Full-cycle deformation monitoring method and system for steel roof based on multi-source sensor fusion

Abstract

The present invention provides a full-cycle deformation monitoring method and system for a steel roof based on multi-source sensor fusion, belonging to the technical field of monitoring structural deformation or stress, including: arranging a multi-source sensor network for synchronously collecting data; determining the stress distribution data of the sliding blocks during the sliding process of the steel structure, and combining the dynamic vibration data to determine the local buckling risk area; dividing the steel roof into several asymmetric sliding blocks, and generating a sliding path correction instruction in combination with the geometric state data of the sliding track; compensating the geometric deviation of the sliding track in real time, and at the same time combining the initial clearance parameters of the embedded support structure to predict the critical deformation threshold in the subsequent construction stage; after the sliding is completed, tracking the steel column falling trajectory data in real time and dynamically adjusting the unloading rate in combination with the critical deformation threshold, and calibrating the unloading path by using the compensated geometric state data of the sliding track until the deformation error converges within the preset tolerance range. It effectively improves the construction accuracy and safety.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring structural deformation or stress, and in particular to a method and system for monitoring full-cycle deformation of a steel roof based on multi-source sensor fusion. Background Art

[0002] As the construction industry continues to increase its requirements for construction accuracy and safety, deformation monitoring and control during the construction of steel structures has become an important technical issue in construction management. In particular, how to monitor and correct deformation in real time during the sliding construction of steel roof structures to ensure construction safety and accuracy has become a key technical challenge in the industry.

[0003] Existing methods for monitoring deformation of steel structures mainly rely on a single type of sensor, such as strain gauges or laser measuring instruments, to collect the strain, geometric state or vibration data of the structure in real time. However, these methods usually only focus on the monitoring of a single factor and lack the fusion and comprehensive analysis of multi-source data, resulting in limited ability to predict and adjust the deformation of the structure during the construction process. In the sliding construction process, the existing technology lacks real-time and dynamic adjustment means for the correction of the geometric deviation of the sliding track and the deformation control in the subsequent construction stage. In addition, the judgment and correction of the buckling risk area during the construction process are not accurate enough, which may lead to safety hazards or delays in construction progress. The existing technical solutions lack dynamic correction of the geometric deviation of the sliding track and accurate identification of the local buckling risk area, resulting in the inability to effectively predict and correct structural deformation during the construction process. Affecting construction accuracy and safety. Summary of the invention

[0004] The present invention provides a full-cycle deformation monitoring method and system for steel roofs based on multi-source sensor fusion, which provides a comprehensive solution for full-cycle deformation monitoring of steel roofs through multi-source sensor fusion. It can collect and analyze data such as the geometric state, strain, vibration, etc. of the sliding track in real time, and accurately monitor the deformation of the steel structure. Compared with the existing technology, it can compensate for track deviations in real time, identify buckling risk areas, and dynamically control the construction process according to the predicted critical deformation threshold to ensure that the deformation error is within the tolerance range, effectively improve the construction accuracy and safety, reduce the impact of deformation on structural stability, and optimize the adjustment strategy during the construction process.

[0005] The present invention provides a full-cycle deformation monitoring method for a steel roof based on multi-source sensor fusion, comprising:

[0006] Step 1: Deploy a multi-source sensor network at the preset positions of the upper and lower chords of the steel truss, the sliding track beams and columns, and the hydraulic booster to synchronously collect the geometric state data of the sliding track, the strain data of the steel structure, the dynamic vibration data, and the initial gap parameters of the embedded support structure;

[0007] Step 2: Determine the stress distribution data of the slip blocks during the slip of the steel structure based on the strain data and the geometric state data of the slip tracks, and determine the local buckling risk area in combination with the dynamic vibration data;

[0008] Step 3: Divide the steel roof into several asymmetric slip blocks based on the local buckling risk area, and generate a slip path correction instruction in combination with the geometric state data of the slip tracks;

[0009] Step 4: During the slip process, compensate for the geometric deviation of the slip tracks in real time based on the slip path correction instruction, and at the same time, in combination with the initial clearance parameters of the embedded support structure, predict the critical deformation threshold in the subsequent construction stage;

[0010] Step 5: After the slip is completed, track the steel column falling trajectory data in real time and dynamically adjust the unloading rate in combination with the critical deformation threshold, and use the compensated geometric state data of the slip tracks to calibrate the unloading path until the deformation error converges within the preset tolerance range.

[0011] Preferably, the multi-source sensor network includes:

[0012] A distributed fiber Bragg grating strain sensor array is arranged on the upper and lower chords of the steel truss for collecting strain data;

[0013] A laser scanning displacement sensor is arranged at the key nodes of the slip tracks for obtaining the geometric state data of the slip tracks, where the geometric state data includes the lateral offset of the tracks, the longitudinal flatness, and the relative spacing between the tracks;

[0014] A three-axis vibration acceleration sensor is arranged at the connection between the steel truss and the slip tracks for monitoring the dynamic vibration data of each key node;

[0015] An embedded gap displacement sensor is arranged at the contact interface between the bottom of the steel column and the embedded support structure for collecting the initial clearance parameters.

[0016] Preferably, determining the stress distribution data of the slip blocks during the slip of the steel structure based on the strain data and the geometric state data of the slip tracks, and determining the local buckling risk area in combination with the dynamic vibration data, includes:

[0017] Collect the strain data of each preset monitoring point of the steel truss through the distributed fiber Bragg grating strain sensor array, and then determine the initial stress value of each preset monitoring point to generate the initial stress distribution of the steel structure;

[0018] Correct the additional stress components caused by track unevenness and offset in the initial stress distribution through the geometric state data of the slip tracks to generate the corrected stress distribution data of the slip blocks;

[0019] Divide the steel truss into several monitoring areas according to the structural grid, and each monitoring area corresponds to several vibration sensors;

[0020] When the main vibration frequency of the monitoring area matches the theoretical buckling frequency, it is marked as a candidate buckling risk area;

[0021] Determine the stress concentration coefficient of each candidate buckling risk area based on the corrected stress distribution data of the slip block;

[0022] Determine the local buckling risk area where the stress concentration coefficient exceeds the preset threshold.

[0023] Preferably, divide the steel roof into several asymmetric slip blocks based on the local buckling risk areas, and generate slip path correction instructions in combination with the geometric state data of the slip tracks, including:

[0024] Divide the steel roof into high-risk blocks and low-risk blocks based on the number of local buckling risk areas per preset unit area;

[0025] Generate basic correction instructions based on the geometric state data of the slip tracks;

[0026] Generate slip path correction instructions based on the block division results and the basic correction instructions.

[0027] Preferably, generate basic correction instructions based on the geometric state data of the slip tracks, including:

[0028] If the lateral offset of the track in the geometric state data of the slip track exceeds the preset tolerance, generate a lateral pressure compensation instruction for the hydraulic jacking system;

[0029] If the longitudinal flatness of the track in the geometric state data of the slip track exceeds the threshold, generate a slip speed adjustment instruction;

[0030] If the relative spacing between the tracks in the geometric state data of the slip track is abnormal, generate a track spacing adjustment instruction.

[0031] Preferably, during the slip process, based on the slip path correction instructions, the geometric deviation of the slip track is compensated in real time, and at the same time, combined with the initial clearance parameters of the embedded support structure, the critical deformation threshold in the subsequent construction stage is predicted, including:

[0032] During the slip process, execute the slip path correction instructions to compensate the geometric deviation of the slip track in real time;

[0033] Determine the contact stress distribution between the steel column and the embedded support structure during the unloading stage based on the initial clearance parameters of the embedded support structure and the compensated slip track state data;

[0034] The geometric state data of the sliding track, the strain data of the steel structure, and the dynamic vibration data during the sliding process are extrapolated to the unloading stage through a preset analysis method, and then combined with the contact stress distribution of the embedded support structure to predict the critical deformation threshold during the construction stage.

[0035] Preferably, after the sliding is completed, the falling trajectory data of the steel column is tracked in real time and the unloading rate is dynamically adjusted in combination with the critical deformation threshold, and the unloading path is calibrated by using the compensated geometric state data of the sliding track until the deformation error converges within the preset tolerance range, including:

[0036] After the sliding is completed, the falling trajectory data of the steel column is tracked in real time based on the laser interferometer and the embedded gap displacement sensor. Among them, the falling trajectory data of the steel column includes: real-time displacement and motion state;

[0037] Compare the real-time displacement with the critical deformation threshold to generate an unloading rate adjustment instruction adapted to the current deformation state;

[0038] Map the corrected geometric state data of the sliding track to the unloading path through a preset correction algorithm, and combine the falling trajectory data of the steel column to correct the unloading direction and positioning accuracy in real time to obtain a calibrated unloading path;

[0039] Unload based on the calibrated unloading path and determine the deviation between the real-time displacement during the unloading process and the target path;

[0040] If the error exceeds the preset tolerance range, iteratively execute the unloading rate adjustment instruction and the unloading path calibration operation. If the error converges within the preset tolerance range, it is determined that the unloading is completed.

[0041] The present invention provides a full-cycle deformation monitoring system for a steel roof based on multi-source sensor fusion, including:

[0042] Multi-source data acquisition module: A multi-source sensor network is arranged at preset positions on the upper and lower chords of the steel truss, the sliding track beam columns, and the hydraulic boosters to synchronously collect the geometric state data of the sliding track, the strain data of the steel structure, the dynamic vibration data, and the initial gap parameters of the embedded support structure;

[0043] Data processing module: Determine the stress distribution data of the sliding block of the steel structure during the sliding process based on the strain data and the geometric state data of the sliding track, and determine the local buckling risk area in combination with the dynamic vibration data;

[0044] Sliding control module: Divide the steel roof into several asymmetric sliding blocks based on the local buckling risk area, and generate a sliding path correction instruction in combination with the geometric state data of the sliding track;

[0045] Prediction and determination module: During the sliding process, based on the sliding path correction instruction, geometric deviations of the sliding track are compensated in real time, and at the same time, combined with the initial clearance parameters of the embedded support structure, the critical deformation threshold in the subsequent construction stage is predicted;

[0046] Unloading control module: After the sliding is completed, the falling trajectory data of the steel column is tracked in real time and the unloading rate is dynamically regulated in combination with the critical deformation threshold, and the unloading path is calibrated using the compensated geometric state data of the sliding track until the deformation error converges within the preset tolerance range.

[0047] Compared with the prior art, the beneficial effects of the present application are as follows:

[0048] Through multi-source sensor fusion, a comprehensive solution for the full-cycle deformation monitoring of the steel roof is provided, which can collect and analyze data such as the geometric state, strain, and vibration of the sliding track in real time, accurately monitor the deformation of the steel structure. Compared with the prior art, it can compensate for track deviations in real time, identify buckling risk areas, and dynamically regulate the construction process according to the predicted critical deformation threshold to ensure that the deformation error is within the tolerance range, effectively improving the construction accuracy and safety, reducing the impact of deformation on the structural stability, and optimizing the adjustment strategy during construction. Description of the Drawings

[0049] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0050] Figure 1 It is a flowchart of the full-cycle deformation monitoring method for the steel roof based on multi-source sensor fusion provided by the embodiment of the present invention.

[0051] Figure 2 It is a structural diagram of the full-cycle deformation monitoring system for the steel roof based on multi-source sensor fusion provided by the embodiment of the present invention. Detailed Embodiments

[0052] To make the purpose, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention in conjunction with the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0053] Embodiment 1:

[0054] The embodiment of the present invention provides a full-cycle deformation monitoring method for steel roofs based on multi-source sensor fusion, as follows Figure 1 shown, including:

[0055] Step 1: Deploy a multi-source sensor network at preset positions on the upper and lower chords of the steel truss, the beam-columns of the sliding track, and the hydraulic boosters to synchronously collect geometric state data of the sliding track, strain data of the steel structure, dynamic vibration data, and initial clearance parameters of the embedded support structure;

[0056] Step 2: Determine the stress distribution data of the sliding blocks of the steel structure during the sliding process based on the strain data and the geometric state data of the sliding track, and determine the local buckling risk area in combination with the dynamic vibration data;

[0057] Step 3: Divide the steel roof into several asymmetric sliding blocks based on the local buckling risk area, and generate a sliding path correction instruction in combination with the geometric state data of the sliding track;

[0058] Step 4: During the sliding process, compensate for the geometric deviation of the sliding track in real time based on the sliding path correction instruction, and at the same time, combine the initial clearance parameters of the embedded support structure to predict the critical deformation threshold in the subsequent construction stage;

[0059] Step 5: After the sliding is completed, track the steel column falling trajectory data in real time and dynamically adjust the unloading rate in combination with the critical deformation threshold, and calibrate the unloading path using the compensated geometric state data of the sliding track until the deformation error converges within the preset tolerance range.

[0060] In this embodiment, the sensor network includes a distributed fiber Bragg grating strain sensor array, a laser scanning displacement sensor, a three-axis vibration acceleration sensor, and an embedded gap displacement sensor.

[0061] In this embodiment, the multi-source sensor network is deployed in a combination of a distributed array and key nodes. For example, a fiber Bragg grating strain sensor is deployed every 3m on the upper chord of the steel truss with a span exceeding 50m, and every 5m on the lower chord; laser scanning displacement sensors (accuracy ±0.1mm) are densely deployed at the inflection points of the beam-columns of the sliding track; pressure sensors (range 0-5000kN) are installed on the piston rods of the hydraulic boosters. All sensors synchronize data in real time through 5G edge computing nodes, and the sampling frequency is set to 200Hz.

[0062] Advantages of the above technical solution: Through multi-source sensor fusion, a comprehensive solution for the full-cycle deformation monitoring of the steel roof is provided. It can collect and analyze data such as the geometric state, strain, and vibration of the sliding track in real time, accurately monitor the deformation of the steel structure. Compared with the existing technology, it can compensate for track deviation in real time, identify buckling risk areas, and dynamically adjust the construction process according to the predicted critical deformation threshold to ensure that the deformation error is within the tolerance range, effectively improving the construction accuracy and safety, reducing the impact of deformation on the structural stability, and optimizing the adjustment strategy during the construction process.

[0063] Embodiment 2:

[0064] The embodiment of the present invention provides a full-cycle deformation monitoring method for a steel roof based on multi-source sensor fusion. The multi-source sensor network includes:

[0065] A distributed fiber Bragg grating strain sensor array is arranged on the upper and lower chords of the steel truss to collect strain data;

[0066] A laser scanning displacement sensor is arranged at the key nodes of the sliding track to obtain the geometric state data of the sliding track. Among them, the geometric state data includes the lateral offset of the track, the longitudinal flatness, and the relative spacing between tracks;

[0067] A three-axis vibration acceleration sensor is arranged at the connection between the steel truss and the sliding track to monitor the dynamic vibration data of each key node;

[0068] An embedded gap displacement sensor is set at the contact interface between the bottom of the steel column and the embedded support structure to collect the initial gap parameters.

[0069] In this embodiment, the method for collecting the geometric state data of the laser scanning displacement sensor is as follows:

[0070] Lateral offset measurement: Based on the laser triangulation principle, with the designed axis of the track as the reference, the lateral deviation is calculated in real time (resolution 0.1 mm); Longitudinal flatness calculation: Based on the elevation data of multiple sections of the track, a reference plane is fitted to calculate the local flatness deviation (error ≤ 0.5 mm / m); Track spacing calibration: A binocular vision sensor is used to assist the laser scanning to achieve synchronous monitoring of the spacing of multiple tracks (accuracy ±1 mm). The fusion algorithm of the laser scanning data and the vision data eliminates the measurement noise caused by environmental vibration through the weighted least squares method.

[0071] In this embodiment, the specific composition of the multi-source sensor network is as follows: The distributed fiber Bragg grating strain sensor array adopts the BOTDA technology, which can realize continuous strain monitoring within a range of 100 km, and the temperature compensation accuracy is ±0.5°C; The laser scanning displacement sensor selects the Trimble CX-100 model, which is equipped with a biaxial inclinometer and can simultaneously measure the lateral offset of the track (range ±50 mm) and the longitudinal flatness (resolution 0.01 mm / m); The three-axis vibration acceleration sensor adopts the PCB 356A17 type, with a frequency response range of 0.5 - 10 kHz and a sensitivity of 100 mV / g;

[0072] The embedded gap displacement sensor adopts a magnetostrictive displacement gauge (model MTS Temposonics), which is bonded to the bottom surface of the steel column through epoxy resin during installation, and the range is 0 - 20 mm.

[0073] The beneficial effects of the above technical solutions: Through the combination of the multi-source sensor network, the comprehensive monitoring of the deformation of the steel roof is realized. Compared with the prior art, the fusion of distributed fiber Bragg grating strain sensors, laser scanning displacement sensors, three-axis vibration acceleration sensors and embedded gap displacement sensors is adopted, which can accurately collect strain, geometric state, vibration data and initial gap parameters. This multi-dimensional data fusion provides more comprehensive and accurate deformation monitoring capabilities, improves the real-time perception and analysis of the structural state, optimizes the deformation prediction and correction strategies, and improves the construction safety and accuracy.

[0074] Embodiment 3:

[0075] The embodiment of the present invention provides a full-cycle deformation monitoring method for a steel roof based on multi-source sensor fusion, which determines the stress distribution data of the sliding blocks of the steel structure during the sliding process based on the strain data and the geometric state data of the sliding track, and determines the local buckling risk area in combination with the dynamic vibration data, including:

[0076] Collect the strain data of each preset monitoring point of the steel truss through the distributed fiber Bragg grating strain sensor array, and then determine the initial stress value of each preset monitoring point to generate the initial stress distribution of the steel structure;

[0077] Correct the additional stress components in the initial stress distribution caused by the unevenness and offset of the track through the geometric state data of the sliding track to generate the corrected stress distribution data of the sliding blocks;

[0078] Divide the steel truss into several monitoring areas according to the structural grid, and each monitoring area corresponds to several vibration sensors;

[0079] When the main vibration frequency of the monitoring area matches the theoretical buckling frequency, it is marked as a candidate buckling risk area;

[0080] Determine the stress concentration factor of each candidate buckling risk area based on the corrected stress distribution data of the slip block.

[0081] Identify the candidate buckling risk areas with stress concentration factors exceeding the preset threshold as local buckling risk areas.

[0082] In this embodiment, the method for correcting the additional stress components caused by track unevenness and offset in the initial stress distribution is as follows: establish a dynamic model of the friction coefficient at the slip track-steel structure contact interface; calculate the additional stress components through finite element contact analysis based on the friction coefficient and track offset. Among them, the dynamic model of the friction coefficient at the slip track-steel structure contact interface is based on contact mechanics theory and mainly includes the following modules: Input parameter layer: Integrate real-time data such as track geometric parameters (lateral offset, longitudinal flatness), contact pressure distribution, relative slip speed, and interface temperature; Friction behavior module: Adopt a modified Coulomb friction model, introduce velocity softening effect and adhesion-slip discrimination criterion to distinguish static friction and dynamic friction states; Dynamic update mechanism: Online correct model parameters through recursive least squares (RLS), and combine neural network to predict the influence of interface pollutants (such as rust, grease) on the friction coefficient; Contact state evaluation: Based on the contact stiffness matrix and energy dissipation rate, judge whether the interface is in elastic contact, plastic deformation or debonding state, and dynamically adjust the value range of the friction coefficient; Finite element coupling interface: Map the friction coefficient field to the contact elements of the finite element model to achieve accurate calculation of additional stress components. This model breaks through the traditional static friction coefficient assumption through multi-physics field coupling and can dynamically reflect the non-linear behavior of the contact interface during the construction process.

[0083] In this embodiment, the theoretical buckling frequency is calculated by the Euler buckling formula:

[0084] ;

[0085] Where: is the theoretical buckling frequency, is the structural stiffness, is the equivalent mass. This formula is used to calculate the critical frequency of a certain structure. In vibration analysis, represents the self-vibration frequency in the structural system, and its value is affected by the structural stiffness and the equivalent mass . The greater the stiffness and the smaller the equivalent mass, the higher the natural vibration frequency of the structure usually is. represents the stiffness of the structure, that is, the ability of the structure to withstand deformation. The greater the stiffness, the smaller the deformation. is the mass of the system. The greater the mass, the stronger the inertia of vibration, and the relatively lower the vibration frequency.

[0086] In this embodiment, the calculation formula of the stress concentration factor:

[0087]

[0088] Wherein: is the corrected local stress value within the monitoring area (usually obtained by scanning the stress distribution in the slip area), is the global average stress value of the steel structure, and the calculation method is the arithmetic mean of the stresses in all monitoring areas, or the nominal stress value determined according to the design load distribution;

[0089] In this embodiment, the stress concentration coefficient represents the ratio of the local stress to the nominal stress. In practical applications, the ratio of the local stress to the global average stress is usually calculated dynamically to meet the transfer requirements and the needs of high-frequency stress fatigue analysis. Dependence on the crosswind area: In structural experiments, the area of the stress concentration coefficient is usually divided into a range of 1.2 - 1.5 as the high-wind risk area (such as the safety risk area of aviation and bridge structures).

[0090] Beneficial effects of the above technical solution: By combining strain data, geometric state data of the slip track, and dynamic vibration data, accurate identification of the stress distribution and buckling risk area during the steel structure sliding process is achieved. Compared with the prior art, it can dynamically correct the additional stress caused by track unevenness or deviation, improving the accuracy of stress analysis. At the same time, vibration data is used to identify potential buckling risk areas, and the local buckling risk areas are further determined through the stress concentration coefficient to ensure safety and structural stability during construction. This method provides more comprehensive and accurate deformation monitoring and warning capabilities.

[0091] Embodiment 4:

[0092] The embodiment of the present invention provides a full-cycle deformation monitoring method for a steel roof based on multi-source sensor fusion. The steel roof is divided into several asymmetric sliding blocks based on the local buckling risk area, and a sliding path correction instruction is generated in combination with the geometric state data of the sliding track, including:

[0093] Dividing the steel roof into high-risk blocks and low-risk blocks based on the number of local buckling risk areas within a preset unit area;

[0094] Generating a basic correction instruction based on the geometric state data of the sliding track;

[0095] Generating a sliding path correction instruction based on the block division result and the basic correction instruction.

[0096] In this embodiment, high-risk block: the area where the buckling risk areas are densely distributed, and the sliding speed needs to be reduced and the jacking force needs to be decreased to avoid the risk of structural instability; low-risk block: the area where the buckling risk areas are sparse or unmarked, and the conventional sliding parameters can be maintained;

[0097] In this embodiment, the determination of high-risk blocks: When the number of buckling risk areas in a unit area (1m×1m grid) is ≥3, it is marked as high-risk; Dynamic segmentation algorithm: Based on Delaunay triangulation, the steel roof is divided into irregular blocks to ensure that high-risk areas are concentrated in the center of the blocks; Correction instruction priority: The priority of the lateral offset compensation instruction is higher than that of the speed adjustment, and an emergency pause signal is triggered when the track spacing is abnormal.

[0098] In this embodiment, a slip path correction instruction is generated based on the block division result and the basic correction instruction: High-risk block control strategy: Superimpose the basic correction instruction and the slip parameter limit conditions (such as maximum speed, minimum jacking force) to generate a dedicated control instruction; Low-risk block control strategy: Only apply the basic correction instruction and maintain the conventional construction parameters; Output: A block-based dynamic control instruction set, that is, the slip path correction instruction (high-risk block restrictive instruction, low-risk block basic instruction); Among them, the setting logic of the slip parameter limit conditions (such as maximum slip speed and minimum jacking force) is as follows: Preset reference value: Based on structural safety analysis and historical construction data, determine the initial limit conditions in the following ways: Structural safety factor: Calculate the safety margin according to the design bearing capacity, material yield strength and buckling critical load of the steel truss, and set the theoretical thresholds of the maximum slip speed and minimum jacking force; Engineering experience: Refer to the construction parameters of similar projects (such as the slip speed range and jacking force safety interval of steel roofs with similar spans), and combine engineering specifications (such as the "Code for Acceptance of Construction Quality of Steel Structure Engineering" GB50205) to set the initial limit values. Dynamic adjustment mechanism: During actual construction, dynamically correct the limit conditions based on real-time monitoring data and buckling risk levels: Based on real-time stress and vibration data: If the stress concentration coefficient or vibration amplitude of the high-risk block is detected to be close to the safety boundary, automatically reduce the upper limit of the slip speed or increase the lower limit of the jacking force; Based on the buckling risk density: When the risk area density in the block increases, further tighten the parameter restrictions; Conversely, it can be appropriately relaxed; Machine learning optimization: Train a prediction model through historical construction data and output dynamic parameter thresholds suitable for different working conditions.

[0099] Beneficial effects of the above technical solution: By dividing based on local buckling risk areas, the steel roof is divided into high- and low-risk blocks, and a slip path correction instruction is generated in combination with the geometric state data of the slip track. Compared with the prior art, it can more precisely identify different risk blocks, realize dynamic adjustment of the slip path, and effectively avoid the occurrence of local buckling risks. By generating correction instructions in real time and optimizing the geometric state of the slip track, the accuracy and safety during construction are significantly improved, ensuring that the deformation error is controlled within an acceptable range.

[0100] Embodiment 5:

[0101] The embodiment of the present invention provides a full-cycle deformation monitoring method for a steel roof truss based on multi-source sensor fusion, generating a basic correction instruction based on the geometric state data of the sliding track, including:

[0102] If the lateral offset of the track in the geometric state data of the sliding track exceeds the preset tolerance, a lateral pressure compensation instruction for the hydraulic jacking system is generated;

[0103] If the longitudinal flatness of the track in the geometric state data of the sliding track exceeds the threshold, a sliding speed adjustment instruction is generated;

[0104] If the relative spacing between the tracks in the geometric state data of the sliding track is abnormal, a track spacing adjustment instruction is generated.

[0105] In this embodiment, the preset tolerance: the maximum allowable lateral offset, such as 5 mm.

[0106] In this embodiment, generating a lateral pressure compensation instruction for the hydraulic jacking system: when the real-time offset > 5 mm, the hydraulic jacking system applies a lateral pressure proportionally (for example: when the offset is 8 mm, the compensation pressure is 10 kN / mm × 3 mm = 30 kN), and pushes the track back within the tolerance range.

[0107] In this embodiment, when the longitudinal flatness of the track exceeds the threshold, the threshold is defined as: if the longitudinal flatness deviation > 2 mm / m, a sliding speed adjustment instruction is generated: when the flatness deviation of a certain section of the track is detected to be 3 mm / m, the sliding speed is reduced from 1.0 m / min to 0.7 m / min to reduce the impact load caused by unevenness.

[0108] In this embodiment, when the relative spacing between the tracks is abnormal, the abnormal determination: the designed spacing is 1.0 m, the measured spacing > 1.05 m or < 0.95 m, a track spacing adjustment instruction is generated: through the servo motor to drive the track fine-tuning mechanism, the spacing is calibrated to 1.0 m ± 0.5 mm with a step accuracy of 0.1 mm to ensure the synchronous sliding of multiple tracks.

[0109] The beneficial effects of the above technical solutions: By generating a basic correction instruction based on the geometric state data of the sliding track, the geometric deviation of the track during construction can be dynamically adjusted. When the lateral offset, longitudinal flatness of the track does not meet the standard, or the track spacing is abnormal, instructions for pressure compensation of the hydraulic jacking system, sliding speed adjustment, or track spacing adjustment can be automatically generated. Compared with the prior art, this dynamic correction mechanism can respond to track deformation in real time, accurately control the sliding path, ensure that the geometric state of the track during construction meets the predetermined standard, and thus improve the construction accuracy and structural safety.

[0110] Example 6:

[0111] The embodiment of the present invention provides a full-cycle deformation monitoring method for a steel roof truss based on multi-source sensor fusion. During the sliding process, the geometric deviation of the sliding track is compensated in real time based on the sliding path correction instruction, and at the same time, combined with the initial clearance parameters of the embedded support structure, the critical deformation threshold in the subsequent construction stage is predicted, including:

[0112] During the sliding process, execute the sliding path correction instruction to compensate the geometric deviation of the sliding track in real time;

[0113] Determine the contact stress distribution between the steel column and the embedded support structure in the unloading stage based on the initial clearance parameters of the embedded support structure and the state data of the compensated sliding track;

[0114] Extrapolate the geometric state data of the sliding track, the strain data of the steel structure, and the dynamic vibration data during the sliding process to the unloading stage through a preset analysis method, and then predict the critical deformation threshold in the construction stage in combination with the contact stress distribution of the embedded support structure.

[0115] In this embodiment, executing the sliding path correction instruction to compensate the geometric deviation of the sliding track in real time includes: lateral deviation compensation: applying lateral pressure through a hydraulic jacking system to offset the lateral offset of the track; longitudinal flatness compensation: adjusting the sliding speed gradient to reduce the impact load in the uneven area; track spacing calibration: adjusting the spacing error of multi-track synchronous sliding; Output: Geometric state data of the compensated sliding track (offset after correction, flatness, spacing).

[0116] In this embodiment, determining the contact stress distribution between the steel column and the embedded support structure in the unloading stage based on the initial clearance parameters of the embedded support structure and the state data of the compensated sliding track includes: determining the contact area when the steel column drops according to the initial clearance parameters; calculating the pressure distribution gradient of the contact interface based on the compensated track geometric state data (such as flatness, spacing);

[0117] Derive the peak contact stress and its distribution range through finite element analysis or empirical formulas. Among them, determining the contact area based on the initial clearance parameter includes: based on the initial clearance parameter of the embedded support structure (the contact radius is calculated by Hertz contact theory for the clearance between the bottom of the steel column and the contact interface of the embedded structure, and the contact area refers to the actual physical contact area between the bottom of the steel column and the embedded support structure at the unloading stage); calculate the average pressure according to the jacking force and the contact area during the slipping stage, and combine the track flatness and spacing data to refine the pressure distribution P(x,y) through finite element analysis or empirical formulas, so that it is non-uniformly distributed within the contact area; output: the pressure distribution P(x,y) of the contact interface (where the jacking force is the driving force applied by the hydraulic jacking system during the slipping process, and its data sources include: direct measurement: the jacking force data is collected in real time through the pressure sensor of the hydraulic system; indirect calculation: the jacking force is deduced based on the lateral pressure compensation amount in the slipping path correction instruction and the system mechanical parameters (such as piston area); the lateral pressure compensation amount of the slipping path correction instruction is obtained by collecting the jacking force data in real time through the pressure sensor of the hydraulic system and recorded as the jacking force history sequence during the slipping stage); based on Hooke's law, convert the pressure distribution P(x,y) into the initial stress distribution σinitial(x,y); adjust the gradient of σinitial(x,y) according to the longitudinal flatness deviation of the track to obtain the corrected contact stress distribution σcorrected(x,y); identify the maximum stress σmax and the high stress area within the contact area based on σcorrected(x,y); output the contact stress distribution during the unloading stage (including the stress peak, distribution range and risk mark).

[0118] In this embodiment, the critical deformation threshold refers to the maximum allowable safe deformation amount (including parameters such as displacement and stress) predicted during the unloading stage. Exceeding this threshold may lead to structural instability or damage. This threshold is obtained through real-time data modeling and extrapolation analysis during the slipping stage, specifically including dynamic safety boundaries such as the displacement upper limit and stress peak.

[0119] The beneficial effects of the above technical solution: By compensating for the geometric deviation of the slipping track in real time and combining the initial clearance parameter, the critical deformation threshold in the subsequent construction stage can be accurately predicted. Compared with the prior art, the system not only dynamically adjusts the slipping path to ensure precise control of the track geometric state, but also predicts the contact stress distribution and deformation threshold during the unloading stage by extrapolating the key data during the slipping process. This method improves the predictability and control ability of the deformation during the construction process, and enhances the safety and construction accuracy of the overall structure.

[0120] Example 7:

[0121] The embodiment of the present invention provides a full-cycle deformation monitoring method for a steel roof based on multi-source sensor fusion. After the slip is completed, the falling trajectory data of the steel column is tracked in real time, and the unloading rate is dynamically adjusted in combination with the critical deformation threshold. The unloading path is calibrated using the compensated geometric state data of the slip track until the deformation error converges within the preset tolerance range, including:

[0122] After the slip is completed, the falling trajectory data of the steel column is tracked in real time based on a laser interferometer and embedded gap displacement sensors. Among them, the falling trajectory data of the steel column includes: real-time displacement and motion state;

[0123] Compare the real-time displacement with the critical deformation threshold to generate an unloading rate adjustment instruction adapted to the current deformation state;

[0124] Map the corrected geometric state data of the slip track to the unloading path through a preset correction algorithm, and combine the falling trajectory data of the steel column to correct the unloading direction and positioning accuracy in real time to obtain a calibrated unloading path;

[0125] Unload based on the calibrated unloading path and determine the deviation between the real-time displacement during unloading and the target path;

[0126] If the error exceeds the preset tolerance range, iteratively execute the unloading rate adjustment instruction and the unloading path calibration operation. If the error converges within the preset tolerance range, determine that the unloading is completed.

[0127] In this embodiment, dynamically adjusting the unloading rate includes: obtaining the data of the embedded gap displacement sensor at the bottom of the steel column in real time, calculating the safe deformation margin of the current unloading stage in combination with the critical deformation threshold; adopting a fuzzy PID control algorithm to dynamically adjust the unloading flow of the hydraulic system according to the deviation between the safe deformation margin and the preset unloading rate; when a mutation point appears in the falling trajectory data of the steel column, trigger a hierarchical unloading strategy to reduce the single unloading amount below the preset threshold until the trajectory returns to a stable state.

[0128] In this embodiment, the unloading rate adjustment instruction adapted to the current deformation state includes: dynamically adjusting the traction rate of the winch. If the real-time displacement is close to the upper limit of the threshold, reduce the traction rate of the winch; if the displacement error converges within the safe range, maintain or moderately increase the rate.

[0129] In this embodiment, if the error exceeds the preset tolerance range, the unloading rate adjustment instruction and the unloading path calibration operation are iteratively executed. If the error converges within the preset tolerance range, determining that the unloading is completed includes: establishing a dynamic error tolerance model and adaptively adjusting the preset tolerance range according to the time series data in the unloading stage; when the deformation errors in three consecutive monitoring periods all exceed the upper tolerance limit, starting a local support preloading program to apply a reaction force through a jack to compensate for the deformation; after the unloading is completed, using a distributed fiber optic grating strain sensor array to verify whether the overall stress of the structure is evenly distributed, and triggering a secondary calibration process if there is a stress mutation area.

[0130] The beneficial effects of the above technical solution: By real-time tracking the falling trajectory of the steel column and dynamically regulating the unloading rate, the deformation error during the unloading process can be accurately controlled to ensure its convergence within the preset tolerance range. Compared with the prior art, the motion state of the steel column is monitored in real time by combining a laser interferometer and a gap displacement sensor, and the unloading path is calibrated using the corrected slip track data to ensure the accuracy of the unloading process, improve the dynamic adjustment ability during the unloading process, optimize the deformation control, and significantly improve the construction accuracy and structural safety.

[0131] Embodiment 8:

[0132] The embodiment of the present invention provides a full-cycle deformation monitoring system for a steel roof truss based on multi-source sensor fusion, including:

[0133] Multi-source data acquisition module: A multi-source sensor network is arranged at preset positions on the upper and lower chords of the steel truss, the beam-columns of the slip track, and the hydraulic boosters to synchronously collect the geometric state data of the slip track, the strain data of the steel structure, the dynamic vibration data, and the initial gap parameters of the embedded support structure.

[0134] Data processing module: Based on the strain data and the geometric state data of the slip track, determine the stress distribution data of the slip blocks of the steel structure during the slipping process, and combine the dynamic vibration data to determine the local buckling risk area.

[0135] Slip control module: Based on the local buckling risk area, divide the steel roof truss into several asymmetric slip blocks, and generate a slip path correction instruction in combination with the geometric state data of the slip track.

[0136] Prediction determination module: During the slipping process, based on the slip path correction instruction, compensate the geometric deviation of the slip track in real time, and at the same time, in combination with the initial gap parameters of the embedded support structure, predict the critical deformation threshold in the subsequent construction stage.

[0137] Unloading control module: After the slipping is completed, real-time track the falling trajectory data of the steel column and dynamically regulate the unloading rate in combination with the critical deformation threshold, and calibrate the unloading path using the compensated geometric state data of the slip track until the deformation error converges within the preset tolerance range.

[0138] Beneficial effects of the above technical solution: Through multi-source sensor fusion, a comprehensive solution for the full-cycle deformation monitoring of the steel roof is provided, which can collect and analyze data such as the geometric state, strain, and vibration of the sliding track in real time, accurately monitor the deformation of the steel structure. Compared with the prior art, it can compensate for track deviation in real time, identify buckling risk areas, and dynamically adjust the construction process according to the predicted critical deformation threshold to ensure that the deformation error is within the tolerance range, effectively improving the construction accuracy and safety, reducing the impact of deformation on the structural stability, and optimizing the adjustment strategy during the construction process.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A full-cycle deformation monitoring method for steel roof trusses based on multi-source sensor fusion, characterized in that, include: Step 1: A multi-source sensor network is deployed at the preset positions of the upper and lower chords of the steel truss, the sliding track beams and the hydraulic boosters to synchronously collect the geometric state data of the sliding track, the strain data of the steel structure, the dynamic vibration data and the initial gap parameters of the embedded support structure, wherein the geometric state data includes the lateral offset of the track, the longitudinal flatness and the relative spacing between the tracks; Step 2: Determine the stress distribution data of the sliding block of the steel structure during the sliding process based on the strain data and the sliding track geometric state data, and determine the local buckling risk area in combination with the dynamic vibration data, including: The strain data of each preset monitoring point of the steel truss is collected by using a distributed fiber grating strain sensor array, and then the initial stress value of each preset monitoring point is determined to generate the initial stress distribution of the steel structure; The additional stress components in the initial stress distribution caused by track unevenness and offset are corrected by using the sliding track geometric state data to generate corrected sliding block stress distribution data; The steel truss is divided into several monitoring areas according to the structural grid, and each monitoring area corresponds to several vibration sensors; When the main vibration frequency of the monitoring area matches the theoretical buckling frequency, it is marked as a buckling risk area to be selected; Determine the stress concentration factor of each buckling risk area to be selected based on the corrected stress distribution data of the slip block; Determine the selected buckling risk area whose stress concentration factor exceeds a preset threshold as a local buckling risk area; Step 3: Divide the steel roof into several asymmetric sliding blocks based on the local buckling risk area, and generate sliding path correction instructions based on the geometric state data of the sliding track; Step 4: During the sliding process, the geometric deviation of the sliding track is compensated in real time based on the sliding path correction instruction, and the critical deformation threshold in the subsequent construction stage is predicted in combination with the initial gap parameters of the embedded support structure; Step 5: After the sliding is completed, the steel column falling trajectory data is tracked in real time and the unloading rate is dynamically controlled in combination with the critical deformation threshold. The unloading path is calibrated using the compensated sliding track geometric state data until the deformation error converges to the preset tolerance range.

2. The method for full-cycle deformation monitoring of a steel roof based on multi-source sensor fusion according to claim 1, characterized in that Multi-source sensor network, including: Distributed fiber Bragg grating strain sensor arrays are placed on the upper and lower chords of the steel truss to collect strain data; Laser scanning displacement sensors are arranged at key nodes of the sliding track to obtain the geometric state data of the sliding track, wherein the geometric state data includes the lateral offset of the track, the longitudinal flatness and the relative spacing between the tracks; The triaxial vibration acceleration sensor is placed at the connection between the steel truss and the sliding track to monitor the dynamic vibration data of each key node; The embedded gap displacement sensor is set at the contact interface between the bottom of the steel column and the embedded supporting structure to collect initial gap parameters.

3. The method for full-cycle deformation monitoring of a steel roof based on multi-source sensor fusion according to claim 2, characterized in that The steel roof is divided into several asymmetric sliding blocks based on the local buckling risk area, and the sliding path correction instructions are generated in combination with the geometric state data of the sliding track, including: The steel roof is divided into high-risk blocks and low-risk blocks based on the number of local buckling risk areas within a preset unit area; Generate basic correction instructions based on the sliding track geometric state data; Generate a slip path correction instruction based on the block division result and the basic correction instruction.

4. The full-cycle deformation monitoring method for steel roofs based on multi-source sensor fusion according to claim 3, characterized in that, Generate a basic correction instruction based on the geometric state data of the slip track, including: If the lateral offset of the track in the geometric state data of the slip track exceeds the preset tolerance, generate a lateral pressure compensation instruction for the hydraulic jacking system; If the longitudinal flatness of the track in the geometric state data of the slip track exceeds the threshold, generate a slip speed adjustment instruction; If the relative spacing between the tracks in the geometric state data of the slip track is abnormal, generate a track spacing adjustment instruction.

5. The method for full-cycle deformation monitoring of a steel roof based on multi-source sensor fusion according to claim 1, characterized in that During the slip process, based on the slip path correction instruction, compensate for the geometric deviation of the slip track in real time, and at the same time, combine the initial clearance parameters of the embedded support structure to predict the critical deformation threshold in the subsequent construction stage, including: During the slip process, execute the slip path correction instruction to compensate for the geometric deviation of the slip track in real time; Determine the contact stress distribution between the steel column and the embedded support structure during the unloading stage based on the initial clearance parameters of the embedded support structure and the compensated slip track state data; Extrapolate the geometric state data of the slip track, the strain data of the steel structure, and the dynamic vibration data during the slip process to the unloading stage through a preset analysis method, and then combine the contact stress distribution of the embedded support structure to predict the critical deformation threshold in the construction stage.

6. The method for full-cycle deformation monitoring of a steel roof based on multi-source sensor fusion according to claim 1, wherein After the slip is completed, track the steel column falling trajectory data in real time and dynamically adjust the unloading rate in combination with the critical deformation threshold, and use the compensated slip track geometric state data to calibrate the unloading path until the deformation error converges within the preset tolerance range, including: After the slip is completed, track the steel column falling trajectory data in real time based on the laser interferometer and the embedded gap displacement sensor. Among them, the steel column falling trajectory data includes: real-time displacement and motion state; Compare the real-time displacement with the critical deformation threshold to generate an unloading rate adjustment instruction adapted to the current deformation state; Map the corrected geometric state data of the slip track to the unloading path through a preset correction algorithm, and combine the steel column falling trajectory data to correct the unloading direction and positioning accuracy in real time to obtain a calibrated unloading path; Perform unloading based on the calibrated unloading path and determine the deviation between the real-time displacement during the unloading process and the target path; If the error exceeds the preset tolerance range, iteratively execute the unloading rate adjustment instruction and the unloading path calibration operation. If the error converges within the preset tolerance range, determine that the unloading is completed.

7. A full-cycle deformation monitoring system for a steel roof truss based on multi-source sensor fusion, which is used to implement the full-cycle deformation monitoring method for a steel roof truss based on multi-source sensor fusion as described in claim 1, characterized in that, Including: Multi-source data acquisition module: Deploy a multi-source sensor network at preset positions on the upper and lower chords of the steel truss, the slip track beam columns, and the hydraulic boosters to synchronously collect the geometric state data of the slip track, the strain data of the steel structure, the dynamic vibration data, and the initial clearance parameters of the embedded support structure; Data processing module: Determine the slip block stress distribution data of the steel structure during the slip process based on the strain data and the geometric state data of the slip track, and combine the dynamic vibration data to determine the local buckling risk area; Slip control module: Divide the steel roof into several asymmetric slip blocks based on the local buckling risk area, and generate a slip path correction instruction in combination with the geometric state data of the slip track; Prediction and determination module: During the sliding process, based on the sliding path correction instruction, geometric deviations of the sliding track are compensated in real time. Meanwhile, combined with the initial clearance parameters of the embedded support structure, the critical deformation threshold in the subsequent construction stage is predicted. Unloading control module: After the sliding is completed, the falling trajectory data of the steel column is tracked in real time, and the unloading rate is dynamically adjusted in combination with the critical deformation threshold. The unloading path is calibrated using the compensated geometric state data of the sliding track until the deformation error converges within the preset tolerance range.

Images