction sequence and transition process at the junction of the sidewalls and the bottom slab; for horseshoe-shaped cross-sections, focus on calibrating the spatial matching relationship between the arch curvature and the tunnel entrance to avoid geometric conflicts. S2.3: Generate an executable integrated construction plan based on the tunnel-channel coupled geometric model. The integrated construction plan includes the single-segment excavation advance length, the spacing of the support structure, the travel speed of the construction equipment, the coordinates of the key control points, and the construction path planning. Establish a mapping rule library between positioning parameters and construction parameters. The positioning parameters include the channel axis coordinates and cross-sectional dimensions. The construction parameters include the excavation angle and the timing of support. S2.4: Spatial registration is performed between the measured geometric data on site and the preset theoretical benchmark in the tunnel-channel coupled geometric model. The actual deviation values ​​at each key location are calculated using point cloud or feature point matching algorithms. Global deviation parameters characterizing the overall construction deviation are constructed, and multi-level deviation judgment thresholds are set according to the construction accuracy level. The multi-level deviation judgment thresholds include early warning thresholds, adjustment trigger thresholds, and emergency intervention thresholds.

[0008] As a preferred embodiment, S3 includes: S3.1: The construction equipment performs excavation and initial support operations in segments along the theoretical positioning axis indicated by the theoretical positioning parameters of the channel axis according to the comprehensive construction plan. The lidar continuously scans the boundary contour of the newly excavated area during the movement of the construction equipment and calculates the spatial deviation between the boundary contour of the newly excavated area and the theoretical positioning axis in real time. S3.2: Geological sensors continuously monitor the deformation response of the surrounding rock under excavation disturbance at a fixed sampling frequency. The deformation response includes radial convergence, axial displacement or crack propagation trend. Environmental sensors simultaneously collect temperature, humidity and dust concentration changes. All monitoring data are transmitted in real time to the tunnel-channel coupled geometric model via wired or wireless network for dynamic updating of the current construction section's status assessment. S3.3: After each section of construction of a set length is completed, the system automatically triggers the positioning verification process, compares the geometric features of the actual construction trajectory with the theoretical positioning parameters of the channel axis, and outputs the real-time deviation value of the current section; when the real-time deviation value does not exceed the preset allowable range, the next construction cycle continues; when the real-time deviation value exceeds the adjustment trigger threshold, the operation of the construction equipment is immediately suspended, and the dynamic adjustment mechanism described in S4 is activated.

[0009] As a preferred embodiment, S4 includes: S4.1: Preset independent adjustment thresholds for positioning deviation and surrounding rock deformation, and determine the type of deviation based on the correlation characteristics of multi-source monitoring data: if the deviation is mainly manifested as abnormal posture parameters of construction equipment while the physical parameters of surrounding rock remain stable, it is determined to be positioning drift; if the deviation is accompanied by a significant decrease in the density of surrounding rock, accelerated stress release or non-uniform deformation, it is determined to be surrounding rock deformation. S4.2: When positioning drift is determined, the pitch angle, roll angle and heading angle of the construction equipment are calculated by combining the attitude angle data output by the inertial navigation module with the laser point cloud registration results, and the spatial coordinates of the theoretical positioning parameters of the channel axis in the tunnel-channel coupled geometric model are updated simultaneously, so that the subsequent construction trajectory returns to the theoretical benchmark. S4.3: When it is determined that the surrounding rock is deformed, the support strategy shall be adjusted immediately. The adjusted support strategy includes shortening the spacing of the support structure, using higher strength anchor bolts or shotcrete materials, increasing temporary support measures, and simultaneously reducing the length of single-section excavation to reduce the intensity of construction disturbance. After the deformation rate of the surrounding rock drops to a stable level, the normal construction parameters shall be gradually restored. S4.4: After completing the above adjustment operations, the system continuously collects multiple monitoring data points to verify whether the deviation is consistently stable within the allowable range; if the verification is successful, the construction suspension state is lifted, continuous operation is resumed, and the updated channel axis coordinates and support parameters are written into the high-precision positioning reference dataset to guide subsequent construction.

[0010] As a preferred embodiment, S5 includes: S5.1: After each construction process is completed, a high-precision 3D laser scanner is used to collect point cloud data covering the inner wall of the formed channel. The point cloud registration, noise reduction and stitching algorithms are used to generate a 3D reconstruction model of the channel that reflects the actual geometric shape. S5.2: Perform a geometric comparison between the three-dimensional reconstruction model of the channel and the theoretical positioning parameters of the channel axis extracted in S1, and calculate the core quality indicators. The core quality indicators include the overall offset of the channel center axis relative to the theoretical axis, the local flatness deviation of the cross-sectional profile, and the error between the longitudinal slope of the channel and the design value. S5.3: If all quality indicators meet the preset acceptance standards, the construction of this section is confirmed to be qualified and the next construction process can begin; if any quality indicator exceeds the allowable range, the system will automatically trace back the positioning deviation evolution trajectory, construction equipment attitude adjustment records, support parameter execution logs and surrounding rock monitoring data of the entire construction section, analyze the root cause of the deviation, and optimize the construction equipment attitude adjustment coefficient, support parameter configuration or construction rhythm control logic accordingly. Then, the positioning calibration and construction operation of this section will be re-executed until the quality meets the standards.

[0011] As a preferred embodiment, S6 includes: S6.1: After all construction sections are completed, the overall quality of the passage shall be inspected and accepted in accordance with the engineering acceptance specifications. A systematic sampling inspection method shall be used to conduct physical measurements of key locations of the passage. The sampling points shall cover different geological conditions, cross-sectional types and construction stages to ensure that the acceptance results are representative. S6.2: Integrate multi-source data generated throughout the construction cycle, including initial positioning benchmark dataset, deviation adjustment records for each segment, construction equipment parameter logs, support execution information, and segmented quality inspection reports, to generate structured, traceable, and standardized technical archives; S6.3: Based on the standardized technical archives, extract the core spatial information required for channel operation and maintenance. The core spatial information includes the definition of the channel's global coordinate system, the coordinates of the three-dimensional control points of key sections, the spatial range of key support sections, and the channel's axis control network, and construct a dedicated operation and maintenance positioning benchmark library. The operation and maintenance positioning benchmark library is used to support subsequent stone slag removal vehicles, inspection robots, or maintenance equipment to achieve high-precision automatic navigation and operation positioning through spatial matching.

[0012] As a preferred embodiment, the tunnel cross-section type includes circular, rectangular, or horseshoe-shaped. During the construction process, the tunnel-channel coupled geometric model automatically identifies the geometric features of the tunnel cross-section and generates corresponding channel excavation angle optimization strategies, support structure spatial layout schemes, and construction path obstacle avoidance plans based on the tunnel cross-section topology, ensuring that the channel and the main tunnel structure do not interfere with each other in space and that construction is feasible.

[0013] As a preferred approach, the determination of the deviation type relies on cross-validation of multi-source sensor data: the typical characteristics of positioning drift are abrupt changes in the attitude parameters of the construction equipment while the physical parameters of the surrounding rock do not change significantly, and the deviation exhibits a rigid translation or rotation mode; the typical characteristics of surrounding rock deformation are significant changes in the density, stress, or water content of the surrounding rock, and the channel geometric deviation exhibits a non-uniform, local uplift, or convergent non-rigid mode; the system automatically classifies the source of deviation by comparing the temporal correlation and spatial distribution characteristics of the two types of data.

[0014] As a preferred embodiment, the operation and maintenance positioning reference library stores the spatial reference information of the channel in digital form. The spatial reference information includes the spatial parameter equation of the channel's central axis, the three-dimensional coordinate set of key mileage sections, the spatial bounding box of the support structure, and a subset of feature point clouds for equipment positioning and matching. The operation and maintenance positioning reference library can be called by the stone slag transportation equipment through a wireless network or local interface to realize automatic navigation based on real-time point cloud matching or coordinate interpolation. The positioning accuracy of the automatic navigation is better than that of traditional manual layout or single measuring instrument guidance methods.

[0015] The present invention can achieve the following beneficial effects: 1. This invention effectively eliminates error drift caused by single measurement methods in dusty and low-light environments by comparing multi-source collaborative measurement (total station, laser pointer and inertial navigation) with digital twin models in real time, thereby achieving high-precision dynamic control of the spatial position of the stone slag channel, avoiding cross-sectional encroachment or deviation, and ensuring reliable connection of the subsequent sealing structure.

[0016] 2. This invention organically integrates the sensing, analysis, decision-making and execution processes, and automatically adjusts the operating parameters of the tunneling or support equipment based on real-time deviations, which greatly reduces manual intervention and rework, and improves construction efficiency and the tightness of process connection.

[0017] 3. This invention integrates surrounding rock deformation monitoring data to dynamically optimize support timing and strength, responding promptly to changes in surrounding rock in soft or jointed strata, preventing local collapse of the tunnel, ensuring smooth transportation and operational safety, and is applicable to long tunnel projects with various cross-sectional forms. Attached Figure Description

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is the deviation difference adjustment logic diagram of the present invention; Figure 3 This is a system architecture diagram of the present invention. Detailed Implementation

[0019] This embodiment is applicable to the construction of pre-reserved rockfill channels in tunnels with various cross-sectional types such as circular, rectangular, and horseshoe shapes, and is particularly suitable for engineering scenarios involving long distances and complex geological conditions (such as well-developed surrounding rock fissures and local stability differences). The core equipment configuration is shown in Table 1. Table 1

[0020] Pre-construction preparations: Collect tunnel design drawings and geological survey reports, and clarify the tunnel design parameters (circular cross-section diameter 3m, rectangular cross-section 2m×2.5m, horseshoe cross-section maximum span 3.2m) and surrounding rock classification (I-V). According to the requirements of S1.1, a set of multimodal sensing units shall be deployed every 4 meters along the tunnel axis. The sensor installation height shall be 1.0-1.5m above the tunnel bottom arch. The spatial coordinate calibration of the sensing unit shall be completed by GPS / BeiDou positioning, and the calibration error shall be ≤3mm. The tunneling machine and support trolley were subjected to joint no-load and load testing to calibrate the mapping relationship between the equipment travel trajectory, the cutter excavation boundary and the positioning reference data, and to ensure that the spatial adaptation error between the equipment action and the positioning command is ≤5mm.

[0021] This method is as follows: Figures 1 to 3 The specific implementation steps are shown below: Step S1: Multimodal data acquisition and collaborative calibration: 1. Multi-source data synchronous acquisition: The multi-modal sensor array is activated, the lidar acquires the three-dimensional point cloud of the inner wall of the channel planning area at a sampling rate of 20Hz, the inertial navigation module synchronously acquires attitude data such as pitch angle, roll angle, and heading angle (sampling rate of 100Hz), the geological sensor detects the density and stress state of the surrounding rock at a sampling rate of 5Hz, and the environmental sensor collects temperature, humidity, and dust concentration data in real time.

[0022] 2. Establishment of spatiotemporal reference: A time synchronization algorithm (based on the PTP precision clock protocol) is used to align the timestamps of data with different sampling frequencies, with a synchronization error of ≤1ms; the global coordinates of each sensor unit are obtained through GPS / BeiDou positioning, and a unified tunnel construction coordinate system is established (with the tunnel starting point as the origin, the axial direction as the Z-axis, and the horizontal and vertical directions as the X and Y axes), and the multi-source data is mapped to the same coordinate system to generate a structured multi-source fusion dataset.

[0023] 3. Noise Suppression Processing: For distance jump noise in laser point clouds, statistical filtering is used (outliers are removed if they are more than 3 times the standard deviation); for attitude drift noise in inertial navigation, zero bias estimation and temperature drift correction algorithms are used; finally, a combination algorithm of "statistical filtering + Kalman filtering" is used to perform three rounds of iterative denoising on the multi-source fusion dataset, outputting a clean dataset with a high signal-to-noise ratio and a data noise level ≤0.5mm.

[0024] 4. Positioning reference extraction: Extract key features from the clean dataset: the three-dimensional theoretical coordinate sequence of the channel center axis (one feature point is taken at 0.5m intervals), the cross-sectional design contour dimensions, and the surrounding rock stability level (determined based on density and stress data) to form a positioning reference dataset; where the theoretical positioning accuracy error of the axis is ≤ ±3mm.

[0025] 5. Equipment Co-calibration: Perform load testing on the tunneling machine and record the equipment attitude response at different travel speeds (0.5-1.2m / min) and cutter speeds (30-50r / min); perform anchor bolt installation and shotcrete operation testing on the support trolley, calibrate the deviation between the support point and the positioning benchmark, and ensure that the execution error of the support parameters is ≤50mm.

[0026] Step S2: Coupled Modeling and Construction Scheme Generation: 1. Coupled Geometric Model Construction: Based on the positioning benchmark dataset, tunnel design drawings, and geological survey data, a coupled geometric model of the tunnel-passage is constructed using BIM technology. The model includes the geological layering structure, surrounding rock mechanical parameters, passage geometric constraints, and construction technology limitations, clarifying the spatial nesting relationship between the passage and the main tunnel (e.g., the center of a circular tunnel passage is 0.8m from the bottom arch, and the rectangular tunnel passage is 0.5m from the sidewall).

[0027] 2. Cross-section adaptation construction strategy: The model automatically identifies the tunnel cross-section type and adapts the strategy accordingly. Circular cross-section: The excavation angle of the bottom arch passage is optimized to 35°, the support spacing is 1.0m, and a ring-shaped anchor bolt arrangement is adopted; Rectangular cross section: Establish the construction sequence of "side walls first, then bottom slab", reserve a 50mm transition section at the junction of the side walls and bottom slab, and maintain a support spacing of 0.9m; Horseshoe-shaped cross-section: calibrate the curvature of the arch to match the spatial alignment of the passage entrance, expand the excavation transition section at the passage entrance by 1.0m, and maintain a support spacing of 0.8m.

[0028] 3. Comprehensive Construction Scheme Output: Generate an executable scheme, specifying the excavation advance per section (2m / section for Class I-III surrounding rock, 1m / section for Class IV-V surrounding rock), the spacing of support structures (0.8-1.2m), the equipment travel speed (0.5-1.0m / min), the coordinates of key control points (such as the start and end mileage of each section, turning nodes), and the construction path planning; establish a rule library for mapping positioning parameters to construction parameters, such as maintaining the original tunneling parameters when the axis offset is ≤3mm, and reducing the tunneling speed to 0.5m / min when the offset is 3-5mm.

[0029] 4. Global Deviation Parameter Calculation: The initial geometric data measured on-site (such as the actual tunnel axis and surrounding rock surface profile) are registered with the model's preset theoretical benchmark using point cloud registration. The Iterative Closest Point (ICP) algorithm is used to calculate the deviation values ​​at each key location, constructing global deviation parameters. Three threshold levels are set: early warning threshold (≤3mm), adjustment trigger threshold (3-5mm), and emergency intervention threshold (>5mm).

[0030] Step S3: Positioning - Construction Collaboration and Monitoring: 1. Segmented construction execution: Construction equipment operates along the theoretical positioning axis according to the comprehensive plan. When the tunneling machine is excavating, the lidar continuously scans the newly excavated boundary and calculates the spatial deviation from the theoretical axis in real time. The support trolley follows synchronously and performs anchor bolt installation and shotcrete operation according to the support points and spacing preset by the model.

[0031] 2. Real-time monitoring and data transmission: Geological sensors monitor surrounding rock deformation (radial convergence and axial displacement) at a sampling rate of 5Hz, while environmental sensors simultaneously collect temperature, humidity, and dust concentration. All data are transmitted in real time to the coupled geometric model via industrial Ethernet to dynamically update the status assessment of the construction section (such as the stability level of the surrounding rock and environmental adaptability).

[0032] 3. Positioning Verification and Threshold Determination: After each construction section (1-2m) is completed, the system automatically triggers positioning verification: extracting the geometric features (axis coordinates, cross-sectional dimensions) of the actual construction trajectory, comparing them with the theoretical positioning parameters, and outputting the real-time deviation value. If the deviation is ≤3mm (early warning threshold), construction continues; if it is 3-5mm (adjustment trigger threshold), construction is paused and dynamic adjustment is initiated; if it is >5mm (emergency intervention threshold), the system is immediately stopped for investigation.

[0033] Step S4: Deviation Differentiation Adjustment: 1. Bias Type Determination: Bias type determination based on multi-source data cross-validation: Positioning drift: Sudden changes in equipment attitude parameters (such as sudden changes in pitch angle > 0.05°), with no significant changes in surrounding rock density and stress, and the deviation exhibits a rigid body translation mode; Surrounding rock deformation: The density of the surrounding rock decreases by more than 10%, stress release accelerates, the geometric deviation of the channel shows non-uniform convergence (local uplift or contraction), and the environmental sensors detect abnormal fluctuations in temperature and humidity.

[0034] 2. Positioning Drift Adjustment: Combining inertial navigation attitude data with laser point cloud registration results, calculate the equipment attitude correction amount (single pitch angle and roll angle correction ≤ 0.05°), and synchronously update the channel axis coordinates in the model (e.g., ΔX = 0.003m, ΔY = -0.002m) so that the subsequent construction trajectory returns to the theoretical benchmark.

[0035] 3. Surrounding Rock Deformation Adjustment: Immediately adjust the support strategy: increase the support spacing to 0.5-0.8m, adopt Φ22 high-strength anchor bolts (originally Φ20), C25 shotcrete (originally C20), and add temporary steel supports; reduce the single-section excavation advance to 0.5m / section and reduce the excavation speed to 0.3-0.5m / min. Once three consecutive sets of monitoring data show that the surrounding rock deformation rate is ≤0.1mm / h (stable level), gradually restore normal construction parameters.

[0036] 4. Verification of adjustment effect: After adjustment, collect 5 sets of monitoring data continuously (with an interval of 0.2m construction distance). If the deviation is stable at ≤3mm, the construction suspension is lifted. Write the updated axis coordinates and support parameters into the positioning reference dataset to guide subsequent construction.

[0037] Step S5: Segmented quality inspection and optimization: 1. Segmented 3D Reconstruction: After the completion of each segment's construction, a Faro Focus S350 3D laser scanner was used to perform a full-coverage scan of the inner wall of the passage, collecting point cloud data (density 100 points / cm²). Through point cloud denoising, registration, and stitching algorithms, an actual 3D reconstruction model of the passage was generated, with a model resolution ≤1mm.

[0038] 2. Quality Indicator Inspection: Compare the actual 3D model with the theoretical positioning parameters and calculate the core indicators: Axis offset: Allowable value ≤ ±5mm; Cross-sectional flatness: Allowable value ≤3mm / m; Longitudinal slope error: allowable value ≤ ±0.3%.

[0039] 3. Compliance Judgment and Optimization: If all indicators meet the standards, proceed to the next stage of construction; if indicators exceed the standards (e.g., a 6mm deviation of the axis in a Class IV surrounding rock section), review the data for that section: the evolution trajectory of positioning deviation, equipment attitude adjustment records, support parameter logs, and surrounding rock monitoring data, and analyze the causes (e.g., failure to adjust support in time due to sudden deformation of the surrounding rock). Optimize the adjustment coefficients (e.g., reduce the support spacing to 0.7m and the tunneling speed to 0.4m / min), and re-execute positioning calibration and construction until the indicators meet the standards.

[0040] Step S6: Completion Acceptance and Establishment of Operation and Maintenance Benchmarks: 1. Completion Acceptance Testing: After the entire section of construction is completed, sampling testing will be conducted according to the engineering acceptance specifications. Sampling points will cover different geological conditions, cross-sectional types, and construction stages, with a sampling ratio of no less than 10% of the total length (e.g., 100 sampling points for a 1000m tunnel). Total stations and laser rangefinders will be used to conduct physical measurements of key points, verifying that the pass rate of core indicators is ≥98%.

[0041] 2. Standardized Technical Archive Generation: Integrating full-cycle data: positioning benchmark dataset, deviation adjustment records for each segment, equipment construction parameter logs, support execution information, and segmented quality inspection reports to generate structured technical archives. The archives include text records, 3D model files, and monitoring data curves, allowing traceability of the entire construction process for each segment.

[0042] 3. Construction of the Operation and Maintenance Location Baseline Library: Extracting core operation and maintenance information from technical archives: Spatial parameter equations for the central axis of the passage; Set of three-dimensional coordinates for key mileage sections (every 50m); Space enclosure box for key support sections (Class IV-V surrounding rock sections); Device location matching feature point cloud subset (one group is extracted at 10m intervals).

[0043] The benchmark library is stored in a digital format and can be accessed by stone slag transportation equipment and inspection robots via wireless network. It supports automatic navigation based on real-time point cloud matching with a navigation accuracy of ≤±10mm.

[0044] This embodiment completes the construction of a reserved rockfill channel in a 1200m tunnel (including circular, rectangular, and horseshoe-shaped cross-section sections) through the above steps, achieving the following effect: 1. The axis offset is ≤ ±4mm, the cross-sectional flatness is ≤ 2.8mm / m, and the longitudinal slope error is ≤ ±0.25%, all of which meet the accuracy requirements in the claims. 2. Compared with traditional methods, the efficiency is improved by 30%, with a construction efficiency of 1.0 m / hour for Class I-III surrounding rock sections and 0.7 m / hour for Class IV-V surrounding rock sections; 3. In the deformation section of Class IV-V surrounding rock, by dynamically adjusting the support parameters and construction rhythm, the positioning accuracy was maintained within ±4.5mm, and no rework was required.

[0045] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A method for positioning and constructing a pre-reserved rockfill channel in a tunnel, characterized in that, The positioning and construction method for the reserved rockfill passage in the tunnel includes the following steps: S1: Deploy a multimodal sensor array to collect three-dimensional geometric data, surrounding rock physical parameters, and environmental perception data of the entire tunnel area. After spatiotemporal alignment and noise suppression processing, generate a high-precision positioning reference dataset and extract theoretical positioning parameters of the tunnel axis. Simultaneously complete the debugging of construction equipment and calibration of the multimodal sensor array to clarify the collaborative reference for positioning and construction. S2: Based on the high-precision positioning reference dataset, construct a tunnel-channel coupled geometric model, and combine the tunnel cross-section type with the construction process parameters to generate a comprehensive construction plan that includes the excavation boundary, support points and construction path; calculate the global deviation parameters by comparing the measured data with the preset theoretical benchmark in the tunnel-channel coupled geometric model. S3: Start the positioning-construction collaborative operation. The construction equipment performs segmented excavation and support along the positioning axis indicated by the theoretical positioning parameters of the channel axis. Simultaneously, the multimodal sensor array is used to monitor the construction trajectory and surrounding rock deformation in real time. If the positioning deviation or surrounding rock deformation exceeds the preset threshold, the dynamic adjustment mechanism is triggered. S4: Perform differentiated adjustments based on the type of deviation: If it is positioning drift, deviation compensation is achieved by adjusting the attitude parameters of the construction equipment and correcting the theoretical positioning parameters of the channel axis; if it is surrounding rock deformation, deformation development is controlled by optimizing support parameters and slowing down the construction pace, and the high-precision positioning reference dataset is updated simultaneously. S5: After the segmented construction is completed, the constructed channel is reconstructed in three dimensions using laser point cloud registration technology. The reconstruction is then compared with the theoretical positioning parameters of the channel axis to detect the axis offset and cross-sectional flatness quality indicators. If the indicators do not meet the standards, the positioning and construction data are backtracked, the parameters are optimized and adjusted, and the construction is restarted. S6: After the completion of the entire section of construction, carry out final acceptance, integrate positioning data, construction records and quality inspection results to generate technical archives, and establish a positioning benchmark for channel operation and maintenance to provide coordinate reference for subsequent stone debris removal and maintenance.

2. The positioning and construction method for the reserved rockfill channel in the tunnel according to claim 1, characterized in that, S1 includes: S1.1: Multiple sets of multimodal sensing units are deployed along the tunnel axis. Each set of multimodal sensing units integrates a lidar for acquiring spatial geometric information, an inertial navigation module for acquiring the motion attitude of the equipment, a geological sensor for detecting the physical state of the surrounding rock, and an environmental sensor for sensing environmental conditions. The lidar, the inertial navigation module, the geological sensor, and the environmental sensor synchronously collect three-dimensional point cloud data of the tunnel planning area, real-time attitude data of the construction equipment, data on the density or stress state of the surrounding rock, and environmental parameters such as temperature, humidity, and dust concentration. S1.2: A time synchronization algorithm is used to align the timestamps of sensor data with different sampling frequencies, and the global coordinates of each multimodal sensor unit in the tunnel space are obtained by combining an external absolute positioning system. A unified spatiotemporal reference framework is established to fuse multi-source heterogeneous data into a structured multi-source fusion dataset. S1.3: Analyze the various noise characteristics in the multi-source fusion dataset. For the range jump noise and attitude drift noise output by the inertial navigation module in the lidar point cloud, a combined filtering algorithm consisting of statistical filtering and Kalman filtering is used for multiple rounds of iterative processing to effectively suppress random and systematic noise and output a clean dataset with high signal-to-noise ratio. S1.4: Extract key geometric and geological features reflecting the channel design intent from the clean dataset, including the three-dimensional theoretical coordinate sequence of the channel's central axis, cross-sectional design profile dimensions, surrounding rock stability level or disturbance sensitivity index, to form a structured positioning benchmark dataset; S1.5: Conduct joint no-load and load testing of the tunneling machine and hydraulic support trolley construction equipment, calibrate the mapping relationship between the construction equipment's travel trajectory, the cutter excavation boundary and the positioning reference dataset, and ensure that the construction equipment's actions and positioning commands are consistent in space to meet the construction coordination accuracy requirements.

3. The positioning and construction method for the reserved rockfill channel in the tunnel according to claim 1, characterized in that, S2 includes: S2.1: Based on the aforementioned positioning benchmark dataset, original tunnel design drawings, and geological survey data, a tunnel-channel coupled geometric model is constructed in the digital modeling platform, which includes geological layering structure, surrounding rock mechanical parameters, channel geometric constraints, and construction technology limitations. The tunnel-channel coupled geometric model clearly expresses the spatial nesting relationship between the channel and the main tunnel. S2.2: Automatically adapt differentiated construction strategies based on tunnel cross-sectional geometry: For circular cross-sections, optimize the excavation angle and support density distribution of the tunnel in the bottom arch area; for rectangular cross-sections, formulate the construction sequence and transition process at the junction of the sidewalls and the bottom slab; for horseshoe-shaped cross-sections, focus on calibrating the spatial matching relationship between the arch curvature and the tunnel entrance to avoid geometric conflicts. S2.3: Generate an executable integrated construction plan based on the tunnel-channel coupled geometric model. The integrated construction plan includes the single-segment excavation advance length, the spacing of the support structure, the travel speed of the construction equipment, the coordinates of the key control points, and the construction path planning. Establish a mapping rule library between positioning parameters and construction parameters. The positioning parameters include the channel axis coordinates and cross-sectional dimensions. The construction parameters include the excavation angle and the timing of support. S2.4: Spatial registration is performed between the measured geometric data on site and the preset theoretical benchmark in the tunnel-channel coupled geometric model. The actual deviation values ​​at each key location are calculated using point cloud or feature point matching algorithms. Global deviation parameters characterizing the overall construction deviation are constructed, and multi-level deviation judgment thresholds are set according to the construction accuracy level. The multi-level deviation judgment thresholds include early warning thresholds, adjustment trigger thresholds, and emergency intervention thresholds.

4. The positioning and construction method for the reserved rockfill channel in the tunnel according to claim 1, characterized in that, S3 includes: S3.1: The construction equipment performs excavation and initial support operations in segments along the theoretical positioning axis indicated by the theoretical positioning parameters of the channel axis according to the comprehensive construction plan. The lidar continuously scans the boundary contour of the newly excavated area during the movement of the construction equipment and calculates the spatial deviation between the boundary contour of the newly excavated area and the theoretical positioning axis in real time. S3.2: Geological sensors continuously monitor the deformation response of the surrounding rock under excavation disturbance at a fixed sampling frequency. The deformation response includes radial convergence, axial displacement or crack propagation trend. Environmental sensors simultaneously collect temperature, humidity and dust concentration changes. All monitoring data are transmitted in real time to the tunnel-channel coupled geometric model via wired or wireless network for dynamic updating of the current construction section's status assessment. S3.3: After each section of construction of a set length is completed, the system automatically triggers the positioning verification process, compares the geometric features of the actual construction trajectory with the theoretical positioning parameters of the channel axis, and outputs the real-time deviation value of the current section; when the real-time deviation value does not exceed the preset allowable range, the next construction cycle continues; when the real-time deviation value exceeds the adjustment trigger threshold, the operation of the construction equipment is immediately suspended, and the dynamic adjustment mechanism described in S4 is activated.

5. The positioning and construction method for the reserved rockfill channel in the tunnel according to claim 1, characterized in that, S4 includes: S4.1: Preset independent adjustment thresholds for positioning deviation and surrounding rock deformation, and determine the type of deviation based on the correlation characteristics of multi-source monitoring data: if the deviation is mainly manifested as abnormal posture parameters of construction equipment while the physical parameters of surrounding rock remain stable, it is determined to be positioning drift; if the deviation is accompanied by a significant decrease in the density of surrounding rock, accelerated stress release or non-uniform deformation, it is determined to be surrounding rock deformation. S4.2: When positioning drift is determined, the pitch angle, roll angle and heading angle of the construction equipment are calculated by combining the attitude angle data output by the inertial navigation module with the laser point cloud registration results, and the spatial coordinates of the theoretical positioning parameters of the channel axis in the tunnel-channel coupled geometric model are updated simultaneously, so that the subsequent construction trajectory returns to the theoretical benchmark. S4.3: When it is determined that the surrounding rock is deformed, the support strategy shall be adjusted immediately. The adjusted support strategy includes shortening the spacing of the support structure, using higher strength anchor bolts or shotcrete materials, increasing temporary support measures, and simultaneously reducing the length of single-section excavation to reduce the intensity of construction disturbance. After the deformation rate of the surrounding rock drops to a stable level, the normal construction parameters shall be gradually restored. S4.4: After completing the above adjustment operations, the system continuously collects multiple monitoring data points to verify whether the deviation is consistently stable within the allowable range; if the verification is successful, the construction suspension state is lifted, continuous operation is resumed, and the updated channel axis coordinates and support parameters are written into the high-precision positioning reference dataset to guide subsequent construction.

6. The positioning and construction method for the reserved rockfill channel in the tunnel according to claim 1, characterized in that, S5 includes: S5.1: After each construction process is completed, a high-precision 3D laser scanner is used to collect point cloud data covering the inner wall of the formed channel. The point cloud registration, noise reduction and stitching algorithms are used to generate a 3D reconstruction model of the channel that reflects the actual geometric shape. S5.2: Perform a geometric comparison between the three-dimensional reconstruction model of the channel and the theoretical positioning parameters of the channel axis extracted in S1, and calculate the core quality indicators. The core quality indicators include the overall offset of the channel center axis relative to the theoretical axis, the local flatness deviation of the cross-sectional profile, and the error between the longitudinal slope of the channel and the design value. S5.3: If all quality indicators meet the preset acceptance standards, the construction of this section is confirmed to be qualified and the next construction process can begin; if any quality indicator exceeds the allowable range, the system will automatically trace back the positioning deviation evolution trajectory, construction equipment attitude adjustment records, support parameter execution logs and surrounding rock monitoring data of the entire construction section, analyze the root cause of the deviation, and optimize the construction equipment attitude adjustment coefficient, support parameter configuration or construction rhythm control logic accordingly. Then, the positioning calibration and construction operation of this section will be re-executed until the quality meets the standards.

7. The positioning and construction method for the reserved rockfill channel in the tunnel according to claim 1, characterized in that, S6 includes: S6.1: After all construction sections are completed, the overall quality of the passage shall be inspected and accepted in accordance with the engineering acceptance specifications. A systematic sampling inspection method shall be used to conduct physical measurements of key locations of the passage. The sampling points shall cover different geological conditions, cross-sectional types and construction stages to ensure that the acceptance results are representative. S6.2: Integrate multi-source data generated throughout the construction cycle, including initial positioning benchmark dataset, deviation adjustment records for each segment, construction equipment parameter logs, support execution information, and segmented quality inspection reports, to generate structured, traceable, and standardized technical archives; S6.3: Based on the standardized technical archives, extract the core spatial information required for channel operation and maintenance. The core spatial information includes the definition of the channel's global coordinate system, the coordinates of the three-dimensional control points of key sections, the spatial range of key support sections, and the channel's axis control network, and construct a dedicated operation and maintenance positioning benchmark library. The operation and maintenance positioning benchmark library is used to support subsequent stone slag removal vehicles, inspection robots, or maintenance equipment to achieve high-precision automatic navigation and operation positioning through spatial matching.

8. The method for positioning and constructing a pre-reserved rockfill passage in a tunnel according to any one of claims 1 to 7, characterized in that, The tunnel cross-section type includes circular, rectangular or horseshoe shape. During the construction process, the tunnel-channel coupled geometric model automatically identifies the geometric features of the tunnel cross-section and generates corresponding channel excavation angle optimization strategies, support structure spatial layout schemes and construction path obstacle avoidance plans based on the tunnel cross-section topology, ensuring that the channel and the main tunnel structure do not interfere with each other in space and that construction is feasible.

9. The method for positioning and constructing a pre-reserved rockfill passage in a tunnel according to any one of claims 1 to 7, characterized in that, The determination of the deviation type relies on cross-validation of multi-source sensor data: the typical characteristics of positioning drift are abrupt changes in the attitude parameters of the construction equipment while the physical parameters of the surrounding rock do not change significantly, and the deviation is in the form of rigid body translation or rotation; the typical characteristics of surrounding rock deformation are significant changes in the density, stress or water content of the surrounding rock, and the channel geometric deviation is in the form of non-uniform, local uplift or convergence non-rigid body mode; the system automatically classifies the source of deviation by comparing the temporal correlation and spatial distribution characteristics of the two types of data.

10. The method for positioning and constructing a pre-reserved rockfill passage in a tunnel according to any one of claims 1 to 7, characterized in that, The operation and maintenance positioning reference library stores the spatial reference information of the channel in digital form. The spatial reference information includes the spatial parameter equation of the channel's central axis, the three-dimensional coordinate set of key mileage sections, the spatial bounding box of the support structure, and the feature point cloud subset used for equipment positioning and matching. The operation and maintenance positioning reference library can be called by stone slag transportation equipment through wireless network or local interface to realize automatic navigation based on real-time point cloud matching or coordinate interpolation. The positioning accuracy of the automatic navigation is better than that of traditional manual layout or single measuring instrument guidance.

Images