Three-dimensional settlement-crack monitoring method and system for static cutting connection area of subway exit

Abstract

The invention belongs to the technical field of urban rail transit civil engineering monitoring, and particularly relates to a subway exit static cutting connection area three-dimensional settlement-crack monitoring method and system. Three groups of monitoring point arrays are arranged, and a total station and a hydrostatic level double link are adopted to collect data; geometric stability check is performed based on redundant data of an external reference group, and a reference system credible mark is generated in combination with double-link cross-chain consistency check; taking a construction event as a time anchor point, matching the main anchor point and the auxiliary anchor point and combining with a linear time mapping model to realize time alignment and generate a corresponding credible mark; and triggering a synchronous analysis window by double marks, judging a collaborative deformation event by combining crack and settlement index space-time correlation characteristics, and outputting a label with an evidence chain. According to the method, the data defect of a single link is made up, the problem of time asynchronization of multiple acquisition ends is solved, the deformation causal relationship is precisely discriminated, a complete evidence chain is formed, the monitoring reliability and preciseness are improved, and precise support is provided for connection area construction safety management and control and risk early warning.

Description

Technical Field

[0001] This invention belongs to the field of urban rail transit civil engineering monitoring technology, specifically relating to a three-dimensional settlement-crack monitoring method and system for the static cutting connection area of ​​a subway exit. Background Technology

[0002] As a crucial node connecting the main structure and ancillary facilities, the structural stability of the subway exit connection area directly affects the safety of subway operations and the surrounding environment. Under the continuous action of static cutting, connection construction, dewatering, and recharge processes, the stress state of the connection area structure will undergo rapid and complex changes, making it highly susceptible to the risk of local differential settlement and crack propagation. Therefore, it is necessary to accurately monitor and promptly capture the structural response.

[0003] In engineering practice, existing monitoring schemes generally adopt a combination of multiple devices and threshold early warning. That is, three-dimensional displacement data are obtained by using an automated total station, continuous elevation changes are collected by static leveling, and crack width is monitored by crack gauge. Then, safety early warnings are triggered based on preset thresholds or simple data trends to achieve preliminary control over the structural status.

[0004] However, the static cutting and connection area at the subway exit is characterized by its narrow space, frequent obstructions, high frequency of measurement point disassembly and assembly, and strong process disturbances. Existing monitoring and interpretation methods are difficult to adapt to the complex needs of this scenario, resulting in three major technical defects that directly lead to insufficient reliability of monitoring results and a high risk of misjudgment. First, the reference frame drifts, causing data distortion. Existing technologies lack an effective self-calibration mechanism for the reference frame. Benchmark points are easily affected by construction vibrations and precipitation, resulting in co-settlement phenomena. The control network may also experience geometric micro-deformation. At the same time, factors such as loose installation of measurement points, aging of single-link equipment, and environmental interference can also cause drift in single-point data or link data, ultimately leading to problems such as "false stability," "false settlement," and "false uplift" in three-dimensional displacement and settlement data, which seriously mislead on-site risk assessment. Second, mismatched timestamps from multiple sources disrupt the coordination relationship. Different monitoring devices and construction event acquisition modules all rely on independent gateways to control clocks. Each gateway not only has initial offsets caused by factory calibration deviations or lack of clock synchronization during installation, but also experiences operational drift due to the accumulation of clock crystal oscillator errors over time. This, coupled with differences in sampling frequencies between different devices (e.g., total stations observe once every 30 minutes, hydrostatic levels sample once every 5 minutes) and communication delays, disrupts the temporal causal relationship between "settlement-cracks," creating "false coupling" (asynchronous changes are judged as coordination risks) or "false decoupling" (synchronous changes are judged as irrelevant), ultimately leading to false alarms, missed alarms, or delayed warnings. Third, the lack of an auditable chain of evidence makes it difficult to verify and trace conclusions. The output results of existing monitoring systems are mostly binary judgments of "abnormal" or "normal" or simple threshold trigger prompts, without recording key quantitative evidence such as the reliability of the reference system, time alignment quality, and cross-chain consistency. When disputes arise regarding warnings, it is impossible to trace the entire data processing process, making it difficult to verify the rationality of the conclusions and failing to meet the auditability requirements of engineering safety monitoring.

[0005] The aforementioned three technical defects collectively reduce the repeatability and interpretability of collaborative monitoring results, making it difficult to meet the high-precision safety monitoring requirements under the conditions of strong disturbance construction in the subway exit connection area. Summary of the Invention

[0006] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution: Firstly, a three-dimensional settlement-crack monitoring method for the static cutting connection area of ​​a subway exit is proposed, including the following steps: An array of monitoring points, including an external stability benchmark group, an on-site working benchmark group, and a structural response group located on the joint or load-bearing component, is set up in the static cutting and connection area of ​​the subway exit; a total station three-dimensional observation link and a static leveling elevation observation link are established to collect data from the monitoring point array through dual links; Geometric stability is checked based on redundant observation data from an external stable benchmark group, and cross-chain consistency is checked for the same monitoring point in both the three-dimensional observation link and the static leveling elevation observation link for elevation increments. When both the geometric stability check index and the cross-chain consistency check index meet the first preset condition determined based on observation uncertainty, a reference system credibility flag is generated. Data on construction events related to structural disturbances are collected, and the occurrence time of the construction events is extracted as the time anchor point. The local timestamp sequences of different collection terminals are mapped to a unified time axis through a time mapping model. When the fitting residual of the time mapping model satisfies the second preset condition determined based on the sampling interval, a time alignment confidence flag is generated. A synchronous analysis window is generated on a unified time axis if and only if both the reference frame confidence flag and the time alignment confidence flag are true. Within the synchronous analysis window, the crack increment index and settlement difference index of the structural response group are calculated. Based on the spatiotemporal correlation characteristics of the crack increment index and settlement difference index, it is determined whether a real coordinated deformation event has occurred, and an evidence chain label containing reference frame check data and time alignment parameters is output.

[0007] Secondly, a three-dimensional settlement-crack monitoring system for the static cutting connection area of ​​a subway exit is proposed, including: The reference system self-calibration module is used to perform geometric stability checks on the external stable benchmark group, as well as cross-link consistency checks between the total station link and the hydrostatic leveling link, and outputs a reference system credibility flag. The construction event anchor point alignment module is used to collect construction events and extract time anchor points, establish a time mapping model to map multi-source data to a unified time axis, and output a time alignment confidence flag based on the fitting residual. The dual-gate collaborative interpretation module is used to generate a synchronous analysis window and calculate crack increment index and settlement difference index to determine collaborative deformation events when both the reference system credibility flag and the time alignment credibility flag are true. The evidence chain archiving and output module is used to store evidence chain metadata containing verification data, alignment parameters and version information, and output interpretation tags.

[0008] Compared with existing technologies, this invention has the following advantages and beneficial effects: By arranging a monitoring point array including an external stable benchmark group, an on-site working benchmark group, and a structural response group, and combining a dual-link acquisition mode of total station three-dimensional observation link and static leveling elevation observation link, it achieves full-dimensional coverage monitoring from benchmark reference to structural response. Compared with traditional single-link monitoring, it effectively compensates for the one-sidedness of single-link data and provides redundant data source support for subsequent data verification and correlation analysis. Through geometric stability checks based on redundant data of the external stable benchmark group and cross-link consistency checks of dual-link elevation increments, combined with setting a first preset condition for observation uncertainty and generating a reference system credibility flag, problems such as benchmark reference system drift and systematic deviation of link data are eliminated from the source, ensuring the benchmark reliability of monitoring data and avoiding subsequent analysis distortion caused by unreliable reference systems. In the time alignment stage, a construction event time anchor point strategy is adopted. Through the main anchor point matching combined with auxiliary anchor point supplementation, and with a linear time mapping model adapted to the linear drift characteristics of the clock, the parameters are solved by the least squares fitting method and the alignment quality is quantified by the fitting residual, generating a time-aligned data. The signal marker effectively solves the problems of asynchronous local time from multiple acquisition terminals and mapping failure caused by insufficient anchor points, achieving precise alignment of data from different links on a unified time axis, laying a temporal foundation for the spatiotemporal correlation analysis of settlement and cracks. By triggering the synchronization analysis window with dual markers of "credible reference system + credible time alignment", collaborative deformation judgment is carried out only when both conditions are met, establishing a strict admission mechanism from the perspective of data validity, filtering out invalid data interference. At the same time, within the synchronization analysis window, the spatiotemporal correlation characteristics of crack increment and settlement difference index are combined to determine the real collaborative deformation events, and output evidence chain labels containing check data and time parameters. This not only achieves accurate identification of the causal relationship between settlement and crack deformation, distinguishes between real collaborative, decoupled and suspected abnormal events, and avoids subjective judgment bias, but also forms a complete evidence chain of monitoring, verification, analysis and judgment, improving the traceability and persuasiveness of monitoring conclusions. It is suitable for the high requirements of monitoring accuracy, data reliability and judgment rigor in the static cutting connection area of ​​the subway exit, providing scientific and accurate technical support for construction safety management and risk warning in the connection area, and reducing construction safety hazards caused by inaccurate deformation monitoring. Attached Figure Description

[0009] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of a three-dimensional settlement-crack monitoring method for a static cutting connection area at a subway exit, provided in Embodiment 1 of the present invention. Detailed Implementation

[0010] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. The illustrative embodiments and descriptions of this invention are for illustrative purposes only and are not intended to limit the invention. The embodiments described below are some, but not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0011] In the following description, numerous specific details are set forth to provide a thorough understanding of the invention. However, it will be apparent to those skilled in the art that these specific details are not necessary to practice the invention. In other embodiments, well-known structures, materials, or methods are not specifically described to avoid obscuring the invention. Unless otherwise specified, the materials, instruments, and reagents used in the following embodiments are commercially available. Unless otherwise specified, the techniques used in the embodiments are conventional methods well known to those skilled in the art.

[0012] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0013] Example 1: The static cutting and connection construction scenario at the subway exit involves strong process disturbances such as static cutting and opening, connection joint treatment, component dismantling and modification, replacement of supports or temporary supports, and rainwater recharge. However, the existing monitoring and interpretation methods have three major technical defects: "data distortion caused by dark drift of the reference system, mismatch of multiple timestamps leading to inverted collaborative relationships, and lack of auditable evidence chain making it difficult to verify and trace the conclusions." This reduces the repeatability and interpretability of collaborative monitoring results, making it difficult to meet the high-precision safety monitoring requirements under the strong disturbance construction conditions of the subway exit connection area.

[0014] To address this, this embodiment provides a three-dimensional settlement-crack monitoring method for the static cutting connection area of ​​a subway exit. By constructing a closed-loop management system encompassing "data acquisition, quality control, collaborative analysis, and output archiving," it achieves full-cycle monitoring of the structural safety of the subway entrance / exit connection area. Specifically, a dual-link monitoring network and a standardized point dictionary are first constructed. Monitoring points are grouped and modeled to clarify the hierarchical relationship between "external stability benchmark group—on-site working benchmark group—structural response group." Simultaneously, a point dictionary is compiled to record the entire lifecycle information of the monitoring points, including identification, installation status, and version changes, providing fundamental support for subsequent data processing, quality verification, and traceability. Based on this, data quality is controlled in advance through two independent gates ("reference system self-calibration gate" and "construction event anchor point alignment gate"): (1) Reference system self-calibration gate, based on the redundancy observation consistency check of the external stable benchmark group, and the cross-chain consistency check of the total station chain (the monitoring link connected by the total stations of each monitoring point) and the static leveling chain (the monitoring link connected by the static leveling collectors of each monitoring point), quantitatively assesses the stability of the reference system; (2) Construction event anchor point alignment gate, by collecting objective event anchor points that are strongly related to the construction process, and by using event sequence matching to establish a time mapping model, the local time of each link is unified to the same time axis, and the alignment quality is quantified by fitting residuals. Only when both gates have passed through can a synchronous analysis window be generated around the event anchor point to calculate core collaborative indicators such as differential settlement, settlement gradient, crack increment, and synchronicity. Then, the results are interpreted according to quantitative rules, and standardized labels and complete evidence fields are output. Simultaneously archived point dictionary versions, data processing parameters, residual statistics, and other full information are recorded to form a traceable and verifiable evidence chain, complete the monitoring closed loop, and meet the auditability requirements of the project.

[0015] Based on the aforementioned closed-loop management system, the three-dimensional settlement-crack monitoring method for the static cutting connection area at the subway exit includes... Figure 1 The following steps are shown: Step 1: Arrange a monitoring point array in the static cutting connection area of ​​the subway exit, including an external stability benchmark group, an on-site working benchmark group, and a structural response group located on the connection joint or load-bearing component; establish a total station three-dimensional observation link and a static leveling elevation observation link, and perform dual-link data acquisition on the monitoring point array.

[0016] The purpose of this step is to build the hardware foundation and data traceability system for multi-source collaborative monitoring, and to provide precise constraints for subsequent reference system self-calibration and collaborative interpretation through hierarchical monitoring point deployment and standardized dictionary compilation, so as to ensure the traceability and reliability of monitoring data.

[0017] The specific implementation method for this step is as follows: First, deploy an array of monitoring points to establish a reliable monitoring system.

[0018] To establish a reliable monitoring system, this embodiment divides all monitoring points into groups A, B, and C according to their functions, and strictly follows the deployment specifications to ensure that the functions of each group of monitoring points are accurately implemented.

[0019] 1. Group A is the external stability benchmark group, used to provide a global absolute stability benchmark and serving as the reference for the entire monitoring system. Each monitoring point in Group A must be located in a stable stratum outside the construction influence zone; for example, each monitoring point in Group A should be located at least 50m outside the construction influence zone. It should be noted that for soft soil areas, due to their poor stratum stability, the deployment distance needs to be increased to at least 100m, prioritizing stable strata or existing stable structures (such as concrete foundations for sidewalks, independent foundations of existing buildings, etc.). Furthermore, the number of monitoring points in Group A should be at least 3, and the monitoring points within the group should be distributed in a triangular pattern to ensure geometric constraint stability. During installation, chemical anchors can be used to fix a special bracket, rigidly connecting the reflective target to the bracket to fundamentally prevent micro-slippage caused by loose fixing. For example, in the static cutting and connection construction of a subway entrance, the three monitoring points in Group A were located at the concrete foundation of a sidewalk 60m outside the construction influence zone, fully meeting the deployment specifications.

[0020] For each monitoring point in Group A: Redundant observations (≥3 rounds, ≥2 repetitions per round) must be conducted periodically (e.g., every 24 hours) at the monitoring point using a total station, collecting data such as the three-dimensional coordinates of the monitoring point, the baseline length between point pairs, and the distance to the geometric centroid of Group A; based on the redundant observation data, the baseline length change Δ should be calculated. L and the change in distance from the center of gravity Δ D ; Change the baseline length by Δ L With the combined uncertainty threshold k × s L In comparison, the change in the distance from the center of gravity Δ D With the combined uncertainty threshold k × s D Comparison; only when Δ L ≤ k × s L And Δ D ≤ k × s D Only when this time is the monitoring point considered stable. k The confidence coefficient (usually taken as ) k =2 or 3), s L The combined uncertainty of the baseline length, s D The combined uncertainty of the distance between the centroids, s Land s D The calculation method will be detailed later.

[0021] Only observation data from stable monitoring points can provide an absolute benchmark for subsequent processing of Group C data, avoiding benchmark drift that could lead to data distortion in Group C (such as false settlement).

[0022] 2. Group B serves as the on-site working baseline group, used to assist total station observations. Each monitoring point in Group B must be placed in a relatively stable location within the construction impact zone. For example, Group B should be placed on non-directly load-bearing components within the construction impact zone, including: the non-cutting area wall outside the entrance / exit, beams far from the joint, etc. Furthermore, the number of monitoring points in Group B should be at least two. For example, in the static cutting and connection construction of a subway entrance / exit, the two monitoring points of Group B are placed on the non-cutting area wall outside the entrance / exit.

[0023] For each monitoring point in Group B: Three-dimensional coordinate and elevation data of the monitoring point need to be collected using a total station. This data will serve as an extension and supplement to the data from the monitoring points in Group A, thereby resolving issues such as observation obstruction and excessive distance that may occur due to the monitoring points in Group A being far from the monitoring area (e.g., 50m outside the construction impact zone). Furthermore, the data from the monitoring points in Group B will provide a nearby stable observation benchmark when subsequently processing the data from Group C.

[0024] To improve the reliability of the benchmark, the data from each monitoring point in Group B should be calibrated with the data from each monitoring point in Group A every 24 hours. This ensures the stability of the working benchmark while guaranteeing the convenience of observation, and avoids deviation from the absolute benchmark provided by each monitoring point in Group A due to slight disturbances on site.

[0025] 3. Group C is the structural response group. Monitoring points in Group C must be precisely deployed around high-risk areas such as joints and corners of openings to ensure timely detection of changes in the response of critical structural components. The monitoring points in Group C are divided into three categories: "three-dimensional displacement monitoring points," "elevation monitoring points," and "crack monitoring points."

[0026] (1) Three-dimensional displacement monitoring points are symmetrically arranged on both sides of the joint, and the spacing between the three-dimensional displacement monitoring points should be controlled within ≤0.3m. When setting up the three-dimensional displacement monitoring points, key locations such as the corner of the opening, the beam end, and the slab span should be covered. At least two three-dimensional displacement monitoring points should be set up in each key location to fully capture the three-dimensional displacement.

[0027] For each three-dimensional displacement monitoring point: the three-dimensional coordinates of the three-dimensional displacement monitoring point are directly obtained by observing the prism target with a total station, providing vertical displacement data for cross-chain consistency verification, and forming data mutual verification with the settlement / elevation measuring points deployed nearby.

[0028] (2) Settlement monitoring points should be arranged along a closed loop or traverse of static leveling; at the same time, settlement monitoring points should be arranged close to the corresponding three-dimensional displacement monitoring points, and the distance between them should be controlled within ≤0.5m.

[0029] It should be noted that: 1) When setting up settlement monitoring points, a closed loop should be used first, with the pipeline connection being firm and leak-free. After filling and venting, ensure that no air bubbles remain, and verify the consistency of the data through the calculation of the closure difference; 2) When the site space is limited and a closed loop cannot be set up, the settlement monitoring points can be set up using a traverse route. The starting point and ending point of the route need to be anchored to an external stable benchmark (Group A) or a field working benchmark (Group B); 3) When setting up a traverse route, the monitoring data of the three-dimensional displacement monitoring points need to be cross-verified with the elevation data corresponding to the total station chain to ensure the reliability of the elevation measurement and provide a data basis for cross-chain consistency verification.

[0030] For each settlement monitoring point: Continuous elevation changes are collected using a hydrostatic leveling instrument based on the principle of liquid connectivity, forming a continuous elevation observation chain. This provides vertical elevation data for cross-chain consistency verification and supports the self-calibration of the reference system.

[0031] (3) The crack monitoring points are displacement crack gauges or resistance crack gauges, which are directly pasted at the joint or on the potential crack development path (such as the corner of the opening at 45°). The two ends of the displacement crack gauge or resistance crack gauge are fixed to both sides of the crack with a distance ≥ 50 mm to ensure that the crack expansion can be accurately captured and to avoid monitoring failure due to improper placement.

[0032] For each crack monitoring point: use a displacement crack gauge or a resistance crack gauge to accurately capture the incremental change Δ in crack width. w This provides crack response data for collaborative interpretation.

[0033] It should be further clarified that the data monitoring operations performed by Groups A and B are essentially self-verifications of baseline stability, rather than direct monitoring of structural response. The monitoring data output by Groups A and B are prerequisites for the validity of the monitoring data output by Group C. Only when the monitoring data output by Groups A and B are stable can the displacement, settlement, and crack data monitored by Group C have a reliable reference, pass the cross-chain consistency check, and ultimately ensure the accuracy of the collaborative interpretation conclusions.

[0034] Furthermore, after the monitoring points are deployed, a "standardized point dictionary" needs to be compiled and full lifecycle version management needs to be implemented.

[0035] The standardized monitoring point dictionary uses a tabular format and includes fields for "monitoring point ID, spatial location, installation method, activation status, version number, and disassembly / removal record" to ensure complete and traceable monitoring point information. The monitoring point ID uses a "group-type-serial number" format, such as A-3D-01, BZ-02, CW-03, ensuring each monitoring point is unique. The spatial location description must include: structural component, relative dimensions, and elevation, facilitating on-site location and verification. The installation method must specify: anchor type, specifications, and installation depth, ensuring construction consistency and stability.

[0036] Full lifecycle version management must strictly follow the principle of "change is iteration". When the monitoring point location shifts, the installation method changes, the belonging structural unit is adjusted, or disassembly and reassembly occur, the version number will automatically increment. The reason for the change, the time of the change, the operator, and the verification results of the reassembly location will be recorded simultaneously. During collaborative analysis, only the data increment within the same version will be calculated, which will effectively avoid disassembly and reassembly errors being misjudged as structural responses and ensure the accuracy of data comparison.

[0037] By compiling a standardized point dictionary, the grouping information, location information, and method of operation of each monitoring point can be clearly defined. Even without initial disassembly and assembly records, this can lay the foundation for subsequent monitoring data traceability and quality verification.

[0038] Step 2: Perform geometric stability checks based on redundant observation data from the external stable benchmark group, and perform cross-chain consistency checks on the same monitoring point for elevation increments in both the three-dimensional observation link and the static leveling elevation observation link; when both the geometric stability check index and the cross-chain consistency check index meet the first preset condition determined based on observation uncertainty, generate a reference system credibility flag.

[0039] This step serves as the "first data access gate," and its purpose is to: quantitatively assess the stability of the reference system through dual consistency checks, automatically identify reference drift, measurement point slippage, and link errors, intercept unreliable data at the source, ensure that the data entering subsequent stages has a reliable reference, and provide data assurance for collaborative interpretation.

[0040] The specific implementation method for this step is as follows: Step 2.1: Perform geometric stability checks on redundant observation data based on the external stable benchmark group, and perform cross-chain consistency checks on the elevation increment for the same monitoring point in the three-dimensional observation link and the static leveling elevation observation link.

[0041] First, establish the initial baseline for Group A. Before construction disturbance begins ( tAt time 0, at least three rounds of redundant observations were conducted on each monitoring point in Group A, with no fewer than two measurements per round. Detailed records were kept of angles, distances, observation timestamps, and environmental parameters (temperature, humidity, air pressure). Based on the observation data, the three-dimensional coordinates of each monitoring point were calculated to determine the initial geometric relationships of Group A. The initial geometric relationships specifically include three aspects: first, the initial baseline length... L 0 — Calculate the baseline length between all monitoring point pairs in group A (e.g., A-3D-01~A-3D-02, A-3D-01~A-3D-03, A-3D-02~A-3D-03). L 0, forming the initial baseline matrix; 2, the initial barycenter coordinates ( X 0, Y 0, Z 0) – The centroid position is calculated based on the initial coordinates of each monitoring point in Group A, serving as the benchmark for subsequent distance change calculations; 3) Initial centroid distance D 0 — Calculate the distance from each monitoring point in Group A to the initial center of gravity, establishing the initial positional relationship between each monitoring point and the center of gravity. This initial data will serve as the core benchmark for subsequent stability checks. For example, in the support replacement construction of another subway entrance / exit connection area, the time before the construction disturbance begins... t 0=2024-06-12 07:00:00, 3 rounds of redundant observations were performed on each monitoring point in Group A, with 2 measurements per round. The baseline length L0 between A-3D-01 and A-3D-02 was calculated to be 25.32145m, and the baseline length between A-3D-01 and A-3D-03 was calculated to be 25.32145m. L Given the baseline length L0 = 28.15672m between A-3D-02 and A-3D-03, L0 = 30.21453m, and the calculated distance from A-3D-01 to the initial centroid... D =0=14.256m, Distance from A-3D-02 to the initial center of gravity D =0=15.321m, Distance from A-3D-03 to the initial center of gravity D 0=14.892m), providing a clear benchmark for subsequent self-calibration.

[0042] Then, calculate the current baseline and uncertainty for group A. S1: At the current time... t 1. Recalculate the baseline length of each monitoring point pair in Group A. L i and the distance from each monitoring point to the current center of gravity D i S2: For each monitoring point pair: Set the baseline length. L i With baseline length L 0 comparison yields the baseline length change Δ L =| Li - L 0|; For each monitoring point: the distance from that monitoring point to the current center of gravity. D i Distance from the monitoring point to the initial center of gravity D 0 comparison yields the corresponding change in center of gravity distance Δ D =| D i - D 0|. S3: Calculate the combined uncertainty of the baseline length according to the error propagation law. s L Combined uncertainty of distance from the center of gravity s D Among them, the combined uncertainty of the baseline length s L The calculation formula is: , This refers to the nominal accuracy of the total station (e.g., distance measurement accuracy of 1mm + 1ppm). Environmental errors (temperature error can be estimated at 0.1 mm / 10℃, humidity and air pressure errors can be calculated according to the formula in the instrument manual). This represents the coordinate calculation error (estimated based on the accuracy of the calculation algorithm, typically taken as 0.5mm). s D The computational logic and s L Consistency is crucial to ensure the rigor of uncertainty calculations. For example, in the static cutting and connection construction of a subway entrance / exit mentioned above, the combined uncertainty of the baseline length was calculated. s L =0.25mm, combined uncertainty of centroid distance s D =0.26mm.

[0043] Next, combine the confidence coefficient k ( k =2 or 3, which can be configured according to the target false alarm rate) to perform the consistency judgment of group A. (1) Only when all monitoring points in group A meet Δ L ≤ k × s L And Δ D ≤ k × s D When, the baseline of group A is determined to be stable; (2) if only one monitoring point in group A shows an abnormality and does not simultaneously meet Δ L ≤ k × s L And Δ D ≤ k × sD The monitoring point is marked as an "unstable candidate point." After removing the unstable candidate point, the centroid and reference frame of group A are recalculated. At the same time, the monitoring data of the unstable candidate point are reduced in weight (weight = s L / Δ L (2) Avoid individual anomalies affecting the overall benchmark; (3) When ≥2 monitoring points are abnormal, directly determine that the reference system is unreliable, output the first label (used to mark reference system drift), and remind the site to check the surrounding environment or control network status of the benchmark point. For example, in the above-mentioned subway entrance and exit connection area replacement construction, the confidence coefficient is taken. k =2, calculate the baseline length change Δ for group A. L The value range is 0.1mm~0.2mm, less than 0.5mm (2×0.25mm), and the change in the center of gravity distance Δ of group A is calculated. D The value range is 0.11mm~0.17mm. If it is less than 0.52mm (2×0.26mm), then the A group benchmark is considered stable; and when the above-mentioned subway entrance and exit connection area replacement support construction reaches... t At 10:00:00 on June 12, 2024, the change in baseline length Δ for Group A was... L (A1-A2=0.62mm, A1-A3=0.58mm, A2-A3=0.65mm) are all > 0.5mm, and the change in the center of gravity distance of group A is Δ D (A1=0.59mm, A2=0.63mm, A3=0.61mm) are all greater than 0.52mm, and the number of abnormal points is greater than or equal to 2, which directly triggers the first tag.

[0044] Finally, to further verify data reliability, a cross-chain consistency check was conducted simultaneously. S1: Select monitoring point pairs from group C, consisting of three-dimensional displacement monitoring points and elevation monitoring points; "nearest neighbors" refers to points in group C with a horizontal distance ≤ 0.5m. S2: For the extracted monitoring point pairs, obtain the total station chain vertical increment Δ within the same time window (e.g., 1 hour). Z TS Vertical increment Δ of static leveling chain Z SL Among them, the vertical increment Δ of the total station chain Z TS The calculation method is as follows: based on the start time of the total station within the time window. t start For redundant observation data of three-dimensional moving monitoring points, calculate the three-dimensional displacement of the monitoring points. t start Moment Z Axis coordinates Z TS,start Similarly, the three-dimensional displacement monitoring points are calculated.t end Z-axis coordinate at time Z TS,end Calculate the vertical increment Δ of the total station chain. Z TS , that is, Δ Z TS = Z TS,end -Z TS,start Similarly, the vertical increment Δ of the static leveling chain Z SL The calculation method is as follows: based on the continuous elevation data of settlement monitoring points obtained by the hydrostatic leveling instrument within the same time window (e.g., 1 hour), after removing abnormal fluctuation values, the starting time is taken. t start and the end time t end The elevation difference is used to obtain the vertical increment Δ of the static leveling chain. Z SL , that is, Δ Z SL = Z SL,end - Z SL,start S3: Based on the vertical increment Δ of the total station chain Z TS Vertical increment Δ of static leveling chain Z SL Calculate the cross-chain difference δZ=Δ Z TS -Δ Z SL and combined uncertainty ,in. s TS Depend on s L The decomposition yields (i.e., the vertical increment Δ of the total station chain) Z TS observation uncertainty s TS It is through the distance measurement uncertainty of the total station s L By combining the spatial geometric relationship of the measuring points with the law of error propagation, the vertical error components obtained through decomposition and calculation are obtained. s TS = s L ×sinθ, where θ is the vertical angle of the baseline). s SL Proportioned by the static leveling closure error (i.e.) , fIn the static leveling chain setup, this represents the maximum permissible deviation between observed values ​​and theoretical closure values. The theoretical closure value is that the sum of the elevation differences of the closed loop should be 0 (where n is the number of monitoring points). S4: If |δ Z |≤ k ×σδ, determine cross-chain consistency, data is trustworthy; if |δ Z |> k ×σδ and the data exhibits single-chain unidirectional drift characteristics (such as Δ). Z TS Continuing to sink and Δ Z SL If the measurement point is stable, output a second label (used to mark the measurement point / link slippage). The data of this measurement point will not be included in the actual settlement accumulation to avoid distorted data interfering with the interpretation. If multiple measurement points show abnormalities at the same time and the A group benchmark is stable, it is necessary to check the link system error, such as the total station not being calibrated, air leakage in the hydrostatic leveling pipeline, or other equipment or construction problems.

[0045] Step 2.2: When both the geometric stability check index and the cross-chain consistency check index meet the first preset condition determined based on observation uncertainty, a reference frame credibility flag is generated.

[0046] The conditions for passing the first data access gate are: (1) Group A consistency check is stable (or only one unstable candidate point has been downgraded), and cross-chain consistency check has no second label (or only a single monitoring point is abnormal and has been removed), at which time the "Reference System Trustworthy" flag is output; (2) If the first label is triggered or multiple measurement points have a second label, it is determined that the access has not been passed, and the corresponding label and Δ are output. L Δ D , s L A complete evidence field composed of quantitative data such as δZ and σδ provides direction for on-site investigation.

[0047] Step 3: Collect construction event data related to structural disturbances, extract the occurrence time of construction events as time anchors, and map the local timestamp sequences of different acquisition terminals to a unified time axis through a time mapping model; when the fitting residual of the time mapping model meets the second preset condition determined based on the sampling interval, a time alignment credibility flag is generated.

[0048] This step, serving as the "second data access gate," aims to resolve the issue of timestamp mismatch in multi-source data, achieve precise time synchronization across all links without external time synchronization such as GPS / NTP, ensure the accuracy of the temporal causal relationship between "settlement and cracks," and provide a unified time reference for subsequent collaborative interpretation. This is crucial for establishing a precise collaborative relationship.

[0049] The specific implementation method for this step is as follows: First, collect objective event anchor points that are strongly related to the construction process. Prioritize independent events that do not depend on structural response data. These events have clear edges and strong measurability, and can accurately correlate the time relationship between structural disturbance and response, avoiding time synchronization errors caused by unreliable anchor points. Specific event collection needs to be accurately configured in combination with the construction scenario to ensure that the anchor points can fully cover the core process - (1) Cutting equipment events can be collected by clamp-type current transformers (range 0-500A, accuracy ±1%). The sampling frequency can be set to ≥1Hz to capture rapidly changing signals; and the 100ms sliding window method is used to identify events. The noise threshold is determined by statistically analyzing the current fluctuation range in the 30 minutes before construction. I 0; Set the trigger threshold I trig =× I 0 (for threshold correction coefficient, typically 1.5~2.0), when the current exceeds [a certain value] for 3~5 consecutive sampling points. I trig When the current is below I0 for 3-5 consecutive sampling points, it is determined as the cutting equipment event start (CutOn). When the current is below I0 for 3-5 consecutive sampling points, it is determined as the cutting equipment event stop (CutOff), ensuring the accuracy of cutting event identification; (2) Rainwater / recharge pump events are collected by the passive contact signal of the pump control cabinet (access gateway digital interface) or flow sensor (range 0-100m³ / h, accuracy ±2%). When the contact is closed or the flow is ≥5m³ / h, it is determined as the rainwater / recharge pump event start (PumpOn). When the contact is open or the flow is ≤1m³ / h, it is determined as the rainwater / recharge pump event stop (PumpOff). At the same time, the validity of the contact signal is verified by the flow data to avoid false triggering of events due to contact failure. (3) Optionally, a hoisting and positioning acceleration switch (range ±5g, resolution 0.01g) is configured. When the absolute value of acceleration is ≥1g, the hoisting and positioning event is recorded to supplement the time anchor type and improve the reliability of time synchronization.

[0050] It should be noted that, in order to achieve accurate correlation between construction operation events and structural response events, ensure data authenticity and reliability, and provide an anchor point for a unified time benchmark, this embodiment uses two independent event acquisition devices deployed in different locations or with different monitoring functions to collect construction events. Furthermore, to ensure the traceability of subsequent event sequence matching and to provide complete and accurate data support for the construction of the time mapping model, all collected events must record full-dimensional metadata, fully covering seven core items: event ID, event type, acquisition terminal ID, local timestamp accurate to milliseconds, event duration, event intensity (recording current amplitude, flow amplitude, or acceleration amplitude depending on the monitoring scenario), and event status (valid or invalid). The metadata must be generated synchronously with event acquisition to ensure that the timestamp does not jump, the intensity parameters are not abnormally missing, and the status label is consistent with the actual validity of the event, preventing sequence matching failures due to incomplete or incorrect metadata.

[0051] After event collection is completed, event sequence matching is carried out following the principle of "primary anchor point priority, secondary anchor point backup." Independent events are used as primary anchor points, and secondary anchor points serve as backups in case the primary anchor point matching fails. First, primary anchor points are grouped according to event type (e.g., cutting event, pumping event) to ensure matching between events of the same type and avoid deviations caused by cross-type matching. The matching process must simultaneously meet two core criteria: first, the sequence consistency criterion, meaning the order in which events occur must be completely consistent between the two collection ends participating in the matching; and second, the duration similarity criterion, meaning the duration difference of paired matching events must satisfy |Δ t |≤10%×max( t 1, t 2) This ensures matching accuracy, among which, t 1. t 2 represents the duration of paired events at the two acquisition ends, Δ t This represents the duration difference between paired events. After matching is complete, at least three successfully matched event points must be obtained. T i , t i ),in, T i This corresponds to the local timestamp of the data acquisition terminal. t i To establish a unified timeframe with the time collected by the cutting equipment as a temporary reference, and to meet specific quantity requirements for different project types, static cutting projects need to obtain multiple pairs of matching points (e.g., 3 pairs of matching points), and support replacement projects need to obtain multiple pairs of matching points (e.g., 4 pairs of matching points). Only after reaching the minimum matching quantity threshold can the subsequent time mapping model construction stage be entered.

[0052] It should be noted that: (1) The purpose of using the time of the cutting equipment acquisition end as a temporary reference benchmark is that: the cutting equipment acquisition end is directly related to the core operation events of the engineering construction (such as cutting start / stop, power change). These events are the direct cause of structural response (displacement, stress change). Using its time as a benchmark is essentially to establish a causal time axis of "construction operation-structural response" to ensure that subsequent analysis can accurately correspond to "when a certain operation occurs and what kind of structural change it causes"; in addition, as the core equipment of construction, the operation events of the cutting equipment have the characteristics of clarity and high identifiability (such as the obvious current amplitude characteristics of the cutting event, which are easy to distinguish from other interference events), which is suitable as a time anchor point; at the same time, the cutting equipment in the same project is usually a single core operation equipment, and the time benchmark is unique, which can avoid the time calibration chaos caused by multiple benchmark conflicts. (2) The static cutting project needs to obtain at least 3 pairs of matching corresponding points. The purpose is that: the event corresponding points ( T i , t i The core function of ) is to construct a linear time mapping model ( t = a · T + b ,in a This is the time scaling factor. b (For time offset), solving for two unknown parameters requires at least two pairs of matching points, but only two pairs of matching points have no redundant verification capability. If one of the pairs has a matching error, it will directly lead to model distortion. Three pairs of matching points are the minimum number to achieve "modeling + verification". Outliers can be eliminated through residual analysis (calculating the deviation between three sets of data and the model) to ensure the accuracy of the mapping model. (3) The support replacement project needs to obtain at least four pairs of matching points. The purpose is that the support replacement project involves multiple stages of operation such as support removal, stress conversion, temporary reinforcement, and permanent support installation. The operating status of construction equipment (such as pumps, cutting machines, and jacks) in different stages varies greatly, and the time characteristics of structural response are complex (such as alternating stress gradual change and sudden change). At this time, a single linear mapping model may not be able to adapt to the time drift law of the whole cycle. Four pairs of matching points can support the construction of a piecewise linear model or the introduction of nonlinear correction terms to improve the time calibration accuracy of different construction stages.

[0053] When the number of corresponding points successfully matched by the main anchor point (independent event) is less than 3 pairs, structural response auxiliary anchor points (such as crack mutation events, elevation mutation events) need to be used as a supplement. Such anchor points need to be selected as structural response characteristic events with complete metadata records and strong correlation with the main anchor point. At the same time, they need to meet any of the following quantitative matching criteria to ensure reliability: (1) Normalized amplitude ratio error of event intensity ≤ 0.2 - The amplitude ratio error after normalization processing of the corresponding event intensity parameters of the two acquisition ends is controlled to ensure the consistency of event intensity characteristics; (2) Correlation coefficient at the time of event occurrence ≥ 0.8 - Based on the linear correlation analysis of time series data, the strong correlation of events in the time dimension is verified. It is necessary to clarify that the auxiliary anchor point is only a supplementary means of the main anchor point and cannot replace the main anchor point. After the matching results are merged with the matching results of the main anchor point, the total number of corresponding points matched must still meet ≥ 3 pairs to provide sufficient data support for the fitting of the time mapping model to ensure accuracy.

[0054] Based on the n pairs (n≥3) of successfully matched event correspondences, and considering the characteristics of linear clock drift in engineering monitoring scenarios, a linear time mapping model is established. t i = a · T i + b .in, t i To standardize the timeline (using the time acquired by the cutting equipment as a temporary reference). T i This is the local timestamp of the data acquisition terminal. a b is the drift coefficient (ideally 1, the degree of deviation reflects the clock's linear drift rate), and b is the initial offset (when...). T i When =0, t i =b, representing the initial deviation between local time and the unified standard time. Further, the model parameters... a The least squares fitting method is used to solve for b, and the parameter matrix X is set as [ a ,b]ᵀ、Observation Vector L =[ t 1, t 2,..., t n Design Matrix A =[[ T 1,1],[ T 2,1],...,[ T n The optimal solution for the parameters is ,1]], X =( A ᵀ A )⁻¹ A ᵀL This method can minimize the fitting error to achieve accurate mapping.

[0055] After fitting is completed, the fitting residuals are calculated. ᵣ es The quantification of time alignment quality is calculated using the following formula: ,in, i For the matching point number, n For the number of matching points, n ≥3, vᵢ For the first i The fitting error for the matching points, vᵢ= t i -( a · T i + b The denominator is taken as n-2 to deduct the influence of the degrees of freedom of the two parameters to be estimated in order to ensure the scientific nature of the calculation. ᵣ es A smaller value indicates better time alignment. If the value exceeds the preset threshold, the validity of the matching points needs to be rechecked or supplemented with data before refitting until the engineering monitoring time accuracy requirements are met. The preset threshold is [value missing]. s res,th , s res,th =α×Δ T (α is the threshold adjustment coefficient, which can be 0.05~0.2, Δ) T (the maximum sampling interval for multi-source devices), if s res ≤ s res,th Determine if the time alignment is reliable, and output the "Time Alignment Reliable" flag and model parameters. a , b ;like s res > s res,th If the number of matching points is insufficient, a third label (used to mark time mismatches) and corresponding evidence fields will be output to provide a basis for on-site investigation of time synchronization issues.

[0056] In summary, the conditions for passing the second data access gate are: (1) If only the main anchor point (independent event) matching is used, the order of events occurring at the two acquisition ends must be completely consistent, and the difference in duration of paired matching events must satisfy |Δ t |≤10%×max( t 1, t2); (2) If the number of main anchor points is insufficient, auxiliary anchor points (structural response events such as crack mutation and elevation mutation) are used to supplement them. The auxiliary anchor points must be selected as feature events with complete metadata and strong correlation with the main anchor points, and must also meet any quantitative matching condition, that is, the normalized amplitude ratio error of the event intensity is ≤0.2, or the correlation coefficient at the time of the event occurrence is ≥0.8. The auxiliary anchor points cannot replace the main anchor points and are only used as a supplementary means. (3) After the main anchor point matching results and the auxiliary anchor point matching results are merged, the total number of successfully matched event corresponding points is ≥3 pairs, which provides a sufficient and effective data foundation for the subsequent construction of the linear time mapping model.

[0057] Step 4: Generate a synchronous analysis window on a unified time axis if and only if both the reference frame confidence flag and the time alignment confidence flag are true; calculate the crack increment index and settlement difference index of the structural response group within the synchronous analysis window; determine whether a real collaborative deformation event has occurred based on the spatiotemporal correlation characteristics of the crack increment index and settlement difference index; and output an evidence chain label containing reference frame check data and time alignment parameters.

[0058] The purpose of this step is to conduct collaborative analysis only on reliable data that has passed through the first and second data access gates, and to output interpretable and verifiable conclusions by establishing quantitative correlations between "differential settlement, crack propagation, and time synchronicity" to provide accurate basis for construction safety decisions.

[0059] The specific implementation method for this step is as follows: First, around the event anchor moments on a unified timeline. t e Generate a synchronous analysis window. The window length must precisely cover the entire event process and the structural response cycle to ensure complete capture of the structural response changes under disturbance. Specifically, follow the formula [ t e -( t pre + t rise ), t e +( t dur + t settle )] Calculation. Among them, t pre The preparation time for the event can be configured according to the type of construction (e.g., 1 to 5 minutes). t rise The duration of the rising edge of the event is obtained from the event source data (e.g., a value of 1 to 3 minutes). t dur The event duration is obtained directly from the paired records of the event anchor. tsettle This represents the structural response convergence time, and its value can be flexibly adjusted according to the structural stiffness (e.g., 5-15 minutes). For example, in a static cutting project, the CutOn event... t e =2024-06-10 10:00:00, Configuration t pre =5 minutes t rise =1 minute t dur =15 minutes (cutting duration) t settle =10 minutes, the synchronous analysis window is set to [2024-06-10 09:54:00, 2024-06-10 10:25:00]; PumpOff event in the support replacement project. t e =2024-06-12 12:30:00, Configuration t pre =5 minutes t rise =2 minutes t dur =4 hours (duration of support change) t settle =10 minutes, analysis window is [2024-06-12 12:23:00, 2024-06-12 13:05:00].

[0060] It should be noted that: Sufficient sample size must be ensured within the synchronous analysis window. Specifically, this requires ≥5 static leveling data points, ≥3 crack data points, and ≥1 total station data point. If the sample size is insufficient, it needs to be expanded. t settle Continue until the sample size meets the requirements.

[0061] Within the simultaneous analysis window, the synergistic index and its corresponding combined uncertainty are calculated. The synergistic index includes: differential settlement Δ... z diff Settlement gradient degree and crack increment Δ w The specific technical methods are as follows: (1) Differential settlement Δ z diff The vertical settlement change Δ is calculated by selecting symmetrically arranged elevation monitoring points on both sides of the joint. z diff =Δ z 1-Δ z 2; where Δ z 1 represents the elevation change of one of the settlement monitoring points, Δ z 2 represents the elevation change at another settlement monitoring point; its combined uncertainty is... (2) Settlement gradient grad is based on differential settlement Δ z diff The horizontal distance L between the elevation monitoring point and the monitoring point is calculated using the formula grad=Δ. z diff / L (Unit: mm / m), Uncertainty s grad = s Δ z / L (3) Crack increment Δ w = w 2- w 1, among which, w 2 represents the crack width at the end of the synchronous analysis time window. w 1 represents the crack width at the start of the synchronous analysis time window; uncertainty. ,in, s d For the resolution of the crack gauge, s e This represents the calibration error of the crack gauge.

[0062] Based on the above collaborative indicators, quantitative interpretation is performed—(1) Interpretation of actual collaborative time. Four interpretation conditions must be met simultaneously: First, Δ w ≥ k ×σΔ w This indicates that the crack has expanded; secondly, Δ z diff ≥ k ×σΔz or degree ≥ k × s grad This indicates the existence of differential settlement or settlement gradient; thirdly, the time difference between the crack propagation time and the settlement change time is ≤ β×Δ T This indicates that the two have a synchronous change relationship, and β is the synchronization coefficient (which can be 2~5); fourth, the cross-chain consistency check is passed (|δ Z |≤ k ×σδ), indicating that the data is reliable. When all four interpretation conditions are met, it can be determined that there is a direct causal relationship between structural differential settlement and crack propagation. If all four interpretation conditions are met, it is determined that there is a direct causal relationship between structural differential settlement and crack propagation, triggering the fourth label (used to allow passage through the first and second data access gates), and including the corresponding data in the valid analysis sample. (2) Decoupling event interpretation. One of the following interpretation conditions must be met: First, Δ w < k ×σΔ w This indicates that the crack is in a stable state, where σΔw is the observation uncertainty of the crack increment; and Δw ≥ k ×σΔ w Δ z diff < k ×σΔ z and degree < k × s grad This indicates that crack propagation is independent of differential settlement, where σΔz is the combined uncertainty of differential settlement. s grad The uncertainty in the observation of the settlement gradient; thirdly, both crack propagation and differential settlement occur, but the time difference between the two is >β×Δ T This indicates that the two are asynchronous changes and have no direct causal relationship. In such cases, the data should be retained for subsequent comprehensive analysis. If any one of the above three conditions is met, it can be determined as a decoupling event. The corresponding data is allowed to pass through the gate but should be marked as "non-causal event" for subsequent comprehensive analysis. (3) Judgment of abnormal coordination time. Abnormal coordination time should output a "requires review" prompt, specifically referring to the fact that both cracks and settlement are significant but cross-chain consistency is abnormal ( k ×σδ<|δ Z |≤( k In scenarios involving +1)×σδ), the corresponding data is temporarily not allowed to pass through the gate. It is necessary to combine on-site inspections to eliminate non-structural disturbance factors (such as instrument offset and environmental interference) before re-performing the quantization interpretation.

[0063] After the interpretation is completed, standardized labels (including: first label, second label, third label and fourth label) and complete evidence fields are output. The first label corresponds to suspected slippage of the measurement point / link, the second label corresponds to time mismatch, the third label corresponds to reference frame drift, and the fourth label corresponds to actual coordination. Each label clearly defines its core meaning and triggering scenario. The evidence fields need to cover five types of information: (1) basic information (point dictionary version number, analysis window start and end time, event anchor type and...). t e (2) Reference system self-calibration data (Group A Δ) L Δ D , s L , s D Cross-chain δ Z (3) Time alignment data (list of event correspondence points, time mapping parameters) a / b , s res , s res,th (4) Collaborative index data (ΔTmax); z diff , s Δ z , degree , s grad , Δw, σΔw, k (5) Judgment basis (judgment rule items that are met / not met, and the reason for the tag triggering). The evidence field ensures that each conclusion is fully quantitatively supported.

[0064] At the same time, the entire evidence chain is archived in JSON format, including the aforementioned evidence fields, original observation data, data processing logs, measurement point disassembly and assembly records, and version change records. The storage period is no less than 5 years after the completion of the project. It supports retrieval by tag type, time range, and measurement point ID, which facilitates subsequent project review, accident tracing, and technical review, and meets the project auditability requirements.

[0065] In summary, this embodiment provides a three-dimensional settlement-crack monitoring method for static cutting connection areas at subway exits. By establishing a reference system self-calibration gate, it automatically identifies and suppresses reference system drift and measuring point slippage, ensuring the baseline reliability of monitoring data from the source and providing a reliable data foundation for subsequent analysis. Then, it establishes a construction event anchor point alignment gate, achieving precise time synchronization of multi-source data even without external time synchronization such as GPS / NTP, completely avoiding the reversal of collaborative relationships caused by time mismatch. Finally, it constructs a dual-gate access collaborative interpretation mechanism, analyzing only reliable data that has passed dual quality control, outputting standardized interpretable labels and a complete chain of evidence, effectively reducing the risk of misjudgment while improving the verifiability and auditability of conclusions, adapting to the safety monitoring needs of highly disturbed construction scenarios.

[0066] Example 2: A three-dimensional settlement-crack monitoring system for the static cutting connection area of ​​a subway exit is provided for performing the monitoring method described in Example 1. The system includes: The reference system self-calibration module is used to perform geometric stability checks on the external stable benchmark group, as well as cross-link consistency checks between the total station link and the hydrostatic leveling link, and outputs a reference system credibility flag. The construction event anchor point alignment module is used to collect construction events and extract time anchor points, establish a time mapping model to map multi-source data to a unified time axis, and output a time alignment confidence flag based on the fitting residual. The dual-gate collaborative interpretation module is used to generate a synchronous analysis window and calculate crack increment index and settlement difference index to determine collaborative deformation events when both the reference system credibility flag and the time alignment credibility flag are true. The evidence chain archiving and output module is used to store evidence chain metadata containing verification data, alignment parameters and version information, and output interpretation tags.

[0067] It should be understood that the terms "system," "device," "unit," and / or "module" as used in this specification are a method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0068] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0069] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0070] It should be noted that the structures, proportions, sizes, etc., illustrated in the accompanying drawings are merely for illustrative purposes to aid those skilled in the art and are not intended to limit the scope of the invention. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in proportions, or adjustments to size, without affecting the effectiveness and purpose of the invention, should still fall within the scope of the disclosed technical content. Furthermore, terms such as "upper," "lower," "left," "right," and "middle" used in this specification are merely for clarity and not intended to limit the scope of the invention. Changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

Claims

1. A method for three-dimensional settlement-crack monitoring of the static cutting connection area at a subway exit, characterized in that, Includes the following steps: An array of monitoring points, including an external stability benchmark group, an on-site working benchmark group, and a structural response group located on the joint or load-bearing component, is set up in the static cutting and connection area of ​​the subway exit. Establish a total station three-dimensional observation link and a static leveling elevation observation link to perform dual-link data acquisition for the monitoring point array; Geometric stability is checked based on redundant observation data from an external stable benchmark group, and cross-chain consistency is checked for the same monitoring point in both the three-dimensional observation link and the static leveling elevation observation link for elevation increments. When both the geometric stability check index and the cross-chain consistency check index meet the first preset condition determined based on observation uncertainty, a reference system credibility flag is generated. Data on construction events related to structural disturbances are collected, and the occurrence time of the construction events is extracted as the time anchor point. The local timestamp sequences of different collection terminals are mapped to a unified time axis through a time mapping model. When the fitting residual of the time mapping model satisfies the second preset condition determined based on the sampling interval, a time alignment confidence flag is generated. A synchronous analysis window is generated on the unified time axis if and only if both the reference frame confidence flag and the time alignment confidence flag are true. Within the synchronous analysis window, the crack increment index and settlement difference index of the structural response group are calculated. Based on the spatiotemporal correlation characteristics of the crack increment index and settlement difference index, it is determined whether a real coordinated deformation event has occurred, and the evidence chain label containing reference frame check data and time alignment parameters is output.

2. The method for three-dimensional settlement-crack monitoring of the static cutting connection area at a subway exit according to claim 1, characterized in that, The methods for geometric stability checks and cross-chain consistency checks are as follows: Calculate the baseline length change Δ at each monitoring point in the external stability benchmark set. L and the change in distance Δ from each monitoring point to the group's centroid D If all measuring points satisfy Δ L ≤ k × σ L And Δ D ≤ k × σ D If the geometric stability check is passed, then the geometric stability check is deemed successful; among which, k , is the confidence coefficient σ L The combined uncertainty of the baseline length, σ D The combined uncertainty of the distance between the centroids; Select nearest neighbor monitoring point pairs consisting of three-dimensional displacement monitoring points and elevation monitoring points, and calculate the cross-chain difference δ within the same time window. Z =Δ Z TS -Δ Z SL If |δ Z |≤ k If ×σδ is considered, then the cross-chain consistency check is considered passed; where Δ Z TS For the vertical increment of the total station chain, Δ Z SL This represents the vertical increment of the static leveling chain.

3. The method for three-dimensional settlement-crack monitoring of the static cutting connection area at a subway exit according to claim 2, characterized in that, The formula for calculating the combined uncertainty is: ;in, σ TS The observation uncertainty is the vertical increment of the total station chain. σ TS = σ L ×sinθ, where θ is the vertical angle of the baseline. σ SL The observation uncertainty is the vertical increment of the static leveling chain. , f The maximum permissible deviation value for static leveling. n This represents the number of monitoring points.

4. The method for three-dimensional settlement-crack monitoring of the static cutting connection area at a subway exit according to claim 1, characterized in that, Before mapping local timestamp sequences from different acquisition terminals to a unified timeline, the following steps are also included: Event sequences from different acquisition terminals are grouped according to event type, and corresponding event points that simultaneously satisfy the sequential consistency criterion and the duration similarity criterion are selected; the duration similarity criterion is: |Δ t |≤10%×max( t 1, t 2); among which, t 1 and t 2 represents the duration of the paired events, Δ t This represents the difference in duration between paired events; Based on multiple successfully matched pairs of event correspondences, a linear time mapping model is constructed using the least squares method: t = a · T + b ;in, t To standardize the timeline, T This is the local timestamp of the data acquisition terminal. a The drift coefficient, b This is the initial offset.

5. A method for three-dimensional settlement-crack monitoring of a static cutting connection area at a subway exit, as described in claim 1, is characterized in that... The second presupposition condition is: the fitting residual σᵣ of the time mapping model. es satisfy σ res ≤ σ res,th ; in, σ res,th =α×Δ T α is the threshold adjustment coefficient, Δ T This is the maximum sampling interval for multi-source devices.

6. The method for three-dimensional settlement-crack monitoring of the static cutting connection area at a subway exit according to claim 1, characterized in that, The method for generating synchronous analysis windows on a unified time axis is as follows: Determine the start time of the analysis window using the aforementioned time anchor point as the center. t start and end time t end ;in, t start = t e -( t pre + t rise ), t end = t e +( t dur + t settle ), t e To unify the anchor point times of events on the timeline, t pre Preparation time for the event t rise The duration of the rising edge of the event. t dur For the duration of the event, t settle Let be the convergence time of the structural response.

7. A method for three-dimensional settlement-crack monitoring of a static cutting connection area at a subway exit, as described in claim 6, is characterized in that... The conditions for determining whether a real collaborative deformation event has occurred are: conditions 1 to 4 must be met simultaneously. Condition 1: Crack increment Δw ≥ k ×σΔw, where σΔw is the observation uncertainty of the crack increment; Condition 2: Differential settlement Δ z diff ≥ k ×σΔz or settlement gradient grad≥ k × σ grad σΔz is the combined uncertainty of differential settlement. σ grad The observation uncertainty of the settlement gradient; Condition 3: The time difference between the crack propagation time and the settlement change time is ≤ β × Δ T β is the synchronization coefficient; Condition 4: Cross-chain consistency check passed.

8. A method for three-dimensional settlement-crack monitoring of a static cutting connection area at a subway exit, as described in claim 1, is characterized in that... The output includes a chain of evidence label containing reference frame verification data and time alignment parameters, specifically including one of the following labels: The first label is used to mark reference frame drift; the output condition for the first label is: reference frame geometric stability check failed or multi-point cross-chain consistency check failed. The second label is used to mark test points or link slippage; the output condition of the second label is: only a single test point fails the cross-link consistency check and exhibits unidirectional drift; The third label is used to mark time mismatches; the output condition for the third label is: the fitting residual of the time mapping model does not meet the second preset condition or the number of matching points is insufficient. The fourth label is used to mark real collaboration; the output condition of the fourth label is: the reference frame is reliable, the time alignment is reliable, and a real collaboration deformation event is determined to have occurred.

9. A method for three-dimensional settlement-crack monitoring of a static cutting connection area at a subway exit, as described in claim 1, is characterized in that... After performing dual-link data acquisition on the monitoring point array, the following steps are also included: A standardized monitoring point dictionary is compiled and full lifecycle version management is implemented. The standardized monitoring point dictionary includes: monitoring point ID, spatial location, installation method, version number, and disassembly / reassembly records. Full lifecycle version management includes: when a monitoring point is disassembled or reassembled, the version number is automatically incremented; when calculating the crack increment index and settlement difference index, only the data increment within the same version number is calculated.

10. A three-dimensional settlement-crack monitoring system for the static cutting connection area of ​​a subway exit, characterized in that, include: The reference system self-calibration module is used to perform geometric stability checks on the external stable benchmark group, as well as cross-link consistency checks between the total station link and the hydrostatic leveling link, and outputs a reference system credibility flag. The construction event anchor point alignment module is used to collect construction events and extract time anchor points, establish a time mapping model to map multi-source data to a unified time axis, and output a time alignment confidence flag based on the fitting residual. The dual-gate collaborative interpretation module is used to generate a synchronous analysis window and calculate crack increment index and settlement difference index to determine collaborative deformation events when both the reference system credibility flag and the time alignment credibility flag are true. The evidence chain archiving and output module is used to store evidence chain metadata containing verification data, alignment parameters and version information, and output interpretation tags.

Images