Deformation monitoring method and device for bidirectional large-span spatial combined steel truss

Abstract

The invention discloses a deformation monitoring method and device for a bidirectional large-span space combined steel truss, and relates to the technical field of structure monitoring, and the method comprises the steps: obtaining the basic structure data and environmental action parameters of the steel truss, constructing a combined steel truss twin model, and simulating the stress-strain response under bidirectional load coupling. And identifying abnormal deformation features and generating a deformation correction coefficient. And adjusting boundary conditions of the model based on the correction coefficient, iteratively optimizing a dynamic monitoring threshold value, and establishing a grading early warning mechanism to realize structural health reminding of the steel truss. The technical problems that an existing large-span steel truss deformation monitoring method is limited in precision and poor in dynamic threshold value adaptability under complex environment coupling loads are solved, and the large-span steel truss deformation monitoring method is obtained by constructing a twinborn model and dividing a dynamic monitoring area in combination with a multi-source sensor data fusion and iterative optimization algorithm. The technical effects of high-precision real-time monitoring of the stress-strain response under the bidirectional load coupling effect and dynamic threshold grading early warning are achieved.

Description

Technical Field

[0001] This application relates to the technical field of structural monitoring, and in particular, to a method and device for monitoring the deformation of a two-way long-span space composite steel grid structure. Background Art

[0002] Existing long-span space steel grid structures are widely used in large public buildings such as stadiums and airport terminals due to their light shape and excellent load-bearing performance. However, such structures are faced with the coupled action of multiple factors such as temperature gradient, wind load, and vibration during the construction and operation stages, which are prone to problems such as stress concentration in members, excessive node displacement, or sudden change in mid-span deflection. Traditional monitoring methods rely on contact sensors and regular inspections, and have the defects of poor real-time performance and insufficient environmental adaptability. For example, the GPS vertical direction accuracy is low, and the laser vibrometer requires strict vibration isolation conditions, while the non-contact method based on images (such as digital image correlation method) is easily affected by light and occlusion interference and depends on preset templates. In addition, most of the existing structural health assessment systems are based on code limits, and it is difficult to integrate finite element simulation and real-time monitoring data to optimize the dynamic threshold, resulting in a lag in early warning of abnormal deformation. Therefore, there is an urgent need for a deformation monitoring method that takes into account real-time performance, accuracy, and environmental robustness.

[0003] In the current related technologies, there are technical problems that the deformation monitoring method of long-span steel grid structures is limited in accuracy under complex environmental coupled loads (temperature, wind, vibration) and has poor adaptability to dynamic thresholds. Summary of the Invention

[0004] This application provides a method and device for monitoring the deformation of a two-way long-span space composite steel grid structure, and solves the technical problems that the existing deformation monitoring method of long-span steel grid structures is limited in accuracy under complex environmental coupled loads (temperature, wind, vibration) and has poor adaptability to dynamic thresholds.

[0005] This application provides a method for monitoring the deformation of a two-way long-span space composite steel grid structure, including: Obtaining the basic structural data of the steel grid structure members, including member geometric parameters, node connection types, and material mechanical property parameters; collecting environmental action parameters including temperature gradient distribution, wind load time history data, and vibration frequency spectrum, and combining the basic structural data to construct a composite steel grid structure twin model to simulate the stress-strain response under the coupled action of two-way loads; at the same time, dividing the dynamic monitoring area in the composite steel grid structure twin model; comparing and analyzing the collected real-time strain sensor data with the simulated stress-strain response to identify the abnormal deformation characteristics of the steel grid structure members and generate a deformation correction coefficient; adjusting the boundary conditions of the composite steel grid structure twin model according to the abnormal deformation characteristics and the deformation correction coefficient, and iteratively optimizing the dynamic monitoring threshold of the steel grid structure members; establishing a hierarchical early warning mechanism with the optimized dynamic monitoring threshold to give a structural health reminder for the steel grid structure members.

[0006] This application provides a deformation monitoring device for a two-way long-span space composite steel grid structure, including: A basic structure data acquisition module, which is used to acquire the basic structure data of the steel grid structure members, including rod geometric parameters, node connection types, and material mechanical property parameters; an environmental action parameter acquisition module, which is used to collect environmental action parameters including temperature gradient distribution, wind load time history data, and vibration frequency spectrum, and combine the basic structure data to construct a composite steel grid structure twin model to simulate the stress-strain response under the coupling action of two-way loads; a dynamic monitoring area division module, which is used to divide the dynamic monitoring area in the composite steel grid structure twin model at the same time; an abnormal deformation feature failure module, which is used to compare and analyze the collected real-time strain sensor data with the simulated stress-strain response, identify the abnormal deformation features of the steel grid structure members, and generate a deformation correction coefficient; a boundary condition adjustment module, which is used to adjust the boundary conditions of the composite steel grid structure twin model according to the abnormal deformation features and the deformation correction coefficient, and iteratively optimize the dynamic monitoring threshold of the steel grid structure members; a hierarchical early warning mechanism construction module, which is used to establish a hierarchical early warning mechanism with the optimized dynamic monitoring threshold to give a structural health reminder for the steel grid structure members.

[0007] It is intended to propose a deformation monitoring method and device for a two-way long-span space composite steel grid structure through this application. First, the basic structure data and environmental action parameters of the steel grid structure are obtained, a composite steel grid structure twin model is constructed to simulate the stress-strain response under the coupling of two-way loads, and the dynamic monitoring area is divided. By comparing the real-time strain sensor data with the simulated response, the abnormal deformation features are identified and the deformation correction coefficient is generated. Based on the correction coefficient, the model boundary conditions are adjusted, the dynamic monitoring threshold is iteratively optimized, and a hierarchical early warning mechanism is established to realize the structural health reminder of the steel grid structure. Through the new material detection technology (probability distribution analysis of steel yield strength) combined with the standardized process (dynamic threshold calibration), the high-precision real-time monitoring of the stress-strain response under the coupling action of two-way loads is achieved, and the technical effect of dynamic threshold hierarchical early warning is realized. Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the application. It should be understood that the operations in the front or below do not necessarily need to be executed precisely in sequence. On the contrary, as needed, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0009] Figure 1 It is a schematic flow chart of a deformation monitoring method for a two-way long-span space composite steel grid provided by an embodiment of the present application; Figure 2 It is a schematic structural diagram of a deformation monitoring device for a two-way long-span space composite steel grid provided by an embodiment of the present application.

[0010] Explanation of reference numerals: Foundation structure data acquisition module 10, environmental action parameter acquisition module 20, dynamic monitoring area division module 30, abnormal deformation feature failure module 40, boundary condition adjustment module 50, hierarchical early warning mechanism construction module 60. Specific embodiments

[0011] The above description is only an overview of the technical solution of the present application. In order to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically given below.

[0012] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0013] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0014] An embodiment of the present application provides a deformation monitoring method for a two-way long-span space composite steel grid, as Figure 1 shown, the method includes: Step S100: Obtain the basic structure data of the steel grid structure components, including the geometric parameters of the members, the node connection types, and the material mechanical property parameters. Specifically, when obtaining the basic structure data of the steel grid structure components, first plan the process, determine the acquisition sequence, tools, personnel arrangements, and time estimates. For the geometric parameters of the members, use total stations, laser rangefinders, etc. to measure the lengths, calipers, 3D laser scanners, etc. to measure the cross-sectional dimensions, take photos and draw sketches to record the shapes, measure long members multiple times, measure special-shaped members in segments or use professional software for processing. When determining the node connection types, observe the welds, bolts, rivets, etc. on-site, consult the design documents to compare the actual situation with the design requirements, and use non-destructive testing to assist in judging the nodes that are difficult to observe. To obtain the material mechanical property parameters, consult the quality certification documents, take samples and send them to a professional laboratory for tensile, compression, hardness, etc. tests. Finally, organize and establish a database containing information such as material types, specifications, batches, properties, and the corresponding positions of the structure components to provide support for subsequent analysis and monitoring.

[0015] In a possible implementation, to obtain the basic structure data of the steel grid structure components, including the geometric parameters of the members, the node connection types, and the material mechanical property parameters, step S100 further includes step S110: Obtain the geometric topology structure of the combined steel grid and extract the length deviation of the members and the offset of the node coordinates. Specifically, when obtaining the geometric topology structure of the combined steel grid and extracting the relevant deviation data, first form a professional team, prepare calibrated measuring tools such as total stations, 3D laser scanners, steel tapes, etc., and collect materials such as design drawings. To obtain the geometric topology structure, the total station measurement method can be used. Set up stations reasonably around the steel grid, accurately measure the three-dimensional coordinates of the nodes, the distances and angles between the nodes, take the average value of multiple measurements, and organize the data to construct the structure; the 3D laser scanning method can also be used. Plan the scanning routes and stations, scan to obtain the point cloud data, and construct the structure after processing such as denoising, stitching, and segmentation. Then compare the data measured on-site and constructed with the design drawings, calculate the length deviation of the members (actual length minus the design length) and the coordinate offsets of the nodes in the X, Y, and Z directions (actual coordinates minus the design coordinates). Finally, organize and analyze the deviation data, and make charts to evaluate the compliance of the structure with the design.

[0016] Step S120: Calibrate the probability distribution function of the yield strength among the material mechanical property parameters based on the geometric topology structure, the deviation of the rod length, and the offset of the node coordinates. Specifically, analyze the geometric topology structure to clarify the connection relationship between each rod and the node, map the rod length deviation and the node coordinate offset to the atlas position, analyze the spatial distribution law and its correlation with the stress characteristics, so as to evaluate the influence mechanism of the structural deviation on the material mechanical properties, and determine the key areas and parameter combinations. Then, based on the mechanical principle and the structural analysis method, use finite element software to simulate the mechanical response of the steel grid under different combinations of structural parameter deviations, and establish a preliminary mathematical relationship model between the yield strength and the structural parameters. Then, select representative material samples from the key areas of the steel grid, conduct tensile tests, accurately measure the stress-strain curve to determine the yield strength, record the corresponding structural parameter data, and repeat the experiment multiple times to obtain sufficient data points. Subsequently, use statistical methods and data fitting techniques to fit the preliminary model, determine the form and parameters of the probability distribution function of the yield strength by comparing the goodness of fit of different distribution functions, and after optimization such as residual analysis, apply it to the actual steel grid structure analysis, compare and verify it with on-site monitoring or actual cases, and if there is a deviation, recheck and correct it, and finally use it for structural reliability assessment and other aspects.

[0017] In a possible implementation manner, when calibrating the probability distribution function of the yield strength among the material mechanical property parameters based on the geometric topology structure, the deviation of the rod length, and the offset of the node coordinates, step S120 further includes step S121, and the probability distribution function of the yield strength ; where is the probability distribution function of the yield strength, S is the yield strength of the steel grid members, is the mean value of the yield strength, is the standard deviation of the yield strength. Specifically, widely sample from different batches, positions, and stress states of the steel grid members, obtain the yield strength values of each sample through standard tensile tests, and calculate the mean value of the yield strength (using the formula ) and the standard deviation (calculated according to ). Then, through data fitting tests, substitute the parameters into the function using statistical software, compare the theoretical curve with the frequency histogram of the actual data, and then use hypothesis testing methods such as chi-square test to determine whether the function is suitable for describing the yield strength distribution of the steel grid members. Finally, in the structural reliability analysis, use the function to calculate the probability that the yield strength of the member is less than the critical value under the given load, and determine a reasonable safety factor according to the function combined with the importance and stress characteristics of the member during the design stage to achieve the economic and safety balance of the steel grid design.

[0018] Step S200: Collect environmental action parameters including temperature gradient distribution, wind load time - history data, and vibration frequency spectrum, and combine with the said basic structure data to construct a combined steel space grid twin model to simulate the stress - strain response under the coupled action of bidirectional loads. Specifically, when constructing a combined steel space grid twin model to simulate the stress - strain response under the coupled action of bidirectional loads, first carry out the work of collecting environmental action parameters. In terms of collecting temperature gradient distribution, select appropriate sensors and arrange them reasonably, record and store data in real - time at a specific sampling frequency, and calibrate the sensors regularly; to collect wind load time - history data, an anemometer and a wind vane need to be installed, collect data at certain intervals, and obtain key parameters through frequency spectrum analysis and statistical processing; for collecting vibration frequency spectrum, arrange vibration sensors at key nodes and members, collect and analyze signals at a high sampling frequency to construct a frequency spectrum. At the same time, sort out, verify and organize the obtained basic structure data of steel space grid members in a certain format. Then enter the model construction link. Select professional finite - element software, create a three - dimensional geometric model according to geometric topology, assign material properties and divide the grid reasonably, then apply bidirectional loads according to environmental parameters, set boundary conditions according to actual supports, submit the calculation task, draw a contour map for analysis of the calculation results, compare with actual experience or specifications, and if there are deviations, correct and recalculate until the results are reasonable.

[0019] In a possible implementation, collect environmental action parameters including temperature gradient distribution, wind load time - history data, and vibration frequency spectrum, and combine with the said basic structure data to construct a combined steel space grid twin model to simulate the stress - strain response under the coupled action of bidirectional loads. Step S200 further includes Step S210: Arrange a distributed temperature sensor array to record the variation law of temperature gradient distribution over time. Specifically, first, according to the monitoring requirements and environment of the steel space grid, select appropriate equipment such as fiber - optic distributed temperature sensors, calculate the quantity according to the scale of the space grid and purchase from reliable suppliers, and carefully check the quality. Then study the space grid structure drawings, plan in a way that combines grid - based and key - area - based methods, arrange at a specific interval in the horizontal and vertical directions, especially densify the parts where temperature concentration is likely to occur or the temperature gradient is large, and draw a detailed layout diagram. Clean the surface of the members at the installation site, prepare tools and train personnel, carefully lay the optical fiber according to the drawing, fix it with clamps, cable ties or glue, fuse the connection parts, connect the data lines and test. Then build a data acquisition system, select a compatible and high - performance acquisition unit, install it in a suitable position, set the parameters, establish a transmission network and debug. Finally, after the system is started, record temperature data in real - time, check regularly, process with data analysis software, construct a temperature distribution matrix, calculate the temperature gradient, draw a curve of variation over time, compare the data at different time periods, seasons and weather conditions, and summarize the variation law to provide support for the analysis and maintenance of the steel space grid structure.

[0020] Step S220: Capture the wind load time history data and extract the dominant vibration frequency components through spectral analysis. Specifically, first select equipment such as ultrasonic anemometers that can accurately measure different wind speed ranges, have high precision, and a measurement range covering the possible maximum wind speed, and pair it with a highly sensitive electronic wind vane. According to the height and shape of the steel grid structure, deploy it at different height levels and multiple orientations, such as setting points in the high, middle, low parts and in the directions of southeast, northwest, etc. Connect the anemometer and the wind vane to a multi-channel data collector with a stable power supply, set parameters such as the sampling frequency and storage format, install data acquisition software and establish a transmission network. After starting the system, the equipment monitors the changes in wind speed and direction in real time, the data collector collects and stores data according to the frequency, regularly checks the integrity and accuracy of the records, and compares them with the data of the surrounding weather stations. Then export the data, use professional software to remove noise, outliers and standardize the data, select spectral analysis methods such as fast Fourier transform, combine with a suitable window function, input the data into the software to generate a spectrogram, find the frequency with the largest energy proportion, and use frequency refinement technology to accurately identify the dominant vibration frequency when necessary, and then judge its rationality in combination with experience and knowledge.

[0021] Step S230: Based on the monitoring sensor network, synchronously collect the multi-point vibration acceleration signals of the composite steel grid structure and construct a vibration frequency spectrum. Specifically, first improve and optimize the monitoring sensor network, determine the installation positions of acceleration sensors according to the steel grid structure at key nodes and main stressed members, check and eliminate interference with other sensors, and ensure that the lines are stable and clear. Then select an acceleration sensor with broadband response and calibration, firmly install it with a special base, etc., ensure that the direction of the sensitive axis is correct and check the connection. Then build a system with a multi-channel synchronous data acquisition card, configure the driver and software in an industrial control computer, set the synchronous trigger and parameters of each channel, and do a good job in power supply and shielding. Start the system to collect data in real time, pay attention to data anomalies, back up regularly, and record environmental information. Finally, import the data into professional software for preprocessing, convert it into a frequency-domain signal using algorithms such as fast Fourier transform, construct a vibration frequency spectrum and visualize it, and analyze the peak frequency to evaluate the condition of the steel grid structure.

[0022] In a possible implementation, environmental parameters including temperature gradient distribution, wind load time history data and vibration frequency spectrum are collected, and combined with the basic structure data to construct a combined steel grid twin model to simulate the stress-strain response under the coupling of bidirectional loads. Step S200 further includes step S240, and an initial model framework is established according to the geometric parameters of the rods and the node connection type. Specifically, first use equipment such as a total station, a laser rangefinder and a caliper to accurately measure the geometric parameters of the steel grid rods, measure the coordinates of the two end points of the straight rods to calculate the length, measure the cross-sectional dimensions multiple times and take the average, and measure the curves or special-shaped rods in sections and then fit them; determine the node connection type through field observation and reference to materials, record relevant details such as welds and bolts, and organize the data into a detailed table. Then select the modeling software according to the complexity of the steel grid and the analysis requirements, such as Tekla_Structures for conventional structures, ANSYS, ABAQUS, etc. for complex structures, open the software and set the unit system, display range, accuracy and coordinate system. Then create nodes according to the sorted data, model according to different connection types, select beam or rod elements to create rods and set cross-section properties. After completion, check the model's geometry and topological relationships from multiple angles, perform simple mechanical analysis when necessary, and make repeated corrections until the initial model architecture accurately reflects the actual structure of the steel grid.

[0023] Step S250, based on the initial model framework, import the material mechanical properties parameters, define nonlinear contact boundary conditions and time-varying load parameters. Specifically, first obtain the steel grid material mechanical properties parameters from the material supplier's documents, standard specifications or laboratory tests, such as the elastic modulus of Q345 steel, etc., and create material definitions in the material library management module of the modeling software after sorting, and assign attributes to the corresponding rods and nodes according to the model framework. Then analyze the steel grid to determine the nonlinear contact area, such as the rod connection, node and support, define the contact pair in the software contact analysis module, estimate the contact stiffness according to the formula, determine the friction coefficient according to the surface conditions, and select a suitable algorithm. Finally, collect meteorological data to determine the temperature load parameters, and use the software "Amplitude" module to create an amplitude curve for application; obtain wind load data from the meteorological department, etc., calculate parameters in combination with the specifications, and apply the measured time history data according to the formula in the time history. If there is no measured data, simulate it according to the wind rose diagram and the specifications; after determining the vibration source, define the vibration load in the corresponding software module according to its frequency, amplitude and other parameters, such as simple harmonic vibration or input seismic waves to simulate seismic effects.

[0024] Step S260: Based on the non - linear contact boundary conditions and time - varying load parameters, perform transient dynamic analysis and output the contour map of the dynamic response of the steel grid under the coupling action of temperature - wind load - vibration. Specifically, when conducting transient dynamic analysis based on non - linear contact boundary conditions and time - varying load parameters for the steel grid and outputting the contour map of the dynamic response, first carefully check the model to confirm that the member, node, and material parameters are accurate, and the non - linear contact and load parameter settings are reasonable. Then, select the transient dynamic analysis module of professional software such as ANSYS and ABAQUS. Next, set an appropriate time step according to the natural vibration period and load frequency of the steel grid. For example, for a grid with a natural vibration period of 0.5 seconds, take about 0.01 seconds. Determine the total time according to the load action duration. For example, for a 30 - minute strong wind action, set 35 minutes. Then select a solution algorithm such as the Newmark method or the Wilson - θ method and set relevant parameters. At the same time, set the convergence tolerance and the maximum number of iterations. For example, the force convergence tolerance is 0.001, the displacement convergence tolerance is 0.0001. For a simple model, the maximum number of iterations is 50 - 100 times, and for a complex model, it is more than 200 - 500 times. After completing the parameter settings, submit the task. During the calculation, pay attention to the software operation, log, and computing resources. After the analysis is completed, extract the dynamic response results such as displacement and stress from the software, and use the post - processing function to generate contour maps of dynamic responses such as displacement, stress, and strain. Animation display can be set to evaluate the mechanical properties and safety of the steel grid.

[0025] Step S300: Meanwhile, divide the dynamic monitoring area in the combined steel grid twin model. Specifically, when dividing the dynamic monitoring area in the combined steel grid twin model, first conduct mechanical analysis and risk assessment on the steel grid under various working conditions through finite element analysis combined with engineering cases and industry standards to clarify the weak links and risks of each part. Based on this, determine the key monitoring parameters such as stress and strain and the accuracy requirements. Then, according to the structural function zoning, divide the grid into areas such as roof load - bearing areas, and set high, medium, and low - risk areas according to the risk level. At the same time, take into account the spatial distribution uniformity to divide the monitoring area. Subsequently, reasonably set monitoring points in each area, and select appropriate sensors such as strain gauges and laser displacement sensors according to the parameters and accuracy. Build a wired or wireless data transmission network, develop data - processing software with functions such as filtering and storage. Finally, regularly calibrate the sensors according to the standards, establish a maintenance system, and check the sensor installation, wiring, and software operation to ensure the long - term reliability of the monitoring system.

[0026] In a possible implementation, at the same time, the dynamic monitoring area is divided in the twin model of the combined steel grid, and step S300 further includes step S310, wherein the dynamic monitoring area includes a high stress concentration area, a node displacement sensitive area, and a mid-span deflection change area. Specifically, the dynamic monitoring area is determined in the twin model of the combined steel grid, and the high stress concentration area is first determined. The model is modeled with the help of finite element software such as ANSYS and ABAQUS, and the parameters are input and various load conditions such as self-weight, wind, snow, and earthquake are applied. The high stress area is initially locked through the stress cloud map, and then the judgment and correction are made in combination with similar engineering case data and expert experience at home and abroad. The nodes are the focus of attention. Then the node displacement sensitive area is determined, and the displacement of each node under different loads is calculated using finite element software. The nodes with large displacement are analyzed and found, such as the edge, the nodes that bear concentrated loads and the key parts of the structural deformation, and the influence of environmental factors such as surrounding vibration sources and actual use conditions such as equipment installation on the node displacement is considered. Finally, the mid-span deflection change zone is determined, the mid-span area of ​​the large-span member is clarified based on the principles of structural mechanics, the deflection is calculated using theoretical formulas, and the finite element model is used for simulation and verification. The range near the mid-span is appropriately expanded considering practical factors such as material unevenness.

[0027] Step S320, based on the stress distribution characteristics of the high stress concentration area, node displacement sensitive area and mid-span deflection change area in the dynamic monitoring area, a monitoring sensor network is arranged, and the monitoring sensor network includes strain sensors, displacement sensors and acceleration sensors. Specifically, when arranging the monitoring sensor network, for the high stress concentration area, foil strain gauge sensors are selected, and 3-4 are evenly arranged every 5-10 cm around the weld of the welding node, 5 cm radius near the bolt hole of the bolt connection node, and every 8-10 cm along the edge of the connection between the node plate and the rod. At the same time, temperature self-compensation strain gauges or double bridge measurement circuits are used to compensate for temperature effects. In the node displacement sensitive area, for each node with large displacement changes, at least one laser displacement sensor or LVDT is installed, and the focus is on the edge, the nodes that bear concentrated loads and the key parts of structural deformation, and the piezoelectric acceleration sensor is installed at a suitable position near the node to monitor vibration acceleration. In the mid-span deflection change area, displacement sensors are arranged every 1-2 meters along the mid-span area of ​​the large-span rod, and acceleration sensors are installed at a suitable position nearby. During deployment, ensure that the sensors are firmly installed and the lines are reliable, and mark them with numbers to build a comprehensive and efficient monitoring sensor network.

[0028] Step S400: Compare and analyze the collected real-time strain sensor data with the simulated stress-strain response to identify the abnormal deformation characteristics of the steel grid structure members and generate a deformation correction coefficient. Specifically, first collect the real-time strain sensor data and preprocess it. At the same time, obtain the simulated stress-strain response data under the same working conditions from the finite element model. Then match the two in space and time, use the strain difference, ratio, etc. as comparison indicators, calculate and plot the comparison curve, and set a threshold to screen out abnormal data points. Then, based on the characteristics of the steel grid structure, judge the type of abnormal deformation, and determine the location of the abnormal deformation according to the sensor position. Next, select a suitable method to calculate the deformation correction coefficient according to the abnormal situation, such as correcting based on the material elastic modulus or structural stiffness, apply the coefficient to the model for re-analysis, and adjust the coefficient until the difference is acceptable after comparing the results. Finally, record the collected data, analysis results, abnormal characteristics, correction coefficient, etc., and generate a report including the analysis process, abnormal description, coefficient calculation, and structural evaluation suggestions to provide support for the health monitoring and maintenance of the steel grid.

[0029] In a possible implementation, when comparing and analyzing the collected real-time strain sensor data with the simulated stress-strain response to identify the abnormal deformation characteristics of the steel grid structure members and generate a deformation correction coefficient, step S400 further includes step S410: perform wavelet packet decomposition on the real-time strain sensor data to extract the energy distribution characteristics of different frequency bands. Specifically, to perform wavelet packet decomposition on the real-time strain sensor data and extract the energy distribution characteristics of different frequency bands, data preparation needs to be done first. The strain sensor should be accurately installed at the key positions of the steel grid, and a stable acquisition system should be used to collect data at an appropriate sampling frequency. Then, filter out the noise, and use min-max or Z-score normalization to eliminate the measurement differences and magnitude effects. Next, perform wavelet packet decomposition. Select a suitable wavelet basis function such as Daubechies according to the data characteristics and analysis purposes, and determine the decomposition level of 3-5 layers according to the data frequency range and analysis requirements. Then complete the decomposition operation. After that, extract the energy distribution characteristics. Divide the frequency band according to equal bandwidth or logarithmic bandwidth, calculate the sum of the squares of the wavelet packet coefficients of each frequency band to obtain the energy value, and then normalize the energy value. Finally, verify the results. Cross-validation or comparison with other methods can be used. Also, analyze the characteristics, observe the energy distribution to judge whether there are abnormalities, and provide a basis for the health monitoring and fault diagnosis of the steel grid.

[0030] Step S420: Perform a matching analysis on the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response, and configure the abnormal deformation feature matrix. Specifically, for the matching analysis of the energy distribution characteristics of different frequency bands and the frequency components of the simulated stress-strain response and the configuration of the abnormal deformation feature matrix, data preparation should be done first. Sort out the energy distribution characteristic data obtained by wavelet packet decomposition and arrange them in the order of frequency bands; perform dynamic analysis on the finite element model of the composite steel grid structure, extract the frequency components and corresponding energy values of the simulated stress-strain response through Fourier transform and sort them. Then carry out the matching analysis. First, align the frequency ranges of the two, and make the ranges consistent through cropping or interpolation; then select methods such as Euclidean distance and cosine similarity to calculate the similarity, judge whether the difference is significant according to the set threshold and record the relevant data. Finally, configure the abnormal deformation feature matrix. Design the matrix structure, with rows representing the monitoring positions and columns representing the frequency bands; fill the matrix with the results of the matching analysis, and set specific values at the places where the difference is not significant; optimize the matrix, such as data smoothing and outlier removal, and verify it by methods such as comparison with known anomalies or cross-validation. If the result is not good, adjust the parameters and configuration methods to ensure that the matrix can effectively identify the abnormal deformation features and provide a basis for the health monitoring and fault diagnosis of the steel grid structure.

[0031] Step S430: Based on the abnormal deformation feature matrix, determine the deformation correction coefficient, which is used to adjust the boundary conditions of the twin model of the composite steel grid structure. Specifically, for the matching analysis of the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response and the configuration of the abnormal deformation feature matrix, the following steps should be taken. First, perform data acquisition and preprocessing. Arrange strain sensors at the key parts of the steel grid structure to collect real-time strain data, and obtain the energy distribution characteristics through wavelet packet decomposition and normalization processing; at the same time, construct a finite element model to simulate the actual working conditions, and perform Fourier transform on the time-domain signal of the simulated stress-strain response to extract the frequency components and energy values. Then carry out the matching analysis. Unify the frequency ranges and resolutions of the two, select methods such as Euclidean distance and cosine similarity to calculate the similarity, judge whether it is abnormal according to the set threshold and record the information. Finally, configure the abnormal deformation feature matrix. The number of rows of the matrix corresponds to the number of monitoring positions, the number of columns corresponds to the number of frequency bands, and the elements are similarity index values. Fill the matrix with the results of the matching analysis, and then perform optimization processing such as data smoothing and outlier removal. Evaluate the matrix by comparison with known anomalies or cross-validation. If the result is not good, adjust the parameters and configuration methods to ensure that the matrix can effectively identify the abnormal deformation features and provide support for structural health monitoring and fault diagnosis.

[0032] In a possible implementation, a matching analysis is performed on the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response, and an abnormal deformation feature matrix is configured. Step S420 further includes step S421, which performs a matching analysis on the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response, and combines a preset matching degree threshold to perform abnormal deformation identification. Specifically, to perform a matching analysis on the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response and combine a preset matching degree threshold to perform abnormal deformation identification, data preparation needs to be done first. The real-time strain data of the steel grid structure is collected by a strain sensor, and the energy distribution characteristics are obtained through wavelet packet decomposition, energy calculation, and normalization; a finite element model is constructed to simulate the stress, and the frequency component data of the simulated stress-strain response is obtained through Fourier transform. Then, a matching analysis is carried out, the frequency ranges and resolutions of the two are unified, and methods such as Euclidean distance and cosine similarity are selected to calculate the similarity and organize the results. The preset matching degree threshold is determined through historical data statistics or simulation tests, and the similarity obtained from the matching analysis is compared with the threshold. If it exceeds the threshold, it is determined that there is abnormal deformation in the corresponding part of the steel grid structure and it is marked and recorded. Finally, methods such as on-site inspection or comparison with other monitoring means are used to verify the results. If there are misjudgments or missed judgments, the threshold is adjusted and optimized to achieve accurate abnormal deformation identification, providing a basis for the health monitoring of the steel grid structure.

[0033] Step S422, through the abnormal deformation identification results, determine the spatio-temporal distribution characteristics of the abnormal deformation area. Specifically, to determine the spatio-temporal distribution characteristics of the abnormal deformation area through the abnormal deformation identification results, data integration and preprocessing need to be carried out first, collect the abnormal deformation identification information, associate the sensor position data, and organize it into an ordered time series. Then, analyze the spatial distribution characteristics, use GIS or 3D modeling software to visualize the abnormal positions, and use clustering and correlation analysis to determine the high-incidence areas and spatial relationships. Then, carry out the analysis of the time distribution characteristics, count the number and frequency of abnormal occurrences, use time series analysis to predict the trend, and analyze the abnormal duration. Then, conduct a spatio-temporal comprehensive analysis, construct a spatio-temporal cube, mine spatio-temporal association rules, and analyze the spatio-temporal evolution trend. Finally, verify the results, compare them with the results of on-site inspections or other monitoring means to ensure accuracy, and generate a detailed report including spatial, temporal, spatio-temporal comprehensive characteristics, and verification results to provide support for the health monitoring and maintenance decision-making of the steel grid structure.

[0034] Step S423: Generate an abnormal deformation feature matrix containing position encoding, deformation magnitude, and evolution trend according to the spatio-temporal distribution characteristics of the abnormal deformation area. Specifically, to determine the spatio-temporal distribution characteristics of the abnormal deformation area through the abnormal deformation identification results, the following steps need to be carried out. First is data collection and integration, summarizing the abnormal deformation identification data, supplementing relevant data such as the steel grid structure, environment, and usage conditions, and cleaning and standardizing the data. Then conduct spatial distribution feature analysis, display the abnormal positions using GIS or visualization tools, and find the high-incidence areas and spatial correlations through clustering and autocorrelation analysis. Next, carry out time distribution feature analysis, draw a time series graph, and use clustering and trend analysis methods to judge the time law and trend of abnormal deformation. After that, conduct spatio-temporal comprehensive analysis, construct a spatio-temporal cube, mine spatio-temporal association rules, and analyze the spatio-temporal evolution trend. Finally, verify the results, adopt cross-validation or comparison with other data sources to ensure accuracy, interpret the results in combination with the actual situation of the steel grid, and provide targeted suggestions for maintenance and management.

[0035] Step S500: Adjust the boundary conditions of the combined steel grid twin model according to the abnormal deformation characteristics and deformation correction coefficients, and iteratively optimize the dynamic monitoring threshold of the steel grid components. Specifically, to adjust the boundary conditions of the combined steel grid twin model and iteratively optimize the dynamic monitoring threshold of the steel grid components according to the abnormal deformation characteristics and deformation correction coefficients, first, data preparation and evaluation need to be done, sorting out the abnormal deformation characteristics and deformation correction coefficients, and evaluating the existing twin model and monitoring threshold. Then determine the boundary condition adjustment strategy based on this data and implement the adjustment, and re-run the model to obtain the response results. After that, compare the model responses before and after the adjustment, evaluate the error after the adjustment, and if the error is large, further analyze and adjust. Then update the monitoring threshold based on the new model results and verify its effectiveness. If it is not ideal, iteratively adjust and optimize. Finally, use multiple methods to verify the finally adjusted twin model and the optimized monitoring threshold, summarize the whole process, and write a report containing analysis results, adjustment steps, verification results, and suggestions, providing a reference for the health monitoring and maintenance management of the steel grid.

[0036] In a possible implementation manner, according to the abnormal deformation characteristics and deformation correction coefficients, the boundary conditions of the combined steel space truss twin model are adjusted, and the dynamic monitoring threshold of the steel space truss members is iteratively optimized. Step S500 further includes step S510 of locally refining the mesh density in the abnormal deformation region based on the deformation correction coefficient. Specifically, to locally refine the mesh density in the abnormal deformation region based on the deformation correction coefficient, data preparation and analysis should be done first. The deformation correction coefficients reflecting the differences between the actual and simulated deformations are collected, the abnormal regions are determined according to the abnormal deformation identifiers and spatio-temporal distribution characteristics, and the correlation between the two is analyzed. Then, a mesh refinement strategy is formulated. The refinement criteria are set according to the deformation correction coefficient, the mesh type is selected in combination with the regional characteristics, and a reasonable transition method is determined to ensure the continuity of the mesh. After that, the model is imported into the finite element analysis software, local mesh refinement is performed on the abnormal region, and then the overall mesh is checked. After the refinement is completed, the model is re-analyzed and calculated, the error between the calculated result and the actual monitoring data is evaluated and the reasons are analyzed. If the error is large, the strategy is adjusted for optimization. Finally, the refinement process and results are recorded, the optimized refined model is applied to the actual monitoring and analysis of the steel space truss, and the model is continuously updated and optimized according to the feedback.

[0037] Step S520, meanwhile, update the confidence interval corresponding to the dynamic monitoring threshold. Specifically, to update the confidence interval corresponding to the dynamic monitoring threshold, the actual monitoring data, deformation correction coefficients, and existing dynamic monitoring threshold and confidence interval information of the abnormal deformation region need to be collected first, and the data is cleaned and integrated. Then, the characteristics of the monitoring data are analyzed, statistical analysis and time series analysis are carried out, the data distribution is judged, and the correlation between the deformation correction coefficient and the monitoring data is investigated. Then, the update method is determined, and methods based on statistical theory or machine learning can be used. After that, the new confidence interval is calculated according to the selected method and the dynamic monitoring threshold is adjusted. Then, the update result is verified and evaluated through historical data and actual monitoring. If the effect is not good, it is readjusted. Finally, the update process and results are recorded in detail, and the updated confidence interval and threshold are applied to the actual monitoring system, and the update is evaluated regularly to adapt to the change of the structural state.

[0038] Step S600: Establish a hierarchical early warning mechanism based on the optimized dynamic monitoring threshold to give structural health alerts for the steel mesh structure members. Specifically, to establish a hierarchical early warning mechanism based on the optimized dynamic monitoring threshold to give structural health alerts for the steel mesh structure members, first, sort out and confirm the optimized dynamic monitoring threshold to ensure its accuracy and reliability. Then design the hierarchical early warning mechanism, divide warning levels such as level 1 (low risk), level 2 (medium risk), level 3 (high risk), etc., clarify the triggering conditions for each level, and formulate corresponding response measures. After that, build and integrate the early warning system, prepare the hardware equipment, develop the software and integrate it with the existing monitoring and management platform, and conduct system testing and debugging. When implementing the structural health alert, conduct real-time monitoring, and promptly issue alerts with detailed information after the warning is triggered, handle it according to the corresponding measures and track the records. Finally, regularly evaluate the accuracy, timeliness, and effectiveness of the response measures of the early warning mechanism, optimize and adjust the triggering conditions and response measures according to the results, and update the dynamic monitoring threshold to ensure that the early warning mechanism adapts to the actual situation.

[0039] The embodiment of this application adopts obtaining the basic structural data and environmental action parameters of the steel grid structure, constructing a combined steel grid structure twin model, simulating the stress-strain response under the coupling of bidirectional loads, and dividing the dynamic monitoring area. By comparing the real-time strain sensor data with the simulated response, identify the abnormal deformation characteristics and generate the deformation correction coefficient. Based on the correction coefficient, adjust the model boundary conditions, iteratively optimize the dynamic monitoring threshold, establish a hierarchical early warning mechanism, and realize the structural health alert of the steel grid structure. Through the new material detection technology (probability distribution analysis of the yield strength of steel) combined with the standardized process (dynamic threshold calibration), it achieves the high-precision real-time monitoring of the stress-strain response under the coupling of bidirectional loads and realizes the technical effect of hierarchical early warning of dynamic thresholds.

[0040] In the above text, with reference to Figure 1 A deformation monitoring method for a bidirectional long-span space combined steel grid structure according to an embodiment of the present invention is described in detail. Next, with reference to Figure 2 A deformation monitoring device for a bidirectional long-span space combined steel grid structure according to an embodiment of the present invention will be described.

[0041] A deformation monitoring device for a two-way long-span space composite steel grid structure according to an embodiment of the present invention solves the technical problems of limited accuracy and poor adaptability of dynamic thresholds in the existing deformation monitoring methods for long-span steel grid structures under complex environmental coupled loads. Through new material detection technology (probability distribution analysis of steel yield strength) combined with a standardized process (dynamic threshold calibration), high-precision real-time monitoring of stress-strain responses under two-way load coupling is achieved, and the technical effect of dynamic threshold classification early warning is realized. A deformation monitoring device for a two-way long-span space composite steel grid structure includes: a basic structure data acquisition module 10, an environmental action parameter acquisition module 20, a dynamic monitoring area division module 30, an abnormal deformation feature failure module 40, a boundary condition adjustment module 50, and a classification early warning mechanism construction module 60.

[0042] The basic structure data acquisition module 10 is used to acquire the basic structure data of the steel grid structure components, including rod geometric parameters, node connection types, and material mechanical property parameters.

[0043] The environmental action parameter acquisition module 20 is used to acquire environmental action parameters including temperature gradient distribution, wind load time history data, and vibration frequency spectrum, and combine with the basic structure data to construct a twin model of the composite steel grid structure to simulate the stress-strain response under two-way load coupling.

[0044] The dynamic monitoring area division module 30 is used to divide the dynamic monitoring area in the twin model of the composite steel grid structure at the same time.

[0045] The abnormal deformation feature failure module 40 is used to compare and analyze the collected real-time strain sensor data with the simulated stress-strain response, identify the abnormal deformation features of the steel grid structure components, and generate a deformation correction coefficient.

[0046] The boundary condition adjustment module 50 is used to adjust the boundary conditions of the twin model of the composite steel grid structure according to the abnormal deformation features and the deformation correction coefficient, and iteratively optimize the dynamic monitoring threshold of the steel grid structure components.

[0047] The classification early warning mechanism construction module 60 is used to establish a classification early warning mechanism with the optimized dynamic monitoring threshold to give a structural health reminder to the steel grid structure components.

[0048] Next, the specific configuration of the basic structure data acquisition module 10 will be described in detail. As described above, the basic structure data of the steel mesh structure members is acquired, including rod geometric parameters, node connection types, and material mechanical property parameters. The basic structure data acquisition module 10 further includes: a geometric topology structure acquisition unit, which is used to acquire the geometric topology structure of the combined steel grid and extract the rod length deviation and node coordinate offset; a yield strength probability distribution function calibration unit, which is used to calibrate the yield strength probability distribution function in the material mechanical property parameters based on the geometric topology structure, rod length deviation, and node coordinate offset.

[0049] Among them, based on the geometric topology structure, rod length deviation, and node coordinate offset, the yield strength probability distribution function in the material mechanical property parameters is calibrated. The yield strength probability distribution function calibration unit further includes: a yield strength probability distribution function subunit, which is used for the yield strength probability distribution function ; where is the probability distribution function of the yield strength, S is the yield strength of the steel mesh structure member, is the mean value of the yield strength, is the standard deviation of the yield strength.

[0050] Next, the specific configuration of the environmental action parameter acquisition module 20 will be described in detail. As described above, environmental action parameters including temperature gradient distribution, wind load time history data, and vibration frequency spectrum are collected, and combined with the basic structure data, a twin model of the combined steel grid is constructed to simulate the stress-strain response under the coupled action of bidirectional loads. The environmental action parameter acquisition module 20 further includes: a change law recording unit, which is used to arrange a distributed temperature sensor array to record the change law of the temperature gradient distribution over time; a dominant vibration frequency component extraction unit, which is used to capture the wind load time history data and extract the dominant vibration frequency component through spectrum analysis; a vibration frequency spectrum construction unit, which is used to synchronously collect the multi-point vibration acceleration signals of the combined steel grid based on the monitoring sensor network and construct a vibration frequency spectrum.

[0051] Among them, the environmental action parameter acquisition module 20 further includes: an initial model architecture construction unit for establishing an initial model architecture according to the geometric parameters of the members and the node connection types; a material mechanical property parameter import unit for importing the material mechanical property parameters based on the initial model architecture, defining non-linear contact boundary conditions and time-varying load parameters; and a transient dynamics analysis unit for performing transient dynamics analysis based on the non-linear contact boundary conditions and time-varying load parameters and outputting a dynamic response contour map of the steel grid under the coupling action of temperature - wind load - vibration.

[0052] Next, the specific configuration of the dynamic monitoring area division module 30 will be described in detail. As described above, at the same time, a dynamic monitoring area is divided in the combined steel grid twin model. The dynamic monitoring area division module 30 further includes: a dynamic monitoring area composition unit for the dynamic monitoring area including high stress concentration areas, node displacement sensitive areas, and mid-span deflection change areas; and a monitoring sensor network layout unit for laying out a monitoring sensor network based on the stress distribution characteristics of the high stress concentration areas, node displacement sensitive areas, and mid-span deflection change areas in the dynamic monitoring area. The monitoring sensor network includes strain sensors, displacement sensors, and acceleration sensors.

[0053] Next, the specific configuration of the abnormal deformation feature failure module 40 will be described in detail. As described above, the collected real-time strain sensor data is compared and analyzed with the simulated stress-strain response to identify the abnormal deformation characteristics of the steel grid members and generate a deformation correction coefficient. The abnormal deformation feature failure module 40 further includes: an energy distribution feature extraction unit for performing wavelet packet decomposition on the real-time strain sensor data to extract the energy distribution characteristics of different frequency bands; a matching analysis unit for performing matching analysis on the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response to configure an abnormal deformation feature matrix; and a deformation correction coefficient determination unit for determining the deformation correction coefficient based on the abnormal deformation feature matrix. The deformation correction coefficient is used to adjust the boundary conditions of the combined steel grid twin model.

[0054] Among them, the energy distribution characteristics of different frequency bands are matched with the frequency components in the simulated stress-strain response to configure an abnormal deformation feature matrix. The matching analysis unit further includes: an abnormal deformation identification subunit, which is used to perform matching analysis on the energy distribution characteristics of different frequency bands and the frequency components in the simulated stress-strain response, and combine a preset matching degree threshold to perform abnormal deformation identification; a spatio-temporal distribution characteristic determination subunit, which is used to determine the spatio-temporal distribution characteristics of the abnormal deformation region through the abnormal deformation identification result; an abnormal deformation feature matrix generation subunit, which is used to generate an abnormal deformation feature matrix including position encoding, deformation magnitude, and evolution trend according to the spatio-temporal distribution characteristics of the abnormal deformation region.

[0055] Next, the specific configuration of the boundary condition adjustment module 50 will be described in detail. As described above, according to the abnormal deformation characteristics and the deformation correction coefficient, the boundary conditions of the combined steel grid twin model are adjusted, and the dynamic monitoring threshold of the steel grid members is iteratively optimized. The boundary condition adjustment module 50 further includes: a local refined grid density unit, which is used to locally refine the grid density in the abnormal deformation region based on the deformation correction coefficient; a confidence interval update unit, which is used to update the confidence interval corresponding to the dynamic monitoring threshold at the same time.

[0056] The deformation monitoring device for a two-way large-span space combined steel grid provided by the embodiments of the present invention can execute the deformation monitoring method for a two-way large-span space combined steel grid provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method.

[0057] Although various references are made to certain modules in the systems according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or the server. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of mutual distinction and do not limit the protection scope of the present invention.

[0058] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A deformation monitoring method for a bidirectional large-span spatial composite steel grid, characterized in that: The method comprises: Obtain basic structural data of steel grid members, including member geometry parameters, node connection types, and material mechanical properties parameters; Collect environmental parameters including temperature gradient distribution, wind load time history data and vibration frequency spectrum, and build a combined steel grid twin model in combination with the foundation structure data to simulate the stress-strain response under the coupling of bidirectional loads; At the same time, a dynamic monitoring area is divided in the combined steel grid twin model; Compare and analyze the collected real-time strain sensor data with the simulated stress-strain response to identify abnormal deformation characteristics of steel grid members and generate deformation correction coefficients; According to the abnormal deformation characteristics and deformation correction coefficients, the boundary conditions of the combined steel grid twin model are adjusted, and the dynamic monitoring threshold of the steel grid components is iteratively optimized; A graded early warning mechanism is established with the optimized dynamic monitoring threshold to provide structural health reminders for the steel grid members.

2. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 1, characterized in that: The dynamic monitoring area includes a high stress concentration area, a node displacement sensitive area and a mid-span deflection change area; Based on the stress distribution characteristics of the high stress concentration area, the node displacement sensitive area and the mid-span deflection change area in the dynamic monitoring area, a monitoring sensor network is deployed, and the monitoring sensor network includes strain sensors, displacement sensors and acceleration sensors.

3. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 2, characterized in that: Collecting environmental action parameters including temperature gradient distribution, wind load time history data and vibration frequency spectrum, the method includes: Arrange a distributed temperature sensor array to record the change of temperature gradient distribution over time; Capture wind load time history data and extract dominant vibration frequency components through spectrum analysis; Based on the monitoring sensor network, multi-point vibration acceleration signals of the combined steel grid are synchronously collected to construct a vibration frequency spectrum.

4. The deformation monitoring method of a bidirectional large-span spatial composite steel grid according to claim 1, characterized in that: Obtaining basic structural data of a steel grid structure, the method comprising: Obtain the geometric topological structure of the composite steel grid and extract the length deviation of the rods and the offset of the node coordinates; Based on the geometric topological structure, rod length deviation, and node coordinate offset, the yield strength probability distribution function in the material mechanical property parameters is calibrated.

5. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 4, characterized in that: The yield strength probability distribution function ; in, is the probability distribution function of yield strength, S is the yield strength of the steel grid member, is the mean yield strength, is the standard deviation of yield strength.

6. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 4, characterized in that: Combining the basic structure data, a combined steel grid twin model is constructed, and the method includes: Establish the initial model framework based on the bar geometry parameters and node connection types; Based on the initial model framework, the material mechanical property parameters are imported, and nonlinear contact boundary conditions and time-varying load parameters are defined; Based on nonlinear contact boundary conditions and time-varying load parameters, transient dynamic analysis is performed to output the dynamic response cloud diagram of the steel grid under the coupling of temperature, wind load and vibration.

7. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 6, characterized in that: Comparing and analyzing the collected real-time strain sensor data with the simulated stress-strain response, identifying abnormal deformation characteristics of the steel grid structure, and generating a deformation correction coefficient, the method further includes: Performing wavelet packet decomposition on the real-time strain sensor data to extract energy distribution characteristics of different frequency bands; Match the energy distribution characteristics of different frequency bands with the frequency components in the simulated stress-strain response and configure the abnormal deformation feature matrix; Based on the abnormal deformation characteristic matrix, the deformation correction coefficient is determined, and the deformation correction coefficient is used to adjust the boundary conditions of the combined steel grid twin model.

8. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 7, characterized in that: Matching and analyzing the energy distribution characteristics of different frequency bands with the frequency components in the simulated stress-strain response, and configuring an abnormal deformation feature matrix, the method includes: The energy distribution characteristics of different frequency bands are matched with the frequency components in the simulated stress-strain response, and abnormal deformation is identified based on the preset matching threshold. Through the abnormal deformation identification results, the temporal and spatial distribution characteristics of the abnormal deformation area are determined; According to the spatiotemporal distribution characteristics of the abnormal deformation area, an abnormal deformation feature matrix including position coding, deformation magnitude and evolution trend is generated.

9. A deformation monitoring method for a bidirectional large-span spatial composite steel grid as claimed in claim 7, characterized in that: Iteratively optimizing a dynamic monitoring threshold of a steel grid member, the method comprising: Based on the deformation correction coefficient, locally refine the mesh density in the abnormal deformation area; At the same time, the confidence interval corresponding to the dynamic monitoring threshold is updated.

10. A deformation monitoring device for a bidirectional large-span spatial composite steel grid, characterized in that: The device is used to implement the deformation monitoring method of a bidirectional large-span spatial composite steel grid according to any one of claims 1 to 9, and the device comprises: A basic structure data acquisition module, which is used to acquire basic structure data of steel grid members, including bar geometry parameters, node connection types, and material mechanical property parameters; An environmental action parameter acquisition module, which is used to collect environmental action parameters including temperature gradient distribution, wind load time history data and vibration frequency spectrum, and to construct a combined steel grid twin model in combination with the foundation structure data to simulate the stress-strain response under the coupling of bidirectional loads; A dynamic monitoring area division module, the dynamic monitoring area division module is used to simultaneously divide the dynamic monitoring area in the combined steel grid twin model; An abnormal deformation feature failure module, which is used to compare and analyze the collected real-time strain sensor data with the simulated stress-strain response, identify the abnormal deformation features of the steel grid structure, and generate a deformation correction coefficient; A boundary condition adjustment module, which is used to adjust the boundary conditions of the combined steel grid twin model according to the abnormal deformation characteristics and the deformation correction coefficient, and iteratively optimize the dynamic monitoring threshold of the steel grid components; A hierarchical early warning mechanism construction module is used to establish a hierarchical early warning mechanism with an optimized dynamic monitoring threshold value to provide structural health reminders for the steel grid structure components.

Images