Deep foundation pit supporting structure deformation prediction method and system based on digital twinning

Abstract

The application provides a deep foundation pit supporting structure deformation prediction method and system based on digital twinning, and relates to the technical field of deep foundation pit engineering.The method comprises the following steps: step 1, displacement sensors, stress sensors and pore water pressure sensors are arranged on the top of the row pile, the midpoint of the internal support and the corner of the surrounding building foundation in the physical entity corresponding to the three-dimensional digital twinning, real-time collection of the horizontal displacement, vertical settlement, stress change and pore water pressure data of the supporting structure is performed, and the collected real-time data is synchronized to the three-dimensional digital twinning, real-time mapping of the physical entity and the digital twinning is realized, and the digital twinning after real-time mapping is obtained.The application realizes accurate prediction and intelligent early warning of the deformation of the deep foundation pit supporting structure.

Description

Technical Field

[0001] This invention relates to the field of deep foundation pit engineering technology, and in particular to a method and system for predicting the deformation of deep foundation pit support structures based on digital twins. Background Technology

[0002] In the construction of urban high-rise buildings, underground utility tunnels and other projects, the deformation stability of deep foundation pit support structures is directly related to construction safety and the normal use of surrounding buildings and underground pipelines. At present, the deformation prediction of deep foundation pit support structures mostly relies on traditional sensor monitoring and empirical formula calculation, which has problems such as poor real-time performance and insufficient prediction accuracy.

[0003] For example, in the construction of a deep foundation pit project in the core area of ​​a city, conventional support methods were adopted. During the construction process, monitoring equipment was only installed on a portion of the support structure. Deformation prediction was made by manually collecting data periodically and combining it with empirical formulas. However, due to the difficulty in achieving real-time synchronization of the physical state, the deformation coordination relationship between the support structure and the surrounding soil was not fully considered, and the stress-related parameters of the support structure were not dynamically optimized and adjusted. This resulted in a deviation between the deformation prediction results and the actual situation, and the abnormal settlement of the support structure was not detected in time, leading to safety hazards such as settlement of the foundations of surrounding buildings. This case exposed the technical defects of traditional deformation prediction methods. They are difficult to synchronize the physical state in real time, lack a comprehensive characterization of deformation characteristics and dynamic correction of stress parameters, and the prediction accuracy is insufficient to meet the high precision and real-time requirements of safety management in deep foundation pit projects. They are also unable to provide accurate and timely guidance for construction adjustments. Summary of the Invention

[0004] This invention provides a method and system for predicting the deformation of deep foundation pit support structures based on digital twins, enabling accurate prediction and intelligent early warning of the deformation of deep foundation pit support structures.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: Firstly, a method for predicting the deformation of deep foundation pit support structures based on digital twins, the method comprising: Step 1: Displacement sensors, stress sensors, and pore water pressure sensors are installed on the top of the piles, the midpoint of the inner support, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. The horizontal displacement, vertical settlement, stress change, and pore water pressure data of the support structure and the surrounding soil are collected in real time. The collected real-time data is synchronized to the 3D digital twin to realize the real-time mapping between the physical entity and the digital twin, and the real-time mapped digital twin is obtained. Step 2: Based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the foundations of the surrounding buildings contained in the real-time mapped digital twin, a spatial deformation feature envelope surface defined by the three deformation data points is constructed in the three-dimensional digital twin; the feature envelope surface is meshed by finite element method to obtain the deformation interpolation of each mesh node, the curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained according to the deviation between the curvature distribution and the preset reference curvature. Step 3: Based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficient, iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin to obtain the corrected mechanical parameters. Step 4: Perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation value of the support structure in the future period; when the predicted deformation value reaches the preset warning threshold, issue a warning signal to guide construction adjustments.

[0006] Furthermore, before step 1, the process also includes: constructing a three-dimensional digital twin containing the deep foundation pit support structure, surrounding buildings, underground pipelines, and site geological conditions using BIM technology, and inputting the support structure parameters, surrounding building parameters, underground pipeline parameters, geological parameters, and groundwater depth information into the three-dimensional digital twin.

[0007] Furthermore, displacement sensors, stress sensors, and pore water pressure sensors are deployed at the top of the piles, the midpoint of the internal supports, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. These sensors collect real-time data on the horizontal displacement, vertical settlement, stress changes of the support structure, and pore water pressure of the surrounding soil. The collected real-time data is then synchronized to the 3D digital twin, achieving real-time mapping between the physical entity and the digital twin. This results in a real-time mapped digital twin, including: Based on the pile position parameters, internal support distribution parameters, and corner coordinate parameters of the surrounding building foundations entered in the 3D digital twin, the layout positions of the top measuring point of the pile, the midpoint measuring point of the internal support, and the corner measuring points of the surrounding building foundations are determined in the corresponding physical entity. Displacement sensors, stress sensors, and pore water pressure sensors are installed at each of the determined measuring points. The displacement sensors collect horizontal displacement data at the top of the pile, vertical settlement data at the midpoint of the internal support, and vertical settlement data at the corners of the foundations of surrounding buildings. The stress sensors collect stress change data of the pile and internal support, and the pore water pressure sensors collect pore water pressure data of the surrounding soil. The collected data on horizontal displacement at the top of the pile, vertical settlement at the midpoint of the internal support, vertical settlement at the corner of the foundation of the surrounding buildings, stress change data, and pore water pressure data are synchronized in real time to the constructed three-dimensional digital twin via a 5G communication network. The real-time data synchronized to the 3D digital twin is fused with the pre-entered support structure parameters, surrounding building parameters, underground pipeline parameters, geological parameters, and groundwater depth information in the 3D digital twin, so that the 3D digital twin reflects the current state of the physical entity in real time, resulting in a real-time mapped digital twin.

[0008] Furthermore, based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the surrounding building foundations contained in the real-time mapped digital twin, a spatial deformation feature envelope defined by three deformation data points is constructed in the three-dimensional digital twin, including: Extract the current horizontal displacement value of the top of the pile, the current vertical settlement value of the midpoint of the inner support, and the current vertical settlement value of the corner of the foundation of the surrounding building from the real-time mapped digital twin, and use the current horizontal displacement value and the current vertical settlement value as the deformation feature parameters of the three feature points; Based on the pre-recorded spatial coordinates of the top of the pile, the midpoint of the inner support, and the corner of the surrounding building foundation in the three-dimensional digital twin, and combined with the deformation characteristic parameters of the three feature points, the offset coordinates of the three feature points in three-dimensional space after deformation are calculated respectively. Using the offset coordinates of three feature points as shape value points, a spatial deformation feature curve passing through the three shape value points is fitted in the three-dimensional digital twin. The spatial deformation feature curve reflects the deformation distribution law along the direction from the top of the pile to the midpoint of the inner support and then to the corner of the surrounding building foundation. Using the spatial deformation characteristic curve as the generatrix, and taking the direction perpendicular to the plane where the piles and internal supports are located as the sweeping direction, a continuous spatial deformation characteristic envelope surface is generated by sweeping in the three-dimensional digital twin.

[0009] Furthermore, the feature envelope surface is meshed using finite element methods to obtain the deformation interpolation of each mesh node. The curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained according to the deviation between the curvature distribution and the preset reference curvature. This includes: Based on the spatial deformation feature envelope, the spatial deformation feature envelope is meshed using finite element methods in a three-dimensional digital twin to generate a discrete mesh covering the spatial deformation feature envelope, and the spatial coordinates of each mesh node in the discrete mesh are obtained. Based on the spatial coordinates of each grid node in the discrete grid, and combined with the deformation distribution law represented by the spatial deformation feature envelope, the deformation interpolation value at each grid node is obtained by interpolation calculation. The deformation interpolation value reflects the amount of deformation of each grid node relative to the initial state. Based on the spatial coordinates of each grid node in the discrete grid and the corresponding deformation interpolation, the local curvature at each grid node is calculated using the discrete curvature estimation method, and the overall curvature distribution of the spatial deformation feature envelope is generated by fitting the local curvature at each grid node. The overall curvature distribution is compared point by point with the preset reference curvature in the three-dimensional digital twin. The curvature deviation at each grid node is calculated, and a weighted integral is performed based on the curvature deviation at all grid nodes to obtain a spatial stiffness correction coefficient that reflects the overall deformation coordination of the current support structure.

[0010] Furthermore, based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficients, the mechanical parameters of the support structure in the three-dimensional digital twin are iteratively corrected to obtain the corrected mechanical parameters, including: Read the initial values ​​of the mechanical parameters of the support structure set at the current time from the three-dimensional digital twin, and extract the real-time monitoring data at the current moment from the real-time mapped digital twin; By comparing real-time monitoring data with historical monitoring data in a three-dimensional digital twin, the deformation rate and stress change rate at the current moment are calculated. Combined with the spatial stiffness correction coefficient, a mechanical parameter correction function reflecting the current stress characteristics of the support structure is constructed. Substitute the initial values ​​of the mechanical parameters of the support structure into the mechanical parameter correction function, and obtain a set of preliminary corrected mechanical parameters through iterative calculation. The iterative calculation aims to reduce the difference between real-time monitoring data and three-dimensional digital twin simulation output data. The initially corrected mechanical parameters are substituted into the three-dimensional digital twin for forward simulation calculation to obtain simulated deformation data. The simulated deformation data is then compared with the real-time monitoring data. If the error exceeds the preset convergence threshold, the initially corrected mechanical parameters are readjusted according to the error until the error between the simulated deformation data and the real-time monitoring data converges to within the convergence threshold, thus obtaining the corrected mechanical parameters.

[0011] Furthermore, deformation prediction calculations are performed based on the corrected mechanical parameters to obtain the predicted deformation values ​​of the support structure in the future period; when the predicted deformation value reaches a preset warning threshold, a warning signal is issued to guide construction adjustments, including: The corrected mechanical parameters are updated in the three-dimensional digital twin, and the real-time monitoring data at the current moment is extracted from the real-time mapped digital twin as the predicted initial state. Based on the construction plan parameters pre-entered in the 3D digital twin, set the load condition sequence for future time periods; The corrected mechanical parameters, predicted initial state, and load condition sequence are input into the three-dimensional digital twin. Time history analysis is performed using the numerical solver built into the three-dimensional digital twin to calculate the predicted deformation value sequence of the support structure at each time node in the future period. Each predicted value in the deformation prediction value sequence is compared with the preset warning threshold in the three-dimensional digital twin. The warning threshold includes the horizontal displacement limit at the top of the pile, the vertical settlement limit at the midpoint of the internal support, and the differential settlement limit at the corner of the foundation of the surrounding buildings. When any predicted value reaches the corresponding warning threshold, the 3D digital twin automatically generates a warning command and sends the warning command and the corresponding deformation prediction value to the construction monitoring terminal to guide the construction unit to adjust the excavation speed, add temporary supports and take dewatering measures.

[0012] Secondly, a deep foundation pit support structure deformation prediction system based on digital twins includes: The mapping module is used to deploy displacement sensors, stress sensors, and pore water pressure sensors on the top of the piles, the midpoint of the internal support, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. It collects data on the horizontal displacement, vertical settlement, stress changes of the support structure, and pore water pressure of the surrounding soil in real time, and synchronizes the collected real-time data to the 3D digital twin to realize the real-time mapping between the physical entity and the digital twin, and obtain the real-time mapped digital twin. The calculation module is used to construct a spatial deformation feature envelope surface defined by three deformation data points in the three-dimensional digital twin based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the foundations of the surrounding buildings contained in the real-time mapped digital twin. The feature envelope surface is meshed by finite element method to obtain the deformation interpolation of each mesh node. The curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained according to the deviation between the curvature distribution and the preset reference curvature. The correction module is used to iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficient, so as to obtain the corrected mechanical parameters. The processing module is used to perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation value of the support structure in the future period; when the predicted deformation value reaches the preset warning threshold, a warning signal is issued to guide construction adjustments.

[0013] Thirdly, a computing device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0014] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0015] The above-described solution of the present invention has at least the following beneficial effects: BIM technology constructs a 3D digital twin encompassing the deep foundation pit support structure, surrounding buildings, underground pipelines, and site geological conditions. Multiple sensors are deployed at key points of the physical entity to achieve real-time data acquisition and state synchronization mapping. A spatial deformation feature envelope is constructed, and curvature analysis is performed to obtain spatial stiffness correction coefficients. Historical and real-time data are combined to iteratively correct the mechanical parameters of the support structure. Simultaneously, deformation prediction and intelligent early warning are achieved based on the corrected parameters. Therefore, this technology overcomes the technical problems of traditional deep foundation pit support structure deformation prediction methods, such as poor real-time performance, difficulty in synchronizing physical entity states, incomplete deformation feature representation, inability to dynamically correct stress parameters, low prediction accuracy, and difficulty in timely guiding construction adjustments. Thus, it achieves accurate prediction and intelligent early warning of deep foundation pit support structure deformation, realistically and dynamically reflecting the state of the support structure and surrounding environment, improving prediction reliability, proactively avoiding construction safety risks, providing accurate and timely guidance for construction adjustments, and significantly improving the technical effect of enhancing the safety and intelligent management level of deep foundation pit engineering construction. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the method for predicting the deformation of deep foundation pit support structures based on digital twins, provided in an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of a deep foundation pit support structure deformation prediction system based on digital twins provided in an embodiment of the present invention. Detailed Implementation

[0018] 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 to provide a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art.

[0019] like Figure 1 As shown, embodiments of the present invention propose a method for predicting the deformation of deep foundation pit support structures based on digital twins. The method includes the following steps: Step 1: Displacement sensors, stress sensors, and pore water pressure sensors are installed on the top of the piles, the midpoint of the inner support, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. The horizontal displacement, vertical settlement, stress change, and pore water pressure data of the support structure and the surrounding soil are collected in real time. The collected real-time data is synchronized to the 3D digital twin to realize the real-time mapping between the physical entity and the digital twin, and the real-time mapped digital twin is obtained. Step 2: Based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the foundations of the surrounding buildings contained in the real-time mapped digital twin, a spatial deformation feature envelope surface defined by the three deformation data points is constructed in the three-dimensional digital twin; the feature envelope surface is meshed by finite element method to obtain the deformation interpolation of each mesh node, the curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained according to the deviation between the curvature distribution and the preset reference curvature. Step 3: Based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficient, iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin to obtain the corrected mechanical parameters. Step 4: Perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation value of the support structure in the future period; when the predicted deformation value reaches the preset warning threshold, issue a warning signal to guide construction adjustments.

[0020] In this embodiment of the invention, displacement sensors, stress sensors, and pore water pressure sensors are deployed at key locations of the physical entity to collect multiple types of data in real time and synchronize them to a three-dimensional digital twin for real-time mapping. Based on the real-time mapping data, a spatial deformation feature envelope is constructed, and a spatial stiffness correction coefficient is obtained through finite element mesh generation, interpolation calculation, and curvature analysis. The mechanical parameters of the support structure are iteratively corrected by combining historical data, real-time data, and the spatial stiffness correction coefficient. Then, deformation prediction is performed based on the corrected mechanical parameters, and an early warning signal is issued when the predicted value reaches the early warning threshold. Therefore, this technique overcomes the limitations of traditional deep foundation pit support structures. The previous method addressed technical issues such as poor real-time performance and insufficient prediction accuracy in structural deformation prediction, difficulty in achieving real-time synchronization between physical entities and digital twins, lack of comprehensive characterization of deformation characteristics, difficulty in dynamically optimizing and adjusting the mechanical parameters of the support structure, and lack of effective early warning mechanisms to guide timely construction adjustments. This method achieves real-time linkage between physical entities and three-dimensional digital twins, accurately characterizes the overall deformation characteristics of the support structure, dynamically optimizes mechanical parameters to improve deformation prediction accuracy, promptly issues early warning signals to guide construction adjustments, effectively avoids construction safety hazards, and ensures the safety of deep foundation pit construction and the normal use of surrounding buildings and underground pipelines.

[0021] In a preferred embodiment of the present invention, the method further includes the following step before step 1: A 3D digital twin containing the deep foundation pit support structure, surrounding buildings, underground pipelines, and site geological conditions is constructed using BIM technology. The parameters of the support structure, surrounding buildings, underground pipelines, geological parameters, and groundwater depth are then entered into the 3D digital twin. Specifically, this involves: comprehensively collecting and organizing basic engineering information. For the target deep foundation pit project, the following parameters are collected sequentially: design parameters of the deep foundation pit support structure, structural and foundation parameters of surrounding buildings, layout parameters of underground pipelines, soil layer parameters of the site, and groundwater depth parameters. Support structure parameters include pile diameter, pile length, spacing, and pile top elevation of the piles; cross-sectional dimensions, location, and span of the internal supports; surrounding building parameters include the building's plan outline dimensions, foundation type, foundation corner coordinates, and foundation depth; underground pipeline parameters include pipeline type, direction, depth, and diameter; and geological parameters include the soil layer name, thickness, and density of each soil layer on the site. The parameters include internal friction angle and cohesion; groundwater depth parameters include groundwater level elevation and water head height; three-dimensional geometric models of deep foundation pit support structure, surrounding buildings, underground pipelines, and site geological bodies are constructed sequentially. During the construction process, the unified three-dimensional spatial coordinate system of the engineering site is used as the reference to locate and calculate the spatial position of each model. The design coordinates of piles, internal supports, foundations of surrounding buildings, and underground pipelines are compared with the measured coordinates on site. The absolute value of the difference is controlled within the preset engineering positioning error limit to ensure that the spatial position of each model component is completely consistent with the physical entity; for the site geological body, three-dimensional models of each soil layer are constructed sequentially according to the actual thickness and distribution range of each soil layer. Then, the soil layer models are vertically superimposed and combined to form a complete three-dimensional geological model of the site. During the superposition process, the difference between the bottom elevation and the top elevation of the soil layer is calculated to determine the actual thickness of each soil layer, ensuring that the geological model is completely consistent with the geological conditions on site.

[0022] After completing the 3D geometric modeling, the various basic parameters collected and organized in the early stage are entered into the corresponding 3D model components one by one. For the support structure components, mechanical parameters such as material strength and elastic modulus of the piles and supports are entered; for the surrounding buildings, parameters such as foundation bearing capacity and allowable settlement value are entered; for underground pipelines, parameters such as pipeline material and allowable deformation value are entered; for the geological body, physical and mechanical parameters such as density, internal friction angle, and cohesion of each layer of soil are entered; for the groundwater part, hydrological parameters such as groundwater level depth and water head height are entered. After the parameters are entered, all parameters are checked for consistency. The difference between the entered parameter values ​​and the original values ​​in the design documents and survey reports is calculated and then divided by the original values ​​to obtain the relative error of parameter entry. The relative error is controlled within the allowable range of the project to ensure the accuracy and completeness of parameter entry. The support structure model, surrounding building model, underground pipeline model, and site geological model with completed parameter entry are integrated to form a 3D digital twin containing the deep foundation pit support structure, surrounding buildings, underground pipelines, and site geological conditions.

[0023] In this embodiment of the invention, a three-dimensional digital twin containing the deep foundation pit support structure, surrounding buildings, underground pipelines, and site geological conditions is constructed using BIM technology. The technical means of completely recording various relevant parameters such as the support structure, surrounding buildings, underground pipelines, geology, and groundwater depth into the three-dimensional digital twin overcomes the technical problems in traditional deep foundation pit deformation prediction, which lack a comprehensive and systematic digital carrier, makes it difficult to integrate various basic parameters of the support structure and the surrounding environment, resulting in a lack of corresponding references for monitoring data and a lack of comprehensive foundation for deformation analysis.

[0024] In a preferred embodiment of the present invention, step 1 above may include: Step 1.1: Based on the pile position parameters, internal support distribution parameters, and surrounding building foundation corner coordinate parameters entered in the 3D digital twin, determine the layout positions of the top measuring points of the piles, the midpoint measuring points of the internal supports, and the corner measuring points of the surrounding building foundations in the corresponding physical entities. Specifically, this includes: extracting the pre-entered pile position parameters, internal support distribution parameters, and surrounding building foundation corner coordinate parameters from the completed 3D digital twin. The pile position parameters include the top spatial coordinates and layout axis position of the piles; the internal support distribution parameters include the coordinates of the two end supports of the internal support and the total support length; and the surrounding building foundation corner coordinate parameters include the 3D spatial coordinates of each corner point of the foundation. Calculate the midpoint coordinates of the internal supports: the X-coordinate of the midpoint is equal to the sum of the X-coordinates of the first and second end supports of the internal support, then divide the result by two; the Y-coordinate of the midpoint is equal to the sum of the Y-coordinates of the first and second end supports of the internal support, then divide the result by two; the Z-coordinate of the midpoint is equal to the sum of the X-coordinates of the first and second end supports of the internal support, then divide the result by two; the Z-coordinate of the midpoint is equal to the sum of the X-coordinates of the first and second end supports of the internal support, then divide the result by two. The theoretical spatial coordinates of the inner support midpoint are obtained by adding the Z-coordinates of the first and second end supports and dividing the result by two. The coordinates of the top of the pile, the midpoint of the inner support, and the corner coordinates of the surrounding building foundations are then matched and converted with the unified construction coordinate system of the project site. The spatial positioning deviation in the X direction is obtained by subtracting the X-coordinate of the corresponding point in the 3D digital twin from the X-coordinate of the origin of the on-site construction coordinate system; the spatial positioning deviation in the Y direction is obtained by subtracting the Y-coordinate of the corresponding point in the 3D digital twin from the Y-coordinate of the origin of the on-site construction coordinate system; and the spatial positioning deviation in the Z direction is obtained by subtracting the Z-coordinate of the corresponding point in the 3D digital twin from the Z-coordinate of the origin of the on-site construction coordinate system. This eliminates the spatial positioning deviation between the physical entity and the 3D digital twin. The converted coordinate values ​​are then transformed into layout and positioning data for the physical entity on-site. Based on this layout and positioning data, precise layout is performed on the physical entity of the deep foundation pit project, sequentially determining the actual placement positions of the measuring points at the top of the pile, the midpoint of the inner support, and the corners of the surrounding building foundations.

[0025] Step 1.2: Deploy displacement sensors, stress sensors, and pore water pressure sensors at the determined measuring points. The displacement sensors collect horizontal displacement data at the top of the pile, vertical settlement data at the midpoint of the internal support, and vertical settlement data at the corners of the surrounding building foundations. The stress sensors collect stress change data of the pile and internal support. The pore water pressure sensors collect pore water pressure data of the surrounding soil. Specifically, this includes: fixing the displacement sensors, stress sensors, and pore water pressure sensors at the corresponding positions of the determined measuring points at the top of the pile, the midpoint of the internal support, and the corners of the surrounding building foundations, ensuring that the sensors are in close contact with the measured structure and soil without loosening or displacement; collecting data using the deployed displacement sensors, collecting horizontal displacement data at the top of the pile, which is obtained from the current horizontal position value of the top of the pile. The following data is calculated by subtracting the initial design horizontal position value: Vertical settlement data of the inner support midpoint is collected at the measuring point at the inner support midpoint, which is calculated by subtracting the initial installation vertical position value from the current vertical position value of the inner support midpoint; Vertical settlement data of the foundation corners is collected at the measuring points at the corners of the surrounding building foundations, which is calculated by subtracting the initial state vertical position value from the current vertical position value of the building foundation corners; Stress change data of the piles and inner supports are collected by deploying stress sensors, with pile stress change data calculated by subtracting the initial stress value from the current stress value of the pile body, and inner support stress change data calculated by subtracting the initial stress value from the current stress value of the inner support members; Pore water pressure data of the soil surrounding the deep foundation pit support structure is collected by deploying pore water pressure sensors, which is a real-time monitoring value of the current pore water pressure inside the surrounding soil.

[0026] Step 1.3 involves synchronizing the collected data on horizontal displacement at the top of the pile, vertical settlement at the midpoint of the internal support, vertical settlement at the corners of the surrounding building foundations, stress changes, and pore water pressure to the constructed 3D digital twin via a 5G communication network in real time. Specifically, this includes: uniformly organizing the data collected by displacement sensors, stress sensors, and pore water pressure sensors on the top of the pile, vertical settlement at the midpoint of the internal support, vertical settlement at the corners of the surrounding building foundations, stress changes, and pore water pressure; completing time-series labeling and point association according to a preset data format; and ensuring that each set of monitoring data... Based on the unique measurement point location and acquisition time, the various types of real-time monitoring data are transmitted at high speed through the 5G communication network. Utilizing the high bandwidth and low latency characteristics of the 5G network, the time lag in data transmission is eliminated, and all real-time monitoring data is synchronized to the corresponding data receiving port of the constructed 3D digital twin without loss or delay. The data receiving port performs real-time analysis and verification on the transmitted monitoring data, eliminating abnormal and erroneous data generated during transmission, ensuring that the real-time data transmitted to the 3D digital twin is authentic and valid, and completing the real-time transmission of physical entity monitoring data to the 3D digital twin.

[0027] Step 1.4 involves fusing real-time data synchronized to the 3D digital twin with pre-recorded parameters in the 3D digital twin, including parameters of the supporting structure, surrounding buildings, underground pipelines, geological conditions, and groundwater depth. This allows the 3D digital twin to reflect the current state of the physical entity in real time, resulting in a real-time mapped digital twin. Specifically, this includes fusing real-time monitoring data synchronized to the 3D digital twin via a 5G network with pre-recorded parameters in the 3D digital twin, including parameters of the supporting structure, surrounding buildings, underground pipelines, geological conditions, and groundwater depth. During the data fusion process, displacement changes are calculated by subtracting the initial displacement value of the corresponding point in the 3D digital twin from the real-time collected displacement value; stress changes are calculated by subtracting the initial stress value of the corresponding point in the 3D digital twin from the real-time collected stress value; and pore water pressure changes are calculated by subtracting the initial pore water pressure value of the corresponding point in the 3D digital twin from the real-time collected pore water pressure value. These changes represent the magnitude of change in the current state of each measuring point compared to its initial state. The changes are correlated and matched with the spatial coordinates, structural attributes, and geological conditions of the corresponding points. The real-time status data after matching is then used to replace the original static parameters in the 3D digital twin. By fusing and calculating the real-time data of all measuring points with the basic parameters one by one, the real-time status of the support structure, surrounding buildings, soil, and groundwater in the 3D digital twin is updated. This ensures that the component positions, stress states, and deformations in the 3D digital twin are completely consistent with the current actual state of the physical entity, eliminating the state deviation between the physical entity and the 3D digital twin. Finally, the digital twin after real-time mapping is obtained.

[0028] In this embodiment of the invention, the locations of measuring points at the top of the piles, the midpoint of the internal support, and the corners of the surrounding building foundations in the physical entity are determined based on relevant parameters entered in the three-dimensional digital twin. Displacement, stress, and pore water pressure sensors are deployed at each measuring point to collect corresponding deformation, stress, and pore water pressure data. The collected real-time data is synchronized to the three-dimensional digital twin via a 5G communication network. The real-time data is then fused with various pre-entered basic parameters in the three-dimensional digital twin, enabling the three-dimensional digital twin to reflect the current state of the physical entity in real time. This overcomes the technical problems of traditional deep foundation pit monitoring, such as lack of scientific basis for measuring point deployment, incomplete and singular data collection, delayed data synchronization, and disconnection between real-time data and basic parameters, which make it difficult to achieve real-time linkage between the physical entity and the three-dimensional digital twin and to accurately reflect the current state of the physical entity. This achieves precise and reasonable measuring point deployment, comprehensive and real-time data collection, and accurate real-time mapping between the physical entity and the three-dimensional digital twin, providing real, comprehensive, and timely data support for subsequent spatial deformation characteristic analysis, mechanical parameter correction, and deformation prediction.

[0029] In a preferred embodiment of the present invention, step 2 above may include: Step 2.1: Extract the current horizontal displacement value of the top of the pile, the current vertical settlement value of the midpoint of the inner support, and the current vertical settlement value of the corner of the surrounding building foundation from the real-time mapped digital twin. Use the current horizontal displacement value and the current vertical settlement value as deformation feature parameters for the three feature points. Specifically, this includes: accurately extracting the real-time deformation data of the three key feature points from the real-time mapped digital twin, namely the current horizontal displacement value of the top of the pile, the current vertical settlement value of the midpoint of the inner support, and the current vertical settlement value of the corner of the surrounding building foundation. The current horizontal displacement value of the top of the pile is the difference between the current horizontal position of the top of the pile and the initial design horizontal position, which has been updated in real time through previous sensor acquisition and data fusion. The current vertical settlement value of the midpoint of the inner support is the difference between the current vertical position of the midpoint of the inner support and the initial installation vertical position. The current vertical settlement value of the corner of the surrounding building foundation is the difference between the current vertical position of the corner of the surrounding building foundation and the initial state vertical position. After extraction, these three real-time deformation data are used as deformation feature parameters for the top of the pile, the midpoint of the internal support, and the corner of the surrounding building foundation, respectively.

[0030] Step 2.2: Based on the pre-recorded spatial coordinates of the top of the pile, the midpoint of the inner support, and the corners of the surrounding building foundations in the 3D digital twin, and combined with the deformation feature parameters of the three feature points, calculate the offset coordinates of the three feature points in 3D space after deformation. Specifically, this includes: based on the extracted deformation feature parameters of the three feature points, extracting the pre-recorded initial spatial coordinates of the three feature points from the real-time mapped digital twin, namely the initial spatial coordinates of the top of the pile, the initial spatial coordinates of the midpoint of the inner support, and the initial spatial coordinates of the corners of the surrounding building foundations. Each initial spatial coordinate contains... direction, direction, The coordinate values ​​in the three dimensions of direction.

[0031] Initial top of pile Coordinates are ,initial Coordinates are ,initial Coordinates are The current horizontal displacement value of the top of the pile is The offset coordinates are ; Initial midpoint of internal support Coordinates are ,initial Coordinates are ,initial Coordinates are The current vertical settlement value at the midpoint of the internal support is The offset coordinates are ; Initial foundation corners of surrounding buildings Coordinates are ,initial Coordinates are ,initial Coordinates are The current vertical settlement value of the corners of the surrounding building foundations is The offset coordinates are .

[0032] Based on the deformation feature parameters of the three feature points, the offset coordinates of each feature point after deformation are calculated. The specific calculation process is as follows: The three-dimensional spatial coordinates of the top of the pile after deformation are obtained through the above calculations. The three-dimensional spatial coordinates of the midpoint of the inner support after deformation are obtained through the above calculations. The above calculations yielded the offset three-dimensional spatial coordinates of the corners of the surrounding building foundations after deformation.

[0033] After the calculation is completed, the offset coordinates of the three feature points are verified. The offset coordinates are compared with the real-time state of the corresponding feature points in the real-time mapped digital twin to ensure that the calculated offset coordinates accurately reflect the actual deformation position of each feature point.

[0034] Step 2.3: Using the offset coordinates of the three feature points as shape points, a spatial deformation feature curve passing through the three shape points is fitted in the 3D digital twin. The spatial deformation feature curve reflects the deformation distribution pattern along the direction from the top of the pile to the midpoint of the inner support and then to the corner of the surrounding building foundation. Specifically, this includes: using the calculated offset coordinates of the top of the pile, the midpoint of the inner support, and the corner of the surrounding building foundation as three core shape points, importing them into the real-time mapped digital twin, and carrying out the fitting of the spatial deformation feature curve; during the fitting process, firstly, the spatial positional relationship of the three shape points is determined, and the spatial distance between two adjacent shape points is calculated: the spatial distance between the top of the pile and the midpoint of the inner support. The calculation formula is: Spatial distance between the midpoint of the internal support and the corner of the foundation of the surrounding building The calculation formula is: Based on the spatial coordinates of the three model points and the spatial distance between adjacent model points, the slope of the line connecting adjacent model points is calculated to determine the approximate direction of the deformation curve. Then, a smooth fitting method is used to supplement the coordinates of the intermediate point between two adjacent model points. Let the coordinates of two adjacent model points be... The coordinates of the midpoint are The calculation formula is: By adding multiple intermediate points, the fitted curve is smoothly connected to the three model points. After fitting, a continuous spatial deformation characteristic curve is generated. This curve strictly passes through the three model points and clearly reflects the deformation distribution pattern along the direction from the top of the pile to the midpoint of the inner support and then to the corner of the surrounding building foundation, clarifying the deformation amplitude and trend of each point in this direction.

[0035] Step 2.4: Using the spatial deformation characteristic curve as the generatrix, and taking the direction perpendicular to the plane where the piles and internal supports are located as the sweeping direction, a continuous spatial deformation characteristic envelope surface is generated in the 3D digital twin. Specifically, this includes: determining the plane where the piles and internal supports are located; extracting the coordinates of multiple points of the piles and internal supports in the real-time mapped digital twin; calculating the normal vector of the plane; the direction of the normal vector is the direction perpendicular to the plane, which is the direction of this sweep; setting the sweeping range, the width of the sweeping range is equal to the lateral layout width of the deep foundation pit support structure, ensuring that the envelope surface generated by the sweeping fully covers the entire support structure and surrounding related areas; during the sweeping process, using the spatial deformation characteristic curve as the generatrix, keeping the shape and position of the generatrix unchanged, and sweeping at a uniform speed along the direction perpendicular to the plane where the piles and internal supports are located. During the sweeping process, every small distance moved generates a cross section with the same shape as the generatrix. Multiple continuous cross sections are connected to form a continuous 3D spatial surface. After the sweeping is completed, a complete spatial deformation feature envelope is obtained. This envelope comprehensively and intuitively represents the overall deformation characteristics of the deep foundation pit support structure and the surrounding buildings. It not only includes the deformation pattern along the direction from the top of the piles to the corner of the foundation of the surrounding buildings, but also covers the distribution of lateral deformation perpendicular to this direction.

[0036] In this embodiment of the invention, by extracting deformation feature parameters of three key feature points from the real-time mapped digital twin, and calculating the offset coordinates after deformation by combining the pre-entered spatial coordinates of each feature point, a spatial deformation feature curve reflecting the deformation distribution law is generated by fitting the offset coordinates as the shape value points. Then, a continuous spatial deformation feature envelope is generated by using this curve as the generatrix and the plane where the piles and internal supports are located as the sweeping direction. Therefore, this method overcomes the technical problem in traditional deformation analysis that relies only on single-point deformation data, making it difficult to comprehensively and intuitively represent the overall deformation distribution law between the support structure and the surrounding buildings, and making it difficult to reflect the spatial correlation of deformation. Thus, it achieves a comprehensive and accurate representation of the overall deformation characteristics of the support structure, clearly presenting the deformation distribution law and spatial correlation from the top of the piles to the midpoint of the internal supports and then to the corner of the foundation of the surrounding buildings.

[0037] In a preferred embodiment of the present invention, step 2 above may include: Step 2.5: Based on the spatial deformation feature envelope, perform finite element meshing on the spatial deformation feature envelope in the 3D digital twin to generate a discrete mesh covering the spatial deformation feature envelope, and obtain the spatial coordinates of each mesh node in the discrete mesh. Specifically, this includes: calling the generated complete spatial deformation feature envelope, and using this envelope as a basis, performing finite element meshing in the real-time mapped digital twin. The core purpose of meshing is to discretize the continuous spatial deformation feature envelope into multiple regular mesh units, realizing refined decomposition and quantitative analysis of the deformation features of the envelope. During the meshing process, set the accuracy parameters of the meshing, based on the size and deformation monitoring of the deep foundation pit support structure. For accuracy requirements, the side length of the mesh cells is determined. The setting of the mesh cell side length must balance computational accuracy and efficiency. The smaller the side length, the higher the computational accuracy, but the greater the computational load. Typically, the side length of the mesh cells is set to 0.1% to 0.5% of the maximum size of the support structure. A quadrilateral meshing method is used to comprehensively divide the spatial deformation feature envelope. During the division process, it is ensured that the mesh cells are evenly distributed, and that adjacent mesh cells are seamlessly connected, without overlap or gaps, completely covering the entire spatial deformation feature envelope without missing any deformation features. After the division is completed, a discrete mesh covering the spatial deformation feature envelope is generated. This discrete mesh consists of multiple regular quadrilateral mesh cells, and the intersection of each mesh cell is the mesh node. Subsequently, the three-dimensional spatial coordinates of all mesh nodes in the discrete mesh are extracted from the real-time mapped digital twin. The spatial coordinates of each mesh node include values ​​in three dimensions: X, Y, and Z. During the extraction process, the coordinates of each node are recorded one by one, establishing a one-to-one correspondence between mesh nodes and coordinate values, forming a list of mesh node coordinates.

[0038] Step 2.6: Based on the spatial coordinates of each grid node in the discrete grid, and combined with the deformation distribution law represented by the spatial deformation feature envelope, the deformation interpolation value at each grid node is obtained through interpolation calculation. The deformation interpolation value reflects the deformation amount of each grid node relative to the initial state. Specifically, based on the obtained spatial coordinates of each grid node in the discrete grid, and combined with the overall deformation distribution law represented by the spatial deformation feature envelope, the deformation interpolation value at each grid node is obtained through interpolation calculation, thereby supplementing the deformation data of the grid nodes and realizing the comprehensive quantification of the deformation characteristics of the envelope. Before the interpolation calculation, the deformation distribution law of the spatial deformation feature envelope is first clarified. This law is determined by the fitted spatial deformation feature curve. Along the direction from the top of the pile to the midpoint of the inner support and then to the corner of the surrounding building foundation, the deformation amount shows a continuous changing trend, and the lateral deformation amount perpendicular to this direction shows a uniform distribution trend. Using the linear interpolation method, the deformation interpolation value of each grid node is calculated based on the feature points of two adjacent known deformation data, the top of the pile, the midpoint of the inner support, the corner of the surrounding building foundation, and the fitted spatial deformation feature curve.

[0039] Let the grid nodes to be solved be The coordinates are Two surrounding reference points were selected as The corresponding deformation is ; node arrive The spatial distance is: ; node arrive The spatial distance is: ; Two reference points and The spatial distance between them is: ; Reference point Interpolation weights: Reference point Interpolation weights: The deformation interpolation value of mesh node P is: Each grid node completes deformation interpolation calculation according to the above method. After the calculation is completed, the deformation interpolation of all grid nodes is verified and the deformation interpolation of adjacent grid nodes is compared to ensure that the trend of deformation interpolation is consistent with the overall deformation law of the spatial deformation feature envelope surface, without abrupt changes or anomalies. The deformation interpolation accurately reflects the deformation of each grid node relative to the initial state.

[0040] Step 2.7: Based on the spatial coordinates of each grid node in the discrete grid and the corresponding deformation interpolation, the local curvature at each grid node is calculated using the discrete curvature estimation method. The overall curvature distribution of the spatial deformation feature envelope surface is then generated by fitting the local curvature at each grid node. Specifically, based on the spatial coordinates of each grid node obtained in Step 2.5 and the deformation interpolation corresponding to each grid node calculated in Step 2.6, the local curvature at each grid node is calculated one by one using the discrete curvature estimation method. Then, the overall curvature distribution of the spatial deformation feature envelope surface is generated by fitting, thereby achieving a refined quantitative analysis of the deformation features of the envelope surface.

[0041] The local curvature at each grid node is calculated. The discrete curvature estimation uses the three-point method. Let the target grid node be... The three adjacent nodes are The spatial distance between each point is calculated by taking the square root of the sum of the squares of the differences in the coordinates of the two points, that is: Depend on Determine the fitting plane and calculate the target nodes. The perpendicular distance d to the plane; take The average spatial distance to the three adjacent nodes is The local curvature of the target node is: The magnitude of local curvature reflects the degree of deformation and bending at the location of the node; the larger the value, the greater the deformation and bending. After calculating the local curvature of all grid nodes, the overall curvature distribution is fitted. The spatial coordinates of all grid nodes are associated with their corresponding local curvatures to establish a correspondence between node coordinates and local curvature. A smooth fitting method is used, based on the local curvature of each grid node and combined with the overall shape of the spatial deformation feature envelope, to fit and generate a continuous overall curvature distribution. During the fitting process, it is ensured that the overall curvature distribution accurately reflects the local curvature change trend of each region, smoothly connects the curvature values ​​of adjacent nodes, and avoids abrupt changes or discontinuities. After fitting, the overall curvature distribution of the spatial deformation feature envelope is obtained.

[0042] Step 2.8 involves comparing the overall curvature distribution with the preset reference curvature in the 3D digital twin point by point, calculating the curvature deviation at each grid node, and performing a weighted integral based on the curvature deviations at all grid nodes to obtain a spatial stiffness correction coefficient reflecting the overall deformation coordination of the current support structure. Specifically, this includes: extracting the preset reference curvature from the real-time mapped digital twin. This reference curvature is the standard curvature distribution corresponding to the spatial deformation feature envelope when the deep foundation pit support structure is in normal working condition and without abnormal deformation. It is pre-calculated and determined by engineering design parameters, geological condition parameters, and mechanical performance parameters of the support structure, serving as a reference standard for evaluating the deformation coordination of the current support structure; comparing the fitted overall curvature distribution with the preset reference curvature point by point, calculating the curvature deviation at each grid node, and assuming the current local curvature of the node is... The reference curvature is Then the curvature deviation is: If the curvature deviation is positive, it indicates that the deformation and bending at that node is greater than the normal state; if the curvature deviation is negative, it indicates that the deformation and bending at that node is less than the normal state; if the curvature deviation is zero, it indicates that the deformation and bending at that node is within the normal range. After the curvature deviation is calculated, a weighted integral is performed based on the curvature deviations at all mesh nodes to obtain the spatial stiffness correction coefficient. Let the weight coefficient of the i-th node be... With the weight of the critical stress area set at 1.2~1.5 and the weight of the secondary area set at 0.8~1.0, the weighted curvature deviation is: Integrating over the entire discrete grid region, the spatial stiffness correction coefficient... for: The spatial stiffness correction coefficient is a quantitative parameter that comprehensively reflects the overall deformation coordination of the current support structure. The closer the coefficient value is to 1, the better the deformation coordination of the support structure and the smaller the deviation from the normal working state. The farther the coefficient value deviates from 1, the worse the deformation coordination of the support structure and the greater the possibility of abnormal deformation.

[0043] In this embodiment of the invention, the spatial deformation feature envelope is divided into finite element meshes to obtain the spatial coordinates of discrete mesh nodes. Combined with the deformation distribution law, the deformation interpolation of each mesh node is obtained by interpolation calculation. The discrete curvature estimation method is used to calculate the local curvature and fit it to generate the overall curvature distribution. Then, the overall curvature distribution is compared with the preset reference curvature point by point, the curvature deviation is calculated, and the spatial stiffness correction coefficient is obtained by weighted integration. Therefore, this method overcomes the technical problems of traditional deformation analysis, such as difficulty in performing fine quantitative calculation of deformation features, difficulty in accurately assessing the overall deformation coordination of the support structure, and lack of reliable stiffness correction parameters. Thus, it achieves fine quantitative analysis of the deformation features of the support structure and accurately obtains the spatial stiffness correction coefficient that reflects the overall deformation coordination of the support structure.

[0044] In a preferred embodiment of the present invention, step 3 above may include: Step 3.1: Read the initial values ​​of the currently set mechanical parameters of the support structure from the 3D digital twin, and extract the real-time monitoring data at the current moment from the real-time mapped digital twin. Specifically, this includes: calling the completed 3D digital twin, accurately reading the initial values ​​of the currently set mechanical parameters of the support structure from the parameter database of the digital twin. These initial values ​​are pre-set based on the design documents, material performance test reports, and geological survey data of the deep foundation pit support structure. The core parameters include the elastic modulus, Poisson's ratio, compressive strength, and shear strength of the piles, and the elastic modulus, moment of inertia, and compressive bearing capacity of the internal supports. Key mechanical parameters, each with a specific numerical value, serve as the basis for subsequent parameter corrections. The real-time mapped digital twin is invoked, and real-time monitoring data at the current moment is extracted from this digital twin. The data all come from sensor acquisition, 5G transmission, and effective data fusion. Specifically, it includes the current horizontal displacement data of the top of the pile, the current vertical settlement data of the midpoint of the internal support, the current vertical settlement data of the corner of the surrounding building foundation, the current stress change data of the pile and internal support, and the current pore water pressure data of the surrounding soil. The real-time and completeness of the data are ensured during the extraction process, and a real-time monitoring data list is established.

[0045] Step 3.2: Compare the real-time monitoring data with the historical monitoring data in the 3D digital twin to calculate the deformation rate and stress change rate at the current moment. Combined with the spatial stiffness correction coefficient, construct a mechanical parameter correction function reflecting the current stress characteristics of the support structure. Specifically, based on the acquired initial mechanical parameters and the current real-time monitoring data, extract recent continuous historical monitoring data from the historical database of the 3D digital twin. The time interval of the historical monitoring data should be consistent with the acquisition interval of the current real-time monitoring data to ensure the rationality of the data comparison. Typically, historical monitoring data from the past 1 to 3 hours is selected, and a set is collected every 10 minutes to form a continuous historical data sequence. Calculate the deformation rate and stress change rate at the current moment. The specific calculation process is as follows: Calculation of deformation rate... , for The most recent set of historical monitoring times is The corresponding historical deformation data is The current time is The real-time deformation data of a certain monitoring indicator (horizontal displacement of the top of the pile, vertical settlement of the midpoint of the internal support, etc.) at the current moment is: The method for calculating the rate of stress change is the same as that for calculating the deformation rate. This is the current rate of stress change.

[0046] After calculating the deformation rate and stress change rate, and combining the obtained spatial stiffness correction coefficient, a mechanical parameter correction function reflecting the current stress characteristics of the support structure is constructed. Spatial stiffness correction factor As a weighting factor, it reflects the influence of the overall deformation coordination of the support structure on the correction of mechanical parameters; deformation rate and rate of change of stress As a core influencing factor, it reflects the dynamic requirements of real-time deformation and stress changes on mechanical parameters; preset correction coefficients This formula is used to adapt to different support structure types and geological conditions, ensuring the applicability of the correction function. Ultimately, the correction amount of mechanical parameters for various support structures is calculated through this formula, achieving accurate matching between mechanical parameters and the current stress state. This ensures that the correction function fully reflects the influence of real-time deformation, stress changes, and the overall deformation coordination on the mechanical parameters.

[0047] Step 3.3: Substitute the initial values ​​of the mechanical parameters of the support structure into the mechanical parameter correction function, and obtain a set of preliminarily corrected mechanical parameters through iterative calculation. The iterative calculation aims to reduce the difference between real-time monitoring data and the output data of the 3D digital twin simulation. Specifically, it includes: substituting the read initial values ​​of the mechanical parameters of the support structure into the constructed mechanical parameter correction function one by one, and carrying out iterative calculation. The core objective of the iterative calculation is to gradually reduce the difference between real-time monitoring data and the output data of the 3D digital twin simulation, and achieve preliminary optimization of the mechanical parameters. The specific iterative calculation process is as follows: First, substitute the initial values ​​of the mechanical parameters into the correction function to calculate the first correction amount of each mechanical parameter. The preliminarily corrected mechanical parameters are equal to the initial values ​​of the mechanical parameters plus the corresponding first correction amount. Then, substitute the preliminarily corrected mechanical parameters into the 3D digital twin for simulation calculation to obtain a set of simulated deformation data and simulated stress data. Compare the simulated data with... The extracted real-time monitoring data is compared, and the difference between the two is calculated. If the difference does not reach the preset iteration threshold, which is set according to the engineering monitoring accuracy requirements and is usually 5% to 10% of the real-time monitoring data, the mechanical parameters after the initial correction are substituted into the correction function again to calculate the second correction amount. The mechanical parameters after the second correction are equal to the mechanical parameters after the initial correction plus the second correction amount. The above process is repeated. After each correction, a simulation calculation and difference comparison are performed until the difference between the simulated data and the real-time monitoring data is reduced to within the preset iteration threshold. The iteration stops, and the mechanical parameters obtained at this time are the mechanical parameters after the initial correction. During the iteration process, the calculation of each correction amount is combined with the current deformation rate, stress change rate, and spatial stiffness correction coefficient to ensure that each correction can fit the actual stress state of the current support structure, gradually optimize the mechanical parameters, and reduce the difference between simulation and actual measurement.

[0048] Step 3.4: Substitute the initially corrected mechanical parameters into the 3D digital twin for forward simulation calculation to obtain simulated deformation data. Compare the simulated deformation data with the real-time monitoring data. If the error exceeds the preset convergence threshold, readjust the initially corrected mechanical parameters according to the error until the error between the simulated deformation data and the real-time monitoring data converges to within the convergence threshold, thus obtaining the corrected mechanical parameters. Specifically, this includes: First, fully substituting the initially corrected mechanical parameters into the parameter system of the 3D digital twin to update the mechanical parameters of the support structure in the digital twin. Perform forward simulation calculation according to the current construction conditions, geological conditions, and load conditions of the deep foundation pit. The simulation calculation process is consistent with the actual construction process, obtaining simulated deformation data under the corresponding conditions, including simulated values ​​of horizontal displacement at the top of the piles, simulated values ​​of vertical settlement at the midpoint of the internal support, and simulated values ​​of vertical settlement at the corners of the surrounding building foundations, etc., which correspond to the extracted real-time monitoring data. After the simulation calculation is completed, the obtained simulation... The deformation data and real-time monitoring data are compared one by one to calculate the error between them. The average of the relative errors of all monitoring indicators is taken as the overall error between the simulation and the actual measurement. The calculated overall error is compared with the preset convergence threshold in the 3D digital twin. The convergence threshold is preset according to the accuracy requirements of safety control in deep foundation pit engineering, usually set to 3% to 5%. If the overall error does not exceed the convergence threshold, it means that the initially corrected mechanical parameters accurately match the actual working conditions. At this time, the initially corrected mechanical parameters are the final corrected mechanical parameters. If the overall error exceeds the preset convergence threshold, the initially corrected mechanical parameters are readjusted according to the magnitude of the error. The adjustment process is as follows: the error adjustment amount is equal to the initially corrected mechanical parameters multiplied by the difference between the overall error and the convergence threshold, and then divided by the convergence threshold. If the simulated data is greater than the real-time monitoring data, it means that the mechanical parameters are too large and the error adjustment amount needs to be subtracted. If the simulated data is less than the real-time monitoring data, it means that the mechanical parameters are too small and the error adjustment amount needs to be added.

[0049] After the adjustment is completed, the adjusted mechanical parameters are substituted back into the three-dimensional digital twin for forward simulation calculation. The above error comparison and parameter adjustment process is repeated until the overall error between the simulated deformation data and the real-time monitoring data converges to within the preset convergence threshold. The adjustment is then stopped, and the corrected mechanical parameters of the support structure are finally obtained.

[0050] In this embodiment of the invention, by reading the initial values ​​of the mechanical parameters of the support structure and real-time monitoring data, and combining them with historical monitoring data to calculate the deformation rate and stress change rate, and by constructing a mechanical parameter correction function based on the spatial stiffness correction coefficient, iterative calculations are performed with the goal of reducing the difference between the monitoring data and the simulation data. Then, through forward simulation, error comparison, and repeated adjustments, the results are converged to obtain the corrected mechanical parameters. Therefore, this method overcomes the technical problems in traditional methods, such as the difficulty in dynamically correcting the mechanical parameters of the support structure, the inconsistency with the actual stress state, the large deviation between the simulation calculation and the measured data, and the difficulty in matching the real working conditions. This achieves precise iterative optimization of mechanical parameters and improves the fit between the parameters and the actual stress state of the structure.

[0051] In a preferred embodiment of the present invention, step 4 above may include: Step 4.1 involves updating the corrected mechanical parameters to the 3D digital twin and extracting the current real-time monitoring data from the real-time mapped digital twin as the initial state for prediction. Specifically, this includes: completely updating the obtained corrected mechanical parameters of the support structure to the parameter database of the 3D digital twin, fully replacing the original uncorrected initial values ​​of the mechanical parameters, and ensuring that the mechanical parameters in the 3D digital twin accurately match the actual stress state of the support structure; the corrected mechanical parameters include all key parameters such as the elastic modulus, Poisson's ratio, compressive strength, and shear strength of the piles, and the elastic modulus, moment of inertia, and compressive bearing capacity of the internal supports. During the update process, the value of each parameter is checked one by one to avoid parameter omissions or input errors, ensuring that the parameter system of the digital twin truly reflects the mechanical performance of the current support structure; and calling the real-time mapped digital twin to extract all the current real-time monitoring data from the digital twin, using it as the initial state for deformation prediction.

[0052] Step 4.2: Based on the pre-entered construction plan parameters in the 3D digital twin, set the load condition sequence for future periods. This includes: extracting the pre-entered construction plan parameters from the 3D digital twin. These parameters, entered by the construction unit according to the engineering construction plan, include the division of excavation periods, excavation depths for each period, construction sequence of support structures, construction equipment arrival plans, load application methods and time nodes, etc., covering key work condition information throughout the construction process, providing a basis for setting the load condition sequence; and setting the load condition sequence for future periods based on the construction plan parameters and the construction characteristics of deep foundation pit engineering. The load condition sequence must be synchronized with the construction progress, clearly defining the load type, load size, and range of action for each time node to ensure the work condition settings are appropriate. In actual construction, the specific setup process is as follows: First, determine the future forecast period, usually set to the next 24 to 72 hours in conjunction with the construction progress, dividing it into 1-hour or 2-hour time nodes to form a continuous time sequence; then, for each time node, determine the load conditions of that node according to the corresponding construction procedures, such as the earth pressure load corresponding to the excavation procedure, the equipment load corresponding to the arrival of construction equipment, and the construction load corresponding to the construction of the support structure; finally, calculate the load magnitude under each condition. The earth pressure load is calculated as the soil weight multiplied by the current excavation depth, the equipment load is determined according to the actual weight of the construction equipment, and the construction load is set according to the construction process requirements. The load type, load magnitude, and location of each time node are compiled into a booklet to form a complete future time period load condition sequence.

[0053] Step 4.3 involves inputting the corrected mechanical parameters, predicted initial state, and load case sequence into the 3D digital twin. The built-in numerical solver within the 3D digital twin performs time history analysis to calculate the predicted deformation sequence of the support structure at each time node in the future period. Specifically, this includes: inputting the updated corrected mechanical parameters, the determined predicted initial state, and the set future load case sequence into the 3D digital twin; starting the built-in numerical solver to perform time history analysis calculations. The core is to simulate the dynamic changes of the load in the future period, combined with the corrected mechanical parameters, to progressively calculate the deformation of the support structure at each time node, thus obtaining the predicted deformation sequence. The numerical solver, starting from the predicted initial state, applies the corresponding loads one by one according to the time nodes of the load case sequence. Within each time period, based on the corrected mechanical parameters such as the elastic modulus of the piles and the moment of inertia of the internal support section, combined with the stress-strain relationship, the deformation increment of the support structure within that time period is calculated. The method for calculating the deformation increment is as follows: The stress change value within a given time period is divided by the corresponding mechanical parameter, such as the elastic modulus, to obtain the strain value. This strain value is then multiplied by the length of the corresponding component of the support structure to obtain the deformation increment for that time period. This deformation increment is added to the deformation prediction value of the previous time node to obtain the deformation prediction value for the current time node. This process is repeated, starting from the first time node and calculating for each time period until the deformation calculation for all time nodes within the future prediction period is completed. This results in a sequence of predicted deformation values ​​for the support structure at each time node in the future period. This sequence covers predicted data for all key monitoring indicators, including the horizontal displacement at the top of the piles, the vertical settlement at the midpoint of the internal support, and the vertical settlement at the corners of the surrounding building foundations. Each predicted value corresponds to a specific time node, clearly reflecting the future deformation trend. The numerical solver verifies the calculation results in real time to ensure that the predicted deformation value at each time node matches the load condition and the corrected mechanical parameters, avoiding calculation errors and ensuring the accuracy and reliability of the deformation prediction value sequence. This provides precise data support for subsequent early warning comparisons.

[0054] Step 4.4 compares each predicted value in the deformation prediction value sequence with the preset warning thresholds in the 3D digital twin. The warning thresholds include the horizontal displacement limit of the top of the pile, the vertical settlement limit of the midpoint of the internal support, and the differential settlement limit of the corner of the foundation of the surrounding buildings. Specifically, it includes: extracting the preset warning thresholds from the 3D digital twin. These warning thresholds are preset according to the deep foundation pit engineering design specifications, geological survey data, and engineering safety control requirements. They are used to determine whether the deformation of the support structure is within a safe range. The core includes three types of warning thresholds: the horizontal displacement limit of the top of the pile, the vertical settlement limit of the midpoint of the internal support, and the differential settlement limit of the corner of the foundation of the surrounding buildings. The calculation method for the differential settlement limit of the corners of the surrounding building foundations is as follows: the difference between the predicted vertical settlement values ​​of two adjacent corners of the surrounding building foundations shall not exceed the allowable differential settlement value of the building foundation. The allowable differential settlement value is set according to the building structure type, and is usually 0.1% to 0.3% of the building height. Each predicted value in the obtained deformation prediction value sequence is compared with the corresponding warning threshold item by item. The comparison process is one-to-one, without omitting any time node and monitoring indicator. For each time node, the predicted horizontal displacement value of the top of the pile is compared with the limit value of the horizontal displacement of the top of the pile, and the predicted vertical settlement value of the midpoint of the internal support is compared with the limit value of the internal support. The vertical settlement limit at the midpoint of the support is compared with the difference between the predicted vertical settlement values ​​of two adjacent corners of the foundation of surrounding buildings and the differential settlement limit of the corners of the foundation of surrounding buildings. At the same time, the deviation between each predicted value and the corresponding warning threshold is calculated. The deviation calculation method is: predicted value minus warning threshold. If the deviation is positive, it means that the predicted deformation has exceeded or reached the warning threshold, and there is a safety hazard. If the deviation is negative, it means that the predicted deformation is within the safe range and there is no safety hazard. The comparison results and deviation values ​​at each time point are recorded one by one to form a complete comparison report, clearly marking the time points where anomalies occur and the corresponding monitoring indicators.

[0055] Step 4.5: When any predicted value reaches the corresponding warning threshold, the 3D digital twin automatically generates a warning command and sends the warning command and the corresponding deformation prediction value to the construction monitoring terminal to guide the construction unit to adjust the excavation speed, add temporary supports, and take dewatering measures. Specifically, after comparison, the 3D digital twin automatically compares and judges the results. When the predicted value of any monitoring indicator at any time node reaches or exceeds the corresponding warning threshold, the digital twin immediately generates a warning command. The warning command clearly includes the warning level, abnormal monitoring indicator, corresponding time node, deformation prediction value, warning threshold, and deviation value, ensuring that construction monitoring personnel can quickly grasp the core information of the abnormal situation. After the warning command is generated, the 3D digital twin sends the warning command, the corresponding deformation prediction value, and the comparison report to the construction monitoring terminal through the wireless communication module. The construction monitoring terminal includes the on-site monitoring room terminal, the mobile phone terminal of the construction management personnel, etc., to ensure that relevant personnel receive the warning information in a timely manner and avoid delays in handling.

[0056] Upon receiving the early warning instruction, the construction unit, based on the early warning information and corresponding deformation prediction values, and in conjunction with the actual on-site construction conditions, promptly takes targeted construction adjustment measures. This includes adjusting the excavation speed, calculating an adjustment coefficient based on the deviation between the deformation prediction value and the early warning threshold (the ratio of the early warning threshold to the deformation prediction value), and multiplying the original excavation speed by the adjustment coefficient to obtain the adjusted excavation speed. A larger deviation requires a smaller adjustment coefficient and a slower excavation speed. Temporary supports are also added. Based on the predicted stress changes of the piles and internal supports, the required number and cross-sectional dimensions of temporary supports are calculated. The cross-sectional area of ​​the temporary supports is calculated by dividing the required stress value by the compressive strength of the temporary support material, ensuring the temporary supports... Effective load sharing and suppression of deformation development; implementation of dewatering measures, calculating the required dewatering depth based on the predicted pore water pressure of the surrounding soil, where the dewatering depth equals the water level height corresponding to the current pore water pressure minus the water level height corresponding to the target pore water pressure, thereby reducing the pore water pressure of the surrounding soil, enhancing soil stability, and reducing deformation of the support structure; after construction adjustments are completed, the 3D digital twin re-extracts the adjusted real-time monitoring data, updates the initial predicted state, and repeats steps 4.2 to 4.4, re-predicting and comparing deformation until all predicted deformation values ​​converge to within the warning threshold, at which point the warning is lifted, ensuring the construction safety of the deep foundation pit support structure and achieving dynamic control of the construction process.

[0057] In this embodiment of the invention, the corrected mechanical parameters are updated, real-time monitoring data is used as the initial state for prediction, and a load condition sequence is set in combination with construction plan parameters. A time history analysis is performed using a numerical solver to obtain a sequence of predicted deformation values. The predicted values ​​are compared with multiple early warning thresholds one by one, and an early warning command is automatically generated and sent to the construction monitoring terminal when the threshold is reached to guide construction adjustments. Therefore, this technical approach overcomes the technical problems of traditional deep foundation pit deformation prediction, such as difficulty in accurately predicting future deformation, delayed early warning, lack of targeted construction control guidance, passive deformation management, and insufficient safety. Thus, it achieves the technical effect of accurate prediction and intelligent early warning of future deformation of the support structure, timely guidance for construction units to take reasonable control measures, effectively avoid construction safety risks, and improve the construction safety and intelligent management level of deep foundation pit projects.

[0058] like Figure 2 As shown, embodiments of the present invention also provide a deformation prediction system for deep foundation pit support structures based on digital twins, comprising: The mapping module is used to deploy displacement sensors, stress sensors, and pore water pressure sensors on the top of the piles, the midpoint of the internal support, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. It collects data on the horizontal displacement, vertical settlement, stress changes of the support structure, and pore water pressure of the surrounding soil in real time, and synchronizes the collected real-time data to the 3D digital twin to realize the real-time mapping between the physical entity and the digital twin, and obtain the real-time mapped digital twin. The calculation module is used to construct a spatial deformation feature envelope surface defined by three deformation data points in the three-dimensional digital twin based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the foundations of the surrounding buildings contained in the real-time mapped digital twin. The feature envelope surface is meshed by finite element method to obtain the deformation interpolation of each mesh node. The curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained according to the deviation between the curvature distribution and the preset reference curvature. The correction module is used to iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficient, so as to obtain the corrected mechanical parameters. The processing module is used to perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation value of the support structure in the future period; when the predicted deformation value reaches the preset warning threshold, a warning signal is issued to guide construction adjustments.

[0059] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0060] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0061] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0062] 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 predicting deformation of a deep foundation pit support structure based on digital twinning, characterized in that, The method includes: Step 1: Displacement sensors, stress sensors, and pore water pressure sensors are installed on the top of the piles, the midpoint of the inner support, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. The horizontal displacement, vertical settlement, stress change, and pore water pressure data of the support structure and the surrounding soil are collected in real time. The collected real-time data is synchronized to the 3D digital twin to realize the real-time mapping between the physical entity and the digital twin, and the real-time mapped digital twin is obtained. Step 2: Based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the foundations of the surrounding buildings contained in the real-time mapped digital twin, a spatial deformation feature envelope surface defined by the three deformation data points is constructed in the three-dimensional digital twin; the feature envelope surface is meshed by finite element method to obtain the deformation interpolation of each mesh node, the curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained according to the deviation between the curvature distribution and the preset reference curvature. Step 3: Based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficient, iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin to obtain the corrected mechanical parameters. Step 4: Perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation value of the support structure in the future period; when the predicted deformation value reaches the preset warning threshold, issue a warning signal to guide construction adjustments. 2.The digital-twin-based deep foundation pit support structure deformation prediction method according to claim 1, characterized in that, Before step 1, the process also includes: constructing a three-dimensional digital twin containing the deep foundation pit support structure, surrounding buildings, underground pipelines and site geological conditions using BIM technology, and inputting the support structure parameters, surrounding building parameters, underground pipeline parameters, geological parameters and groundwater depth information into the three-dimensional digital twin. 3.The digital-twin-based deep foundation pit support structure deformation prediction method according to claim 2, characterized in that, Step 1 above includes: Based on the pile position parameters, internal support distribution parameters, and corner coordinate parameters of the surrounding building foundations entered in the 3D digital twin, the layout positions of the top measuring point of the pile, the midpoint measuring point of the internal support, and the corner measuring points of the surrounding building foundations are determined in the corresponding physical entity. Displacement sensors, stress sensors, and pore water pressure sensors are installed at each of the determined measuring points. The displacement sensors collect horizontal displacement data at the top of the pile, vertical settlement data at the midpoint of the internal support, and vertical settlement data at the corners of the foundations of surrounding buildings. The stress sensors collect stress change data of the pile and internal support, and the pore water pressure sensors collect pore water pressure data of the surrounding soil. The collected data on horizontal displacement at the top of the pile, vertical settlement at the midpoint of the internal support, vertical settlement at the corner of the foundation of the surrounding buildings, stress change data, and pore water pressure data are synchronized in real time to the constructed three-dimensional digital twin via a 5G communication network. The real-time data synchronized to the 3D digital twin is fused with the pre-entered support structure parameters, surrounding building parameters, underground pipeline parameters, geological parameters, and groundwater depth information in the 3D digital twin, so that the 3D digital twin reflects the current state of the physical entity in real time, resulting in a real-time mapped digital twin.

4. The digital-twin-based deep foundation pit support structure deformation prediction method according to claim 3, characterized in that, Step 2: Based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the surrounding building foundations contained in the real-time mapped digital twin, construct a spatial deformation feature envelope defined by three deformation data points in the three-dimensional digital twin, including: Extract the current horizontal displacement value of the top of the pile, the current vertical settlement value of the midpoint of the inner support, and the current vertical settlement value of the corner of the foundation of the surrounding building from the real-time mapped digital twin, and use the current horizontal displacement value and the current vertical settlement value as the deformation feature parameters of the three feature points; Based on the pre-recorded spatial coordinates of the top of the pile, the midpoint of the inner support, and the corner of the surrounding building foundation in the three-dimensional digital twin, and combined with the deformation characteristic parameters of the three feature points, the offset coordinates of the three feature points in three-dimensional space after deformation are calculated respectively. Using the offset coordinates of three feature points as shape value points, a spatial deformation feature curve passing through the three shape value points is fitted in the three-dimensional digital twin. The spatial deformation feature curve reflects the deformation distribution law along the direction from the top of the pile to the midpoint of the inner support and then to the corner of the surrounding building foundation. Using the spatial deformation characteristic curve as the generatrix, and taking the direction perpendicular to the plane where the piles and internal supports are located as the sweeping direction, a continuous spatial deformation characteristic envelope surface is generated by sweeping in the three-dimensional digital twin.

5. The digital-twin-based deep foundation pit support structure deformation prediction method according to claim 4, characterized in that, The feature envelope surface is meshed using finite element methods to obtain deformation interpolation for each mesh node. The curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes. The spatial stiffness correction coefficient is obtained based on the deviation between the curvature distribution and the preset reference curvature, including: Based on the spatial deformation feature envelope, the spatial deformation feature envelope is meshed using finite element methods in a three-dimensional digital twin to generate a discrete mesh covering the spatial deformation feature envelope, and the spatial coordinates of each mesh node in the discrete mesh are obtained. Based on the spatial coordinates of each grid node in the discrete grid, and combined with the deformation distribution law represented by the spatial deformation feature envelope, the deformation interpolation value at each grid node is obtained by interpolation calculation. The deformation interpolation value reflects the amount of deformation of each grid node relative to the initial state. Based on the spatial coordinates of each grid node in the discrete grid and the corresponding deformation interpolation, the local curvature at each grid node is calculated using the discrete curvature estimation method, and the overall curvature distribution of the spatial deformation feature envelope is generated by fitting the local curvature at each grid node. The overall curvature distribution is compared point by point with the preset reference curvature in the three-dimensional digital twin. The curvature deviation at each grid node is calculated, and a weighted integral is performed based on the curvature deviation at all grid nodes to obtain a spatial stiffness correction coefficient that reflects the overall deformation coordination of the current support structure. 6.The digital-twin-based deep foundation pit support structure deformation prediction method according to claim 5, characterized in that, Step 3: Based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficients, iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin to obtain the corrected mechanical parameters, including: Read the initial values ​​of the mechanical parameters of the support structure set at the current time from the three-dimensional digital twin, and extract the real-time monitoring data at the current moment from the real-time mapped digital twin; By comparing real-time monitoring data with historical monitoring data in a three-dimensional digital twin, the deformation rate and stress change rate at the current moment are calculated. Combined with the spatial stiffness correction coefficient, a mechanical parameter correction function reflecting the current stress characteristics of the support structure is constructed. Substitute the initial values ​​of the mechanical parameters of the support structure into the mechanical parameter correction function, and obtain a set of preliminary corrected mechanical parameters through iterative calculation. The iterative calculation aims to reduce the difference between real-time monitoring data and three-dimensional digital twin simulation output data. The initially corrected mechanical parameters are substituted into the three-dimensional digital twin for forward simulation calculation to obtain simulated deformation data. The simulated deformation data is then compared with the real-time monitoring data. If the error exceeds the preset convergence threshold, the initially corrected mechanical parameters are readjusted according to the error until the error between the simulated deformation data and the real-time monitoring data converges to within the convergence threshold, thus obtaining the corrected mechanical parameters.

7. The method for predicting the deformation of deep foundation pit support structures based on digital twins according to claim 6, characterized in that, Step 4: Perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation values ​​of the support structure in the future period. When the predicted deformation value reaches the preset warning threshold, a warning signal is issued to guide construction adjustments, including: The corrected mechanical parameters are updated in the three-dimensional digital twin, and the real-time monitoring data at the current moment is extracted from the real-time mapped digital twin as the predicted initial state. Based on the construction plan parameters pre-entered in the 3D digital twin, set the load condition sequence for future time periods; The corrected mechanical parameters, predicted initial state, and load condition sequence are input into the three-dimensional digital twin. Time history analysis is performed using the numerical solver built into the three-dimensional digital twin to calculate the predicted deformation value sequence of the support structure at each time node in the future period. Each predicted value in the deformation prediction value sequence is compared with the preset warning threshold in the three-dimensional digital twin. The warning threshold includes the horizontal displacement limit at the top of the pile, the vertical settlement limit at the midpoint of the internal support, and the differential settlement limit at the corner of the foundation of the surrounding buildings. When any predicted value reaches the corresponding warning threshold, the 3D digital twin automatically generates a warning command and sends the warning command and the corresponding deformation prediction value to the construction monitoring terminal to guide the construction unit to adjust the excavation speed, add temporary supports and take dewatering measures.

8. A deep foundation pit support structure deformation prediction system based on digital twins, wherein the system implements the method as described in any one of claims 1 to 7, characterized in that, include: The mapping module is used to deploy displacement sensors, stress sensors, and pore water pressure sensors on the top of the piles, the midpoint of the internal support, and the corners of the surrounding building foundations in the physical entity corresponding to the 3D digital twin. It collects data on the horizontal displacement, vertical settlement, stress changes of the support structure, and pore water pressure of the surrounding soil in real time, and synchronizes the collected real-time data to the 3D digital twin to realize the real-time mapping between the physical entity and the digital twin, and obtain the real-time mapped digital twin. The calculation module is used to construct a spatial deformation feature envelope defined by three deformation data points in the three-dimensional digital twin based on the horizontal displacement data of the top of the piles, the vertical settlement data of the midpoint of the internal support, and the differential settlement data of the corners of the foundations of the surrounding buildings contained in the real-time mapped digital twin. The feature envelope surface is meshed using finite element methods to obtain the deformation interpolation of each mesh node. The curvature distribution of the feature envelope surface is calculated based on the discrete mesh nodes, and the spatial stiffness correction coefficient is obtained based on the deviation between the curvature distribution and the preset reference curvature. The correction module is used to iteratively correct the mechanical parameters of the support structure in the three-dimensional digital twin based on historical monitoring data, real-time data contained in the real-time mapped digital twin, and spatial stiffness correction coefficient, so as to obtain the corrected mechanical parameters. The processing module is used to perform deformation prediction calculations based on the corrected mechanical parameters to obtain the predicted deformation value of the support structure in the future period; when the predicted deformation value reaches the preset warning threshold, a warning signal is issued to guide construction adjustments.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement 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 stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images