Micro-disturbance construction measurement and control system across subway underground works

Abstract

The application discloses a cross-metro underground engineering micro-disturbance construction measurement and control system and relates to the field of construction safety management.The system comprises a laser shape measurement module, a collection module and a model.The laser shape measurement module is used for controlling laser measurement devices arranged in the contour of an existing metro structure and an excavation disturbance influence area, and is used for acquiring a multi-point three-dimensional deformation displacement sequence of a stratum, support and existing structure in real time.The collection module is used for synchronously collecting real-time axial force values of adjustable supports in an excavation area, seepage pressure values of a seepage prevention curtain and stress deformation values of a pile foundation protection structure, and is used for generating a construction state parameter set.The application is coupled with a model through monitoring data, recursively predicts subsequent deformation, generates a regulation and control instruction according to the prediction and a limit value, feeds back an execution effect to correct the model, adjusts support axial force, grouting and pile foundation sequences, takes into account deformation rate and cumulative amount, and makes parameters self-adapt to working conditions, so as to guarantee safety, reduce shutdown and reinforcement, and improve construction continuity.

Description

Technical Field

[0001] This invention relates to the field of construction safety management technology, specifically to a micro-disturbance construction monitoring and control system for underground engineering projects spanning subway lines. Background Technology

[0002] With urban underground space development reaching saturation, it has become commonplace for new underground projects to be constructed adjacent to existing subway lines. The stress release and ground deformation caused by construction inevitably transfer to nearby subway structures, potentially inducing segment cracking or joint opening. Micro-disturbance construction technology emphasizes ensuring subway safety by controlling construction speed and deformation thresholds. Currently, laser measurement methods such as 3D laser scanning and automatic total stations are used for real-time monitoring of tunnel deformation and support displacement; adjustable steel supports, interlocking pile seepage prevention walls, and pile foundation isolation protection are also widely used in close-proximity construction.

[0003] The invention patent application with application number 202410962285.0 discloses a system and method for monitoring, predicting, and warning the safety of construction of structures close to existing subway lines. This application comprehensively improves the monitoring, prediction, and warning capabilities during the construction process of structures close to subway lines by integrating a distributed automated monitoring scheme, a hybrid heuristic algorithm, finite element numerical simulation analysis, and a multi-level early warning mechanism. This ensures the safety and stability of the construction process and meets the relevant requirements for full-process construction safety monitoring and linkage deformation control, including the direct contact surfaces on both sides and directly above the tunnel. It can comprehensively guarantee the construction safety, structural safety, and operational safety during the construction process of structures close to existing subway lines.

[0004] In this type of construction scenario, the existing monitoring data and numerical simulations are mostly offline and fragmented applications for the spatiotemporal superposition of disturbances in multiple processes (excavation, support replacement, and pile foundation construction). It is difficult to match the disturbance model with real-time measured data to dynamically control the process, resulting in insufficient precision in soil deformation control and a persistently high risk of structural damage. Summary of the Invention

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a micro-disturbance construction monitoring and control system for underground engineering across subway lines, which can effectively solve the problems of the existing technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions;

[0007] This invention discloses a micro-disturbance construction monitoring and control system for underground engineering projects spanning subway lines, comprising:

[0008] The laser shape measurement module is used to control the laser measurement devices deployed in the existing subway structure outline and the area affected by excavation disturbance to acquire multi-point three-dimensional deformation displacement sequences of the strata, support and existing structure in real time.

[0009] The data acquisition module is used to simultaneously acquire the real-time axial force value of the adjustable support in the excavation area, the seepage pressure value of the anti-seepage curtain, and the stress and deformation of the pile foundation protection structure, and generate a set of construction status parameters.

[0010] The modeling module is used to take the three-dimensional deformation displacement sequence and construction state parameter set as input to construct and correct online a numerical disturbance model that reflects the spatiotemporal superposition effect of multiple processes of excavation, support replacement and pile foundation construction.

[0011] The simulation module is used to call the numerical disturbance model to calculate the process data of the deformation of the existing subway structure caused by the next construction sequence as the process steps accumulate.

[0012] The decision module is used to compare process data with preset deformation control limits and generate adaptive control instructions. The adaptive control instructions are used to adjust the axial force target of the adjustable support in the excavation area, trigger the reinforcement action of the seepage prevention curtain, and rearrange the operation sequence of the pile foundation construction.

[0013] The verification module is used to distribute the adaptive control instructions to the execution terminals corresponding to the construction of adjustable supports, seepage prevention curtains and pile foundations within the excavation area, and to collect construction execution status data and send it to the modeling module to correct the numerical disturbance model.

[0014] The laser shape measurement module is interconnected with a data acquisition module and a modeling module via a local area network. The data acquisition module and the modeling module are interconnected via a local area network. The modeling module is interconnected with a deduction module and a decision-making module via a local area network. The deduction module is interconnected with the decision-making module via a local area network. The modeling module and the decision-making module are interconnected with a verification module via a local area network.

[0015] Furthermore, the laser shape measurement module includes multiple transverse measurement sections arranged at equal intervals along the longitudinal extension direction of the existing subway structure. In each transverse measurement section, laser reflective targets are fixedly installed on the surface of the surrounding rock at the arch top, arch bottom, side walls, and the boundary of the excavation disturbance area of ​​the existing subway structure.

[0016] The laser measuring device operates at a scanning frequency. Pulsed laser beams are sequentially emitted towards each of the laser-reflecting targets, and echo signals are received. The time difference between the pulse emission time and the echo reception time is used to determine the optimal timing. Calculate the spatial distance of each target relative to the laser measurement device. The initial spatial coordinates of each target deployment and the change in spatial distance measured in each scan Generate a deformation displacement sequence for each target in three-dimensional space, wherein the deformation displacement sequence contains the deformation displacement sequence of each target at each scan time. The three-dimensional displacement components.

[0017] Furthermore, the acquisition module includes several acquisition sub-units respectively deployed corresponding to the adjustable support, the seepage prevention curtain, and the pile foundation protection structure;

[0018] The acquisition module triggers each acquisition subunit to perform the same sampling at the same time using a synchronous clock signal. Parallel acquisition of real-time axial force values ​​of adjustable supports The seepage pressure value of the seepage barrier curtain and the stress deformation of the pile foundation protection structure in the horizontal and vertical directions. ;

[0019] The acquisition module has a built-in data cleaning unit, which sequentially performs median filtering and gross error removal on the acquired real-time axial force values, seepage pressure values, and stress deformation values, and arranges the cleaned valid data in the order of sampling time to form the construction state parameter set. , where the subscript j represents the j-th synchronous sampling period;

[0020] When performing median filtering for denoising, for each type of collected parameter sequence, a sliding filter window of length L is set, where L is an odd number not less than 5. The values ​​of each sampling time in the window are sorted in ascending order, and the value in the middle position after sorting is taken as the filtered output value of the center time of the window. The window is slid point by point with a step size of 1 to traverse all sampling times and obtain the denoised parameter sequence.

[0021] Gross error removal is performed based on the Laida criterion: For each class of parameter sequences after median filtering, the arithmetic mean of the sequence is first calculated. And the standard deviation s, determine each data point one by one. Does it meet the requirements? If the condition is met, the data point is identified as a gross error and removed. Then, the arithmetic mean and standard deviation are recalculated for the remaining data points, and the above identification and removal process is repeated until all remaining data points meet the condition. The remaining data points will be output as valid data.

[0022] Furthermore, the modeling module includes an initial model building unit and an online correction unit;

[0023] The initial model building unit establishes an initial three-dimensional finite element numerical model covering the excavation area, existing subway structure and surrounding rock medium based on the engineering geological survey report and construction design documents, and sets the time sequence of excavation, support replacement and pile foundation construction.

[0024] The online correction unit uses the deformation displacement sequence and construction state parameter set obtained in each synchronous sampling period as measured boundary conditions to construct the control equations of the numerical disturbance model:

[0025] ;

[0026] The geotechnical parameter vector is then iteratively updated using the following formula:

[0027] ;

[0028] In the formula: K(p) is a vector of soil and rock parameters to be corrected. The overall stiffness matrix; u is the calculated displacement field vector; F is the equivalent nodal load vector obtained by mapping based on the construction state parameter set; J is the calculated displacement sensitivity matrix to soil and rock parameters; β is the damping factor; I is the identity matrix; This is the measured displacement vector; This is the calculated displacement vector under the current parameters;

[0029] The online correction unit iterates repeatedly until... Less than the preset threshold The Euclidean norm is used to represent the updated parameters. The updated parameters are then substituted into the control equations to complete the model correction, resulting in the corrected numerical disturbance model.

[0030] Furthermore, the deduction module extracts the construction parameters of each of the k process steps included in the predetermined work plan for the next construction sequence as the input excitation of the numerical disturbance model, and iteratively calculates the estimated cumulative deformation at each target position on the existing subway structure after each process step is completed:

[0031] ;

[0032] In the formula: The current moment; The time of completion of the k-th process step; M is the total number of actions in the excavation and support replacement process; Let m be the spatial influence coefficient of the m-th action; Let m be the displacement discontinuity caused by the m-th action within the surrounding rock. Let be the spatiotemporal decay factor for the m-th action; is the start time of the m-th action; Q is the total number of hammer blows or static pressure segments for a single pile penetration in the pile foundation construction process; Let q be the unit load disturbance coefficient for the qth penetration. This represents the increment of additional stress in the soil around the pile caused by the qth penetration. Let q be the duration of the q-th penetration. The consolidation characteristic time of the surrounding rock;

[0033] The simulation module arranges the cumulative deformation estimates calculated at each target location in the order of the process steps to form the process data.

[0034] Furthermore, the decision module compares the cumulative deformation estimate of each target position after each process step in the process data with the preset deformation control limit point by point. If the cumulative deformation estimate of any target position exceeds the preset deformation control limit, the adjustment command generation process is triggered.

[0035] The control instruction generation process determines the target control amount for each execution terminal according to the following formula:

[0036] ;

[0037] In the formula: To set the target value of the parameter to be adjusted on the execution terminal; The current parameter value of the execution terminal; The preset deformation control limit value corresponding to the target location where the limit is exceeded; This is the estimated cumulative deformation of the target location under the current process step; This represents the sensitivity gradient of the cumulative deformation to the control parameters of the execution terminal in the numerical perturbation model. This is the estimated cumulative deformation of the target location after the completion of the previous process step; This is an over-limit trend correction factor;

[0038] The decision module generates the adaptive control instruction based on the calculated target control amount of each execution terminal. The adaptive control instruction includes at least the target axial force value of the adjustable support, the pressure and grout volume parameters of the anti-seepage curtain grouting reinforcement, and the operation sequence identifier of each pile position in the pile foundation construction.

[0039] Furthermore, the verification module includes an instruction distribution unit and an execution status recovery unit;

[0040] The instruction distribution unit encodes the adaptive control instructions into the corresponding control signal format according to the type of the execution terminal, and sends them to the adjustable support hydraulic servo valve group, the grouting pump group of the seepage prevention curtain, and the pile driver controller for pile foundation construction, respectively.

[0041] After the adaptive control command is completed, the execution status recovery unit collects the execution status data fed back by each execution terminal in the next synchronous sampling cycle. The execution status data includes the measured axial force value after the adjustable support is executed, the measured seepage pressure value after the anti-seepage curtain is executed, and the measured stress deformation after the pile foundation construction is executed.

[0042] Furthermore, during the decision-making module's operational phase, the simulation module simultaneously performs a risk classification assessment on the cumulative deformation prediction values ​​for each target location. This risk classification assessment is based on early warning indicators. Make a judgment:

[0043] ;

[0044] In the formula: R is the total number of targets that require risk assessment; This represents the estimated cumulative deformation of the r-th target position after the completion of the current process step. This is the preset deformation control limit value corresponding to the r-th target position; It is a non-linear sensitivity index; The time-varying rate of the cumulative deformation estimate at the r-th target position; Let be the critical deformation rate at the r-th target position;

[0045] when When the first threshold is exceeded, the early warning module sends a signal to the decision-making module to accelerate the generation process of the control command; when When the second threshold is exceeded, the early warning module outputs a work stoppage alarm to the construction site. The second threshold is greater than the first threshold, and both thresholds are based on user-defined settings on the system side.

[0046] Furthermore, the system uses a synchronous sampling period The process is executed sequentially at cyclic intervals: data acquisition by the acquisition module, model correction by the modeling module, process data calculation by the inference module, risk classification assessment by the early warning module, and adaptive control instruction generation by the decision-making module.

[0047] The verification module completes instruction distribution and execution status collection in the next cycle interval, and feeds back the collected execution status data to the model correction in the next cycle interval to form a multi-level hierarchical closed-loop control link.

[0048] The synchronous sampling period Based on the geological characteristic frequency of the excavation area Confirmed, satisfied .

[0049] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0050] This invention dynamically couples real-time acquired three-dimensional deformation with construction status parameters, and uses an online-corrected numerical model to recursively predict structural deformation caused by subsequent processes. Then, it generates control instructions based on the comparison between the prediction results and the limit values, and feeds the execution effect back to the model to form a closed loop. This system can implement measured-driven dynamic control of the stratum and structural response under multiple overlapping processes.

[0051] Meanwhile, by continuously updating the model parameters, the predictions are synchronized with the site conditions, thereby adjusting the axial force of the support, the timing of grouting, and the sequence of pile foundations. This avoids the limitations of static indicators or empirical formulas, and takes into account the deformation rate and cumulative amount, while also considering the safety margin of different parts. The construction parameters are adaptively corrected in the rolling process of each procedure, thereby reducing downtime and reinforcement while ensuring safety, and improving the continuity and safety of construction in complex environments. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0053] Figure 1 A schematic diagram of the micro-disturbance construction monitoring and control system for underground engineering projects spanning subway lines;

[0054] Figure 2 This is a geometric schematic diagram of the three-dimensional finite element numerical model of the subway crossing project in this invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0056] The present invention will be further described below with reference to embodiments.

[0057] Example:

[0058] The micro-disturbance construction monitoring and control system for underground engineering projects spanning subway lines in this embodiment, such as Figure 1 As shown, it includes:

[0059] The laser shape measurement module is used to control the laser measurement devices deployed in the existing subway structure outline and the area affected by excavation disturbance to acquire multi-point three-dimensional deformation displacement sequences of the strata, support and existing structure in real time.

[0060] The laser shape measurement module includes multiple transverse measurement sections arranged at equal intervals along the longitudinal extension direction of the existing subway structure. In each transverse measurement section, laser reflective targets are fixedly installed on the surface of the surrounding rock at the arch top, arch bottom, side walls, and the boundary of the excavation disturbance area of ​​the existing subway structure.

[0061] The spacing between adjacent transverse measurement sections is determined comprehensively based on the segment ring width of the existing subway structure and the geological conditions. For shield tunnels, the measurement section spacing is 1 to 3 times the segment ring width, and does not exceed 3m; for cut-and-cover structures, the measurement section spacing is 2m to 5m.

[0062] In each transverse measurement section, one laser reflective target is set up at the arch top, arch bottom and two side walls of the existing subway structure, for a total of 4 targets; at least 3 laser reflective targets are evenly set up along the cross section direction on the surface of the surrounding rock at the boundary of the excavation disturbance area.

[0063] When there is a sudden change in geological conditions in the excavation area or when there is recorded structural damage to the existing subway structure, the spacing between the measurement sections will be increased to 50% of the lower limit of the above spacing in the abrupt or damaged sections.

[0064] The total number of measurement sections shall not be less than three, to ensure that sufficient deformation distribution information is obtained along the longitudinal direction;

[0065] Laser measuring device with scanning frequency Pulsed laser beams are sequentially emitted towards each laser-reflecting target, and echo signals are received. The time difference between the pulse emission time and the echo reception time is used to determine the optimal timing. Calculate the spatial distance of each target relative to the laser measurement device. The initial spatial coordinates of each target deployment and the change in spatial distance measured in each scan Generate a deformation displacement sequence for each target in three-dimensional space. The deformation displacement sequence contains the deformation displacement sequence of each target at each scan time. The three-dimensional displacement components;

[0066] Among them, scanning frequency The sampling frequency should be dynamically adjusted according to the principle of being positively correlated with the excavation advance rate and the pile impact frequency, and should not be lower than the minimum sampling frequency required to capture the transient deformation response of the soil.

[0067] Scan frequency Determined according to the following rules:

[0068] ,in This represents the current excavation advance rate (unit: m / h). The impact frequency (unit: Hz) during pile foundation construction. Use 0.5~2.0 Hz·h / m. Take a value of 0.3 to 1.0. The minimum sampling frequency required to capture the transient deformation response of soil. Based on the shear wave velocity of the surrounding rock Equivalent characteristic size of the excavation disturbance area according to Calculate, and Not lower than 0.2Hz;

[0069] When the excavation advance rate changes or the pile foundation impact frequency changes, the system recalculates and adjusts the scanning frequency according to the above formula in the next synchronous sampling cycle;

[0070] The data acquisition module is used to simultaneously acquire the real-time axial force value of the adjustable support in the excavation area, the seepage pressure value of the anti-seepage curtain, and the stress and deformation of the pile foundation protection structure, and generate a set of construction status parameters.

[0071] The data acquisition module includes several data acquisition sub-units that are respectively deployed to correspond to the adjustable support, the seepage prevention curtain, and the pile foundation protection structure;

[0072] The acquisition module triggers each acquisition subunit to perform the same sampling at the same time using a synchronous clock signal. Parallel acquisition of real-time axial force values ​​of adjustable supports The seepage pressure value of the seepage barrier curtain and the stress deformation of the pile foundation protection structure in the horizontal and vertical directions. ;

[0073] The acquisition module has a built-in data cleaning unit, which performs median filtering and gross error removal on the acquired real-time axial force, seepage pressure, and stress deformation values. The cleaned valid data are then arranged in the order of sampling time to form a set of construction status parameters. , where the subscript j represents the j-th synchronous sampling period;

[0074] When performing median filtering for denoising, for each type of collected parameter sequence, a sliding filter window of length L is set, where L is an odd number not less than 5. The values ​​of each sampling time in the window are sorted in ascending order, and the value in the middle position after sorting is taken as the filtered output value of the center time of the window. The window is slid point by point with a step size of 1 to traverse all sampling times and obtain the denoised parameter sequence.

[0075] Specifically, for the real-time axial force value sequence of adjustable supports, the sliding filter window length L is 7; for the seepage pressure value sequence of the seepage prevention curtain, L is 5; and for the stress deformation value sequence of the pile foundation protection structure, L is 9.

[0076] The filter window length for each parameter sequence is fixed at the above values ​​during system initialization. If the number of sampling points for a certain type of parameter sequence is less than the corresponding window length, the window length for that type of parameter is adjusted to the actual number of sampling points (which must be an odd number not less than 3; if the actual number of sampling points is even, then decrease by 1).

[0077] Within three consecutive sampling periods following a significant change in the construction status (such as the excavation face crossing the stratum interface or the pile foundation construction entering a new stratum), the filter window length of the corresponding parameter sequence is temporarily shortened to 50% of the original value (taking an odd number not less than 3) to enhance the tracking sensitivity of abrupt signals.

[0078] Gross error removal is performed based on the Laida criterion: For each class of parameter sequences after median filtering, the arithmetic mean of the sequence is first calculated. And the standard deviation s, determine each data point one by one. Does it meet the requirements? If the condition is met, the data point is identified as a gross error and removed. Then, the arithmetic mean and standard deviation are recalculated for the remaining data points, and the above identification and removal process is repeated until all remaining data points meet the condition. The remaining data points will be output as valid data.

[0079] The iteration termination condition for gross error removal is as follows:

[0080] The iteration is terminated when no new gross error data points are detected in two consecutive iterations.

[0081] If the number of remaining data points after iterative elimination is less than 60% of the original number of data points, the data collected in this batch is deemed unreliable as a whole, all data in this batch is discarded, and the data is collected again in the next sampling period.

[0082] If the number of remaining data points is between 60% and 80% of the original number of data points, then the median of the remaining data points is used to replace the value at the position of the removed data point as the valid data output.

[0083] If the number of remaining data points exceeds 80% of the original number of data points, the remaining data points will be used directly as valid data output. If the proportion of abandoned data exceeds 10% of the total data collected by all acquisition sub-units in the same sampling period, the system will output a data anomaly alarm signal to the monitoring terminal.

[0084] The modeling module is used to take the three-dimensional deformation displacement sequence and construction state parameter set as input to construct and correct online a numerical disturbance model that reflects the spatiotemporal superposition effect of multiple processes of excavation, support replacement and pile foundation construction.

[0085] The modeling module includes an initial model building unit and an online correction unit;

[0086] The initial model building unit establishes an initial three-dimensional finite element numerical model covering the excavation area, existing subway structure and surrounding rock medium based on the engineering geological survey report and construction design documents, and sets the time sequence of excavation, support replacement and pile foundation construction.

[0087] The online correction unit uses the deformation displacement sequence and construction state parameter set acquired in each synchronous sampling period as measured boundary conditions to construct the governing equations of the numerical disturbance model:

[0088] ;

[0089] This formula uses the finite element stiffness balance relationship to associate the surrounding rock parameters with the structural displacement, transforms the multi-source measured construction parameters into time-varying equivalent nodal loads, and replaces the traditional fixed parameter and static load modeling method. Based on the field monitoring data, it solves the subway structure displacement field under the superposition of multiple processes in real time, providing the basic control equations for online inversion and correction of the model.

[0090] The geotechnical parameter vector is then iteratively updated using the following formula:

[0091] ;

[0092] In the formula: K(p) is a vector of soil and rock parameters to be corrected. The overall stiffness matrix; u is the calculated displacement field vector; F is the equivalent nodal load vector obtained by mapping based on the construction state parameter set; J is the calculated displacement sensitivity matrix to soil and rock parameters; β is the damping factor; I is the identity matrix; This is the measured displacement vector; This is the calculated displacement vector under the current parameters;

[0093] Among them, the real-time axial force value of the adjustable support Mapping to equivalent nodal loads according to the following rules: Determine the coordinates of the contact nodes between each adjustable support and the surrounding rock or support structure in the numerical disturbance model, and map the axial force values... A concentrated force is applied along the axial direction of the support at the contact node, with the direction of the support under compression being positive. Each support is mapped independently, forming an axial force load sub-vector. ;

[0094] seepage pressure value of the seepage barrier Map according to the following rules: Locate the spatial plane where the seepage barrier is located in the numerical perturbation model, and map the seepage pressure value. The surface force is applied perpendicularly to the curtain surface at each element node on the surface, with the magnitude of the surface force being uniformly distributed, forming a sub-vector of seepage pressure load. ;

[0095] Stress deformation of pile foundation protective structure It is not used as a direct load boundary condition, but rather as a verification basis for model correction, and does not participate in the mapping of the equivalent nodal load vector. Equivalent nodal load vector F= + The two sub-vectors are directly superimposed on their corresponding degrees of freedom. The mapping is performed once before the model correction is completed in each synchronous sampling cycle. The node coordinates and orientation parameters used for the mapping are automatically extracted from the grid information of the current numerical perturbation model.

[0096] The above formula is based on Gauss-Newton inverse analysis with the addition of a damping regularization term. It relies on the sensitivity matrix to reflect the influence of soil and rock parameters on deformation, and uses the difference between measured and calculated displacement to drive parameter iteration. The damping factor is adaptively adjusted with the residual, and the identity matrix avoids matrix singularity. It adapts to the dynamic evolution characteristics of soil parameters and completes continuous self-calibration of the model.

[0097] The online correction unit iterates repeatedly until... Less than the preset threshold The Euclidean norm is used to represent the updated parameters. The updated parameters are then substituted into the control equations to complete the model correction and obtain the corrected numerical perturbation model.

[0098] In the initial three-dimensional finite element numerical model, based on the stratigraphic information and geophysical parameters in the engineering geological survey report, and the geometric dimensions and support layout scheme of the existing subway structure in the construction design documents, the geometry of the surrounding rock medium, the existing subway structure, and the foundation pit support structure are established in three-dimensional space. Each geometry is assigned a corresponding elastic modulus, Poisson's ratio, density, and initial ground stress field. Tetrahedral or hexahedral elements are used for meshing. Normal displacement constraints are applied to the model boundary, and corresponding unloading and loading boundary conditions are applied to the excavation face and pile foundation location. After the initial ground stress balance is achieved, the action sequence of excavation, support replacement, and pile foundation penetration is generated according to the construction organization design, thereby establishing the initial three-dimensional finite element numerical model and setting the trigger time and duration of each process.

[0099] The material parameters of each geometric body are extracted and assigned values ​​from the engineering geological survey report according to the following rules:

[0100] The elastic modulus and Poisson's ratio of the surrounding rock medium are taken as the arithmetic mean of the indoor geotechnical test results of each soil layer in the exploration report, and the density is taken as the weighted average of the in-situ test results.

[0101] For model areas where no exploration boreholes have been laid out, the soil parameters corresponding to the three boreholes closest to the area are determined by weighted interpolation based on the inverse distance.

[0102] The initial geostress field is determined by superimposing the self-weight stress field and the tectonic stress field. The vertical principal stress is calculated based on the self-weight of the overlying soil layer, and the horizontal principal stress is determined by multiplying the vertical principal stress by the lateral pressure coefficient. The lateral pressure coefficient is assigned layer by layer based on the test results of the static lateral pressure coefficient of each soil layer in the exploration report; when test data is unavailable, it is calculated as follows: ( Estimate the internal friction angle of the soil layer;

[0103] The elastic modulus and Poisson's ratio of the existing subway structure are determined based on the concrete grade and reinforcement ratio in the structural design documents.

[0104] This represents the elastic modulus of the surrounding rock. Represents Poisson's ratio. This represents the geostress balance coefficient. Based on the measured geostress data in the engineering geological survey report, the initial value is the ratio of the average horizontal principal stress to the average vertical principal stress in the model area. This value is then used for optimization and adjustment along with the soil and rock parameter vector during the iterative update of the online correction unit.

[0105] The overall stiffness matrix K(p) is determined by the initial three-dimensional finite element numerical model through element discretization, element stiffness integration, assembly according to nodal degree of freedom numbering and displacement constraint conditions, and is recalculated as p is updated;

[0106] u through the control equation Given the current geotechnical parameter vector p and the current construction state parameter set The equivalent nodal load vector F is obtained by numerical solution (using preprocessed conjugate gradient method or direct sparse solver). This solution process is executed in each iteration of the online correction element to obtain the calculated displacement values ​​at each node in the global domain under the current parameters, and the calculated displacement components corresponding to the positions of each laser reflection target are extracted to form the matrix. ;

[0107] F is formed by mapping the excavation unloading volume, the measured value of the support axial force, the measured value of the seepage pressure, and the pile penetration load to the corresponding nodes and then superimposing them;

[0108] Each element of the matrix in J In the formula, i is the displacement component index of each target position. Let be the calculated displacement component at the i-th target position. Let l be the l-th parameter in the soil and rock parameter vector p. This matrix is ​​obtained by processing each parameter in the soil and rock parameter vector p. Apply a preset perturbation amount The governing equations were then solved again using the difference scheme. Approximate determination, wherein the preset perturbation amount Take it as 0.1% to 1.0% of the current value of the parameter (for example, take 0.5% of the current value for the elastic modulus E, and Poisson's ratio). Since it is a dimensionless decimal, it is taken as 0.5% of the current value (the geostress balance coefficient λ is taken as 0.5% of the current value), and the parameters after perturbation are guaranteed. Within the physically reasonable range of this parameter (such as Poisson's ratio) Located between 0 and 0.5, this ratio range can be pre-calibrated according to the surrounding rock type and parameter sensitivity, and fixed as the default perturbation step size of each parameter during system initialization to ensure that the difference approximation has sufficient numerical accuracy and does not produce calculation oscillations;

[0109] The above formula obtains the partial derivative of displacement with respect to soil parameters using the finite difference method. After applying a small perturbation to each parameter within the appropriate physical range, the displacement is recalculated. The sensitivity matrix elements are obtained through the difference ratio, eliminating the need for complex analytical derivation. This reduces computational overhead while ensuring calculation accuracy and accurately quantifies the influence weight of each parameter on the deformation of the subway measuring point.

[0110] The initial value of β is preset to be one-thousandth of the sum of the squares of the initial magnitudes of the components of the soil and rock parameter vector. During the iteration process, it is determined based on the ratio of the residual norms of two consecutive iterations. Dynamic adjustment, if If <0.25, then β will be reduced to 0.5 times its current value. If the value is greater than 0.75, then β will be increased to twice its current value; otherwise, β will remain unchanged.

[0111] I and the geotechnical parameter vector pp have the same dimension, i.e., a dimension of 3×3, with diagonal elements being 1 and off-diagonal elements being 0. This is used in matrix inversion operations. Full rank is reversible;

[0112] The simulation module is used to call the numerical disturbance model to calculate the process data of the deformation of the existing subway structure caused by the next construction sequence as the process steps accumulate.

[0113] For the k steps included in the pre-planned operation scheme of the next construction sequence, the deduction module extracts the construction parameters of each step as the input excitation of the numerical disturbance model, and iteratively calculates the estimated cumulative deformation at each target position on the existing subway structure after the completion of each step.

[0114] ;

[0115] In the formula: The current moment; The time of completion of the k-th process step; M is the total number of actions in the excavation and support replacement process; Let m be the spatial influence coefficient of the m-th action; Let m be the displacement discontinuity caused by the m-th action within the surrounding rock. Let be the spatiotemporal decay factor for the m-th action; is the start time of the m-th action; Q is the total number of hammer blows or static pressure segments for a single pile penetration in the pile foundation construction process; is the unit load disturbance coefficient for the qth penetration, and its physical dimension is the ratio of the displacement dimension to the stress dimension, which characterizes the instantaneous elastic displacement response of the soil around the pile caused by a unit increase in additional stress. This represents the increment of additional stress in the soil around the pile caused by the qth penetration. Let q be the duration of the q-th penetration. The consolidation characteristic time of the surrounding rock;

[0116] The above formula calculates the deformation separately for two types of disturbances: excavation unloading and pile penetration. The integral term represents the instantaneous elastic decay deformation during excavation, and the summation term reflects the soil consolidation creep caused by the pile foundation. It uniformly superimposes the spatiotemporal disturbance effects of multiple processes, which is different from the limitations of single-process prediction formulas and fully restores the entire process of cumulative deformation of existing subway structures.

[0117] The simulation module arranges the cumulative deformation estimates calculated at each target location in the order of process steps to form process data;

[0118] in, Using the numerical perturbation model as a baseline platform, a unit displacement discontinuity is applied at the boundary node corresponding to the m-th action. And extract the response displacement generated at each target location. ,but When there are multiple target locations, Take the weighted average of all target response displacements, with the weights determined by the reciprocal of the spatial distance between the target and the action center;

[0119] Spatial distance refers to the three-dimensional Euclidean distance between the target position and the center point of the action;

[0120] When multiple excavation or support replacement actions occur simultaneously, the spatial influence coefficient of each action is... The coefficient, obtained by linear superposition, serves as the comprehensive spatial influence coefficient at that moment. The superposition formula is as follows: However, the combined spatial impact coefficient after superposition does not exceed 1.0;

[0121] Let m be the displacement discontinuity caused by the m-th action within the surrounding rock. For the excavation action, The instantaneous unloading displacement at the center point of the excavation face is obtained by the numerical disturbance model after applying the excavation unloading boundary conditions; for the support replacement action... Take the displacement difference at the point of application of the support before and after its removal;

[0122] The determination method is as follows: In the numerical disturbance model, a unit additional stress (1 kPa) is applied at the pile penetration location, the displacement response at each target location around the pile is calculated, and the average value of the displacement response of each target is taken as the mean value. Before each penetration, the parameters of the soil around the pile are recalculated. ;

[0123] The increment of additional stress in the soil around the pile caused by the qth penetration is calculated using the Mindlin stress solution: ,in q represents the hammer impact energy or static pressure load of the qth penetration (unit: kN), h represents the current burial depth of the pile tip (unit: m), and z represents the burial depth of the target position (unit: m). These represent the spatial distances (in meters) from the target location to the pile tip and to the mirror image point at the top of the pile. The Poisson's ratio of the surrounding rock;

[0124] according to Calculation, where The formation energy dissipation coefficient is obtained by converting the equivalent damping ratio of the surrounding rock medium in the numerical perturbation model. For the shear wave velocity of the surrounding rock, Let be the spatial distance between the center point of the m-th action and the corresponding target position. For uniformly distributed unloading actions without a clearly defined center point... Take the absolute value of the difference between the equivalent radius of the excavation face and the target burial depth;

[0125] according to Confirmed, among which The maximum drainage path length of the saturated soil within the influence zone of the seepage barrier curtain is determined by the relative relationship between the depth of the seepage barrier curtain and the groundwater level in the excavation area. Let be the consolidation coefficient of saturated soil, and , The soil permeability coefficient, The soil compression modulus, The specific gravity of water;

[0126] The decision module is used to compare process data with preset deformation control limits and generate adaptive control instructions. The adaptive control instructions are used to adjust the axial force target of adjustable supports in the excavation area, trigger the reinforcement action of the seepage prevention curtain, and rearrange the operation sequence of pile foundation construction.

[0127] The decision module compares the cumulative deformation estimate of each target position after each process step in the process data with the preset deformation control limit point by point. If the cumulative deformation estimate of any target position exceeds the preset deformation control limit, the control instruction generation process is triggered.

[0128] The control instruction generation process determines the target control amount for each execution terminal according to the following formula:

[0129] ;

[0130] In the formula: To set the target value of the parameter to be adjusted on the execution terminal; The current parameter value of the execution terminal; The preset deformation control limit value corresponding to the target location where the limit is exceeded; This is the estimated cumulative deformation of the target location under the current process step; This represents the sensitivity gradient of the cumulative deformation to the control parameters of the execution terminal in the numerical perturbation model. This is the estimated cumulative deformation of the target location after the completion of the previous process step; This is an over-limit trend correction factor;

[0131] Among them, sensitivity gradient The finite difference method is used for calculation:

[0132] For the axial force control parameters of the adjustable support, the current axial force value is used in the numerical disturbance model. and ( Using 0.01 as the boundary condition, two forward calculations were performed, and the cumulative deformation value at the target location was extracted. ;

[0133] The same finite difference scheme is used for the grouting pressure of the seepage prevention curtain and the construction sequence parameters of the pile foundation. The sensitivity gradient is determined by taking values ​​of 0.02 and the change in pile location number, respectively. The sensitivity gradient is recalculated after each model correction and remains constant within the current synchronous sampling period until the next model correction is completed.

[0134] The above formula combines the deformation limit deviation and deformation sensitivity to obtain the basic control amount. The control amplitude is dynamically adjusted by superimposing the deformation rate correction term. When the deformation approaches the limit, the control is increased; when the deformation falls back, the control is slightly adjusted. This changes the fixed amplitude control mode and can curb the risk of excessive deformation of the subway structure in advance.

[0135] The decision module generates adaptive control instructions based on the calculated target control values ​​of each execution terminal. The adaptive control instructions include at least the target axial force value of the adjustable support, the pressure and grout volume parameters of the anti-seepage curtain grouting reinforcement, and the operation sequence identifier of each pile position in the pile foundation construction.

[0136] The work sequence for each pile location during pile foundation construction is rearranged according to the following risk priority rules:

[0137] For each pile location involved in the next construction sequence, taking that pile location as the current unique construction action, the simulation module is invoked to calculate the estimated cumulative deformation at each target location. And calculate the risk impact factor for each pile location. , where R is the total number of targets and α is the nonlinear sensitivity index;

[0138] All pile locations to be constructed are arranged as follows The values ​​are sorted in descending order of magnitude and used as the new priority order for the tasks. When two stake positions... When the values ​​differ by no more than 5%, priority is given to the target position that is farther away from the existing subway structure.

[0139] If there are already pile positions under construction but not yet completed in the current process, the current operation status of the pile position will remain unchanged and will not be involved in the reordering. The operation sequence identifiers of the remaining pile positions that have not yet started will be regenerated according to the above rules.

[0140] The preset deformation control limit is determined in advance based on the service safety standards and operation and maintenance levels of the existing subway structure. For the r-th target position, the preset deformation control limit is... according to Settings, in which This is the allowable deformation benchmark value at the target location, determined based on existing subway structural design drawings and structural durability requirements. This is a safety margin reduction factor. The value is taken in the range of 0.60 to 0.85, and for the target positions located on the arch and side walls of the existing subway structure, Take the lower limit of this range for the target position located at the bottom of the arch. Take the upper limit of this range, and at the same time, consider the existing subway structure's operational lifespan and historical structural health monitoring data. Dynamic corrections will be made whenever the operational lifespan increases by five years or the cumulative deformation rate in the historical structural health monitoring data exceeds a preset rate threshold. Multiply the current value by a regression factor of 0.95, and the corrected value is... Not less than 0.5;

[0141] , Take the difference between the completion time of the current process step and the completion time of the previous process step. The reference deformation rate is predetermined for the corresponding target location. The reference deformation rate is determined by the allowable deformation rate of the existing subway structure. The positive or negative sign indicates the direction of the deformation trend under the current process step. If the cumulative deformation rate increases towards the preset deformation control limit, then... If the cumulative deformation rate deviates from the preset deformation control limit, then the adjustment range is increased to a positive value. A negative value would cause the target adjustment amount to be moderately withdrawn;

[0142] The above formula uses the ratio of the deformation rate at the process interval to the baseline rate to obtain the trend correction base value. By trunculating the correction factor range with upper and lower limits, the control intensity is amplified when the deformation rate exceeds the standard, and the adjustment amount is reduced when the deformation is reversed, so as to prevent the control parameters from fluctuating drastically and ensure that the construction adjustment is stable without secondary disturbance.

[0143] The verification module is used to distribute adaptive control instructions to the execution terminals corresponding to the construction of adjustable supports, seepage prevention curtains and pile foundations within the excavation area, and to collect construction execution status data and send it to the modeling module to correct the numerical disturbance model.

[0144] The verification module includes an instruction distribution unit and an execution status recovery unit;

[0145] The instruction distribution unit encodes the adaptive control instructions into the corresponding control signal format according to the type of the execution terminal, and sends them to the adjustable support hydraulic servo valve group, the grouting pump group of the seepage prevention curtain, and the pile driver controller for pile foundation construction, respectively.

[0146] After the adaptive control command is completed, the execution status recovery unit collects the execution status data fed back by each execution terminal in the next synchronous sampling cycle. The execution status data includes the measured axial force value after the adjustable support is executed, the measured seepage pressure value after the anti-seepage curtain is executed, and the measured stress deformation after the pile foundation construction is executed.

[0147] The execution status recovery unit sends the execution status data along with the deformation displacement sequence at the corresponding time to the online correction unit of the modeling module as input data for the next round of numerical perturbation model correction.

[0148] During the decision-making module's operational phase, the simulation module simultaneously assesses the risk level of the cumulative deformation estimate at each target location, based on early warning indicators. Make a judgment:

[0149] ;

[0150] In the formula: R is the total number of targets that require risk assessment; This represents the estimated cumulative deformation of the r-th target position after the completion of the current process step. This is the preset deformation control limit value corresponding to the r-th target position; It is a non-linear sensitivity index; The time-varying rate of the cumulative deformation estimate at the r-th target position; Let be the critical deformation rate at the r-th target position;

[0151] The above formula integrates the relative proportion of deformation at each measuring point with the real-time deformation rate to construct a global risk indicator. The nonlinear exponent amplifies the risk of points near the limit value, and the arch sidewall is configured with a higher sensitive weight. It simultaneously distinguishes between slow accumulation and rapid abrupt deformation, providing a unified quantitative basis for graded early warning, accelerated control, and work stoppage alarm.

[0152] when When the first threshold is exceeded, the early warning module sends a signal to the decision-making module to accelerate the generation process of control instructions; when When the second threshold is exceeded, the early warning module outputs a work stoppage alarm to the construction site. The second threshold is greater than the first threshold, and both thresholds are based on user-defined settings on the system side.

[0153] Preferably, the default value of the first threshold is 1.0, and the default value of the second threshold is 2.5;

[0154] The correspondence between the first and second thresholds and the safety levels of existing subway structures is as follows:

[0155] For sections with a safety level of Level 1 (severe structural damage consequences), the first threshold is 0.7 and the second threshold is 1.8; for sections with a safety level of Level 2, the first threshold is 1.0 and the second threshold is 2.5; for sections with a safety level of Level 3 (relatively minor structural damage consequences), the first threshold is 1.3 and the second threshold is 3.2. The safety level is determined comprehensively based on the design service life of the existing subway structure, the current technical condition assessment level, and the safety management regulations of the operating unit.

[0156] Users can view the above default values ​​and corresponding security levels in the system initialization interface, and manually adjust them within ±30% of the default values ​​according to the actual project situation;

[0157] in, ,in The structural safety level coefficient is used for the arch and side walls of existing subway structures. Take 1.5, for the arch base position, Take 0.8, and The value range is limited to 1.0 to 2.5;

[0158] The above formula sets differentiated safety level coefficients according to the structural parts of the subway, amplifies the risk weight of weak areas, increases the sensitivity index nonlinearly with the increase of deformation ratio and limits the value range, strengthens the identification ability of key measuring points near the limit, and improves the early warning and prediction effect.

[0159] The system uses a synchronous sampling period The process is executed sequentially at cyclic intervals: data acquisition by the acquisition module, model correction by the modeling module, process data calculation by the inference module, risk classification assessment by the early warning module, and adaptive control instruction generation by the decision-making module.

[0160] The verification module completes instruction distribution and execution status collection in the next cycle interval, and feeds back the collected execution status data to the model correction in the next cycle interval to form a multi-level hierarchical closed-loop control link.

[0161] Synchronous sampling period Based on the geological characteristic frequency of the excavation area Confirmed, satisfied Among them, the characteristic frequencies of the formation It is calculated from the ratio of the shear wave velocity of the surrounding rock medium to the equivalent characteristic size of the excavation area in the numerical perturbation model;

[0162] Equivalent feature size of excavation area Determine as follows:

[0163] For rectangular excavation areas Take the geometric mean of the maximum horizontal dimension and the maximum vertical dimension of the excavated area, that is... ,in The maximum span of the excavated area on the horizontal plane (unit: m). Vertical depth of the excavation area (unit: m);

[0164] For circular or near-circular excavation areas Determine the excavation diameter;

[0165] When the excavation area contains multiple interconnected sub-regions, Take the weighted average of the equivalent feature sizes of each sub-region, with the weights determined according to the volume proportion of each sub-region;

[0166] Formation characteristic frequencies ,in, The shear wave velocity of the surrounding rock medium (unit: m / s) is determined by the material parameters of the surrounding rock medium in the numerical disturbance model. The shear wave velocity of each soil layer within the excavation influence range is taken as the thickness-weighted average value.

[0167] The laser shape measurement module is interconnected with the acquisition module and the modeling module via a local area network. The acquisition module and the modeling module are interconnected via a local area network. The modeling module is interconnected with the deduction module and the decision module via a local area network. The modeling module and the decision module are interconnected via a local area network. A verification module is also interconnected via a local area network.

[0168] In this embodiment, the system uses the synchronous sampling cycle as the cyclic benchmark. At the beginning of each cycle, the laser shape measurement module and the acquisition module start in parallel. The laser shape measurement module sets up multiple transverse measurement sections along the longitudinal direction of the existing subway structure. Reflective targets are fixed at the arch top, arch bottom, side walls, and surrounding rock surface of each section. The three-dimensional deformation displacement sequence of each target is obtained through pulsed laser ranging. The acquisition module triggers each acquisition subunit with a synchronous clock and acquires the real-time axial force of the adjustable support, the seepage pressure of the seepage prevention curtain, and the stress and deformation of the pile foundation protection structure in parallel. After data cleaning, the data is integrated into a construction state parameter set. The two types of data are acquired strictly synchronously in time, providing high-fidelity input for subsequent model correction.

[0169] The modeling module uses the measured deformation displacement sequence and the construction state parameter set as boundary conditions, and substitutes them into the initial three-dimensional finite element model pre-established based on the geological survey report and construction design. The online correction unit continuously adjusts the elastic modulus, Poisson's ratio and geostress balance coefficient of the surrounding rock through iterative calculation, so that the model's calculated displacement gradually approaches the measured displacement until the residual converges, completing the online correction of the numerical disturbance model. After this dynamic calibration, the model's calculation parameters are synchronized with the actual working conditions, improving the accuracy of subsequent predictions.

[0170] The simulation module takes over the revised model and extracts the construction parameters of each step in the next construction sequence as input excitation. It iteratively calculates the cumulative deformation estimate of each target position of the existing subway structure under the spatiotemporal superposition effect of multiple actions such as excavation unloading, support replacement, and pile penetration. The data is arranged in the order of the procedures to form process data. The decision module compares these estimates with the preset deformation control limits of each target point by point. Once an exceedance is found, the target control amount of each execution terminal is calculated based on the sensitivity of the cumulative deformation to the control parameters of the execution terminal and the current deformation exceedance trend. An adaptive control instruction is generated, which includes the support axial force target, curtain grouting reinforcement parameters, and pile foundation operation sequence. At the same time, the simulation module performs risk classification assessment on each target position. The risk level is determined by the deformation ratio and the rate of change. When the risk index exceeds the first threshold, the control process is accelerated. When it exceeds the second threshold, a work stoppage alarm is output to the construction site, adding an active defense line for construction safety.

[0171] The verification module then converts the adaptive control instructions into control signals corresponding to each execution terminal, and sends them to the hydraulic servo valve group of the adjustable support, the grouting pump group of the seepage prevention curtain, and the pile driver controller of the pile foundation construction. After execution, the verification module collects the execution status data such as the measured axial force, seepage pressure, and stress deformation from each terminal in the next synchronous sampling cycle, and sends them back to the modeling module along with the deformation displacement sequence collected at the same time for the next round of model correction. The entire process is based on the synchronous sampling cycle, and from data acquisition, model update, advanced simulation, risk assessment, control decision to execution feedback, the links are interlocked to form a closed control link, ensuring that the system maintains continuous perception, prediction and control capability for the deformation of the existing subway structure throughout the construction process.

[0172] See Figure 2 As shown in the figure, the numerical disturbance model is built and corrected online in the modeling module. The model includes layered surrounding rock, existing subway lining, excavation area, support, seepage prevention curtain and pile foundation protection structure. The model is divided into grids and ground stress and displacement constraints are applied. The geometric points correspond to the laser targets on site. The model can be connected to measured displacement, support axial force and seepage pressure data to iteratively calibrate the soil and rock parameters. After calibration, the model provides a simulation base for the deduction module to calculate the cumulative deformation of the subway structure under the superposition of multiple processes, and supports the back-end risk assessment and construction control calculation.

[0173] In summary, this system dynamically couples real-time acquired three-dimensional deformation with construction status parameters, and uses an online-corrected numerical model to recursively predict structural deformations caused by subsequent processes. It then generates control commands based on the comparison between the prediction results and limits, and feeds the execution effect back to the model to form a closed loop. This system can implement measured-driven dynamic control of the stratum and structural response under multiple overlapping processes. Simultaneously, by continuously updating model parameters, it synchronizes predictions with the actual site conditions, thereby adjusting the support axial force, grouting timing, and pile sequence. This avoids the limitations of static indicators or empirical formulas, and incorporates deformation rate and cumulative amount, taking into account the safety margins of different parts. It adaptively corrects construction parameters during the rolling process, thereby reducing downtime and reinforcement while ensuring safety, and improving the continuity and safety of construction in complex environments.

[0174] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A micro-disturbance construction monitoring and control system for underground engineering projects spanning subway lines, characterized in that, include: The laser shape measurement module is used to control the laser measurement devices deployed in the existing subway structure outline and the area affected by excavation disturbance to acquire multi-point three-dimensional deformation displacement sequences of the strata, support and existing structure in real time. The data acquisition module is used to simultaneously acquire the real-time axial force value of the adjustable support in the excavation area, the seepage pressure value of the anti-seepage curtain, and the stress and deformation of the pile foundation protection structure, and generate a set of construction status parameters. The modeling module is used to take the three-dimensional deformation displacement sequence and construction state parameter set as input to construct and correct online a numerical disturbance model that reflects the spatiotemporal superposition effect of multiple processes of excavation, support replacement and pile foundation construction. The simulation module is used to call the numerical disturbance model to calculate the process data of the deformation of the existing subway structure caused by the next construction sequence as the process steps accumulate. The decision module is used to compare process data with preset deformation control limits and generate adaptive control instructions. The adaptive control instructions are used to adjust the axial force target of the adjustable support in the excavation area, trigger the reinforcement action of the seepage prevention curtain, and rearrange the operation sequence of the pile foundation construction. The verification module is used to distribute the adaptive control instructions to the execution terminals corresponding to the construction of adjustable supports, seepage prevention curtains and pile foundations within the excavation area, and to collect construction execution status data and send it to the modeling module to correct the numerical disturbance model. The laser shape measurement module includes multiple transverse measurement sections arranged at equal intervals along the longitudinal extension direction of the existing subway structure. In each transverse measurement section, laser reflective targets are fixedly installed on the surface of the surrounding rock at the arch top, arch bottom, side walls, and the boundary of the excavation disturbance area of ​​the existing subway structure. The laser measuring device operates at a scanning frequency. Pulsed laser beams are sequentially emitted towards each of the laser-reflecting targets, and echo signals are received. The time difference between the pulse emission time and the echo reception time is used to determine the optimal timing. Calculate the spatial distance of each target relative to the laser measurement device. The initial spatial coordinates of each target deployment and the change in spatial distance measured in each scan Generate a deformation displacement sequence for each target in three-dimensional space, wherein the deformation displacement sequence contains the deformation displacement sequence of each target at each scan time. The three-dimensional displacement components; The simulation module extracts the construction parameters of each of the k steps in the pre-planned operation scheme for the next construction sequence as the input excitation of the numerical disturbance model, and iteratively calculates the estimated cumulative deformation at each target position on the existing subway structure after each step is completed. ； In the formula: The current moment; The time of completion of the k-th process step; M is the total number of actions in the excavation and support replacement process; Let m be the spatial influence coefficient of the m-th action; Let m be the displacement discontinuity caused by the m-th action within the surrounding rock. Let be the spatiotemporal decay factor for the m-th action; is the start time of the m-th action; Q is the total number of hammer blows or static pressure segments for a single pile penetration in the pile foundation construction process; Let be the unit load disturbance coefficient for the qth penetration. This represents the increment of additional stress in the soil around the pile caused by the qth penetration. Let q be the duration of the q-th penetration. The consolidation characteristic time of the surrounding rock; The simulation module arranges the cumulative deformation estimates calculated at each target location in the order of process steps to form the process data. The spacing between adjacent transverse measurement sections is determined comprehensively based on the segment ring width of the existing subway structure and the geological conditions. For shield tunnels, the measurement section spacing is 1 to 3 times the segment ring width, and does not exceed 3m; for cut-and-cover structures, the measurement section spacing is 2m to 5m.

2. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, The acquisition module includes several acquisition sub-units respectively deployed corresponding to the adjustable support, the seepage prevention curtain and the pile foundation protection structure; The acquisition module triggers each acquisition subunit to perform the same sampling at the same time using a synchronous clock signal. Parallel acquisition of real-time axial force values ​​of adjustable supports The seepage pressure value of the seepage barrier curtain and the stress deformation of the pile foundation protection structure in the horizontal and vertical directions. ; The acquisition module has a built-in data cleaning unit, which sequentially performs median filtering and gross error removal on the acquired real-time axial force values, seepage pressure values, and stress deformation values, and arranges the cleaned valid data in the order of sampling time to form the construction state parameter set. , where the subscript j represents the j-th synchronous sampling period; When performing median filtering for denoising, for each type of collected parameter sequence, a sliding filter window of length L is set, where L is an odd number not less than 5. The values ​​of each sampling time in the window are arranged in ascending order, and the value in the middle position after sorting is taken as the filtered output value of the center time of the window. The window is slid point by point with a step size of 1 to traverse all sampling times and obtain the denoised parameter sequence. Gross error removal is performed based on the Laida criterion: For each class of parameter sequences after median filtering, the arithmetic mean of the sequence is first calculated. And the standard deviation s, determine each data point one by one. Does it meet the requirements? If the condition is met, the data point is identified as a gross error and removed. Then, the arithmetic mean and standard deviation are recalculated for the remaining data points, and the above identification and removal process is repeated until all remaining data points meet the condition. The remaining data points will be output as valid data.

3. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, The modeling module includes an initial model building unit and an online correction unit; The initial model building unit establishes an initial three-dimensional finite element numerical model covering the excavation area, existing subway structure and surrounding rock medium based on the engineering geological survey report and construction design documents, and sets the time sequence of excavation, support replacement and pile foundation construction. The online correction unit uses the deformation displacement sequence and construction state parameter set obtained in each synchronous sampling period as measured boundary conditions to construct the control equations of the numerical disturbance model: ； The geotechnical parameter vector is then iteratively updated using the following formula: ； In the formula: K(p) is a vector of soil and rock parameters to be corrected. The overall stiffness matrix; u is the vector for calculating the displacement field; F is the equivalent nodal load vector obtained by mapping based on the construction state parameter set; J is the sensitivity matrix of displacement to soil and rock parameters. β is the damping factor; I is the identity matrix; This is the measured displacement vector; This is the calculated displacement vector under the current parameters; The online correction unit iterates repeatedly until... Less than the preset threshold The Euclidean norm is used to represent the updated parameters. The updated parameters are then substituted into the control equations to complete the model correction, resulting in the corrected numerical disturbance model.

4. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, The decision module compares the cumulative deformation estimate of each target position after each process step in the process data with the preset deformation control limit point by point. If the cumulative deformation estimate of any target position exceeds the preset deformation control limit, the control instruction generation process is triggered. The control instruction generation process determines the target control amount for each execution terminal according to the following formula: ； In the formula: To set the target value of the parameter to be adjusted on the execution terminal; The current parameter value of the execution terminal; The preset deformation control limit value corresponding to the target location where the limit is exceeded; This is the estimated cumulative deformation of the target location under the current process step; This represents the sensitivity gradient of the cumulative deformation to the control parameters of the execution terminal in the numerical perturbation model. This is the estimated cumulative deformation of the target location after the completion of the previous process step; This is an over-limit trend correction factor; The decision module generates the adaptive control instruction based on the calculated target control amount of each execution terminal. The adaptive control instruction includes at least the target axial force value of the adjustable support, the pressure and grout volume parameters of the anti-seepage curtain grouting reinforcement, and the operation sequence identifier of each pile position in the pile foundation construction.

5. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, The verification module includes an instruction distribution unit and an execution status recovery unit; The instruction distribution unit encodes the adaptive control instructions into the corresponding control signal format according to the type of the execution terminal, and sends them to the adjustable support hydraulic servo valve group, the grouting pump group of the seepage prevention curtain, and the pile driver controller for pile foundation construction, respectively. After the adaptive control command is completed, the execution status recovery unit collects the execution status data fed back by each execution terminal in the next synchronous sampling cycle. The execution status data includes the measured axial force value after the adjustable support is executed, the measured seepage pressure value after the anti-seepage curtain is executed, and the measured stress deformation after the pile foundation construction is executed.

6. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, During the decision-making module's operation phase, the simulation module simultaneously performs risk classification assessment on the cumulative deformation prediction values ​​at each target location. This risk classification assessment is based on early warning indicators. Make a judgment: ； In the formula: R is the total number of targets that require risk assessment; This represents the estimated cumulative deformation of the r-th target position after the completion of the current process step. This is the preset deformation control limit value corresponding to the r-th target position; It is a non-linear sensitivity index; The time-varying rate of the cumulative deformation estimate at the r-th target position; Let be the critical deformation rate at the r-th target position; when When the first threshold is exceeded, the early warning module sends a signal to the decision-making module to accelerate the generation process of the control command; when When the second threshold is exceeded, the early warning module outputs a work stoppage alarm to the construction site. The second threshold is greater than the first threshold, and both thresholds are based on user-defined settings on the system side.

7. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, The system uses a synchronous sampling period The process is executed sequentially at cyclic intervals: data acquisition by the acquisition module, model correction by the modeling module, process data calculation by the inference module, risk classification assessment by the early warning module, and adaptive control instruction generation by the decision-making module. The verification module completes instruction distribution and execution status collection in the next cycle interval, and feeds back the collected execution status data to the model correction in the next cycle interval to form a multi-level hierarchical closed-loop control link. The synchronous sampling period Based on the geological characteristic frequency of the excavation area Confirmed, satisfied .

8. The micro-disturbance construction monitoring and control system for underground engineering projects across subway lines according to claim 1, characterized in that, The laser shape measurement module is interconnected with a data acquisition module and a modeling module via a local area network. The data acquisition module and the modeling module are interconnected via a local area network. The modeling module is interconnected with a deduction module and a decision-making module via a local area network. The deduction module is interconnected with the decision-making module via a local area network. The modeling module and the decision-making module are interconnected with a verification module via a local area network.

Images