Intelligent monitoring system and method for complex environment pipe jacking construction based on digital twinning

Abstract

The application provides a complex environment pipe jacking construction intelligent monitoring system and method based on digital twinning, the system comprises a perception layer, a transmission layer, a digital twinning core layer, an intelligent decision-making layer and a control execution layer; the perception layer comprehensively collects construction multi-parameters, which are transmitted to the digital twinning core layer through the transmission layer, the digital twinning core layer constructs a virtual mirror image of a physical construction scene and realizes real-time synchronization of the virtual mirror image and the physical construction scene, the intelligent decision-making layer completes data screening, posture stress linkage analysis and risk prediction grading, and the control execution layer realizes dynamic regulation and control of construction parameters. The application integrates five intelligent analysis logics of data screening, posture stress linkage analysis, rate comparison, structure health evaluation and risk grading prediction, can automatically identify various safety hazards such as stress concentration, deformation overrun and posture deviation, replaces manual research and judgment with intelligent analysis, greatly improves the risk response speed and judgment accuracy, and effectively enhances the risk pre-control ability of pipe jacking construction in complex environments.

Description

Technical Field

[0001] This invention belongs to the field of building construction technology, and in particular relates to an intelligent monitoring system and method for pipe jacking construction in complex environments based on digital twins. Background Technology

[0002] With the acceleration of urbanization, underground space development has become a key way to alleviate land resource shortages and improve urban traffic conditions. As a highly efficient and environmentally friendly trenchless construction technology, pipe jacking technology plays an increasingly important role in urban underground engineering construction due to its advantages such as minimal impact on surface traffic and low environmental disturbance. This technology uses jacking equipment to precisely push prefabricated pipe sections underground and utilizes a pressure balancing system within the soil chamber to effectively control groundwater and soil pressure, thereby significantly reducing the impact of construction on the surrounding environment.

[0003] When urban pipe jacking projects face extremely complex conditions such as deep overburden, large cross-sections, long distances, proximity to buildings and structures, and water-rich sand layers, the construction environment is harsh, the load coupling is intense, and the risks are highly concentrated. As the core construction equipment, the key load-bearing parts of the pipe jacking machine, such as the breast plate, stiffening plate, and main load-bearing frame, are structurally weak points. Under the long-term alternating jacking force, uneven soil pressure, high groundwater seepage and erosion, and the coupled effects of temperature stress and stratum disturbance, they are prone to hidden damage such as stress concentration, local plastic deformation, fatigue cracking, sealing failure, and structural instability. Such structural damage is highly concealed in its early stages and cannot be identified in advance by conventional manual inspections and monitoring. As construction continues, the damage gradually expands, which will directly lead to major safety accidents such as water and sand leakage in the excavation chamber, pipe section cracking and leakage, ground subsidence, tilting and cracking of surrounding buildings, and damage to underground pipelines, causing delays in the construction period, economic losses, environmental damage, and adverse social impacts.

[0004] In existing technologies, real-time monitoring of pipe jacking construction mainly relies on various distributed sensor networks. These sensors can collect multiple key parameters, including total jacking force and cutterhead torque. However, in water-rich sandy geological conditions, the weak load-bearing structure of the pipe jacking machine is more prone to fatigue damage and excessive deformation in large-section rectangular pipe jacking machines, leading to a chain of accidents such as water leakage, sand leakage, and excessive ground settlement. Existing monitoring systems do not cover the health status of the key load-bearing structures of the pipe jacking machine and lack digital twin-driven virtual-real synchronization, multi-physics simulation, and quantitative intelligent decision-making, making it difficult to meet the safety, efficiency, and intelligent management requirements of pipe jacking construction in complex environments. Therefore, there is an urgent need to develop an intelligent monitoring system and method for pipe jacking construction that can achieve real-time structural health monitoring, accurate digital twin mapping, intelligent analysis and early warning, and closed-loop automatic control. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent monitoring system and method for pipe jacking construction in complex environments based on digital twins. It can automatically identify various safety hazards such as stress concentration, excessive deformation, and attitude deviation, replacing manual judgment with intelligent analysis, significantly improving the speed of risk response and the accuracy of judgment, and effectively strengthening the risk prevention and control capabilities of pipe jacking construction in complex environments.

[0006] The present invention achieves the above-mentioned technical objectives through the following technical means.

[0007] A digital twin-based intelligent monitoring system for pipe jacking construction in complex environments includes a perception layer, a transmission layer, a digital twin core layer, an intelligent decision-making layer, and a control execution layer. The perception layer is used to collect multi-source data from the entire pipe jacking construction process in real time. The transmission layer is used to ensure stable data transmission. The digital twin core layer is used to construct a virtual mirror of the physical construction scene and achieve real-time synchronization between the virtual and real environments. The intelligent decision-making layer is used to complete data screening, attitude and force linkage analysis, and risk prediction and classification. The control execution layer is used to realize dynamic control of construction parameters.

[0008] Furthermore, the sensing layer includes: a breast plate stress gauge, a stiffening plate stress gauge, and a breast plate temperature acquisition device for monitoring the structural health of the pipe jacking machine; a jacking force sensor, an excavation chamber earth pressure sensor, a jacking speed sensor, an odometer, an earth chamber liquid level sensor, a screw conveyor torque sensor, and a grouting pressure sensor for monitoring the jacking parameters and excavation face status of the pipe jacking machine; a pipe section axial stress sensor, a pipe perimeter earth pressure sensor, a mud pressure sensor, and a pipe section joint strain sensor for monitoring the stress on the pipe section and the contact pressure with the surrounding rock; a dual-axis tilt sensor, a three-dimensional attitude sensor, a laser target positioning device, an intelligent total station, and a target prism for measuring the attitude data of the pipe jacking machine and pipe sections; and surface settlement monitoring points, building settlement monitoring points, building tilt monitoring devices, groundwater level gauges, pipeline settlement measuring points, and crack observation instruments for environmental monitoring.

[0009] Furthermore, the core layer of the digital twin includes a digital twin modeling module, a virtual-real synchronization module, a simulation analysis module, and a structural health visualization module. The digital twin modeling module, based on BIM model lightweighting technology, detailed geological survey data, and construction design data, constructs geometric, physical, and mechanical behavior models of the pipe jacking machine, pipe sections, and geological environment. The virtual-real synchronization module receives data preprocessed by the intelligent decision-making layer from the perception layer and calculates the spatiotemporal registration error. , ,when When the thickness is ≤10mm, the virtual model's state is updated in milliseconds, achieving a complete mapping between the physical construction site and the virtual model. , , These represent the spatial coordinate deviations in the X, Y, and Z directions, respectively. This refers to the time difference between the physical entity and the virtual model. The time-space conversion coefficient is used; the simulation analysis module performs construction process deduction, parameter sensitivity analysis and risk prediction simulation based on the virtual model; the structural health visualization module connects the stress, deformation and temperature data of the breast plate and stiffening plate to the platform to realize remote online viewing, historical data playback and three-dimensional visualization display.

[0010] Furthermore, the intelligent decision-making layer includes a data screening module, an attitude-force linkage analysis module, a rate comparison module, a structural health assessment module, and a risk classification prediction module. The data screening module is used to perform synchronization determination, integrity determination, anomaly removal, and time sequence reconstruction on the data transmitted by the perception layer to ensure the quality of construction data and structural monitoring data. Specifically, the synchronization determination is achieved by calculating the sensor response offset. Remove Data longer than 5ms; integrity is determined by calculating the validity of time-paired sequences. ,Will <95% of the time series segments were marked as missing data segments and interpolated to complete them; This indicates that the system meets the requirements after synchronous verification within a single monitoring period. The number of qualified data entries in the ≤5ms synchronization standard. This indicates the total number of raw data entries collected by the sensor within a single monitoring period.

[0011] Furthermore, the attitude force linkage analysis module is used to compare the synchronicity between attitude angle changes and propulsion force and soil pressure, utilizing a synchronicity coefficient. Determine the attitude offset type and output the attitude offset determination flag, where 0.3 ≤ When the value is ≤3, it is considered a synchronous change. <0.3 or A value greater than 3 is considered an independent attitude shift. Indicates the rate of change of attitude angle. This indicates the change in jacking force. Indicates a time interval;

[0012] The rate comparison module compares the direction and duration of the jacking force change rate with the attitude change rate for the marked attitude deviation period. When the two rates are continuously unbalanced and the change directions are opposite, the module outputs a trajectory deviation response signal, marks the start and end time of the imbalance interval and the imbalance level, and uploads it to the risk classification module.

[0013] Furthermore, the structural health assessment module analyzes the stress, deformation, and temperature monitoring data of the breast plate and stiffening plate in real time to calculate the structural health index. And classified by level: ≥80 is excellent, 60≤ <80 is considered good. A difference of <60 is used to achieve a quantitative assessment of the structural safety status. This represents the actual stress value of the currently monitored breast plate / stiffening plate. This represents the allowable stress value for the breast plate / stiffening plate. To monitor the deformation value of the breast plate in real time, The maximum allowable deformation value is designed for the breast plate. To monitor the temperature of the chest plate in real time, The limit for the permissible operating temperature range of the breast plate.

[0014] Furthermore, the risk grading prediction module is based on the attitude deviation characteristics output by the attitude force linkage analysis module, the rate imbalance output by the rate comparison module, and the structural health index output by the structural health assessment module. The simulation results from the simulation analysis module in the core layer of the digital twin, combined with preset standards, output a three-level risk level sequence (yellow, orange, and red) and predictive warning information:

[0015] (Green, safe): ≥80, normal posture, balanced speed → routine monitoring, no warning;

[0016] (Yellow, attention): 60≤ <80, slight attitude deviation, minor rate imbalance and duration <30s → platform pop-up window + yellow light warning;

[0017] (Orange, Warning): 60≤ <80, obvious attitude deviation, continuous rate imbalance and duration ≥30s → sound and light linkage orange warning, push notification for handling;

[0018] (Red, Danger): <60, structural stress / deformation exceeds limits, severe deviation, simulation prediction of sand inrush / collapse → highest level red audio-visual warning;

[0019] And form a risk level sequence for advancement. This enables dynamic risk classification and real-time early warning. , , They represent the 1st, 2nd, and 3rd times, respectively. The data collection time corresponding to this risk assessment. express The risk level determined at any given time.

[0020] Furthermore, the execution layer includes a controller, a main jack adjustment unit, a grouting adjustment unit, a deviation correction adjustment unit, and a structural safety protection unit. The controller is communicatively connected to the intelligent decision-making layer, receives decision commands, and controls the operation of the main jack adjustment unit, the grouting adjustment unit, and the deviation correction adjustment unit, thereby controlling the jacking force of the main jack, the opening of the grouting adjustment valve, and the action of the deviation correction jack to achieve dynamic adjustment of construction parameters. When the stress or deformation of the breast plate or stiffening plate exceeds the limit and triggers the highest level of risk, the structural safety protection unit intervenes urgently and adjusts the jacking parameters in a coordinated manner.

[0021] A method for intelligent monitoring of pipe jacking construction in complex environments, utilizing the aforementioned intelligent monitoring system based on digital twins, includes the following steps:

[0022] Step 1: Preliminary Modeling and System Deployment; The digital twin modeling module of the core layer of the digital twin is used to build a full-element digital twin model of the pipe jacking construction, including geometric model, physical model, and mechanical behavior model. The model is then validated, and the model coordinates are aligned with the actual measured coordinates on site to ensure that the virtual scene and the physical site spatial reference are consistent. Then, the installation, wiring, and debugging of all sensors in the perception layer and the transmission layer equipment are completed at the preset locations.

[0023] Step 2: Real-time acquisition of multi-source data across the entire area; upon initiating pipe jacking construction, the sensing layer continuously acquires data, and all acquired raw data is transmitted to the transmission layer in real time;

[0024] Step 3: Data transmission, screening, and timing reconstruction; the transmission layer stably transmits multi-source data to the digital twin core layer and intelligent decision-making layer; the data screening module of the intelligent decision-making layer performs data preprocessing: first, it calculates the sensor response offset. Determine data synchronization and remove [data items]. Data longer than 5ms; then calculate the effectiveness of time-paired sequences. To determine data integrity, <95% of the time series segments are marked as missing data segments; then outlier removal and missing value interpolation are performed, and the data is reordered in ascending order by timestamp to finally generate a standardized time series correction set. This provides effective data for subsequent analysis;

[0025] Step 4: Digital Twin Virtual-Real Synchronization and Simulation; The virtual-real synchronization module of the core layer of the digital twin receives the corrected data and calculates the spatiotemporal registration error. ,when When the thickness is ≤10mm, the virtual model is updated in milliseconds, achieving a complete mapping between the physical construction site and the virtual model. The structural health visualization module of the digital twin core layer synchronously displays the stress cloud map, deformation cloud map, and temperature distribution of the breast plate and stiffening plate, while simultaneously providing a three-dimensional display of the real-time posture and axis deviation of the pipe jacking machine and pipe sections. The simulation analysis module of the digital twin core layer combines real-time data to conduct sensitivity analysis of construction parameters and multi-physics field simulation, predicting changes in ground subsidence and groundwater seepage in the near future, predicting the stress evolution trend of the breast plate and stiffening plate of the pipe jacking machine, and analyzing the sensitivity of construction parameters, providing simulation basis for intelligent decision-making.

[0026] Step 5: Multi-dimensional intelligent analysis and structural health assessment;

[0027] The intelligent decision-making layer's attitude-force linkage analysis module compares the synchronicity of attitude angle changes with jacking force and soil pressure, uses the synchronicity coefficient to determine the attitude deviation type, distinguishes between normal synchronous deviation and independent abnormal deviation, and outputs an attitude deviation judgment label. The intelligent decision-making layer's rate comparison module compares the direction and continuity of the jacking force change rate with the attitude change rate for the marked attitude deviation time period. If the two rates are continuously unbalanced and the change directions are opposite, it outputs a trajectory deviation response signal, marks the start and end time of the imbalance interval and the imbalance level, and uploads it to the risk classification module. The intelligent decision-making layer's structural health assessment module calculates the structural health index and classifies it according to the level, thereby realizing a quantitative assessment of the structural safety status.

[0028] Step 6: Risk classification, early warning and command issuance; The risk classification prediction module integrates attitude deviation judgment results, rate imbalance characteristics, structural health level and digital twin simulation prediction results, and generates a dynamic risk level sequence by comparing with the preset risk level classification; According to different risk levels, corresponding audible and visual early warnings and platform pop-up warnings are triggered, and control commands are issued to the controller of the control execution layer.

[0029] Step 7: Parameter dynamic adjustment and closed-loop control; The controller receives risk commands and drives the main jack adjustment unit, grouting adjustment unit, correction adjustment unit, and structural safety protection unit to perform closed-loop control.

[0030] Furthermore, the specific scheme for closed-loop control in step 7 is as follows:

[0031] When the risk level is (Safety Status): The system maintains the original routine monitoring and construction parameters, carries out normal construction, and cyclically executes the process from step 2 to step 7 to achieve continuous intelligent monitoring of the entire construction process;

[0032] When the risk level is (Slight attitude deviation / slight structural deterioration): Activate the correction adjustment unit to fine-tune the correction jacks and slightly correct the attitude deviation of the pipe jacking machine; activate the grouting adjustment unit to simultaneously fine-tune the grouting pressure, stabilize the mud sleeve around the pipe, and reduce ground disturbance; no need to reduce speed or jacking force, routine on-site inspections are sufficient.

[0033] When the risk level is (Significant deviation / deterioration of structural condition): Activate the main jack adjustment unit to synchronously reduce the jacking force of the main jack, reduce the jacking speed, and jack up at a low speed and smoothly; activate the correction adjustment unit to continuously and dynamically correct the deviation and gradually correct the axis deviation; activate the grouting adjustment unit to dynamically adjust the pressure in real time according to the pipe circumferential pressure; increase the monitoring sampling frequency to continuously track changes in structural stress and attitude.

[0034] When the risk level is (Structural overrun / water / sand inrush / severe displacement): The structural safety protection unit is activated first for emergency intervention, significantly reducing the jacking force, and then the jacking operation is suspended; the entire system is monitored at high frequency, with a focus on monitoring the stress of the breast plate and stiffening plate and the liquid level in the soil chamber; the grouting adjustment unit is urgently pressurized to reinforce the strata around the pipe and seal the seepage channels; on-site personnel immediately conduct hazard investigation; jacking can only be resumed step by step after the hazards are completely eliminated and the structural health index returns to the safe level.

[0035] The present invention has the following beneficial effects:

[0036] The monitoring dimensions are more comprehensive. In addition to traditional monitoring of pipe jacking construction parameters, pipe section stress, attitude, and surrounding environment, specialized structural health monitoring of the pipe jacking machine's breast plate and stiffening plate has been added. This achieves full coverage from "construction status monitoring" to "equipment structural safety monitoring," filling the technological gap of lacking real-time monitoring of key load-bearing structures of pipe jacking machines in complex geological formations. The system relies on two types of algorithms—data synchronization judgment and data integrity judgment—to complete the preprocessing of raw data, effectively ensuring the validity and reliability of multi-source monitoring data. Simultaneously, this invention constructs a full-element digital twin model covering the pipe jacking machine, pipe sections, geological formation, breast plate, and stiffening plate, and uses a spatiotemporal registration error formula to conduct virtual-real synchronous verification, achieving millisecond-level dynamic mapping between the physical construction site and the virtual model. The platform can intuitively display breast plate stress, deformation, and temperature cloud maps. Compared to traditional two-dimensional data monitoring methods, the visualization effect is stronger and the information dimensions are richer, facilitating managers to have a comprehensive grasp of construction dynamics and equipment structural safety. In addition, the system integrates five layers of intelligent analysis logic, including data screening and calibration, attitude and stress linkage analysis, rate comparison, structural health assessment, and risk classification prediction. It can automatically identify various safety hazards such as stress concentration, deformation exceeding limits, and attitude deviation. By replacing manual judgment with intelligent analysis, it significantly improves the speed of risk response and the accuracy of judgment, and effectively strengthens the risk prevention and control capabilities of pipe jacking construction in complex environments. Attached Figure Description

[0037] Figure 1 This is an architecture diagram of the intelligent monitoring system for pipe jacking construction in complex environments based on digital twins, as described in this invention.

[0038] Figure 2 This is a flowchart of the intelligent decision-making layer data processing described in this invention;

[0039] Figure 3 This is a flowchart of the intelligent monitoring method for pipe jacking construction in complex environments as described in this invention. Detailed Implementation

[0040] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, but the scope of protection of the present invention is not limited thereto.

[0041] like Figure 1 As shown, the intelligent monitoring system for pipe jacking construction in complex environments based on digital twins according to the present invention includes a perception layer, a transmission layer, a digital twin core layer, an intelligent decision-making layer, and a control execution layer.

[0042] The perception layer is used to achieve comprehensive acquisition of multiple construction parameters; the transmission layer is used to ensure stable data transmission; the digital twin core layer is used to construct a virtual mirror of the physical construction scene and achieve real-time synchronization between the virtual and real; the intelligent decision-making layer is used to complete data screening, attitude and force linkage analysis, and risk prediction and classification; and the control execution layer is used to achieve dynamic adjustment of construction parameters.

[0043] The sensing layer comprises a first sensing module, a second sensing module, a third sensing module, a fourth sensing module, and a fifth sensing module. The first sensing module is used for structural health monitoring of the pipe jacking machine; the second sensing module is used to monitor the real-time jacking force, excavation chamber pressure, jacking speed, and jacking distance of the pipe jacking machine; the third sensing module is used to monitor the axial end force, axial internal force, circumferential pressure distribution, and circumferential mud pressure of the pipe section; the fourth sensing module is used to monitor the attitude data of the pipe jacking machine and the pipe section, including tilt angle, rotation angle, three-dimensional coordinates, and deviation from the design axis; the fifth sensing module is used to monitor the surrounding environment, including ground settlement, building settlement, and tilt.

[0044] The first sensing module includes five breast plate stress gauges and three stiffening plate stress gauges installed at key locations where the breast plate deformation is greatest and stress concentration is concentrated. Temperature acquisition equipment is also installed to simultaneously monitor the operating temperature of the breast plate. The second sensing module includes a jacking force sensor, an excavation chamber earth pressure sensor, a jacking speed sensor, an odometer, an earth chamber liquid level sensor, a screw conveyor torque sensor, and a grouting pressure sensor, enabling comprehensive acquisition of jacking parameters and excavation face conditions. The third sensing module includes a pipe section axial stress sensor, a pipe perimeter earth pressure sensor, a mud pressure sensor, and a pipe section joint strain sensor, enabling real-time monitoring of the stress on the pipe section and the contact pressure with the surrounding rock. The fourth sensing module includes a dual-axis tilt sensor, a three-dimensional attitude sensor, a laser target positioning device, an intelligent total station, and a target prism, enabling high-precision measurement of the pipe jacking machine's attitude, rotation angle, horizontal deviation, and elevation deviation. The fifth sensing module includes surface settlement monitoring points, building settlement monitoring points, building tilt monitoring devices, groundwater level gauges, pipeline settlement measuring points, and crack observation instruments, covering comprehensive environmental monitoring of surrounding structures, roads, underground pipelines, and surface deformation around the foundation pit.

[0045] The transmission layer includes a wireless transmission module, a signal amplifier, and a signal receiver. The wireless transmission module supports WiFi, Ethernet, 4G / 5G, and GPRS communication, and is used to establish communication connections between the perception layer, the digital twin core layer, the intelligent decision-making layer, and the control execution layer. The signal amplifier is installed in the middle of the main jack with its center aligned with the center of the opening, and the signal receiver is located on the ground to ensure the stability of data transmission in complex environments. Structural health monitoring data such as those from the breast plate and stiffening plates are transmitted via an RS-485 bus.

[0046] The core layer of the digital twin includes a digital twin modeling module, a virtual-real synchronization module, a simulation analysis module, and a structural health visualization module. The digital twin modeling module, based on BIM model lightweighting technology, detailed geological survey data, and construction design data, constructs geometric, physical, and behavioral models of the pipe jacking machine, pipe sections, and geological environment. The virtual-real synchronization module receives data collected from the perception layer and updates the virtual model's status in real time, achieving millisecond-level mapping between the physical construction scene and the virtual model. The simulation analysis module performs construction process simulations, parameter sensitivity analysis, and risk prediction simulations based on the virtual model. The structural health visualization module integrates stress, deformation, and temperature data of the breast plate and stiffening plates into the platform, enabling remote online viewing, historical data playback, and 3D visualization.

[0047] The virtual-real synchronization module uses spatiotemporal registration error... The virtual model status is updated in real time, using the following formula:

[0048]

[0049] in, , , These represent the spatial coordinate deviations in the X, Y, and Z directions, in mm. The time difference between the physical entity and the virtual model, in milliseconds; This is the spacetime transformation coefficient, with a value of 0.2 mm / ms. When... When the difference is ≤10mm, the virtual and real phases are considered to be effective, ensuring that the construction posture, jacking force, structural stress, and ground deformation are accurately mapped in real time.

[0050] like Figure 2 As shown, the intelligent decision-making layer includes a data screening module, an attitude-stress linkage analysis module, a rate comparison module, a structural health assessment module, and a risk classification prediction module. The data screening module performs integrity checks, outlier removal, and time-series reconstruction on synchronized data to ensure the quality of construction data and structural monitoring data. The attitude-stress linkage analysis module compares the synchronicity of attitude angle changes with propulsion force and soil pressure to identify independent attitude deviations. The rate comparison module analyzes the rate balance and continuity of propulsion force and attitude changes. The structural health assessment module analyzes the stress, deformation, and temperature data of the breast plate and stiffening plate in real time to determine whether the structure is in a safe state and identify anomalies such as stress concentration and excessive deformation. The risk classification prediction module, based on deviation characteristics, structural health status, and digital twin simulation results, outputs a three-level risk level sequence (yellow, orange, and red) and predictive warning information, combined with preset standards.

[0051] The data screening and calibration module first synchronously calibrates the multi-source sensing data, using the sensor response offset formula. Perform synchronization judgment when A time interval of ≤5ms is considered a valid synchronous acquisition; then, the validity of the time-paired sequence is determined using the formula... Perform integrity determination when A time series is considered acceptable when the accuracy is ≥95%; subsequently, outlier removal, missing value completion, and time series reconstruction are performed to generate a time series integrity correction set. The single data structure in the calibration set includes: time, jacking force, earth pressure, attitude angle, and breast plate stress, thus providing high-quality time series data for subsequent analysis; This indicates that the system meets the requirements after synchronous verification within a single monitoring period. The number of qualified data entries in the ≤5ms synchronization standard. This indicates the total number of raw data entries collected by the sensor within a single monitoring period. , , , These represent the corrected first, second, and third articles, respectively. One valid monitoring sample.

[0052] The attitude-force linkage analysis module compares the synchronicity between attitude angle changes and jacking force and soil pressure, and uses the attitude-force synchronicity coefficient to determine the synchronicity. Make a judgment when 0.3 ≤ When the value is ≤3, it is considered a synchronous change. <0.3 or If the value is greater than 3, it is determined to be an independent attitude deviation, and an attitude deviation determination flag is output. This represents the rate of change of attitude angle, expressed in ° / ms. This represents the change in jacking force, in kN. This indicates a time interval, measured in milliseconds (ms).

[0053] The rate comparison module analyzes the balance and continuity of the jacking force change rate and the attitude change rate, determines the trend imbalance interval, and outputs the trajectory deviation response signal.

[0054] The structural health assessment module analyzes the stress, deformation, and temperature monitoring data of the breast plate and stiffening plate in real time and calculates the structural health index. , And classified by level: ≥80 is excellent, 60≤ <80 is considered good. A difference of <60 is used to quantitatively assess the structural safety status. Where: The values ​​represent the actual stress values ​​of the breast plate / stiffening plate currently being monitored, in MPa. The allowable stress value (design maximum allowable stress) for the breast plate / stiffening plate, in MPa. To monitor the deformation value of the breast plate in real time, The maximum allowable deformation value is designed for the breast plate. To monitor the temperature of the chest plate in real time, The limit for the permissible operating temperature range of the breast plate.

[0055] The risk grading prediction module generates a risk level code set based on offset features, structural health status, and digital twin simulation results, according to preset standards. The risk levels are divided into four levels:

[0056] (Green, safe): ≥80, normal posture, balanced speed → routine monitoring, no warning;

[0057] (Yellow, attention): 60≤ <80, slight attitude deviation, minor rate imbalance and duration <30s → platform pop-up window + yellow light warning;

[0058] (Orange, Warning): 60≤ <80, obvious attitude deviation, continuous rate imbalance and duration ≥30s → sound and light linkage orange warning, push notification for handling;

[0059] (Red, Danger): <60, structural stress / deformation exceeds limits, severe deviation, simulation prediction of sand inrush / collapse → highest level red audio-visual warning;

[0060] And form a risk level sequence for advancement. This enables dynamic risk classification and real-time early warning. , , They represent the 1st, 2nd, and 3rd times, respectively. The data collection time corresponding to this risk assessment. express The risk level determined at any given time.

[0061] The execution layer includes a controller, a main jack adjustment unit, a grouting adjustment unit, a correction adjustment unit, and a structural safety protection unit. The controller communicates with the intelligent decision-making layer, receives decision commands, and controls the jacking force of the main jack, the opening of the grouting adjustment valve, and the action of the correction jack, realizing dynamic adjustment of construction parameters. When the stress or deformation of the breast plate or stiffening plate exceeds the limit and triggers an early warning, the structural safety protection unit adjusts the jacking parameters, reduces the jacking speed, and reduces the jacking force in conjunction, effectively reducing the risk of structural damage caused by local overload and ensuring the safety of the breast plate structure of the pipe jacking machine.

[0062] Reference Figure 3 As shown, the intelligent monitoring method for complex environment pipe jacking construction using the intelligent monitoring system for complex environment pipe jacking construction based on digital twins described in this invention includes the following processes:

[0063] Step 1: Preliminary modeling and system deployment.

[0064] Step 1.1: Combining the engineering survey report, pipe jacking machine design drawings, pipe section parameters, geological distribution, surrounding buildings and pipeline data, BIM lightweight technology is adopted to complete the construction of a full-element digital twin model through the digital twin modeling module. The model includes three types: geometric model, physical model, and mechanical behavior model.

[0065] The geometric model recreates the three-dimensional geometry of the pipe jacking machine (breast plate, stiffening plate, soil chamber, screw conveyor), prefabricated pipe sections, launching shaft, receiving shaft, strata, surface buildings, and underground pipelines at a 1:1 scale. The physical model assigns physical parameters such as stratum unit weight, internal friction angle, permeability coefficient, steel elastic modulus, and pipe section concrete strength. The mechanical behavior model embeds the mechanical coupling relationship between jacking force, earth pressure, structural stress, and stratum disturbance, providing support for subsequent simulation and deduction.

[0066] Step 1.2: Perform model verification, align the model coordinates with the actual measured coordinates on site, and ensure that the virtual scene and the physical site spatial reference are consistent.

[0067] Step 1.3: System Deployment; Complete the installation, wiring and debugging of all sensors in the sensing layer and transmission layer equipment at the preset locations such as the pipe jacking machine, pipe sections, ground surface, and buildings, and complete the communication link connectivity test to ensure normal signal transmission of RS-485 bus and wireless communication module.

[0068] Step 2: Real-time collection of multi-source data across the entire domain.

[0069] Once pipe jacking construction commences, the five modules of the sensing layer simultaneously conduct 24 / 7 continuous data acquisition: The first sensing module collects stress, deformation, and temperature data for the breast plate and stiffening plate; the second sensing module collects construction parameters such as jacking force, excavation chamber pressure, jacking speed, mileage, soil chamber liquid level, torque, and grouting pressure; the third sensing module collects data on axial stress of the pipe section, soil pressure around the pipe, mud pressure, and joint strain; the fourth sensing module collects attitude data such as the inclination angle, rotation angle, three-dimensional coordinates, and axis deviation of the pipe jacking machine; and the fifth sensing module collects data on surface, building, and pipeline settlement, inclination, groundwater level, and cracks. All raw data is transmitted to the transmission layer in real time.

[0070] Step 3: Data transmission, screening, and timing reconstruction.

[0071] The transmission layer stably transmits multi-source data to the digital twin core layer and intelligent decision-making layer via a wireless module, signal amplifier, and RS-485 bus; the data screening module of the intelligent decision-making layer performs data preprocessing: firstly, based on... To determine data synchronization, discard data with a synchronization deviation greater than 5ms; then apply the timing validity formula. Determine data integrity. <95% of the time series segments are marked as missing data segments; then outlier removal and missing value interpolation are performed, and the data is reordered in ascending order by timestamp to finally generate a standardized time series correction set. This provides useful data for subsequent analysis.

[0072] Step 4: Digital twin virtual-real synchronization and simulation.

[0073] The virtual-real synchronization module of the digital twin core layer receives the corrected data and calculates the spatiotemporal registration error. , ,when When the thickness is ≤10mm, the virtual model is updated in milliseconds, achieving a complete mapping between the physical construction site and the virtual model;

[0074] The structural health visualization module simultaneously displays stress cloud maps, deformation cloud maps, and temperature distribution of the breast plate and stiffening plate, using different colors to distinguish areas of high and low stress; it also provides a 3D display of the real-time posture and axis deviation of the pipe jacking machine and pipe sections; it supports historical data playback and fixed-point query, and management personnel can view it remotely.

[0075] The simulation analysis module combines real-time data to conduct sensitivity analysis of construction parameters and multi-physics field simulation, predict changes in ground subsidence and groundwater seepage in the near future, predict the stress evolution trend of the pipe jacking machine's breast plate and stiffening plate, and analyze the sensitivity of parameters such as jacking speed, jacking force, and grouting pressure, providing simulation basis for intelligent decision-making.

[0076] Step 5: Multi-dimensional intelligent analysis and structural health assessment.

[0077] The intelligent decision-making layer sequentially performs multi-level analyses (including attitude stress analysis, velocity comparison, and structural health assessment):

[0078] The attitude-force linkage analysis module compares the synchronicity between attitude angle changes and jacking force and soil pressure, and uses the synchronicity coefficient. ( Determine the type of attitude offset and distinguish between normal synchronization offset (0.3≤ ≤3) and independent anomalous offsets ( <0.3 or >3), and output the attitude offset determination flag;

[0079] The rate comparison module compares the direction and duration of the jacking force change rate with the attitude change rate for the marked attitude deviation period. If the two rates are continuously unbalanced and the change directions are opposite, the module outputs a trajectory deviation response signal, marks the start and end time of the imbalance interval and the imbalance level, and uploads it to the risk classification module.

[0080] The structural health assessment module calculates the structural health index. ( ), and classified by level: ≥80 is excellent, 60≤ <80 is considered good. A value less than 60 indicates a difference, thus enabling a quantitative assessment of the structural safety status.

[0081] Step 6: Risk classification, early warning and instruction issuance.

[0082] The risk grading and prediction module integrates attitude deviation judgment results, velocity imbalance characteristics, structural health level, and digital twin simulation prediction results, and generates a dynamic risk level sequence by comparing it with the preset risk level classification. ( Based on different risk levels, corresponding audible and visual warnings and platform pop-up warnings are triggered, and control commands are sent to the controller in the control execution layer.

[0083] Step 7: Parameter dynamic adjustment and closed-loop control.

[0084] The controller receives risk commands and drives the four main units—main jack, grouting, deviation correction, and structural protection—to execute closed-loop control, with four levels of response plans, as detailed below:

[0085] When the risk level is (Green and safe status): The system maintains the original routine monitoring and construction parameters, carries out normal construction, and cyclically executes the process from step 2 to step 7 to achieve continuous intelligent monitoring of the entire construction process;

[0086] When the risk level is (Yellow indicates risk, slight attitude deviation / slight structural deterioration): Activate the correction adjustment unit to fine-tune the correction jacks and slightly correct the attitude deviation of the pipe jacking machine; activate the grouting adjustment unit to simultaneously fine-tune the grouting pressure by ±5%, stabilize the mud sleeve around the pipe, and reduce ground disturbance; no need to reduce speed or jacking force, routine on-site inspections are sufficient.

[0087] When the risk level is (Orange warning of risk, significant deviation / deterioration of structural condition): Activate the main jack adjustment unit, simultaneously reduce the jacking force of the main jack by 15%~20%, reduce the jacking speed to 3~5mm / min, and jack up at a low speed and smoothly; activate the deviation correction adjustment unit to continuously and dynamically correct the deviation and gradually correct the axial deviation; activate the grouting adjustment unit to dynamically adjust the pressure in real time according to the pressure around the pipe to prevent the formation from losing water and collapsing; increase the monitoring sampling frequency to continuously track changes in structural stress and attitude;

[0088] When the risk level is (Red danger risk, structural over-limit / water and sand inrush / severe displacement): The structural safety protection unit is activated first for emergency intervention, significantly reducing the jacking force, and then the jacking operation is suspended; the entire system is monitored at high frequency (50Hz), with a focus on monitoring the stress of the breast plate and stiffening plate and the liquid level in the soil chamber; the grouting adjustment unit is urgently pressurized to reinforce the strata around the pipe and seal the seepage channels; on-site personnel immediately conduct a hidden danger investigation: check the breast plate welds, seals, and soil chamber for water and sand leakage; jacking can only be resumed step by step after the hidden dangers are completely eliminated and the structural health index returns to a good level or above.

[0089] The embodiments described above are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Any obvious improvements, substitutions or modifications that can be made by those skilled in the art without departing from the essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A digital twin-based intelligent monitoring system for pipe jacking construction in complex environments, characterized in that, It includes a perception layer, a transmission layer, a digital twin core layer, an intelligent decision-making layer, and a control execution layer. The perception layer is used to collect multi-source data from the entire field during the pipe jacking construction process in real time. The transmission layer is used to ensure stable data transmission. The digital twin core layer is used to construct a virtual mirror of the physical construction scene and achieve real-time synchronization between the virtual and real. The intelligent decision-making layer is used to complete data screening, attitude and force linkage analysis, and risk prediction and classification. The control execution layer is used to realize dynamic control of construction parameters.

2. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins as described in claim 1, characterized in that, The sensing layer includes: breast plate stress gauges, stiffening plate stress gauges, and breast plate temperature acquisition devices for monitoring the structural health of the pipe jacking machine; jacking force sensors, excavation chamber earth pressure sensors, jacking speed sensors, odometers, soil chamber liquid level sensors, screw conveyor torque sensors, and grouting pressure sensors for monitoring the jacking parameters and excavation face status of the pipe jacking machine; pipe section axial stress sensors, pipe perimeter earth pressure sensors, mud pressure sensors, and pipe section joint strain sensors for monitoring the stress on the pipe section and the contact pressure with the surrounding rock; dual-axis tilt sensors, three-dimensional attitude sensors, laser target positioning devices, intelligent total stations, and target prisms for measuring the attitude data of the pipe jacking machine and pipe sections; and surface settlement monitoring points, building settlement monitoring points, building tilt monitoring devices, groundwater level gauges, pipeline settlement measuring points, and crack observation instruments for environmental monitoring.

3. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins as described in claim 1, characterized in that, The core layer of the digital twin includes a digital twin modeling module, a virtual-real synchronization module, a simulation analysis module, and a structural health visualization module. The digital twin modeling module is used to construct geometric, physical, and mechanical behavior models of the pipe jacking machine, pipe sections, and geological environment. The virtual-real synchronization module receives data collected by the perception layer after preprocessing by the intelligent decision-making layer and calculates the spatiotemporal registration error. , When the thickness is ≤10mm, the virtual model status is updated to achieve a complete mapping between the physical construction site and the virtual model. , , These represent the spatial coordinate deviations in the X, Y, and Z directions, respectively. This refers to the time difference between the physical entity and the virtual model. This is the spatiotemporal conversion coefficient; the simulation analysis module performs construction process deduction, parameter sensitivity analysis, and risk prediction simulation based on the virtual model; The structural health visualization module connects the stress, deformation, and temperature data of the breast plate and stiffening plate to the platform, enabling remote online viewing, historical data playback, and 3D visualization.

4. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins as described in claim 1, characterized in that, The intelligent decision-making layer includes a data screening module, an attitude and force linkage analysis module, a rate comparison module, a structural health assessment module, and a risk classification and prediction module. The data screening module is used to perform synchronization determination, integrity determination, anomaly removal, and time sequence reconstruction on the data transmitted by the perception layer. Synchronization determination is achieved by calculating the sensor response offset. Remove Data longer than 5ms; integrity is determined by calculating the validity of time-paired sequences. ,Will <95% of the time series segments were marked as missing data segments and interpolated to complete them; This indicates that the system meets the requirements after synchronous verification within a single monitoring period. The number of qualified data entries in the ≤5ms synchronization standard. This indicates the total number of raw data entries collected by the sensor within a single monitoring period.

5. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins according to claim 4, characterized in that, The attitude force linkage analysis module is used to compare the synchronicity between attitude angle changes and thrust and soil pressure, using a synchronicity coefficient. Determine the type of attitude deviation and output the attitude deviation determination flag; 0.3≤ When the value is ≤3, it is considered a synchronous change. <0.3 or A value greater than 3 is considered an independent attitude shift. Indicates the rate of change of attitude angle. This indicates the change in jacking force. Indicates a time interval; The rate comparison module compares the direction and duration of the jacking force change rate with the attitude change rate for the marked attitude deviation period. When the two rates are continuously unbalanced and the change directions are opposite, the module outputs a trajectory deviation response signal, marks the start and end time of the imbalance interval and the imbalance level, and uploads it to the risk classification module.

6. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins according to claim 4, characterized in that, The structural health assessment module analyzes the stress, deformation, and temperature monitoring data of the breast plate and stiffening plate in real time and calculates the structural health index. And classified by level: ≥80 is excellent, 60≤ <80 is considered good. A score <60 indicates a difference, enabling a quantitative assessment of the structural safety status; This represents the actual stress value of the currently monitored breast plate / stiffening plate. This represents the allowable stress value for the breast plate / stiffening plate. To monitor the deformation value of the breast plate in real time, The maximum allowable deformation value is designed for the breast plate. To monitor the temperature of the chest plate in real time, The limit for the permissible operating temperature range of the breast plate.

7. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins according to claim 4, characterized in that, The risk grading prediction module is based on the attitude deviation characteristics output by the attitude force linkage analysis module, the rate imbalance output by the rate comparison module, and the structural health index output by the structural health assessment module. The simulation results from the simulation analysis module in the core layer of the digital twin, combined with preset standards, output a risk level sequence and predictive warning information: (Safety): ≥80, normal posture, balanced speed → routine monitoring, no warning; (Attention): 60≤ <80, slight attitude deviation, minor rate imbalance and duration <30s → platform pop-up window + yellow light warning; (Warning): 60≤ <80, obvious attitude deviation, continuous rate imbalance and duration ≥30s → sound and light linkage orange warning, push notification for handling; :(Danger): <60, structural stress / deformation exceeds limits, severe deviation, simulation prediction of sand inrush / collapse → highest level red audio-visual warning.

8. The intelligent monitoring system for pipe jacking construction in complex environments based on digital twins according to claim 1, characterized in that, The execution layer includes a controller, a main jack adjustment unit, a grouting adjustment unit, a deviation correction adjustment unit, and a structural safety protection unit. The controller is connected to the intelligent decision-making layer, receives decision commands, and controls the main jack adjustment unit, grouting adjustment unit, and deviation correction adjustment unit to achieve dynamic adjustment of construction parameters. When the stress or deformation of the breast plate or stiffening plate exceeds the limit and triggers the highest level of risk, the structural safety protection unit intervenes urgently and adjusts the jacking parameters in a coordinated manner.

9. A method for intelligent monitoring of pipe jacking construction in complex environments using the intelligent monitoring system for complex environment pipe jacking construction based on digital twins as described in claim 1, characterized in that, The process includes the following: Step 1: Preliminary modeling and system deployment; through the digital twin modeling module of the core layer of the digital twin, complete the construction and verification of the geometric model, physical model and mechanical behavior model of the pipe jacking machine, pipe section and geological environment; then complete the installation, wiring and debugging of all sensors of the sensing layer and transmission layer equipment at the preset points; Step 2: Real-time acquisition of multi-source data across the entire domain; upon initiating pipe jacking construction, the sensing layer continuously acquires data, and all acquired raw data is transmitted to the transmission layer in real time; Step 3: Data transmission, screening, and timing reconstruction; the transmission layer stably transmits multi-source data to the digital twin core layer and intelligent decision-making layer; the data screening module of the intelligent decision-making layer first calculates the sensor response offset. Determine data synchronization and eliminate [data items]. Data longer than 5ms was then used to calculate the validity of time-paired sequences. To determine data integrity, <95% of the time series segments were marked as missing data segments. Then, outlier deletion and missing value interpolation were performed. The data were then rearranged in ascending order by timestamp to generate a standardized time series correction set, providing effective data for subsequent analysis. Step 4: Digital twin virtual-real synchronization and simulation; The virtual-real synchronization module of the digital twin core layer receives the corrected data and calculates the spatiotemporal registration error. ,when When the thickness is ≤10mm, the virtual model status is updated to achieve a complete mapping between the physical construction site and the virtual model. The structural health visualization module of the digital twin core layer synchronously displays the stress cloud map, deformation cloud map, and temperature distribution of the breast plate and stiffening plate, while simultaneously providing a three-dimensional display of the real-time posture and axis deviation of the pipe jacking machine and pipe sections. The simulation analysis module of the digital twin core layer combines real-time data to conduct sensitivity analysis of construction parameters and multi-physics field simulation, predict changes in ground subsidence and groundwater seepage in the near future, predict the stress evolution trend of the breast plate and stiffening plate of the pipe jacking machine, analyze the sensitivity of construction parameters, and provide simulation basis for intelligent decision-making. Step 5: Multi-dimensional intelligent analysis and structural health assessment; The attitude force linkage analysis module of the intelligent decision layer compares the synchronicity of attitude angle changes with jacking force and soil pressure, uses the synchronicity coefficient to determine the attitude deviation type, distinguishes between normal synchronous deviation and independent abnormal deviation, and outputs attitude deviation judgment label. The rate comparison module of the intelligent decision layer compares the direction and duration of the jacking force change rate with the attitude change rate for the marked attitude deviation period. If the two rates are continuously unbalanced and the change directions are opposite, the module outputs a trajectory deviation response signal, marks the start and end time of the imbalance interval and the imbalance level, and uploads it to the risk classification module. The structural health assessment module of the intelligent decision layer calculates the structural health index and classifies it according to the level, thereby realizing a quantitative assessment of the structural safety status. Step 6: Risk classification, early warning and instruction issuance; The risk classification prediction module integrates attitude deviation judgment results, rate imbalance characteristics, structural health level and digital twin simulation prediction results, and generates a dynamic risk level sequence by comparing with the preset risk level classification. Based on different risk levels, corresponding audible and visual warnings and platform pop-up warnings are triggered, and control commands are sent to the controller in the control execution layer. Step 7: Parameter dynamic adjustment and closed-loop control; The controller receives risk commands and drives the main jack adjustment unit, grouting adjustment unit, correction adjustment unit, and structural safety protection unit to perform closed-loop control.

10. The intelligent monitoring method for pipe jacking construction in complex environments according to claim 9, characterized in that, The specific scheme for closed-loop control in step 7 is as follows: When the risk level is (Safety Status): The system maintains the original monitoring and construction parameters and executes steps 2 to 7 in a loop to achieve continuous intelligent monitoring of the entire construction process; When the risk level is (Slight attitude deviation / Slight structural deterioration): Activate the correction adjustment unit to fine-tune the correction jacks and slightly correct the attitude deviation of the pipe jacking machine; Start the grouting adjustment unit and simultaneously fine-tune the grouting pressure to stabilize the mud sleeve around the pipe and reduce formation disturbance; there is no need to reduce speed or jacking force, and routine on-site inspections are sufficient. When the risk level is (Significant deviation / deterioration of structural condition): Activate the main jack adjustment unit, synchronously reduce the jacking force of the main jack, reduce the jacking speed, and jack up at a low speed and smoothly; The correction adjustment unit is activated to continuously and dynamically correct the axis deviation. Start the grouting adjustment unit and dynamically adjust the pressure in real time according to the pipe circumferential pressure; increase the monitoring sampling frequency and continuously track structural stress and attitude changes; When the risk level is (Structural over-limit / water / sand inrush / severe displacement): The structural safety protection unit is activated first, and emergency intervention is carried out to significantly reduce the jacking force, and then the jacking operation is suspended; The entire system is monitored at high frequency, with a focus on the stress of the breast plate and the reinforced plate, as well as the liquid level in the soil chamber. The grouting adjustment unit is pressurized urgently to reinforce the strata around the pipe and seal the seepage channels. On-site personnel immediately conduct a hazard investigation, and the jacking can only be resumed step by step after the hazards have been completely eliminated.

Images