A method and system for engineering supervision and management

Abstract

This invention provides a method and system for engineering supervision, relating to the field of engineering supervision technology. The method includes: collecting multi-dimensional heterogeneous supervision data through multiple sensors deployed at the construction site; wherein the multiple sensors are arranged at preset spatial reference points on a large-span prestressed concrete load-bearing structure; the preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key cross-section of the structure and the center point of the prestressed anchorage end; based on the preset spatial reference points, a spatial reference control body covering a part or the whole of the structure is virtually constructed in the edge computing nodes. This invention can effectively solve the problems of systematic data deviation, misjudgment of risk warnings, and lack of accurate support for supervision decisions.

Description

Technical Field

[0001] This invention relates to the field of engineering supervision technology, and in particular to an engineering supervision and management method and system. Background Technology

[0002] In the construction supervision of long-span prestressed concrete structures (such as bridges and large stadiums), real-time control of structural load-bearing safety and construction quality relies on the accurate collection and effective analysis of diverse and heterogeneous supervision data. These engineering structures are characterized by large spans, complex stresses, and long construction periods. The monitoring data of key indicators such as cross-sectional deformation and prestress transfer effect directly determine the rationality of supervision decisions.

[0003] In existing engineering supervision methods, the technical solution of deploying sensor clusters to collect data at the construction site is commonly used. However, there are the following technical defects: the raw measurement data collected by the sensors lacks a dynamic correlation calibration mechanism with the overall geometric deformation of the structure. That is, a unified data calibration benchmark based on the changes in the spatial morphology of the structure has not been established. Data processing is only based on the static calibration parameters of the sensors themselves or a single-dimensional local correction model.

[0004] This deficiency leads to the following related consequences: Because large-span prestressed concrete structures are prone to overall geometric deformation under construction loads and environmental changes, the installation reference of each sensor will shift with the structural deformation. Existing methods do not incorporate this overall deformation into the data calibration logic, resulting in systematic deviations in the original sensor measurement data, which cannot accurately reflect the actual stress and deformation state of the structure. Fusion analysis and risk assessment based on this deviation data will generate a chain of errors, leading to misjudgments of risk warning signals (such as underreporting critical structural hazards or falsely reporting normal deformation). Consequently, supervisory decisions lack accurate data support, and the closed-loop supervision mechanism for the construction process is difficult to implement effectively. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide an engineering supervision and management method that can provide accurate and reliable technical support for the construction supervision of large-span prestressed concrete structures, and ensure construction quality and structural safety.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] Firstly, an engineering supervision and management method, the method comprising:

[0008] Multi-dimensional and heterogeneous supervision data are collected by various sensors deployed at the construction site; among them, various sensors are deployed at the preset spatial reference points of the large-span prestressed concrete load-bearing structure; the preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key section of the structure and the center point of the prestressed anchorage end.

[0009] Based on preset spatial reference points, a spatial reference control volume covering the local or overall structure is virtually constructed in the edge computing nodes. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on multi-dimensional heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume in continuous time series.

[0010] Based on the volume change rate, the original measurement data of the relevant sensors collected in advance are corrected for geometric consistency, and the calibrated supervision data and sensor calibration status parameters are generated simultaneously.

[0011] The calibrated supervision data is cross-validated to form validated supervision data; the validated supervision data is then uploaded to the cloud-based intelligent supervision center and encrypted and stored using an anti-tampering evidence storage engine.

[0012] In the cloud-based intelligent supervision center, the verified supervision data is fused and analyzed by combining sensor calibration status parameters to generate fusion analysis results; based on the fusion analysis results, a visual supervision interface that is linked with the BIM digital twin model is constructed.

[0013] Based on the results of the integrated analysis, risk warnings and supervision decisions are made with the assistance of a visual monitoring interface, so as to achieve closed-loop supervision of the entire construction process.

[0014] Furthermore, diverse and heterogeneous monitoring data are collected through multiple sensors deployed at the construction site; these sensors are positioned at pre-defined spatial reference points on the prestressed concrete box girder; these pre-defined spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key structural sections and the center point of the prestressed anchorage end, including:

[0015] Identify and determine several key monitoring sections of large-span prestressed concrete load-bearing structures;

[0016] On the web centerline of each key monitoring section, locate the one-quarter, one-half, and three-quarter positions in the span direction as spatial reference points for deformation monitoring.

[0017] In the central region of the prestressed anchoring system at both ends of a large-span prestressed concrete load-bearing structure, the geometric center point of the anchor plate is positioned as the spatial reference point for load transfer.

[0018] The deformation monitoring spatial reference point and the load transfer spatial reference point are jointly defined as the preset spatial reference point;

[0019] At each of the preset spatial reference points, a corresponding sensor cluster is deployed to synchronously collect multi-dimensional heterogeneous supervision data.

[0020] Furthermore, based on preset spatial reference points, a spatial reference control volume covering a local or overall structure is virtually constructed in the edge computing nodes. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on the multi-element heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume over a continuous time series, including:

[0021] Select at least four non-coplanar reference points from the preset spatial reference points as target vertices;

[0022] In the edge computing node, all target vertices are connected according to a preset topology connection order to generate an irregular polyhedron composed of target vertices, and the irregular polyhedron is defined as a spatial reference control volume.

[0023] Real-time spatial coordinate monitoring data of each target vertex is obtained through edge computing nodes;

[0024] Based on the spatial coordinate monitoring data, the instantaneous volume of the spatial reference control volume at each acquisition moment is calculated;

[0025] The rate of change of volume of the space reference control volume is calculated based on a series of instantaneous volumes over a continuous time series.

[0026] Furthermore, based on the volume change rate, the pre-collected raw measurement data from relevant sensors are geometrically corrected for consistency, and calibrated monitoring data and sensor calibration status parameters are generated simultaneously, including:

[0027] Based on the volume change rate of the spatial reference control volume, the structural geometric deformation correction coefficient under the corresponding time series is calculated.

[0028] Based on the structural geometric deformation correction coefficient and the pre-established mapping relationship between sensor measurements and spatial deformation, the required geometric consistency correction amount for the original measurement data of each relevant sensor is determined;

[0029] The geometric consistency correction amount is used to correct the original measurement data of the relevant sensors and generate calibrated supervision data.

[0030] During the data correction process, the magnitude and application time of the geometric consistency correction are recorded simultaneously. The magnitude and application time of the geometric consistency correction are then encapsulated together with the identification information of the corresponding sensor to generate sensor calibration status parameters.

[0031] Furthermore, the calibrated supervision data is cross-validated to form validated supervision data; the validated supervision data is then uploaded to the cloud-based intelligent supervision center and encrypted and stored using an anti-tampering evidence storage engine, including:

[0032] The calibrated supervision data is subjected to protocol compliance verification to verify whether the data collection and transmission process follows the preset secure communication protocol, and the first verification result is generated.

[0033] The calibrated supervision data is cross-referenced with the static quality certification documents from the material supplier to verify the logical consistency of the data and generate a second verification result.

[0034] Based on the results of the first and second verifications, a data verification conclusion is generated, and the verified supervision data is formed according to the data verification conclusion.

[0035] The verified supervision data will be uploaded to the cloud-based intelligent supervision center in the predetermined format.

[0036] The anti-tampering evidence storage engine within the cloud-based intelligent supervision center is invoked to perform encryption operations and timestamp solidification on the verified supervision data, generating legally valid on-chain supervision evidence.

[0037] Furthermore, within the cloud-based intelligent supervision center, the verified supervision data is fused and analyzed by combining sensor calibration status parameters to generate fusion analysis results. Based on these results, a visualized supervision interface linked to the BIM digital twin model is constructed, including:

[0038] After the on-chain supervision evidence is generated, the verified supervision data and corresponding sensor calibration status parameters are imported into the cloud-based intelligent supervision center.

[0039] Based on the sensor calibration status parameters, the reliability of the verified supervision data is quantitatively evaluated to form data weighting coefficients;

[0040] Based on the data weighting coefficients, the verified supervision data is subjected to weighted fusion analysis to identify contradictions and abnormal patterns and generate fusion analysis results.

[0041] The key indicators and spatial positioning information contained in the fusion analysis results are mapped to the corresponding components and coordinate positions of the BIM digital twin model to drive the synchronous update of the BIM digital twin model status and generate the updated BIM digital twin model.

[0042] Based on the updated BIM digital twin model and the fusion analysis results, a visual monitoring interface is constructed and rendered. The visual monitoring interface integrates dynamic data layers, anomaly alarm panels, and an overview of structural safety status.

[0043] Furthermore, based on the results of the integrated analysis, and with the assistance of a visual monitoring interface, risk warnings and supervisory decisions are made to achieve closed-loop monitoring of the entire construction process, including:

[0044] Based on the fusion analysis results, identify the abnormal indicators that exceed the preset threshold and their spatial distribution;

[0045] For the identified abnormal indicators, the severity and spatial clustering characteristics of the abnormal indicators are analyzed, and a graded risk warning signal is generated based on the severity and spatial clustering characteristics of the abnormal indicators.

[0046] The tiered risk warning signals are pushed to the abnormal alarm panel of the visual monitoring interface for alarm purposes. At the same time, the monitoring interface receives supervisory decision inputs for the tiered risk warning signals.

[0047] By combining the updated BIM digital twin model, an impact simulation analysis is performed on the input of supervision decisions to generate structured supervision instructions;

[0048] Structured supervision instructions are issued to the construction site terminals, and the execution status of the structured supervision instructions and the related supervision data feedback are tracked simultaneously to complete the closed-loop supervision of the entire construction process.

[0049] Secondly, an engineering supervision and monitoring system, which executes the method, including:

[0050] A data acquisition module is deployed to collect multi-dimensional and heterogeneous supervision data through various sensors deployed at the construction site. These sensors are deployed at preset spatial reference points on the large-span prestressed concrete load-bearing structure. The preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key structural sections and the center point of the prestressed anchorage end.

[0051] A calibration module is constructed to virtually build a spatial reference control volume covering a local or overall structure in the edge computing node based on preset spatial reference points. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on multi-dimensional heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume in a continuous time series.

[0052] The data verification module is used to perform geometric consistency correction on the raw measurement data of relevant sensors collected in advance based on the volume change rate, and simultaneously generate calibrated supervision data and sensor calibration status parameters; cross-validate the calibrated supervision data to form verified supervision data; upload the verified supervision data to the cloud-based intelligent supervision center, and encrypt and store it through an anti-tampering evidence storage engine;

[0053] The fusion analysis module is used to perform fusion analysis on multi-source supervision data in the cloud-based intelligent supervision center by combining sensor calibration status parameters, and generate fusion analysis results; based on the fusion analysis results, a visual supervision interface that is linked with the BIM digital twin model is constructed.

[0054] The early warning and decision-making module is used to make risk warnings and supervision decisions based on the results of the integrated analysis and with the assistance of a visual monitoring interface, so as to achieve closed-loop supervision of the entire construction process.

[0055] Thirdly, a computing device including a memory and a processor;

[0056] The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in the first aspect.

[0057] Fourthly, a computer-readable storage medium for storing a computer program for performing the method as described in the first aspect.

[0058] The above-described solution of the present invention has at least the following beneficial effects:

[0059] By deploying multiple sensors at preset spatial reference points in large-span prestressed concrete load-bearing structures to collect diverse and heterogeneous supervision data; constructing a spatial reference control volume through edge computing nodes to calculate the volume change rate under continuous time series, achieving dynamic calibration of data geometric consistency; performing multi-dimensional cross-validation on the calibrated supervision data and encrypting and storing it through an anti-tampering evidence storage engine; and conducting weighted fusion analysis in the cloud-based intelligent supervision center in conjunction with sensor calibration status parameters, linking with the BIM digital twin model to construct a visual supervision interface, and implementing hierarchical risk warning and full-process closed-loop supervision based on the fusion analysis results, this approach overcomes the core problem of data calibration being disconnected from the overall geometric deformation of large-span prestressed concrete structures in existing engineering supervision methods. It also solves the derivative problems of systematic deviations in original sensor measurement data and misjudgments of risk warning signals, improving the current situation where closed-loop supervision mechanisms are difficult to implement effectively during construction. This results in accurate and reliable supervision data and accurate and efficient risk warnings, providing reasonable supervision technology support for the construction of large-span prestressed concrete structures and effectively ensuring construction quality and structural safety. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of a project supervision and management method.

[0061] Figure 2 This is a schematic diagram of an engineering supervision and monitoring system;

[0062] Figure 3 This is a schematic diagram of a computing device. Detailed Implementation

[0063] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0064] An embodiment of the present invention proposes an engineering supervision and monitoring method, the method comprising:

[0065] Step 1: Collect multi-dimensional heterogeneous supervision data by deploying multiple sensors at the construction site; among which, multiple sensors are deployed at the preset spatial reference points of the large-span prestressed concrete load-bearing structure; the preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key section of the structure and the center point of the prestressed anchorage end.

[0066] Step 2: Based on preset spatial reference points, a spatial reference control volume covering the local or overall structure is virtually constructed in the edge computing nodes. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on the multi-dimensional heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume in continuous time series.

[0067] Step 3: Perform geometric consistency correction on the pre-collected raw measurement data of relevant sensors based on the volume change rate, and simultaneously generate calibrated supervision data and sensor calibration status parameters.

[0068] Step 4: Cross-validate the calibrated supervision data to form validated supervision data; upload the validated supervision data to the cloud-based intelligent supervision center and encrypt and store it through the anti-tampering evidence storage engine.

[0069] Step 5: In the cloud-based intelligent supervision center, the verified supervision data is fused and analyzed in combination with the sensor calibration status parameters to generate fusion analysis results; based on the fusion analysis results, a visual supervision interface that is linked with the BIM digital twin model is constructed.

[0070] Step 6: Based on the results of the fusion analysis, risk warning and supervision decisions are made with the assistance of a visual monitoring interface to achieve closed-loop supervision of the entire construction process.

[0071] In this embodiment of the invention, by deploying sensors at specific preset spatial reference points, diverse and heterogeneous supervision data can be accurately collected, providing a reliable foundation for subsequent analysis. By constructing a spatial reference control volume using edge computing nodes and calculating the volume change rate, real-time dynamic calibration of the data is achieved, effectively avoiding systematic deviations in the original measurement data. Cross-validation and encrypted storage of the calibrated data ensures its reliability and security, while also possessing legal validity. Fusion analysis is performed by combining sensor calibration status parameters, and a visual interface is constructed in conjunction with the BIM digital twin model, making the supervision process more intuitive and the analysis results more accurate. The visual interface assists in risk warning and supervision decision-making, achieving closed-loop supervision throughout the construction process and enabling the timely detection of structural safety hazards.

[0072] In a preferred embodiment of the present invention, step 1 above may include:

[0073] Step 1.1: Identify and determine several key monitoring sections of the large-span prestressed concrete load-bearing structure. Specifically, this includes: first, collecting a complete set of design data for the large-span prestressed concrete load-bearing structure, including structural construction drawings, stress calculation reports, structural safety level descriptions, and seismic fortification intensity documents, to fully understand the structure's design parameters and core stress requirements; second, combining the structure's design span, support system form, and prestressed steel strand arrangement, analyzing the main stress paths and stress distribution patterns during the construction phase, focusing on the mid-span area, support constraint area, adjacent area of ​​prestressed anchorage ends, and key connection points of main load-bearing components, as these areas are prone to stress concentration or significant deformation due to load action; third, referring to the key process arrangements in the construction organization design, such as the concrete pouring sequence, prestressing batch tensioning scheme, and formwork support removal sequence, judging the degree of influence of each process on the local and overall stress state of the structure; fourth, from the key parts analyzed above, selecting sections that comprehensively cover the key stress nodes of the structure without redundancy, numbering each selected section and clearly marking it on the design drawings, and finally determining these sections as key monitoring sections.

[0074] Step 1.2: On the web centerline of each critical monitoring section, locate the one-quarter, one-half, and three-quarters span points as spatial reference points for deformation monitoring. Specifically, for each numbered critical monitoring section, first clean the section surface to expose the complete web outline; combining the design dimensions of the section with the actual cast-in-place web shape, determine the actual centerline position of the web using a pre-set measurement procedure, and make a clear and indelible mark on the web surface. That is, along the longitudinal axis of the structure, measure the total span length corresponding to both ends of the critical monitoring section, record the measurement results, and perform two repeated verifications to ensure the accuracy of the span length data; based on the verified total span length, calculate the specific distance according to the proportional relationship in the span direction, divide the total span length into four equal parts, and take the distance corresponding to the one-quarter span position at the end of the first part. The distance corresponding to the half-span position is taken from the middle dividing point, and the distance corresponding to the three-quarters span position is taken from the end of the third segment. Starting from one end of the web centerline, the position is located segment by segment according to the calculated distance values. After each preliminary position is determined, a unified positioning process is initiated: The process is based on the determined global spatial reference control points of the structure and the verified span length data. First, the actual spatial coordinates of the preliminary position are obtained, and then the theoretical calculated coordinates of the corresponding position are retrieved. The two sets of coordinates are compared dimension by dimension, and the deviation value is calculated. If the deviation exceeds the allowable range, the preliminary position is adjusted until the deviation meets the requirements to ensure accurate positioning. At each accurately positioned position, a non-wear-resistant mark is fixed on the web surface as a spatial reference point for deformation monitoring, and the number of each reference point, the associated information of the corresponding key monitoring section, and the spatial coordinate data are recorded in detail.

[0075] Step 1.3: In the central area of ​​the prestressed anchorage system at both ends of the large-span prestressed concrete load-bearing structure, locate the geometric center point of the anchor plate as the spatial reference point for load transfer. Specifically, this includes: first, clearly defining the pre-set installation areas of the prestressed anchorage system at both ends of the large-span prestressed concrete load-bearing structure; after arriving at the corresponding area on the construction site, clearing away obstructions to expose the complete structure of the anchorage system; accurately identifying and locating the solid structure of the anchor plate from the anchorage system; observing the appearance of the anchor plate to distinguish its body from the boundaries of surrounding connecting components; meticulously examining the outer edge of the anchor plate segment by segment to clarify its actual contour and edge range; initiating the pre-set dimension confirmation process; selecting three different points at both ends and the middle along the long side of the contour for length measurement; similarly selecting three different points at both ends and the middle along the short side for width measurement; recording the measurement results at each point; and repeatedly verifying the obtained length and width data. Multiple measurements in one dimension are compared to calculate the deviation. If the deviation exceeds the allowable range, the measurements are remeasured and adjusted until the measurements in the same dimension are consistent to ensure accurate dimensional data. The geometric center coordinates are calculated based on the verified actual dimensions. The specific process is as follows: First, using one long side of the anchor plate as a reference, determine the endpoints of the long side and take the midpoint of the line connecting the two points as the midpoint of the long side. Then, using one short side of the anchor plate as a reference, determine the endpoints of the short side and take the midpoint of the line connecting the two points as the midpoint of the short side. Draw a straight line perpendicular to the long side through the midpoint of the long side and a straight line perpendicular to the short side through the midpoint of the short side. The intersection of these two lines is the geometric center of the anchor plate. Mark this geometric center point on the surface of the anchor plate. After marking, verify whether the center point coincides with the preset axis of prestress transfer to ensure accurate positioning. The marked point is the load transfer spatial reference point, and the reference point number and its association information with the corresponding anchoring system are recorded simultaneously.

[0076] Step 1.4 defines the deformation monitoring spatial reference points and load transfer spatial reference points as preset spatial reference points. This includes: First, compiling all deformation monitoring spatial reference point information, recording the unique number of each reference point, the corresponding key monitoring section number, spatial coordinate data, and the installation location characteristics of the markers. Classifying and organizing these reference points by key monitoring sections, a mapping relationship is established between reference points and their corresponding structural components. This means that by establishing a one-to-one correspondence between reference point numbers and key monitoring section numbers, the specific structural section to which each deformation monitoring reference point belongs and its location within the overall structure is clarified. Next, comprehensively collecting all load transfer spatial reference point information, recording in detail the unique number of each reference point, the corresponding anchoring system number, spatial coordinate data, and the specific location characteristics of the markers. Classifying and organizing these reference points by the distribution area of ​​the anchoring systems, a mapping relationship is also established between reference points and their corresponding anchoring systems. This mapping relationship is then established by linking the reference point number with the anchoring system. The system is numbered and bound to clearly define the specific anchoring structure corresponding to each load transfer reference point. Next, the coordinate system processing flow is initiated, using the global spatial reference set at the construction site as a unified reference. The original spatial coordinate data of both types of reference points are converted into the coordinate format corresponding to this global reference. During the conversion process, the integrity and accuracy of the coordinate data are checked simultaneously to avoid subsequent data correlation deviations due to coordinate system differences. After the coordinate unification is completed, the two types of standardized reference points are included in the scope of preset spatial reference points. A unique global number is assigned to each preset spatial reference point, and its reference point type (deformation monitoring or load transfer) and corresponding structural association information are marked. Finally, a complete list is formed, including the global number, type, associated structural part number, associated structural part name, spatial coordinates under the unified coordinate system, and marking characteristics of all preset spatial reference points. At the same time, corresponding coordinate files are established and properly stored.

[0077] Step 1.5: Deploy corresponding sensor clusters at each of the preset spatial reference points to synchronously collect multi-dimensional heterogeneous monitoring data. Specifically, this includes: Positioning the deformation monitoring spatial reference points according to their monitoring functions in the preset spatial reference point list; configuring displacement sensors and strain sensors for the deformation monitoring spatial reference points. Displacement sensors capture longitudinal, lateral, and vertical displacement changes in the structure, while strain sensors monitor the stress and strain of the web. For the load transfer spatial reference points, configuring pressure sensors and stress sensors. Pressure sensors collect pressure data on the anchor plates, while stress sensors monitor stress changes during prestress transfer. All types of sensors together form a sensor cluster. Then, securely install each sensor at its corresponding preset spatial reference point mark, ensuring the sensor's sensing surface is completely flush with the reference point mark without gaps, and that the sensor's installation direction meets the requirements for collecting monitoring parameters. After all sensors are installed, perform unified debugging, setting the same data acquisition frequency and data transmission protocol. Calibrate the acquisition accuracy of each sensor. Once the sensor communication connection is stable and data acquisition is normal, start the sensor cluster to work synchronously, collecting multi-dimensional heterogeneous monitoring data such as structural deformation, stress, and pressure in real time.

[0078] In this embodiment of the invention, key monitoring sections of large-span prestressed concrete load-bearing structures are accurately identified, clarifying the core monitoring range; specific span locations are positioned at the centerline of the web of the key monitoring sections, making the selection of deformation monitoring benchmarks more targeted; the geometric center point of the anchor plate is located in the central area of ​​the prestressed anchoring system, accurately capturing key nodes of load transfer; two types of benchmarks are integrated to form preset spatial benchmarks, achieving comprehensive coverage of monitoring dimensions; sensor clusters are deployed at each preset spatial benchmark to ensure the synchronous acquisition of diverse and heterogeneous supervision data; the overall design makes the layout of monitoring points more reasonable, and the collected data more closely matches the core characteristics of structural stress and deformation.

[0079] In a preferred embodiment of the present invention, step 2 above may include:

[0080] Step 2.1: Select at least four non-coplanar reference points from the preset spatial reference points as target vertices. The selection range of target vertices is determined according to the current supervision and monitoring requirements: if the monitoring target is the overall geometric deformation of the structure, select from all preset spatial reference points; if the monitoring target is a key local area of ​​the structure, select from the preset spatial reference points within that local area. Specifically, this includes: first, retrieving the complete list of preset spatial reference points and their corresponding coordinate files, extracting the core information of each reference point in the list, including the unique global number, reference point type (deformation monitoring or load transfer), associated structural part number and name, and three-dimensional spatial coordinates (x, y, z) under a unified global coordinate system, and organizing this information into a standardized data table; next, clarifying the coverage of the spatial reference control body according to the specific requirements of this supervision and monitoring; if the monitoring target is a key local area of ​​the structure, such as the mid-span load-bearing section or the support connection section, selecting all preset spatial reference points within that local area from the data table; if the monitoring target is the entire structure, retaining all preset spatial reference points in the table; then, starting... The process for selecting non-coplanar reference points begins by randomly selecting three reference points as an initial candidate group. The spatial relationship between these three points is then used to determine if they are coplanar: Connect each pair of the three candidate points to form two intersecting line segments. If these two line segments perfectly align on the same virtual plane without spatial misalignment, the three points are considered coplanar and a new group needs to be selected; otherwise, they are considered non-coplanar, thus completing the initial group selection. Next, a fourth candidate point is selected from the remaining reference points, and its spatial position is compared with the three points in the initial candidate group: It is observed whether this candidate point can align perfectly with the three points in the initial group. On a virtual plane, if not, the four points are determined to be non-coplanar, completing the initial screening. If a polyhedron with more vertices needs to be constructed, the subsequent non-coplanar vertices are selected according to the same logic. During the screening process, it is necessary to ensure that the selected target vertices cover both deformation monitoring and load transfer reference points to avoid the control volume being unable to fully reflect the relationship between structural stress and deformation due to a single type of reference point. Finally, at least four non-coplanar reference points are determined as target vertices, and their information is organized into a dedicated data table for target vertices, recording the global number, coordinates, and type information of each vertex.

[0081] Step 2.2: In the edge computing node, connect all target vertices according to a preset topology connection order to generate an irregular polyhedron composed of target vertices. Simultaneously, define the irregular polyhedron as a spatial reference control volume. Specifically, this includes: First, importing a compiled target vertex-specific data table into the edge computing node, simultaneously loading spatial distribution characteristic information such as the main force path and component connection relationships of the large-span prestressed concrete load-bearing structure. Based on this, a preset topology connection order is established: vertices are connected sequentially first along the main force direction of the structure to ensure that the connected three-dimensional structure conforms to the actual shape of the structure and accurately captures volume changes during stress deformation; then, starting... The spatial structure construction process begins by inputting the three-dimensional coordinates of the target vertices into the node's data processing unit in a preset order, forming an ordered sequence of vertex coordinates. This means that the input order strictly follows the preset topological connection logic, and each coordinate is recorded in the format of vertex global number + three-dimensional coordinate value to ensure that the vertex order is not confused during subsequent connections. Then, following the topological connection order, spatial connection relationships between adjacent vertices are established one by one to form continuous spatial line segments. At the same time, the two endpoint numbers of each line segment are recorded to ensure that the connection relationship is traceable. After all line segments are generated, closed spatial surfaces are formed by connecting the beginning and end of the line segments. Adjacent spatial surfaces are seamlessly connected, ultimately forming a complete closed three-dimensional solid structure.

[0082] Since the target vertices satisfy the non-coplanar condition, when there are four vertices, an irregular tetrahedron is directly formed. When there are more than four vertices, a preset tetrahedron partitioning rule is followed: using a core vertex as a reference, the remaining vertices are grouped according to their spatial distribution. Each group of vertices combines with the core vertex to form a sub-tetrahedron, ensuring that each sub-tetrahedron does not overlap and collectively covers the entire closed structure. The closed structure is decomposed into multiple non-overlapping irregular tetrahedrons, and the collection of all sub-tetrahedrons constitutes a complete irregular polyhedron. Then, this closed three-dimensional solid structure is defined as a spatial reference control volume, and a control volume is established in the edge computing nodes, linking it to the target vertex. A point association mapping table clearly records the global number, the number of its sub-tetrahedron, the numbers of adjacent vertices, and the corresponding spatial coordinates of each vertex, clarifying the spatial ownership and association of each vertex in the control volume. Finally, the closure and integrity of the control volume are verified by checking whether all spatial line segments are completely connected end to end without breaks, whether spatial surfaces completely cover the structural outline without gaps or overlapping areas, and confirming that all target vertices are included in the control volume without redundant vertices. If there are non-closed areas, missing vertices, or overlapping issues, the topology connection order is readjusted and the construction process is repeated until the control volume meets the usage requirements.

[0083] Step 2.3 involves acquiring real-time spatial coordinate monitoring data for each target vertex via edge computing nodes. This includes: proactively initiating communication connection requests to the deployed sensor cluster; establishing dedicated communication links with the sensors corresponding to each target vertex using a stable communication protocol; synchronously performing encryption authentication during link establishment to ensure data transmission security; performing time synchronization calibration with the sensor cluster by sending time synchronization commands based on its own time to ensure the sensor's acquisition clock remains consistent, with synchronization errors controlled within a preset allowable range to ensure accurate time matching of subsequent data; and issuing acquisition commands to the sensor cluster at a unified acquisition frequency, with the sensors synchronously acquiring real-time spatial coordinate data of the corresponding target vertex and transmitting the data back. When transmitting and receiving data, a unique millisecond-level timestamp is added to each set of target vertex coordinate data to ensure that different vertex data are accurately associated with the same acquisition time. A preliminary validity check is performed on the received data, and a check threshold is set according to the reasonable coordinate range of the area. If the data exceeds the threshold, it is marked as abnormal and the vertex number and acquisition time are recorded. If data is missing, a supplementary acquisition mechanism is immediately triggered to re-request acquisition from the corresponding sensor. Supplementary acquisition is limited to no more than 3 times. If it is still invalid, the missing status is recorded and an alarm is triggered. Valid data that has passed the check is stored in the local high-speed cache in the order of timestamps. At the same time, a receiving log is generated to record the acquisition time, number of vertices, amount of valid data, and abnormal or missing situations to ensure data integrity, timeliness, and traceability.

[0084] Step 2.4: Based on the spatial coordinate monitoring data, calculate the instantaneous volume of the spatial reference control volume at each acquisition moment. Specifically, this includes: calculating the instantaneous volume of the spatial reference control volume at each acquisition moment based on the geometric principle of volume calculation for irregular tetrahedrons. When there are four target vertices, an irregular tetrahedron is directly constructed. When there are more than four target vertices, the polyhedron is decomposed into multiple non-overlapping irregular tetrahedrons, and the volumes are calculated separately and then summed. Taking the volume calculation of a single irregular tetrahedron as an example, the core principle is: constructing vectors using the spatial coordinates of the four vertices of the tetrahedron and solving for the volume using a combination of vector cross product and dot product. Let the spatial coordinates of the four target vertices be... , , , Where x, y, and z represent the coordinate values ​​of the three orthogonal dimensions along the unified global coordinate system, corresponding to the vertical, horizontal, and longitudinal directions of the structure; Building from the base point Vectors pointing to the other three vertices , , The components of a vector are calculated using the difference between the coordinates of two points, as shown in the following expression:

[0085] ;

[0086] First, for vectors and Performing the cross product operation yields the normal vector perpendicular to the plane containing these two vectors. The result of the cross product operation middle, ,in They respectively refer to the normal vector obtained after the cross product operation of vectors. The components in the three orthogonal dimensions of the unified global coordinate system; then the normal vector with vector Perform a dot product operation to obtain the scalar value. The volume V of the irregular tetrahedron is obtained by dividing the absolute value of the scalar value T by 6. The formula for calculating the volume V is:

[0087] ;

[0088] For a spatial reference control volume composed of multiple tetrahedrons, the volume of each sub-tetrahedron is calculated one by one according to the above method. Then, the volumes of all sub-tetrahedrons are summed to obtain the instantaneous volume of the spatial reference control volume at the acquisition time. The instantaneous volume is then associated with and stored with the timestamp of the corresponding acquisition time.

[0089] Step 2.5: Calculate the volume change rate of the spatial reference control volume based on a series of instantaneous volumes in a continuous time series. This specifically includes: first, extracting instantaneous volume data from the local cache module of the edge computing node for all acquisition times, sorting them in ascending order of timestamps to form a set of instantaneous volumes in a continuous time series. Where V0 represents the instantaneous volume at the initial acquisition time (i.e., the time of the first effective acquisition after monitoring starts), This represents the instantaneous volume at the first acquisition moment, and so on. The instantaneous volume at the nth acquisition time is represented; simultaneously, the corresponding acquisition time sequence is extracted. ,in This represents the specific time of the i-th acquisition, and the time interval between adjacent acquisition moments is determined by a preset acquisition frequency. For example, when the acquisition frequency is 10Hz, the time interval is 0.1 seconds. The core principle of calculating the volume change rate is: by comparing the instantaneous volume difference between two consecutive acquisition moments with the time difference between these two moments, the degree of volume change of the spatial reference control volume per unit time is obtained, thereby quantifying the real-time geometric deformation rate of the structure. The specific calculation formula is as follows: ,in This represents the rate of volume change from the (i-1)th to the ith sampling time. Its positive or negative value reflects the trend of volume increase or decrease, i.e., positive value indicates volume increase and negative value indicates volume decrease; the absolute value reflects the rate of change. It represents the instantaneous volume difference between two adjacent data collection times, directly reflecting the volume change. This represents the time difference between two adjacent acquisition moments, i.e., the data acquisition time interval. Following the formula above, the volume change rate is calculated sequentially for every two adjacent acquisition moments in the continuous time series, generating the corresponding volume change rate sequence. The sequence is then associated with and stored with the corresponding timestamp sequence to form a real-time updated volume change rate dataset. This dataset can dynamically reflect the volume change law of the spatial reference control body over time, thereby indirectly mapping the overall geometric deformation state of the large-span prestressed concrete load-bearing structure.

[0090] In this embodiment of the invention, at least four non-coplanar reference points are selected as target vertices, providing a reliable spatial foundation for the construction of the spatial reference control body and ensuring that the control body has a complete three-dimensional shape. An irregular polyhedron is constructed as the spatial reference control body according to a preset topological connection order, which can conform to the actual shape of the structure, accurately cover the monitoring range, and effectively correlate the overall spatial characteristics of the structure. Real-time acquisition of spatial coordinate monitoring data for each target vertex ensures the timeliness and synchronization of the data, providing an accurate data source for subsequent volume calculation. Instantaneous volume is calculated based on the coordinate data, transforming discrete vertex coordinates into intuitive control body volume parameters, clearly reflecting the spatial state of the monitoring area. The volume change rate is calculated through the instantaneous volume of a continuous time series, dynamically capturing the geometric deformation rate of the structure and establishing a direct correlation between data changes and the overall deformation of the structure. The overall process effectively overcomes the problem of data disconnection from the overall deformation of the structure in existing methods, improving the matching degree between the supervision data and the actual state of the structure.

[0091] In a preferred embodiment of the present invention, step 3 above may include:

[0092] Step 3.1: Based on the volume change rate of the spatial reference control body, calculate the structural geometric deformation correction coefficient under the corresponding time series. Specifically, this includes: firstly, extracting the complete volume change rate sequence of the spatial reference control body generated in Step 2.5, and simultaneously extracting the millisecond-level timestamps corresponding to each volume change rate to form a data group associated with the volume change rate and timestamps, to ensure the integrity and traceability of the time dimension; clarifying the core correlation logic between the volume change rate and the overall geometric deformation of the structure, that is, when the structural region wrapped by the spatial reference control body undergoes three-dimensional deformation, it will directly cause a change in the volume of the control body. The absolute value of the volume change rate is positively correlated with the structural deformation, and the positive and negative signs correspond to the stretching and contraction trend of the deformation. Positive values ​​are dominated by stretching deformation, and negative values ​​are dominated by compression deformation; introducing a spatial affine transformation algorithm as the core support, decomposing the volume change rate at each moment into three independent affine transformation components according to the algorithm principle: translation deformation component, rotation deformation component, and scaling deformation component.

[0093] The derivation process for the translational deformation component is as follows: First, the center-of-gravity coordinates of the spatial reference control body at each acquisition moment are calculated. Then, these coordinates are compared with the center-of-gravity coordinates of the control body at the initial monitoring moment to obtain the offset of the center of gravity in three-dimensional space. Combining the trend of the offset with time, after eliminating the interference of random fluctuations, the translational deformation component reflecting the linear positional offset of the overall structure is finally derived. The derivation process for the rotational deformation component is as follows: Record the positional changes of each target vertex of the control body relative to the initial state at each moment, analyze the angle changes of the line connecting any two adjacent vertices, verify the consistency through the angle change patterns of multiple sets of adjacent vertices, eliminate the influence of local deformation, and calculate the rotational deformation component reflecting the spatial attitude deflection of the structure based on the verified angle change trend. The derivation process for the scaling deformation component is as follows: Extract the instantaneous volume of the control body at each moment and the original volume of the initial state, calculate the ratio between the two, and then combine the range of the structural region enclosed by the control body to convert the volume ratio into the degree of scaling in the corresponding dimension, finally obtaining the scaling deformation component that directly reflects the scaling state of the structure.

[0094] Based on the design stiffness distribution data, material elastic modulus parameters, and deformation sensitivity levels of different locations in large-span prestressed concrete load-bearing structures, appropriate weighting coefficients are assigned to the three affine transformation components. Regions with lower stiffness and greater deformation sensitivity, such as the mid-span area, have higher weights for the scaling deformation component, while constrained areas like supports have higher weights for the rotational deformation component. The sum of the weighting coefficients is 1. For example, in the mid-span area, the translational deformation component can have a weight of 0.2, the rotational deformation component 0.3, and the scaling deformation component 0.5. A weighted fusion calculation is then performed on the three components based on these weighting coefficients, using a simple weighted summation formula. The final correction coefficient is calculated as follows: Final structural geometric deformation correction coefficient = Translation deformation component × Translation weight + Rotation deformation component × Rotation weight + Scaling deformation component × Scaling weight. In the specific calculation, each component is first multiplied by its corresponding weight coefficient to obtain the intermediate correction coefficient for each component. Then, the three intermediate correction coefficients are added together to obtain the final structural geometric deformation correction coefficient. Simultaneously, each correction coefficient is bound to a corresponding timestamp to ensure precise alignment with the subsequent sensor data acquisition time, ultimately forming a complete data sequence of timestamp, translation component, rotation component, scaling component, and final correction coefficient.

[0095] Step 3.2: Based on the structural geometric deformation correction coefficient and the pre-established mapping relationship between sensor measurements and spatial deformation, determine the required geometric consistency correction amount for the original measurement data of each relevant sensor. Specifically, this includes: first, retrieving the pre-constructed mapping relationship between sensor measurements and spatial deformation from a pre-set database. The construction process of this mapping relationship encompasses preliminary multi-scenario structural simulation tests and on-site calibration tests. The simulation tests simulated the affine transformation deformation of the structure under different loads and environmental conditions, including translation, rotation, and scaling, and recorded the changes in the measurement values ​​of various sensors. The on-site calibration tests applied controllable force to typical parts of the actual structure. The deformation is verified and the simulation data is corrected, ultimately forming a mapping relationship system covering all sensor types and installation locations. The system is stored in a three-level classification according to sensor type, installation location, and monitoring parameters. Each classification contains quantitative correspondence rules between different affine transformation components and sensor measurement values. The timestamp, affine transformation component, and correction coefficient data sequence generated in step 3.1 are matched one by one to the sensor working scene at the corresponding acquisition time according to the timestamp. At the same time, based on the unique identification information of the sensor to be corrected, the global number, installation location number, and monitoring parameter type, the appropriate sub-mapping relationship is accurately retrieved from the three-level classification system.

[0096] When calculating the correction amount based on the preset quantization rules in the sub-mapping relationship, the core content of the quantization rules is first clarified, namely, the pre-defined range of structural geometric deformation correction coefficients, the corresponding table of basic correction amounts, and the corresponding table of the proportion of each affine transformation component and the correction adjustment coefficient. First, based on the current value of the structural geometric deformation correction coefficient, the corresponding basic correction amount is matched in the table. For example, when the correction coefficient is in the range of 0.3 to 0.5, the basic correction amount is 0.2 mm. Then, the proportion data of each affine transformation component in step 3.1 is extracted, and the corresponding adjustment coefficient for each component is matched from the adjustment coefficient table according to the proportion. For example, a 60% proportion of the translation component corresponds to an adjustment coefficient of 1.1, a 30% proportion of the rotation component corresponds to an adjustment coefficient of 0.9, and a 10% proportion of the scaling component corresponds to an adjustment coefficient of 0.8. Finally, the proportion of each component is multiplied by the corresponding adjustment coefficient. The component adjustment weights are obtained, and then the basic correction amount is summed with the adjustment weights of each component to finally calculate the geometric consistency correction amount required for the original measurement data of the sensor. If the scaling deformation component is the main component, the correction amount is adjusted with reference to the correspondence rules between scaling deformation and measurement value in the mapping relationship. If the proportion of translation or rotation deformation is higher, the quantization rules of the corresponding component are matched first. The direction of the correction amount must strictly correspond to the deformation direction of the affine transformation component. For example, the longitudinal translation deformation of the structure corresponds to the same direction correction of the longitudinal displacement sensor measurement value; the rotation deformation of the structure around the transverse axis corresponds to the reverse correction of the transverse strain sensor measurement value. After the calculation is completed, the complete correlation between the correction amount and the correction coefficient, the proportion of the affine transformation component, and the sensor information is recorded to form a correction amount data set of sensor identifier, timestamp, correction amount, and associated components.

[0097] Step 3.3 involves correcting the original measurement data of the relevant sensors using the geometric consistency correction amount to generate calibrated supervision data. Specifically, this includes: first, organizing the collected original measurement data of the relevant sensors, classifying and categorizing them according to sensor identification information and timestamps to form a structured data set of sensor identification, timestamps, and original measurement values; then, precisely matching the determined geometric consistency correction amount to the corresponding original measurement data according to the sensor identification and timestamp, ensuring that each sensor's original data at each acquisition time corresponds to a unique correction amount; and finally, performing data correction calculations based on the inverse transformation logic of the spatial affine transformation algorithm: for displacement sensors, if translational deformation is dominant, directly superimposing the original measurement value with the translational correction component to obtain the calibration value; if rotational deformation is present, then... The original measured values ​​are adjusted according to the geometric offset corresponding to the rotation angle, and then the correction is added. For strain sensors, the angle deviation between the sensor measurement direction and the structural deformation direction is corrected first according to the angle change corresponding to the rotation deformation component, and then the correction corresponding to the scaling deformation is added. For pressure sensors, the focus is on adjusting the pressure transmission deviation in the original measured values ​​caused by structural expansion and contraction, combined with the scaling deformation correction component. After the correction is completed, the validity of each set of calibration data is verified. The calibration data is compared with the reasonable measurement range allowed by the structural design and the preset error threshold (the threshold is finely adjusted according to the structural stress state at different acquisition times). If it exceeds the range, the matching of the correction and mapping relationship is checked until the calibration data meets the requirements. Finally, the post-calibration supervision data with sensor identification and timestamp is generated.

[0098] Step 3.4: During the data correction process, the magnitude and application time of the geometric consistency correction are recorded simultaneously. The magnitude and application time of the geometric consistency correction are then encapsulated along with the corresponding sensor identification information to generate sensor calibration status parameters. Specifically, this includes: simultaneously recording the magnitude of each geometric consistency correction in real time, including the total correction and the corrections for each affine transformation component after decomposition, with the magnitude recording retaining a preset precision to ensure data accuracy; simultaneously recording the application time of the correction, using a millisecond-level timestamp format consistent with the post-calibration supervision data to ensure accurate correlation in the time dimension; extracting the unique identification information of the corresponding sensor, including the global number, installation location number, and monitoring parameter type, and integrating the sensor identification information, the magnitude of the geometric consistency correction (total correction + component correction), the application time, and the corresponding structural geometric deformation correction coefficient; encapsulating the integrated information according to a set standardized format to generate sensor calibration status parameters. Each sensor generates an independent calibration status parameter sequence, which is stored synchronously with the post-calibration supervision data and the original measurement data by timestamp, providing a complete geometric deformation calibration basis for subsequent data traceability, sensor performance evaluation, and mapping relationship optimization.

[0099] In a preferred embodiment of the present invention, step 4 above may include:

[0100] Step 4.1: Perform protocol compliance verification on the calibrated supervision data to verify whether the data acquisition and transmission process follows the preset secure communication protocol and generate the first verification result. This specifically includes: first, extracting the calibrated supervision data; simultaneously retrieving the preset secure communication protocol specification, i.e., clarifying the core verification items required by the protocol, including data packet header format, field length, encryption algorithm identifier, checksum generation rules, and transmission timing requirements; initiating the protocol compliance verification process, first parsing the transmission protocol header information of the calibrated supervision data, checking whether the header fields are complete and whether the format is consistent with the specification; if there are missing fields or format deviations, they are directly marked. The protocol is deemed non-compliant. Next, the encrypted fields during data transmission are verified to ensure the encryption algorithm used is consistent with the specifications. The integrity of the decryption of the encrypted fields is also verified to avoid data leakage risks due to incomplete encryption. Subsequently, the verification code for the calibrated supervisory data is recalculated and compared with the verification code carried in the data. If the two are inconsistent, it is determined that the data transmission process may have been tampered with. During the verification process, each verification result is recorded in real time, including qualified items, unqualified items, and anomaly details. Finally, the first verification result is generated, clearly indicating whether it is compliant or non-compliant. For non-compliant items, the specific violation type and corresponding data segment identifier must be provided.

[0101] Step 4.2 involves cross-referencing the calibrated supervision data with the static quality certification documents from the material suppliers to verify data logical consistency and generate a second verification result. Specifically, this includes: first, retrieving the static quality certification documents from the material suppliers for the corresponding large-span prestressed concrete load-bearing structure. These documents cover core static parameters such as material strength grade, elastic modulus, fatigue resistance, allowable design stress, and preset deformation range. The documents are then categorized and organized according to material type and component location to form a standardized static parameter comparison table. Next, key monitoring parameters are extracted from the calibrated supervision data, including dynamic monitoring data such as stress, strain, displacement, and deformation rate for various structural components. These are categorized by component location and monitoring time to establish a comparison with the static data. The process involves mapping the parameters to a reference table; initiating a cross-comparison process to match the corresponding material static parameter thresholds or reasonable ranges from the static parameter reference table for each dynamic monitoring data item. For example, the monitored stress data of concrete components is compared with the allowable stress of the material, and the monitored deformation data is compared with the allowable range of elastic deformation of the material. In addition to threshold comparison, the consistency of the data logical trend must also be verified. For example, when the material is subjected to increased stress, the monitored stress data should show a synchronous upward trend, and the increase should not exceed the theoretical range derived from the static parameters. After the comparison is completed, the comparison results of each set of data are recorded, including the status of consistency, deviation within the allowable range, and serious deviation. The corresponding static parameter basis must be marked for the deviation data, and finally, a second verification result is generated.

[0102] Step 4.3: Based on the first and second verification results, generate data verification conclusions and form verified supervision data according to the data verification conclusions. Specifically, this includes: first, integrating the first and second verification results, establishing a verification result association table, and clarifying the identifiers of the same batch of calibrated supervision data corresponding to the two results; setting the judgment rules for the data verification conclusions: if the first verification result is compliant and the second verification result shows data consistency or deviation within the allowable range, the verification conclusion is judged as valid; if the first verification result is non-compliant, regardless of the second verification result, it is judged as invalid and marked as a violation of the agreement; if the first verification result is compliant but the second verification result shows non-compliance... If there is a serious deviation, it is determined to be pending review, and the consistency between the calibration process and the static quality documents needs to be further verified. For data determined to be pending review, the review process is initiated to re-examine the calibration calculation logic and the accuracy of the correction quantity matching. At the same time, the validity and relevance of the static quality certification documents are verified to eliminate anomalies caused by document mismatch or calibration deviation. The data is organized according to the final judgment result. For data with valid verification conclusions, complete calibration information and verification results are retained to form verified supervision data. For invalid data, the reasons for invalidity are recorded and isolated for storage. Finally, a data verification report is generated, which includes the verification batch, the amount of valid data, the amount of invalid data, the amount of data pending review, and details of various anomalies.

[0103] Step 4.4 involves uploading the verified supervision data to the Cloud-based Intelligent Supervision Center according to a predetermined format. This includes: first, clarifying the core positioning of the Cloud-based Intelligent Supervision Center, which is a core platform integrating centralized storage, intelligent analysis, and full-process traceability management of supervision data. It is responsible for receiving the verified supervision data transmitted from the front end, integrating multi-dimensional monitoring information, and providing data support and technical empowerment for structural safety assessment, risk warning, and engineering supervision decision-making; then, retrieving the data upload format specification preset by the center. The specification clearly defines the sorting order, data type, numerical precision, encoding method, and file encapsulation requirements of the data fields. For example, it requires that the fields be arranged in the order of sensor identifier, monitoring time, verified data value, and data verification conclusion, with numerical precision retained to three decimal places, and encapsulated into a JSON file using UTF8 encoding format.

[0104] The data formatting process is initiated, adjusting the field order, correcting data types, and calibrating numerical precision of the verified supervision data according to specifications, while supplementing auxiliary fields such as data source identifiers and verification batch numbers. The formatted data undergoes an integrity check to ensure no missing fields or data format errors; if any issues are found, it is reformatted. A secure communication link is then established with the Yunji Intelligent Supervision Center, with identity authentication and encryption negotiation completed during link establishment to ensure data upload security. The formatted verification data is then uploaded in batches, awaiting a reception confirmation signal from the center after each batch. If no confirmation is received or a transmission error signal is received, the batch is re-uploaded, with a maximum of three retransmissions. After all uploads are complete, an upload log is generated, recording upload time, batch quantity, data volume per batch, upload status, etc., and a data reception receipt is obtained from the Yunji Intelligent Supervision Center. The receipt is stored in association with the upload log to ensure data upload traceability.

[0105] Step 4.5: Invoke the anti-tampering evidence storage engine within the Cloud-based Intelligent Supervision Center to perform encryption operations and timestamp solidification on the verified supervision data, generating legally valid on-chain supervision evidence. Specifically, this includes: first, clarifying the core function of the anti-tampering evidence storage engine, which is a core security component built into the Cloud-based Intelligent Supervision Center, specifically used for anti-tampering evidence storage of supervision data. Relying on asymmetric encryption algorithms and blockchain distributed storage, it can achieve data encryption protection, accurate timestamp solidification, and immutable on-chain storage, ensuring the legal validity and traceability of the evidence data; then, invoking the anti-tampering evidence storage engine to initiate an evidence storage processing request, carrying the verified supervision data identifier and upload batch information; after receiving the request, first performing encryption operations on the verified supervision data, encrypting the data using the engine's built-in private key to generate encrypted data blocks, and then generating the corresponding public key for data decryption and verification.

[0106] The process involves synchronously executing timestamp solidification. The engine synchronizes with an authoritative time source to obtain a standard timestamp accurate to the millisecond level. The timestamp is then bound to an encrypted data block, generating a timestamp verification code during the binding process to ensure the timestamp cannot be tampered with. Subsequently, the encrypted data block, the bound timestamp, the timestamp verification code, the data identifier, and the uploaded batch information are integrated and packaged into a notarization data unit according to the standard format of blockchain notarization. The notarization data unit is then uploaded to a designated blockchain node, which verifies the notarization data unit. Once verified, it is written to the blockchain, generating a unique on-chain notarization identifier. Finally, a legally valid on-chain supervisory notarization is generated, which includes the notarization identifier, the encrypted data block, the solidified timestamp, the timestamp verification code, and related associated information. Simultaneously, a notarization write confirmation is obtained from the blockchain node and stored in association with the on-chain supervisory notarization.

[0107] In a preferred embodiment of the present invention, step 5 above may include:

[0108] Step 5.1: After the on-chain supervision certificate is generated, the verified supervision data and corresponding sensor calibration status parameters are imported into the cloud-based intelligent supervision hub. Specifically, this includes: after the on-chain supervision certificate is generated, the verified supervision data of the corresponding batch is retrieved from the cloud-based intelligent supervision hub, and the sensor calibration status parameter sequence is extracted to ensure that both types of data carry complete sensor identifiers, timestamps, and associated batch information; the two types of data are preprocessed for association, and a one-to-one correspondence between the verified supervision data and the sensor calibration status parameters is established with the sensor identifier + timestamp as the core association key, forming an integrated data group of sensor identifier, timestamp, verification data, and calibration status parameters; then the integrated data group is format-validated to check the field completeness, data type consistency, and the rationality of the association relationship. If there is a format deviation, it is returned for correction and re-validation; after the validation is passed, the integrated data group is imported in batches. During the import process, the import progress, the number of successful entries, and the details of failures are recorded simultaneously. After the import is completed, the data is classified and stored in the structured database of the hub according to the monitoring batch, component location, and sensor type, and a data index is established to improve the efficiency of subsequent query and analysis.

[0109] Step 5.2: Based on the sensor calibration status parameters, quantitatively evaluate the reliability of the verified supervision data to form data weighting coefficients. Specifically, this includes: first, extracting core information of the sensor calibration status parameters from the imported integrated data set, including the magnitude of the geometric consistency correction, correction frequency, the correction ratio of each affine transformation component, and deformation correction coefficient, etc., clarifying the correlation logic between these parameters and data reliability: the smaller the correction magnitude and the lower the correction frequency, the less the sensor measurement reference is affected by structural deformation, and the higher the data reliability; conversely, the lower the reliability. Establish a multi-dimensional quantitative evaluation system, setting three core evaluation dimensions: correction stability dimension, correction... The system employs two dimensions: logical rationality and sensor adaptability. Each dimension corresponds to different evaluation indicators and scoring standards. For example, the correction stability dimension uses the fluctuation range of the correction amount over 10 consecutive data acquisition times as the indicator. A fluctuation range ≤5% earns a full score of 10 points, with larger fluctuation ranges resulting in lower scores. The logical rationality dimension verifies the matching degree between the correction direction and the structural deformation direction. A perfect match earns a full score of 8 points, a partial match earns 4 points, and a mismatch earns 0 points. Based on the scoring results of each dimension, a weighted summation method is used to calculate the comprehensive reliability score of each sensor verification data. The weight allocation is as follows: correction stability 40%, logical rationality 35%, and sensor adaptability 25%. Data weight coefficients are generated based on the comprehensive score, with the coefficient range set from 0 to 1. A score ≥9 points earns a weight coefficient of 1.0, indicating high reliability; 7 to 8.9 points earns 0.8, indicating relatively high reliability; 5 to 6.9 points earns 0.6, indicating medium reliability; and <5 points earns 0.3, indicating low reliability. This ultimately forms a quantitative evaluation result of sensor identification, data entries, reliability scores, and weight coefficients.

[0110] Step 5.3: Based on the data weighting coefficients, perform weighted fusion analysis on the verified supervision data to identify contradictions and anomalies, and generate fusion analysis results. Specifically, this includes: first, dividing the verified supervision data into multiple analysis units based on the structural component locations and monitoring parameter types. Each analysis unit corresponds to multi-sensor data of the same component location and the same type of monitoring parameter, ensuring the targeted nature of the fusion analysis; for each analysis unit, calling the data weighting coefficients generated in Step 5.2, and using a weighted fusion algorithm to integrate the verification data within the unit: using the weighting coefficients corresponding to each data item as weights, multiplying the data value by the weighting coefficients, summing the results, and then dividing by the sum of the weighting coefficients to obtain the fused comprehensive data value; during the fusion process, contradiction and anomaly analysis are performed simultaneously. Normal pattern recognition and contradiction recognition compare the fusion deviation of data from different sensors within the same analysis unit. If the deviation of a sensor's data from the fused composite value exceeds a preset threshold, such as 20%, it is determined to be a data contradiction, and the sensor data and its corresponding weight coefficient are marked. Abnormal pattern recognition analyzes the time series trend of the fused data values, sets a trend change threshold and a static threshold. If the data suddenly changes, such as a change amplitude > 30% or continuously exceeds the reasonable range allowed by the structural design, it is determined to be an abnormal pattern, and the time of occurrence, data value, and corresponding component location are recorded. After fusion and recognition are completed, a fusion analysis result is generated, which includes the composite data value of each analysis unit, details of data contradictions (sensor identification, deviation amplitude), abnormal pattern type, and abnormal location information.

[0111] Step 5.4 maps the key indicators and spatial positioning information contained in the fusion analysis results to the corresponding components and coordinate positions in the BIM digital twin model to drive the synchronous update of the BIM digital twin model's state, generating a state-updated BIM digital twin model. Specifically, this includes: first, clarifying the core attributes of the BIM digital twin model, which is a full-size digital mapping model of a large-span prestressed concrete load-bearing structure, containing geometric information, material information, design parameters, and precise spatial coordinates of the components, and supporting real-time data-driven state updates; second, clarifying its core application scenarios, covering real-time supervision and monitoring during the structural construction phase, dynamic safety status assessment during the operation phase, emergency response assistance in abnormal situations, and trend prediction for long-term operation and maintenance: during the construction phase, the model can be used to synchronously monitor data in real time, intuitively controlling construction quality and structural stress state; during the operation phase, the model can be used to continuously track changes in structural performance and identify potential safety hazards in advance; when an anomaly occurs, the model can accurately locate the abnormal part, providing spatial guidance for emergency repairs; in long-term operation and maintenance, combining historical data and model analysis can provide data support for structural maintenance and reinforcement decisions.

[0112] Subsequently, the construction and training of the BIM digital twin model were completed. The construction process closely integrated the core information of the large-span prestressed concrete load-bearing structure monitored by this invention: First, the detailed design drawings, component parameter list, and material performance data of the structure were retrieved. These data were consistent with the core parameters of the static quality certification documents of the material supplier in step 4.2, ensuring that the material properties of the model matched the actual structure. Then, the generated preset spatial reference point coordinate files were integrated, and the reference points were accurately implanted into the corresponding spatial positions of the model as the core anchor points for subsequent data mapping. Subsequently, based on the component decomposition logic of the structure, a full-size three-dimensional model was built according to the core components such as beams, slabs, columns, and supports, restoring the component connection relationship and overall spatial layout. At the same time, the installation position and monitoring range of the deployed sensors were associated with the corresponding components of the model, forming a three-dimensional association system of components, reference points, and sensors, thus completing the basic construction of the model.

[0113] The training process uses the supervision data collected in the early stages of this invention as the core training sample: multiple sets of complete data links containing original measurement data, calibration state parameters, and verification data are selected as the training set, and the training objective is to enable the model to accurately map the correspondence between monitoring data and structural state; the model parameters are optimized through iterative learning so that the model can automatically match the theoretical state of the component according to different input monitoring data (such as stress and strain), and at the same time, the structural geometric deformation law obtained in step 2 is combined, including the volume change rate and affine transformation components; the model's response logic to structural deformation is optimized to ensure that the model's sensitivity to abnormal data is consistent with the actual engineering safety requirements. After training, the model accuracy is verified through multiple sets of verification data until the requirement of accurate mapping between data and model state is met.

[0114] After model construction and training are completed, key indicators and spatial positioning information are extracted from the fusion analysis results. Key indicators include fused data such as stress, strain, displacement, and deformation rate. Spatial positioning information is linked to the coordinates of preset spatial reference points in the model, unique component numbers, and sensor installation locations. A mapping rule is established between the fusion analysis results and the BIM model. Target components in the model are matched using their unique component numbers, and monitoring points on the components are precisely located using 3D coordinates, ensuring accurate correspondence between data and model elements. The extracted key indicators are written into the attribute fields of the corresponding components and monitoring points in the BIM model, triggering a model status synchronization update process: based on the magnitude and range of the indicator values, the visualization status of the corresponding components is automatically adjusted. For example, when stress data exceeds the warning threshold, the component model is displayed in red; within the allowable range, it is displayed in green; and between the warning and allowable ranges, it is displayed in yellow. Simultaneously, real-time data values ​​are marked on the corresponding monitoring points of the sensors. During the update process, the accuracy of data mapping and the consistency of model status display are checked simultaneously. If there are mapping errors or display anomalies, the position is recalibrated or the index threshold is corrected by combining the preset spatial reference points. After correction, the update is repeated to finally generate the BIM digital twin model with updated status. This model can directly serve the aforementioned core application scenarios, clearly presenting the real-time monitoring status, abnormal parts and precise spatial location of each component, providing intuitive and accurate digital support for supervision decisions.

[0115] Step 5.5: Based on the updated BIM digital twin model and the fusion analysis results, construct and render a visual monitoring interface. This interface integrates dynamic data layers, anomaly alarm panels, and a structural safety status overview. Specifically, this includes: first, determining the core design principles of the visual monitoring interface, aiming to intuitively present the structural safety status and quickly locate abnormal areas; adopting a BIM model as the core, with multiple layers and interconnected panels; constructing the interface's base layer based on the updated BIM digital twin model, rendering the model at a 1:1 scale in the central area of ​​the interface, preserving the model's component hierarchy and spatial relationships, and supporting interactive operations such as scaling, translation, and rotation; overlaying a dynamic data layer on the base layer, overlaying key indicators from the fusion analysis results onto the corresponding monitoring points using numerical labels, color gradients, and dynamic curves; and then allowing users to click to monitor. Detailed time-series data and reliability assessment results can be viewed by clicking; an anomaly alarm panel is designed and integrated into the right side of the interface. The panel displays anomaly information categorized by alarm level (urgent, important, general). Each alarm message includes the abnormal component number, anomaly type, occurrence time, current indicator value, and warning threshold. It supports the filtering, sorting, and processing status marking of alarm information, such as unprocessed, processing, and resolved; a structural safety status overview panel is integrated at the top of the interface, which intuitively displays the safety score of the entire structure (the safety score is calculated based on the fusion analysis results), the number of alarms at each level, the proportion of normal or abnormal components, and other core overview indicators through dashboards, progress bars, etc.; after completing the construction and rendering of each module of the interface, interactive logic testing is carried out to ensure smooth layer switching, timely panel linkage, and real-time data updates, and finally, a visual monitoring interface that supports real-time monitoring and interactive queries is generated.

[0116] In a preferred embodiment of the present invention, step 6 above may include:

[0117] Step 6.1: Based on the fusion analysis results, identify the abnormal indicators exceeding the preset thresholds and their spatial distribution. Specifically, this includes: first, retrieving the preset abnormal indicator threshold system. This system is based on the design specifications for large-span prestressed concrete load-bearing structures, the allowable parameters in the static quality certification documents of material suppliers, and is formulated in conjunction with engineering supervision experience. It is subdivided according to the monitoring indicator type (stress, strain, displacement, deformation rate, etc.) and the criticality of the components (core load-bearing components, secondary connecting components, etc.). Each sub-item has two threshold levels: a warning threshold approaching the safety upper limit and a danger threshold exceeding the safety upper limit. For example, the stress warning threshold for the core load-bearing beam is set to 80% of the material's allowable stress, and the danger threshold is set to 95%. From the fusion analysis results generated in Step 5.3... In this process, the comprehensive data values, corresponding unique component numbers, and three-dimensional coordinate information of each analysis unit are extracted and compared with the threshold system one by one according to the index type and component location. If the comprehensive data value exceeds the warning threshold but does not reach the danger threshold, it is marked as a warning-level anomaly; if it exceeds the danger threshold, it is marked as a danger-level anomaly. At the same time, the specific value of the abnormal index, the magnitude of exceeding the threshold, and the number of continuous collection times are recorded. Based on the unique component number and three-dimensional coordinates, the marked abnormal indexes are mapped to the spatial layout of the structure to sort out the distribution pattern of the anomalies, distinguish between single-point scattered anomalies and multi-point concentrated anomalies, and finally form an abnormal index list and corresponding spatial distribution map containing the abnormal index type, value, magnitude of exceeding, duration, component number, three-dimensional coordinates, and anomaly level.

[0118] Step 6.2: For the identified abnormal indicators, analyze their severity and spatial clustering characteristics, and generate graded risk warning signals based on these characteristics. Specifically, this includes: first, conducting an analysis of the severity of the abnormal indicators and establishing a two-dimensional evaluation standard. Dimension 1 is the magnitude of exceeding the threshold: exceeding the warning threshold by less than 10% is considered a slight exceedance, 10% to 20% is a moderate exceedance, and more than 20% is a severe exceedance. Dimension 2 is the duration of the abnormality: 3 or fewer consecutive collection periods are considered short-term abnormalities, 4 to 10 collection periods are considered medium-term abnormalities, and more than 10 periods are considered long-term abnormalities. The severity is then comprehensively determined by combining the two dimensions. For example, severe exceedance plus long-term abnormality is considered extremely severe, moderate exceedance plus medium-term abnormality is considered relatively severe, and slight exceedance plus short-term abnormality is considered moderately severe. Next, spatial clustering characteristic analysis is performed. By statistically analyzing the number of abnormalities in the component numbers and regional coordinates of each component in the abnormal indicator list, a clustering threshold is set. The presence of 3 or more abnormal indicators within the same component or a radius of 5 meters constitutes spatial clustering. Simultaneously, the core coordinates of the clustered area, the number of components involved, and the main abnormality types are recorded. A grading system is established based on severity and spatial clustering characteristics. Generally severe with no clustering is a Level 1 warning signal, moderately severe or slight clustering is a Level 2 warning signal, severe with moderate clustering is a Level 3 warning signal, and extremely severe with heavy clustering is a Level 4 warning signal. Each level of signal corresponds to one of four visual identifiers: blue, yellow, orange, and red. The final result is a graded risk warning signal that includes the warning level, the affected area, anomaly details, and the basis for judgment.

[0119] Step 6.3: Push the graded risk warning signal to the anomaly alarm panel of the visual monitoring interface for alarm activation. Simultaneously, receive supervisory decision input for the graded risk warning signal through the visual monitoring interface. Specifically, this includes: first, pushing the graded risk warning signal to the anomaly alarm panel of the visual monitoring interface, with the pushed content simultaneously including a spatial distribution map of the anomaly indicators, the BIM model location link of the corresponding component, and a list of anomaly details; after receiving the signal, the alarm panel automatically triggers the corresponding alarm mechanism according to the warning level. Level 1 warnings only display a blue indicator and text prompt; Level 2 warnings display a yellow indicator accompanied by a slight alarm sound; Level 3 warnings display... The system displays an orange indicator and a continuous audible alert. Level four warnings display a red indicator, a continuous audible and visual alarm, and the interface automatically redirects to the BIM model view of the abnormal area. Simultaneously, the visual monitoring interface provides an interactive entry point, allowing supervisors to input supervisory decisions based on the graded risk warning signals. Decision input options include viewing detailed analysis reports, issuing on-site verification instructions, initiating emergency response procedures, and requesting data review. After selecting the corresponding option, supervisors can add supplementary text descriptions, such as verification focus and handling requirements. The interface automatically records the decision input time, operator, and specific content, forming a decision input log that is associated with and stored in relation to the corresponding warning signal.

[0120] Step 6.4: Combining the updated BIM digital twin model, conduct an impact simulation analysis on the supervisor's decision inputs to generate structured supervisor instructions. Specifically, this includes: first, calling the updated BIM digital twin model to convert the supervisor's input decision schemes, such as on-site verification, suspension of construction, component reinforcement, and load adjustment, into simulation parameters recognizable by the model; second, based on the current structural state data, the updated BIM digital twin model deduces the structural stress distribution, deformation trend, and safety factor changes after the decision schemes are implemented. For example, simulating whether abnormal stress indices can return to a reasonable range after reinforcement of components in a certain concentrated area, and whether new stress concentrations will occur in surrounding components. After the simulation is completed, an impact analysis report is generated, which clarifies the feasibility, expected effects, and potential risks of the solution. If the solution has risks, such as the reinforcement measures causing other components to be overloaded, the feedback is given to the supervisor to re-enter the decision. If the solution is feasible, a structured supervision instruction is constructed based on the decision and the impact analysis results. The structured instruction is written according to the preset format specifications, including the instruction category, the target of execution, the construction site team, the equipment number, the core requirements, the execution time limit, the technical standards, the acceptance indicators, and the safety precautions. For example, the component reinforcement instruction must clearly specify the component number to be reinforced, the standard of the reinforcement material used, the reinforcement thickness requirement, and the stress acceptance threshold after reinforcement, to ensure that the instruction is clear and implementable.

[0121] Step 6.5 involves sending structured supervision instructions to the construction site terminal and simultaneously tracking the execution status of these instructions and related supervision data feedback to achieve closed-loop supervision of the entire construction process. Specifically, this includes: first, establishing a secure communication link between the cloud-based intelligent supervision hub and the construction site terminal to encrypt and send the structured supervision instructions; simultaneously performing identity authentication and data integrity verification during the sending process to ensure the instructions are not tampered with or leaked; after receiving the instructions, the construction site terminal automatically displays an alarm and the full text of the instructions; the terminal operator must confirm receipt and provide feedback on the receipt status, simultaneously recording the instruction sending time, receiving time, and receiving personnel information; subsequently, the instruction execution status tracking process is initiated, with the construction site terminal uploading execution progress every 30 minutes according to a preset cycle, such as reinforcement materials... The progress information, such as material delivery completion, reinforcement construction in progress, and acceptance completion, can be viewed in real time by the supervisor through a visual monitoring interface. Simultaneously, sensor monitoring data from the corresponding areas of the construction site is retrieved to track changes in the structural state during the execution of instructions and verify the execution effect. After the instructions are executed, acceptance data, such as stress monitoring values ​​after reinforcement and construction video data, are uploaded. The results of the BIM model simulation are combined with the acceptance data for effect evaluation. If the acceptance data meets the requirements, the instruction execution is deemed qualified, completing closed-loop supervision. If it does not meet the requirements, a rectification prompt is pushed to the terminal, requiring re-execution and re-acceptance. The entire process of instruction issuance, execution, and acceptance is recorded and integrated with corresponding supervision data, early warning signals, and decision logs for archiving, forming a complete closed-loop construction supervision file.

[0122] An engineering supervision and monitoring system includes:

[0123] A data acquisition module is deployed to collect multi-dimensional and heterogeneous supervision data through various sensors deployed at the construction site. These sensors are deployed at preset spatial reference points on the large-span prestressed concrete load-bearing structure. The preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key structural sections and the center point of the prestressed anchorage end.

[0124] A calibration module is constructed to virtually build a spatial reference control volume covering a local or overall structure in the edge computing node based on preset spatial reference points. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on multi-dimensional heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume in a continuous time series.

[0125] The data verification module is used to perform geometric consistency correction on the raw measurement data of relevant sensors collected in advance based on the volume change rate, and simultaneously generate calibrated supervision data and sensor calibration status parameters; cross-validate the calibrated supervision data to form verified supervision data; upload the verified supervision data to the cloud-based intelligent supervision center, and encrypt and store it through an anti-tampering evidence storage engine;

[0126] The fusion analysis module is used to perform fusion analysis on multi-source supervision data in the cloud-based intelligent supervision center by combining sensor calibration status parameters, and generate fusion analysis results; based on the fusion analysis results, a visual supervision interface that is linked with the BIM digital twin model is constructed.

[0127] The early warning and decision-making module is used to make risk warnings and supervision decisions based on the results of the integrated analysis and with the assistance of a visual monitoring interface, so as to achieve closed-loop supervision of the entire construction process.

[0128] The monitoring system according to embodiments of the present invention can correspond to the execution of the methods described in the embodiments of the present invention, and the above and other operations and / or functions of each module of the monitoring system are respectively for implementing Figure 1 The corresponding process of the method in the illustrated embodiment will not be described in detail here for the sake of brevity.

[0129] This application also provides a computing device. This computing device can utilize a server.

[0130] like Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0131] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0132] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0133] Communication interface 703 is used for external communication. Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD). Executable code is stored in memory 704, and processor 702 executes this executable code to perform the aforementioned engineering supervision and monitoring method.

[0134] Specifically, in implementing the engineering supervision and monitoring system described in the above embodiments, and where each module or unit of the engineering supervision and monitoring system described in the above embodiments is implemented by software, the software or program code required to execute the functions of each module / unit in the engineering supervision and monitoring system described in the above embodiments can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704 to execute the aforementioned engineering supervision and monitoring method.

[0135] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium capable of being stored by a computing device, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned engineering supervision and monitoring method.

[0136] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0137] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0138] When the computer program product is executed by a computer, the computer executes any of the aforementioned engineering supervision and management methods. The computer program product can be a software installation package; when any of the aforementioned engineering supervision and management methods needs to be used, the computer program product can be downloaded and executed on the computer.

[0139] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for engineering supervision and monitoring, characterized in that, The method includes: Multi-dimensional and heterogeneous supervision data are collected by various sensors deployed at the construction site; among them, various sensors are deployed at the preset spatial reference points of the large-span prestressed concrete load-bearing structure; the preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key section of the structure and the center point of the prestressed anchorage end. Based on preset spatial reference points, a spatial reference control volume covering the local or overall structure is virtually constructed in the edge computing node. The spatial reference control volume is formed by connecting several preset spatial reference points as vertices to form an irregular polyhedron. The volume change rate of the spatial reference control volume under continuous time series is calculated by performing real-time dynamic calibration on multi-dimensional heterogeneous supervision data. Based on the volume change rate, the original measurement data of the relevant sensors collected in advance are corrected for geometric consistency, and the calibrated supervision data and sensor calibration status parameters are generated simultaneously. The calibrated supervision data is cross-validated to form validated supervision data; the validated supervision data is uploaded to the cloud-based intelligent supervision center and encrypted and stored through an anti-tampering evidence storage engine. In the cloud-based intelligent supervision center, the verified supervision data is fused and analyzed by combining sensor calibration status parameters to generate fusion analysis results; based on the fusion analysis results, a visual supervision interface that is linked with the BIM digital twin model is constructed. Based on the results of the integrated analysis, risk warnings and supervisory decisions are made with the assistance of a visual regulatory interface.

2. The engineering supervision and management method according to claim 1, characterized in that, Collect diverse and heterogeneous supervision data through multiple sensors deployed at the construction site; Multiple sensors are deployed at preset spatial reference points on the prestressed concrete box girder; The pre-set spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key structural sections, and the center point of the prestressed anchorage end, including: Identify and determine several key monitoring sections of large-span prestressed concrete load-bearing structures; On the web centerline of each key monitoring section, locate the one-quarter, one-half, and three-quarter positions in the span direction as spatial reference points for deformation monitoring. In the central region of the prestressed anchoring system at both ends of a large-span prestressed concrete load-bearing structure, the geometric center point of the anchor plate is positioned as the spatial reference point for load transfer. The deformation monitoring spatial reference point and the load transfer spatial reference point are jointly defined as the preset spatial reference point; At each of the preset spatial reference points, a corresponding sensor cluster is deployed to synchronously collect multi-dimensional heterogeneous supervision data.

3. The engineering supervision and management method according to claim 2, characterized in that, Based on preset spatial reference points, a spatial reference control volume covering a local or overall structure is virtually constructed in the edge computing nodes. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on multi-dimensional heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume over a continuous time series, including: Select at least four non-coplanar reference points from the preset spatial reference points as target vertices; In the edge computing node, all target vertices are connected according to a preset topology connection order to generate an irregular polyhedron composed of target vertices, and the irregular polyhedron is defined as a spatial reference control volume. Real-time spatial coordinate monitoring data of each target vertex is obtained through edge computing nodes; Based on the spatial coordinate monitoring data, the instantaneous volume of the spatial reference control body at each acquisition moment is calculated; The rate of change of volume of the space reference control volume is calculated based on a series of instantaneous volumes over a continuous time series.

4. The engineering supervision and monitoring method according to claim 3, characterized in that, Based on the volume change rate, the raw measurement data of the relevant sensors collected in advance are geometrically corrected for consistency, and the calibrated supervision data and sensor calibration status parameters are generated simultaneously, including: Based on the volume change rate of the spatial reference control volume, the structural geometric deformation correction coefficient under the corresponding time series is calculated. Based on the structural geometric deformation correction coefficient and the pre-established mapping relationship between sensor measurements and spatial deformation, the required geometric consistency correction amount for the original measurement data of each relevant sensor is determined. The geometric consistency correction amount is used to correct the original measurement data of the relevant sensors and generate calibrated supervision data. During the data correction process, the magnitude and application time of the geometric consistency correction are recorded simultaneously. The magnitude and application time of the geometric consistency correction are then encapsulated together with the identification information of the corresponding sensor to generate sensor calibration status parameters.

5. The engineering supervision and management method according to claim 4, characterized in that, The calibrated supervision data is cross-validated to form validated supervision data; the validated supervision data is then uploaded to the cloud-based intelligent supervision center and encrypted and stored using an anti-tampering evidence storage engine, including: The calibrated supervision data is subjected to protocol compliance verification to verify whether the data collection and transmission process follows the preset secure communication protocol, and the first verification result is generated. The calibrated supervision data is cross-referenced with the static quality certification documents from the material supplier to verify the logical consistency of the data and generate a second verification result. Based on the results of the first and second verifications, a data verification conclusion is generated, and the verified supervision data is formed according to the data verification conclusion. The verified supervision data will be uploaded to the cloud-based intelligent supervision center in the predetermined format. The anti-tampering evidence storage engine within the cloud-based intelligent supervision center is invoked to perform encryption operations and timestamp solidification on the verified supervision data, generating legally valid on-chain supervision evidence.

6. The engineering supervision and monitoring method according to claim 5, characterized in that, In the cloud-based intelligent supervision center, the verified supervision data is fused and analyzed by combining the sensor calibration status parameters to generate fusion analysis results; A visualized monitoring interface, linked to the BIM digital twin model, is constructed based on the results of the fusion analysis, including: After the on-chain supervision evidence is generated, the verified supervision data and corresponding sensor calibration status parameters are imported into the cloud-based intelligent supervision center. Based on the sensor calibration status parameters, the reliability of the verified supervision data is quantitatively evaluated to form data weighting coefficients; Based on the data weighting coefficients, the verified supervision data is subjected to weighted fusion analysis to identify contradictions and abnormal patterns and generate fusion analysis results. The key indicators and spatial positioning information contained in the fusion analysis results are mapped to the corresponding components and coordinate positions of the BIM digital twin model to drive the synchronous update of the BIM digital twin model status and generate the updated BIM digital twin model. Based on the updated BIM digital twin model and the fusion analysis results, a visual monitoring interface is constructed and rendered. The visual monitoring interface integrates dynamic data layers, anomaly alarm panels, and an overview of structural safety status.

7. The engineering supervision and management method according to claim 6, characterized in that, Based on the fusion analysis results, and with the assistance of a visual monitoring interface, risk warnings and supervisory decisions are made to achieve closed-loop monitoring of the entire construction process, including: Based on the fusion analysis results, identify the abnormal indicators that exceed the preset threshold and their spatial distribution; For the identified abnormal indicators, the severity and spatial clustering characteristics of the abnormal indicators are analyzed, and a graded risk warning signal is generated based on the severity and spatial clustering characteristics of the abnormal indicators. The tiered risk warning signals are pushed to the abnormal alarm panel of the visual monitoring interface for alarm purposes. At the same time, the monitoring interface receives supervisory decision inputs for the tiered risk warning signals. By combining the updated BIM digital twin model, an impact simulation analysis is performed on the input of supervision decisions to generate structured supervision instructions; Structured supervision instructions are issued to the construction site terminals, and the execution status of the structured supervision instructions and the related supervision data feedback are tracked simultaneously to complete the closed-loop supervision of the entire construction process.

8. An engineering supervision and monitoring system, characterized in that, The system performs the method as described in any one of claims 1 to 7, comprising: A data acquisition module is deployed to collect multi-dimensional and heterogeneous supervision data through various sensors deployed at the construction site. Among them, various sensors are deployed at preset spatial reference points on the large-span prestressed concrete load-bearing structure. The preset spatial reference points include at least one-quarter, one-half, and three-quarters of the web centerline of the key section of the structure and the center point of the prestressed anchorage end. A calibration module is constructed to virtually build a spatial reference control volume covering a local or overall structure in the edge computing node based on preset spatial reference points. The spatial reference control volume is formed by connecting several selected preset spatial reference points as vertices to form an irregular polyhedron. Real-time dynamic calibration is performed on multi-dimensional heterogeneous supervision data to calculate the volume change rate of the spatial reference control volume in a continuous time series. The data verification module is used to perform geometric consistency correction on the raw measurement data of relevant sensors collected in advance based on the volume change rate, and simultaneously generate calibrated supervision data and sensor calibration status parameters; cross-validate the calibrated supervision data to form verified supervision data; upload the verified supervision data to the cloud-based intelligent supervision center, and encrypt and store it through an anti-tampering evidence storage engine; The fusion analysis module is used to perform fusion analysis on multi-source supervision data in the cloud-based intelligent supervision center by combining sensor calibration status parameters, and generate fusion analysis results; based on the fusion analysis results, a visual supervision interface that is linked with the BIM digital twin model is constructed. The early warning and decision-making module is used to make risk warnings and supervision decisions based on the results of the integrated analysis and with the assistance of a visual monitoring interface, so as to achieve closed-loop supervision of the entire construction process.

9. A computing device, characterized in that, Including memory and processor; The memory stores one or more computer programs, the one or more computer programs including instructions; when the instructions are executed by the processor, the computing device performs the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program for performing the method as described in any one of claims 1 to 7.

Images