Large steel structure hoisting precision real-time monitoring method and system

Abstract

The invention relates to the technical field of data processing, and discloses a large steel structure hoisting precision real-time monitoring method and system. The method comprises the following steps: scanning and collecting the edge of a steel structure through a distributed sensor network, and obtaining structure data to obtain an original monitoring data set; extracting and analyzing structure transient response features based on the original monitoring data set, and generating a state feature matrix; analyzing a coupling relationship between local deformation and overall torsion to obtain a dynamic response prediction matrix; performing space-time registration and data fusion on the monitoring data, and constructing a state evaluation data set; carrying out optimization calculation on the hoisting parameters to generate a hoisting control data packet; and performing intelligent reasoning analysis in combination with historical case data, and outputting a control instruction sequence. Real-time monitoring of the coupling relation between local deformation and overall torsional deformation in the hoisting process of the large steel structure is achieved, and the technical defect that in the prior art, the structural deformation development trend cannot be accurately recognized and predicted is overcome.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular, to a method and system for real-time monitoring of the hoisting accuracy of large steel structures. Background Art

[0002] In the construction of large steel structure projects, hoisting is a key construction link. At present, the hoisting accuracy monitoring of large steel structures mainly relies on traditional measurement means, such as total stations, levels and other optical measurement devices for fixed-point measurement, or local strain sensors for local deformation monitoring. At the same time, some engineering projects have begun to adopt monitoring methods based on computer vision, by deploying cameras to track the hoisting process in real time, and combining image processing technology to analyze the displacement and deformation of the structure. These monitoring means have accumulated rich engineering experience and technical data in practical applications.

[0003] However, the existing monitoring methods have obvious deficiencies. First, traditional optical measurement devices can only obtain discrete measuring point data, and it is difficult to achieve continuous monitoring of the overall deformation of the structure; second, although local strain monitoring can obtain continuous data, it is difficult to reflect the overall deformation characteristics of the structure; third, the monitoring method based on computer vision is easily affected by environmental factors, such as light changes, occlusion, etc. will affect the accuracy of monitoring; most importantly, the existing monitoring methods generally lack the ability to analyze the coupling relationship between local deformation and overall torsional deformation of the structure, and it is difficult to accurately predict and control the development trend of deformation during the hoisting process. Summary of the Invention

[0004] This application provides a method and system for real-time monitoring of the hoisting accuracy of large steel structures, which is used to realize the real-time monitoring of the coupling relationship between local deformation and overall torsional deformation during the hoisting process of large steel structures, and overcome the technical defect that the development trend of structural deformation cannot be accurately identified and predicted in the prior art.

[0005] In a first aspect, the present application provides a method for real-time monitoring of the hoisting accuracy of large steel structures. The method for real-time monitoring of the hoisting accuracy of large steel structures includes: scanning and collecting the edges of the steel structure through a distributed sensing network to obtain structural angular velocity data, strain data, and load distribution data, and obtaining an original monitoring data set; extracting and analyzing the transient response characteristics of the structure according to the original monitoring data set to generate a structural state characteristic matrix including local deformation data and overall torsion data; calculating and analyzing the coupling relationship between local deformation and overall torsion based on the structural state characteristic matrix to obtain torsion deformation evolution data and stress distribution prediction data, and integrating them to obtain a dynamic response prediction matrix; based on the dynamic response prediction matrix, performing spatio-temporal registration and data fusion on the real-time collected monitoring data to construct a state evaluation data set including torsion deformation indexes and stress distribution indexes; performing dynamic optimization calculation on the hoisting parameters according to the state evaluation data set to generate a hoisting control data packet including hoisting point position parameters, hoisting moment parameters, and compensation control parameters; using the hoisting control data packet and combining historical case data for intelligent reasoning analysis to output a real-time control instruction sequence to achieve the accuracy monitoring and control of the hoisting process.

[0006] In a second aspect, the present application provides a system for real-time monitoring of the hoisting accuracy of large steel structures. The system for real-time monitoring of the hoisting accuracy of large steel structures includes: A collection module for scanning and collecting the edges of the steel structure through a distributed sensing network to obtain structural angular velocity data, strain data, and load distribution data, and obtaining an original monitoring data set; An extraction module for extracting and analyzing the transient response characteristics of the structure according to the original monitoring data set to generate a structural state characteristic matrix including local deformation data and overall torsion data; A coupling module for calculating and analyzing the coupling relationship between local deformation and overall torsion based on the structural state characteristic matrix to obtain torsion deformation evolution data and stress distribution prediction data, and integrating them to obtain a dynamic response prediction matrix; A fusion module for performing spatio-temporal registration and data fusion on the real-time collected monitoring data based on the dynamic response prediction matrix to construct a state evaluation data set including torsion deformation indexes and stress distribution indexes; An optimization module for performing dynamic optimization calculation on the hoisting parameters according to the state evaluation data set to generate a hoisting control data packet including hoisting point position parameters, hoisting moment parameters, and compensation control parameters; A control module for using the hoisting control data packet and combining historical case data for intelligent reasoning analysis to output a real-time control instruction sequence to achieve the accuracy monitoring and control of the hoisting process.

[0007] In the technical solution provided by this application, by using a distributed sensing network to scan and collect the edges of the steel structure, multi-source heterogeneous data collection of structural angular velocity data, strain data, and load distribution data is achieved, solving the problem that it is difficult for traditional single sensors to comprehensively monitor the structural state; by extracting and analyzing the transient response characteristics of the original monitoring data set, a structural state feature matrix containing local deformation data and overall torsion data is generated, achieving a comprehensive characterization of the structural deformation characteristics; using a coupling relationship calculation and analysis method, the correlation between local deformation and overall torsion is deeply analyzed to obtain torsion deformation evolution data and stress distribution prediction data, and these data are integrated to form a dynamic response prediction matrix, providing a reliable basis for structural deformation prediction; based on the dynamic response prediction matrix, spatio-temporal registration and data fusion are performed on the real-time collected monitoring data to construct a state evaluation data set containing torsion deformation indicators and stress distribution indicators, improving the spatio-temporal consistency of the monitoring data; according to the state evaluation data set, a dynamic optimization algorithm is used to calculate the lifting parameters in real time, generating a lifting control data packet containing hoisting point position parameters, lifting moment parameters, and compensation control parameters, making the generation of control parameters more scientific; by combining historical case data for intelligent reasoning and analysis, a real-time control instruction sequence is output, realizing the closed-loop optimization of the accuracy monitoring and control during the lifting process. In terms of the application of artificial intelligence algorithms, this invention innovatively combines deep learning with traditional mechanical models. By mining and learning historical data, a prediction model more in line with engineering reality is established. At the same time, an intelligent reasoning algorithm is used to dynamically optimize the control strategy, significantly improving the accuracy and real-time performance of the control instructions, making the entire monitoring and control scheme more adaptable and robust. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0009] Figure 1 It is a schematic diagram of an embodiment of the real-time monitoring method for the lifting accuracy of a large steel structure in an embodiment of this application; Figure 2 It is a timing diagram for extracting and analyzing the transient response characteristics of the structure in an embodiment of this application; Figure 3 It is a schematic diagram of an embodiment of the real-time monitoring system for the lifting accuracy of a large steel structure in an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0010] The embodiments of the present application provide a method and system for real-time monitoring of the hoisting accuracy of large steel structures. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and the above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data used can be interchanged under appropriate circumstances, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "including" or "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0011] For ease of understanding, the specific process of the embodiments of the present application will be described below. Please refer to Figure 1 , an embodiment of the method for real-time monitoring of the hoisting accuracy of large steel structures in the embodiments of the present application includes: Step S101: Scan and collect the edges of the steel structure through a distributed sensing network to obtain structural angular velocity data, strain data, and load distribution data, and obtain an original monitoring data set; Step S102: Extract and analyze the structural transient response characteristics according to the original monitoring data set, and generate a structural state characteristic matrix including local deformation data and overall torsion data; Step S103: Calculate and analyze the coupling relationship between local deformation and overall torsion based on the structural state characteristic matrix, obtain torsion deformation evolution data and stress distribution prediction data, and integrate them to obtain a dynamic response prediction matrix; Step S104: Based on the dynamic response prediction matrix, perform spatio-temporal registration and data fusion on the real-time collected monitoring data, and construct a state evaluation data set including torsion deformation indicators and stress distribution indicators; Step S105: Dynamically optimize and calculate the hoisting parameters according to the state evaluation data set, and generate a hoisting control data packet including hoisting point position parameters, hoisting moment parameters, and compensation control parameters; Step S106: Use the hoisting control data packet to perform intelligent reasoning analysis in combination with historical case data, and output a real-time control instruction sequence to realize the accuracy monitoring and control of the hoisting process.

[0012] It can be understood that the execution entity of the present application can be a system for real-time monitoring of the hoisting accuracy of large steel structures, or a terminal or a server. Specifically, it is not limited here. The embodiments of the present application will be described by taking the server as the execution entity as an example.

[0013] Specifically, the real-time monitoring method for the hoisting accuracy of large steel structures mainly focuses on the accuracy control problem of large steel structure components during the hoisting process. In the specific implementation process, data collection is carried out through a distributed sensing network, where the distributed sensing network includes fiber Bragg grating sensors, gyroscope arrays, and dynamic strain sensors. The fiber Bragg grating sensors are arranged in a spiral pattern along the edge of the steel structure at intervals of 0.5 meters to perform high-frequency sampling of the strain at the structure edge. The initial strain signal is obtained through wavelength demodulation analysis, and then the wavelet transform is used to perform multi-scale decomposition on the signal. The decomposed signals in each frequency band are processed through threshold processing and reconstruction to generate the strain data at the structure edge. The gyroscope array collects the three-axis angular velocity of key nodes at a sampling frequency of 200 Hz. The collected original data is screened through a band-pass filter to remove high-frequency noise and low-frequency drift, and then the Kalman filter is used for state estimation to obtain the angular velocity data. The dynamic strain sensor monitors the stress points of the hoisting equipment. The collected load time-series data is transformed into the frequency domain through the fast Fourier transform, and after identifying and eliminating the interference frequency components, the inverse transform is performed to obtain the load distribution data.

[0014] When extracting the transient response characteristics of the data in the original monitoring dataset, data segments within the key time window are extracted through time-domain segmentation to generate time-series characteristic data. Spatial correlation analysis is performed on the data of each measuring point, and the correlation coefficient matrix between the measuring points is calculated and the key measuring point combinations are screened. The data of the key measuring point combinations are decomposed in multiple dimensions to extract the displacement, velocity, and acceleration components to obtain the local response characteristics. Through data reconstruction, the local response characteristics are spatially mapped to establish the node displacement field to obtain the local deformation data. According to the local deformation data, overall coordination analysis is carried out, the deformation gradient and deformation rate of each region are calculated, and the overall torsion data is constructed. The local deformation data and the overall torsion data are organized and arranged according to the spatio-temporal correspondence relationship to generate the structure state characteristic matrix. When performing coupling relationship analysis based on the structure state characteristic matrix, the matrix is segmented according to the time series, and the local deformation and overall torsion data segments within each time window are extracted. The local deformation data segments are subjected to spatial decoupling analysis, the relative displacement and deformation rate of each monitoring point are calculated, and the local deformation characteristic vector is generated. The overall torsion data segments are used to calculate the torsion angle distribution and establish the torsion center trajectory to obtain the overall torsion characteristic vector. Through tensor operation, the coupling measurement of the two characteristic vectors is carried out to establish the stress-deformation mapping relationship to obtain the torsion deformation evolution data. Based on the local stress concentration coefficient, the stress-deformation mapping relationship is corrected to construct the stress distribution field and generate the stress distribution prediction data. The torsion deformation evolution data and the stress distribution prediction data are combined and associated according to the spatio-temporal correspondence principle to form the dynamic response prediction matrix.

[0015] When performing data fusion based on the dynamic response prediction matrix, the matrix is sampled in time series, and the reference time information is extracted to obtain the time reference sequence. The time stamps of the real-time collected monitoring data are compared to establish a mapping relationship with the time reference sequence, and the time-registered data is obtained. The time-registered data is spatially mapped using the monitoring point coordinate information, and the relative position deviation of each node is calculated to obtain the space-registered data. The time-registered data and the space-registered data are subjected to correlation analysis with the dynamic response prediction matrix, the deviation distribution is calculated, and the torsional deformation index is generated. The monitoring area is divided into grids according to the torsional deformation index, the stress distribution characteristics at the grid nodes are calculated, and the stress distribution index is constructed. The torsional deformation index and the stress distribution index are organized according to the topological relationship to obtain the multi-dimensional feature data table, which constitutes the state evaluation data set. When performing parameter optimization based on the state evaluation data set, the gradient analysis of the torsional deformation index in the data set is carried out, the spatial distribution of the structurally deformed sensitive area is calculated, and the deformation sensitivity distribution data is obtained. Based on the deformation sensitivity distribution data, the force analysis of the structural stress points is carried out, the stress concentration degree of each node is calculated, and the stress distribution weight data is obtained. The deformation sensitivity distribution data and the stress distribution weight data are numerically superimposed to evaluate and analyze the force state of the hanging point, and the hanging point position parameters are generated. The moment balance calculation is carried out for the hanging point position parameters, the force balance relationship between the hanging points is established, and the hoisting moment parameters are constructed. According to the hoisting moment parameters, the deformation trend during the hoisting process is predicted, the displacement compensation amount of each key point is calculated, and the compensation control parameters are obtained. The hanging point position parameters, the hoisting moment parameters, and the compensation control parameters are integrated according to the control priority to generate the hoisting control data packet.

[0016] The feature analysis of the hoisting control data packet is carried out, the numerical range of each parameter is extracted, and the control parameter feature data is obtained. The control parameter records under similar working conditions are extracted from the historical case data, the parameter correspondence relationship is established, and the historical parameter mapping data is generated. The similarity calculation is carried out between the control parameter feature data and the historical parameter mapping data, the historical control strategies with high matching degrees are screened out, and the control strategy database is constructed. The time series analysis of the control parameters in the control strategy database is carried out, the change trend of each control parameter is calculated, and the parameter evolution sequence is obtained. Based on the parameter evolution sequence, the prediction and inference of the control parameters are carried out, the value of the control parameter at the next moment is calculated, and the control parameter update data is generated. The control parameter update data is organized in time series according to the control execution order to generate the real-time control instruction sequence.

[0017] For example, during the hoisting process of large steel structures, the initial strain signals collected by fiber Bragg grating sensors are demodulated by wavelength to obtain the strain value distribution. The angular velocity data collected by the gyroscope array reflects the inclination state of the structure after filtering. The load data collected by dynamic strain sensors shows the force conditions of each hoisting point after frequency domain analysis. After feature extraction and analysis of these data, a state feature matrix reflecting local deformation and overall torsion is obtained. Through the coupled analysis of the feature matrix, the deformation trend and stress distribution of the structure are predicted. Based on these prediction results, combined with real-time monitoring data, a state assessment is carried out to generate optimized hoisting control parameters, including hoisting point positions, hoisting moments, and compensation control amounts. According to the experience of historical successful cases, a control instruction sequence is generated to guide the precise implementation of the hoisting operation.

[0018] In the embodiment of the present application, by scanning and collecting the edges of the steel structure through a distributed sensing network, multi-source heterogeneous data collection of structural angular velocity data, strain data, and load distribution data is achieved, solving the problem that it is difficult for traditional single sensors to comprehensively monitor the structural state. By extracting and analyzing the transient response characteristics of the original monitoring data set, a structural state feature matrix containing local deformation data and overall torsion data is generated, achieving a comprehensive characterization of the structural deformation characteristics. Using the coupled relationship calculation and analysis method, the correlation between local deformation and overall torsion is deeply analyzed to obtain torsional deformation evolution data and stress distribution prediction data, and these data are integrated to form a dynamic response prediction matrix, providing a reliable basis for structural deformation prediction. Based on the dynamic response prediction matrix, spatio-temporal registration and data fusion of the real-time collected monitoring data are carried out to construct a state assessment data set containing torsional deformation indicators and stress distribution indicators, improving the spatio-temporal consistency of the monitoring data. According to the state assessment data set, a dynamic optimization algorithm is used to calculate the hoisting parameters in real time to generate a hoisting control data packet containing hoisting point position parameters, hoisting moment parameters, and compensation control parameters, making the generation of control parameters more scientific. By combining historical case data for intelligent reasoning and analysis, a real-time control instruction sequence is output, realizing the closed-loop optimization of the precision monitoring and control of the hoisting process. In the application of artificial intelligence algorithms, the present invention innovatively combines deep learning with traditional mechanical models. By mining and learning historical data, a prediction model more in line with engineering reality is established. At the same time, an intelligent reasoning algorithm is used to dynamically optimize the control strategy, significantly improving the accuracy and real-time performance of control instructions, making the entire monitoring and control scheme more adaptable and robust.

[0019] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Arrange fiber Bragg grating sensors in a spiral pattern along the edge of the steel structure at a spacing of 0.5 meters, perform high-frequency sampling on the structure edge, conduct wavelength demodulation analysis at each sampling point to obtain the initial strain signal, then perform multi-scale decomposition on the initial strain signal through wavelet transform, perform threshold processing and reconstruction on the decomposed signals in each frequency band to generate the strain data of the structure edge; (2) Continuously sample the key nodes of the steel structure by the gyroscope array at a sampling frequency of 200 Hz, collect the raw triaxial angular velocity data, screen the frequency band of the raw data through a band-pass filter to eliminate high-frequency noise and low-frequency drift, and then use the Kalman filtering algorithm to perform state estimation and prediction on the processed data to obtain the structure angular velocity data; (3) Install dynamic strain sensors at the force application points of the lifting equipment, collect the load time-series data, convert the time-domain data to the frequency domain through fast Fourier transform, identify and eliminate the interference frequency components, and then use the inverse Fourier transform to reconstruct the signal to obtain the load distribution data with high signal-to-noise ratio; (4) Perform temperature compensation and linear calibration on the strain data of the structure edge to remove the influence of environmental factors, then calculate the strain change rate through the sliding window method, and perform feature extraction in combination with the strain amplitude to generate the edge strain feature data; (5) Perform dimensionality reduction processing on the structure angular velocity data through principal component analysis, extract the main change features, and then perform resampling and synchronization processing on the data through the spline interpolation algorithm to obtain the angular velocity feature data; (6) Normalize the maximum and minimum values of the load distribution data, perform standardization processing based on the statistical characteristics of historical data, and then eliminate short-term fluctuations through the moving average method to construct the load feature data; (7) Align the edge strain feature data, angular velocity feature data, and load feature data according to the time reference, establish data association through timestamp matching, then perform interpolation completion and outlier processing on the aligned data, and integrate the three types of feature data into the structure state description matrix through the data fusion algorithm to generate the original monitoring data set.

[0020] It should be noted that the fiber Bragg grating sensor is a sensor in which a Bragg grating is inscribed in the fiber core. When the external strain or temperature changes, the grating period will change, resulting in a change in the reflected light wavelength. These sensors are arranged spirally along the edge of the steel structure at a fixed spacing of 0.5 meters to obtain a continuous monitoring network. Each grating sensing point corresponds to a specific central wavelength. By using a demodulator to detect the change value of the reflected light wavelength in real time, the initial strain signal can be obtained. Wavelet transform multi-scale decomposition is performed on these signals to decompose the signals into wavelet coefficients of different frequency bands. Corresponding thresholds are set for the coefficients of different frequency bands to eliminate the influence of noise. Finally, the signals are reconstructed through inverse wavelet transform to obtain the strain data at the edge of the structure. At the key node positions of the structure, a gyroscope array is arranged for angular velocity monitoring. Each gyroscope measures the angular velocities in the X, Y, and Z directions simultaneously, and the sampling frequency is set to 200 Hz to ensure accurate capture of the transient response of the structure. The collected raw data is processed by a band-pass filter to filter out high-frequency noise above 100 Hz and low-frequency drift signals below 0.1 Hz. Then, the Kalman filtering algorithm is used to perform state estimation on the filtered data. Through the iteration of the prediction step and the update step, the Kalman filter continuously optimizes the state estimation value to obtain the angular velocity data of the structure.

[0021] Dynamic strain sensors are installed at the main stress points of the hoisting equipment, and these sensors continuously collect load time-series data. The collected data is transformed into the frequency domain through fast Fourier transform, and various frequency components can be clearly identified in the frequency domain. By analyzing the spectral characteristics, the interfering frequency components are identified, the amplitudes corresponding to these frequencies are set to zero, and then the signal is transformed back to the time domain through inverse Fourier transform to obtain the load distribution data after removing the interference. Temperature compensation and linear calibration are performed on the obtained strain data at the edge of the structure. According to the real-time temperature value measured by the temperature sensor, the original strain value is corrected using the temperature-strain relationship. Then, the strain value is corrected using the linear calibration coefficient determined through the calibration test to eliminate the systematic error. The sliding window method is used to calculate the strain change rate. The window length is determined according to the data sampling rate, and the strain change rate is obtained by calculating the difference in strain values within the window. Combining the absolute value of the strain and the change rate information, characteristic parameters are extracted to generate the edge strain characteristic data.

[0022] The structural angular velocity data contains a large amount of redundant information and is processed by dimensionality reduction using the principal component analysis method. The data is standardized, the covariance matrix is calculated, the eigenvalues and eigenvectors are solved, and several principal components with the largest contribution rate are selected as the main features. Then, the cubic spline interpolation algorithm is used to resample the dimensionality-reduced data to make the data sampling interval uniform. The load distribution data is normalized, and the data range is mapped to the interval [0, 1]. The mean and standard deviation are calculated based on historical data and standardized to make the data conform to the standard normal distribution. The moving average method is used to smooth the standardized data to remove the influence of short-term random fluctuations.

[0023] Synchronize the time of three types of characteristic data (edge strain characteristic data, angular velocity characteristic data, and load characteristic data). Determine the time reference and align according to the timestamps of each data. For data points with inconsistent sampling times, the missing values are filled by interpolation. Detect and process the outliers in the data to ensure the continuity and reliability of the data. Finally, through the data fusion algorithm, the three types of characteristic data are integrated into a matrix according to the physical meaning and spatial relationship to obtain the original monitoring data set.

[0024] For example, taking the hoisting process of a 30-meter-long steel structure crossbeam as an example, fiber Bragg grating sensors are arranged at intervals of 0.5 meters along the edge of the beam, gyroscopes are arranged at the four endpoints of the beam, and dynamic strain sensors are installed at the four lifting points. When the crossbeam starts to be hoisted, the fiber optic sensor network detects the edge strain distribution, and after temperature compensation and wavelet noise reduction, it reflects the real-time deformation state of the structure. The angular velocity data detected by the gyroscope is filtered and state-estimated to reveal the spatial attitude change of the structure. The load data at the lifting points is analyzed in the frequency domain and reconstructed to show the force balance at each lifting point. After these data are feature-extracted and fused, the monitoring data set is obtained.

[0025] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Intercept the data of the original monitoring data set through time-domain segmentation, select the data segments within the key time window, and generate time-series characteristic data; (2) Conduct spatial correlation analysis on the data of each measuring point in the time-series characteristic data, calculate the correlation coefficient matrix between the measuring points, and screen to obtain the key measuring point combinations; (3) Decompose the data of the key measuring point combinations in multiple dimensions, extract the displacement component, velocity component, and acceleration component to obtain local response characteristics; (4) Perform spatial mapping on the local response characteristics through data reconstruction, establish the node displacement field, and obtain local deformation data; (5) Conduct overall coordination analysis based on the local deformation data, calculate the deformation gradient and deformation rate of each region, and construct the overall torsion data; (6) Organize and arrange the local deformation data and the overall torsion data according to the spatio-temporal correspondence relationship to generate a structural state feature matrix.

[0026] Specifically, as Figure 2 shown, in the embodiment of the present application, it is a time sequence diagram for extracting and analyzing the structural transient response characteristics. The original monitoring data set is segmented in the time domain to obtain time sequence feature data; then, spatial correlation analysis is performed on the time sequence feature data to screen and obtain a key measuring point combination; multi-dimensional decomposition is performed on the key measuring point combination to extract local response characteristics; through spatial mapping reconstruction, the local response characteristics are converted into local deformation data; overall coordination analysis is performed on the local deformation data to construct overall torsion data; finally, the local deformation data and the overall torsion data are organized and arranged according to the spatio-temporal correspondence relationship to generate a structural state feature matrix.

[0027] Time domain segmentation refers to the process of dividing continuous time series data into several data segments according to a specific time length. In specific operations, a time window with a fixed length is selected to intercept the data, and the time window length is determined according to the dynamic characteristics of the structure. Generally, a time length that can cover a complete vibration cycle of the structure is selected. The sliding window technique is adopted during the data interception process, and there is a certain overlap between adjacent windows to ensure the continuity of the data. The data within each time window is extracted to obtain independent data segments, and these data segments contain the response characteristics of the structure at different time periods, constituting the time sequence feature data.

[0028] Perform spatial correlation analysis on the obtained time-series feature data. The purpose of spatial correlation analysis is to identify the degree of association between measurement points and find the key measurement points that are representative of the structural deformation characteristics. The specific approach is to calculate the correlation coefficient matrix between different measurement points, and the correlation coefficient matrix reflects the linear correlation degree between the data of each measurement point. The Pearson correlation coefficient method is used to calculate the correlation coefficient. For the data sequences of any two measurement points, calculate their covariance and divide it by the product of their respective standard deviations to obtain the correlation coefficient. The value range of the correlation coefficient is between -1 and 1, and the larger the absolute value, the stronger the correlation. By setting the correlation coefficient threshold, select the measurement point combinations with significant correlations, and these measurement point combinations can better reflect the overall deformation characteristics of the structure. Perform multi-dimensional decomposition on the data of the selected key measurement point combinations to extract the component information of different physical quantities. Multi-dimensional decomposition refers to the process of converting the displacement time-history data of the measurement points into velocity and acceleration information through numerical differentiation methods. Perform numerical differentiation on the displacement data, use the central difference format to calculate the velocity component, and then perform differentiation on the velocity component to obtain the acceleration component. The displacement component reflects the deformation state of the structure, the velocity component reflects the development trend of the deformation, and the acceleration component reflects the dynamic response characteristics of the structure. These three types of component data complement each other and jointly constitute the complete information describing the local response characteristics of the structure.

[0029] Reconstruct the local response feature data into a continuous nodal displacement field through a spatial mapping method. The spatial mapping process uses shape function interpolation technology to estimate the continuous displacement distribution within the entire structural domain using the data of a finite number of key measurement points. The shape function is a set of interpolation functions defined based on spatial coordinates, which can ensure the continuity and smoothness of the interpolation results. Through the position coordinates and response data of the key measurement points, combined with shape function interpolation, calculate the displacement value at any position to construct the nodal displacement field, thereby obtaining the local deformation data.

[0030] Perform overall coordination analysis based on the local deformation data. The core of this analysis is to calculate the deformation gradient and deformation rate. The deformation gradient refers to the rate of change of the displacement field in space, which is obtained by calculating the displacement difference between adjacent nodes. The deformation rate is the rate of change of displacement with time, which is obtained by calculating the displacement difference between consecutive time instants. These two parameters jointly reflect the torsional deformation characteristics of the structure. By establishing the motion trajectory and torsional angle distribution of the torsional center, construct the overall torsional data.

[0031] Organize and arrange the local deformation data and the overall torsional data according to the spatio-temporal correspondence relationship. The spatio-temporal correspondence relationship refers to the mapping relationship of the data in the time dimension and the space dimension, and it is necessary to ensure that different types of data are synchronized in time and corresponding in space. Organize all the data according to the format to obtain a multi-dimensional array structure, where different dimensions correspond to time, spatial position, and physical quantity type respectively, and finally generate the structural state feature matrix.

[0032] Taking the hoisting process of a large steel structure member as an example: The original monitoring data is divided into time windows of 10 seconds, with an overlap of 5 seconds between windows, to obtain a series of time-series data segments. The correlation coefficients between measurement points are calculated for these data segments, and it is found that the measurement points located at the four corners of the structure have strong correlations. These measurement points are combined as key measurement points. The differential calculation is performed on the data of the key measurement points to obtain the response characteristics in three dimensions of displacement, velocity, and acceleration. The discrete measurement point data is reconstructed into a continuous displacement field through shape function interpolation, and the deformation distribution of each region is calculated. The spatial gradient and time variation rate of the deformation field are analyzed to identify the position of the torsional center of the structure and the variation law of the torsional angle. Finally, all the data is integrated into a matrix, where the rows of the matrix represent different moments, the columns represent different spatial positions, and the depth direction represents different physical quantities, to obtain the state characteristic description.

[0033] In a specific embodiment, the process of performing step S103 may specifically include the following steps: (1) Perform segmented processing on the structural state characteristic matrix according to the time series, divide the continuous data stream into multiple time windows, and extract the local deformation and overall torsion data segments within each window; (2) Perform spatial decoupling analysis on the local deformation data segments, calculate the relative displacement and deformation rate of each monitoring point, and generate local deformation feature vectors; (3) Calculate the torsional angle distribution for the overall torsion data segments, establish the trajectory of the torsional center, and obtain the overall torsion feature vectors; (4) Perform coupling measurement on the local deformation feature vectors and the overall torsion feature vectors through tensor operations, establish the stress-deformation mapping relationship, and obtain the torsional deformation evolution data; (5) Perform correction calculation on the stress-deformation mapping relationship based on the local stress concentration coefficient, construct the stress distribution field, and generate stress distribution prediction data; (6) Combine and correlate the torsional deformation evolution data and the stress distribution prediction data according to the principle of spatio-temporal correspondence, establish the time-varying feature mapping matrix, and form the dynamic response prediction matrix.

[0034] Specifically, perform time-domain segmentation on the structural state characteristic matrix, and use the sliding window method to divide the continuous data stream into multiple time windows. The length of each time window is determined according to the structural characteristic frequency, usually taking 3-5 times the main vibration period of the structure, and a 50% overlap rate is maintained between adjacent windows to ensure data continuity. Within each window, the data segments of local deformation and overall torsion are respectively extracted, and these data segments contain the deformation characteristic information of the structure within a specific time period.

[0035] For the spatial decoupling analysis of local deformation data segments, it involves calculating the relative displacement and deformation rate. The calculation formula for the relative displacement is as follows: ; Where, represents the relative displacement of the monitoring point (i,j), is the weight coefficient of the k-th measuring point, is the measured value at the current position, is the measured value at the reference position, is the spatial attenuation coefficient, and n is the number of reference points.

[0036] For the calculation of the overall torsion angle distribution and the establishment of the torsion center trajectory, the following formula is used: ; Where, represents the torsion angle in the xy plane, is the measuring point weight matrix, and are the coordinates of the measuring points, is the initial phase angle of the measuring point, and M and N are the numbers of measuring points in the x and y directions respectively.

[0037] The coupling measurement of the local deformation eigenvector and the overall torsion eigenvector uses tensor operations, and its mathematical expression is: ; Where, is the coupling measurement tensor, is the weight tensor, is the local deformation eigenvector, is the overall torsion eigenvector, is the coupling coefficient tensor, and I, J, and K represent the sizes of the three dimensions respectively.

[0038] The results calculated above are combined and processed, and the stress-deformation mapping relationship is corrected based on the local stress concentration coefficient. The local stress concentration coefficient is a dimensionless parameter that reflects the non-uniform stress distribution caused by local geometric shape changes in the structure. By correcting the stress distribution with this coefficient, stress distribution prediction data is obtained. The torsional deformation evolution data and the stress distribution prediction data are combined and correlated according to the spatio-temporal correspondence principle. The spatio-temporal correspondence principle requires that all data must strictly correspond in the time and space dimensions. By establishing a spatio-temporal coordinate system, different types of data are integrated into the same matrix framework.

[0039] Taking the hoisting process of a 30-meter-span steel structure truss as an example, the collected data is segmented according to a 5-second time window, with a 2.5-second overlap between windows. Within each time window, local deformation data and overall torsion data at key nodes are extracted. For the local deformation data, the relative displacement is obtained by calculating the displacement difference of each monitoring point relative to the reference point; the deformation rate is calculated by calculating the change rate of displacement over time. For the overall torsion data, by analyzing the displacement vector directions of each measuring point, the torsion angle distribution is calculated, and the movement trajectory of the torsion center is obtained by least squares fitting. These two sets of data are coupled and analyzed through tensor operations to establish a mapping relationship between local deformation and overall torsion. According to the stress concentration coefficient determined by the structural geometric characteristics, the stress distribution is corrected, and finally a dynamic response prediction matrix is obtained.

[0040] In a specific embodiment, the process of performing step S104 may specifically include the following steps: (1) Align the data of the dynamic response prediction matrix through time series sampling, extract the reference time information of each monitoring point, and generate a time reference sequence; (2) Compare the time stamps of the real-time collected monitoring data, establish a mapping relationship with the time reference sequence, and obtain time-registered data; (3) Perform spatial mapping on the time-registered data using the monitoring point coordinate information, calculate the relative position deviation of each node, and obtain spatially registered data; (4) Perform a correlation analysis on the time-registered data and the spatially registered data with the dynamic response prediction matrix, calculate the deviation distribution, and generate a torsional deformation index; (5) Divide the monitoring area into grids according to the torsional deformation index, calculate the stress distribution characteristics at the grid nodes, and construct a stress distribution index; (6) Organize and sort the torsional deformation index and the stress distribution index according to the topological relationship to obtain a multi-dimensional feature data table, which constitutes a state evaluation data set.

[0041] Specifically, in the process of real-time monitoring of the hoisting accuracy of large steel structures, time series sampling is a method for processing data in the time domain. When aligning the data of the dynamic response prediction matrix, a reference time point is determined, and starting from this time, the data of each monitoring point is extracted at a fixed sampling interval. The following formula is used in the time series sampling process: ; Among them, is the data matrix after sampling, is the sampling weight coefficient, and are the up-sampling and down-sampling operators respectively, is the time synchronization factor, and M and N are the sampling levels. This process generates a time reference sequence.

[0042] When comparing the timestamps of the real-time collected monitoring data, it is necessary to establish a time correspondence. The real-time collected monitoring data usually contains data streams with different sampling frequencies, and it is necessary to unify them to the same time reference through difference calculation. The specific calculation formula is: ; where is the data matrix after time registration, is the registration weight coefficient, and are the forward and backward time registration operators respectively, is the time alignment factor, and A and B are the registration levels.

[0043] When performing spatial mapping on the time-registered data using the monitoring point coordinate information, the following calculation formula is adopted: ; where is the spatial registration data matrix, is the spatial mapping weight, and are the mapping operators in the x-y plane and y-z plane respectively, is the spatial correction factor, and I and J are the mapping scale parameters.

[0044] When performing correlation analysis, calculate the deviation distribution between the time-registered data and the spatial-registered data and the dynamic response prediction matrix. This process involves the comparison and matching between multiple data sets. By calculating the correlation coefficient and deviation value, a torsional deformation index is generated. According to these indexes, the monitoring area is divided into grids, and the size and density of the grids are determined according to the structural characteristics. At each grid node, calculate the stress distribution characteristics, including principal stress, shear stress, and equivalent stress, and construct a stress distribution index.

[0045] Organize and arrange the torsional deformation index and the stress distribution index according to the topological relationship. The topological relationship reflects the spatial connection relationship between the various parts of the structure. By establishing a data structure, different types of index data are integrated into a multi-dimensional feature data table to form a state evaluation data set.

[0046] For example: During the hoisting process, data sampling is performed on the dynamic response prediction matrix at a sampling frequency of 100 Hz, and the response data of each monitoring point at each moment is extracted. Among the real-time collected monitoring data, the sampling frequency of the strain data is 200 Hz, and the sampling frequency of the angular velocity data is 50 Hz. These data are unified to a time base of 100 Hz through time registration. These unified data are mapped according to the spatial coordinate information of the monitoring points, and the deviation values relative to the design position are calculated. These deviation data are compared with the prediction matrix to calculate the torsional deformation index. According to the torsional deformation index, the monitoring area is divided into several grid cells, and the stress distribution characteristics are calculated at each grid node. These index data are finally integrated into a multi-dimensional data table to obtain a state evaluation data set, reflecting the real-time state information of the structure during the hoisting process.

[0047] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Perform gradient analysis on the torsional deformation index in the state evaluation data set, calculate the spatial distribution of the structure deformation sensitive area, and obtain the deformation sensitivity distribution data; (2) Based on the deformation sensitivity distribution data, perform a force analysis on the structure stress points, calculate the stress concentration degree of each node, and obtain the stress distribution weight data; (3) Numerically superimpose the deformation sensitivity distribution data and the stress distribution weight data, evaluate and analyze the force state of the lifting points, and generate the lifting point position parameters; (4) Perform moment balance calculation for the lifting point position parameters, establish the force balance relationship between each lifting point, and construct the hoisting moment parameters; (5) Predict the deformation trend during the hoisting process according to the hoisting moment parameters, calculate the displacement compensation amount of each key point, and obtain the compensation control parameters; (6) Integrate the lifting point position parameters, the hoisting moment parameters, and the compensation control parameters according to the control priority to generate a hoisting control data packet.

[0048] Specifically, the gradient analysis is achieved by calculating the deformation difference between adjacent monitoring points in space. The specific calculation formula is: ; Among them, represents the deformation gradient tensor, is the gradient weight coefficient, and are the gradient operators in the j-k plane and the k-l plane respectively, is the spatial sensitivity factor, and D and E are the dimension parameters for gradient calculation. This calculation process performs a scanning analysis on the entire structure area to determine the sensitive areas with larger deformation gradient values and obtain the deformation sensitivity distribution data.

[0049] Based on the deformation sensitivity distribution data, perform stress state analysis of the structural stress points. The calculation formula is as follows: ; Among them, is the stress distribution weight tensor, is the stress weight coefficient, and are the stress operators in the u-v plane and the v-w plane respectively, is the stress concentration factor, and F and G are the scale parameters for stress calculation. This calculation takes into account the geometric characteristics and material properties of the structure, and reflects the influence of local geometric shape changes on the stress distribution through the stress concentration factor.

[0050] When performing superposition analysis of deformation sensitivity and stress distribution weight data, the following formula is used: ; Among them, is the suspension point position parameter tensor, is the position weight coefficient, and are the position operators in the a-b plane and the b-c plane respectively, is the position correction factor, and H and I are the spatial dimensions of the position parameters. This calculation process comprehensively evaluates the applicability of each candidate suspension point position through the weighted average method.

[0051] For the calculation of the suspension point moment balance, the following formula is used: ; Among them, is the hoisting moment parameter tensor, is the moment weight coefficient, and are the moment operators in the r-s plane and the s-t plane respectively, is the moment balance factor, and M and N are the direction parameters for moment calculation. This calculation ensures the force balance state of all suspension points.

[0052] The calculation of the displacement compensation amount adopts: ; Among them, is the compensation control parameter tensor, is the compensation weight coefficient, and are the compensation operators in the x-y plane and the y-z plane respectively, is the compensation correction factor, and P and Q are the hierarchical parameters for compensation calculation.

[0053] The actual application process of these formulas is as follows: For a large steel structure crane girder, through the gradient analysis of the torsional deformation index, it is found that there are large deformation gradients at the ends and mid - spans of the girder. These positions are marked as deformation - sensitive areas, specifically, the gradient value at the end position reaches 2 - 3 times that of the middle. In the stress analysis at the joints, the stress concentration factor at the connection is significantly higher than other positions. After superimposing these two types of data, four optimal hoisting point positions are selected. These positions not only avoid high - stress areas but also ensure the overall force balance. Through moment balance calculation, the specific hoisting moment values required for each hoisting point are obtained, and the structural deformation trend under these moments is predicted. According to the prediction results, the displacement compensation amounts required for each key point are calculated. These parameters are integrated in the order of priority of stability influence, accuracy influence, and auxiliary control to obtain a hoisting control data packet.

[0054] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Conduct feature analysis on the hoisting control data packet, extract the numerical ranges of the hoisting point position parameters, hoisting moment parameters, and compensation control parameters to obtain control parameter feature data; (2) Extract the control parameter records under similar working conditions from the historical case data, establish parameter correspondence relationships, and generate historical parameter mapping data; (3) Calculate the similarity between the control parameter feature data and the historical parameter mapping data, screen out historical control strategies with high matching degrees, and construct a control strategy database; (4) Conduct time - series analysis on the control parameters in the control strategy database, calculate the change trends of each control parameter, and obtain parameter evolution sequences; (5) Based on the parameter evolution sequences, perform prediction and inference on the control parameters, calculate the control parameter values at the next moment, and generate control parameter update data; (6) Organize the control parameter update data in time - series according to the control execution order to generate a real - time control instruction sequence.

[0055] Specifically, extract the lifting point position parameters, including the spatial coordinate values and allowable deviation ranges of each lifting point; secondly, analyze the lifting moment parameters, including the magnitude and direction of the forces on each lifting point; finally, extract the compensation control parameters, including the displacement compensation amount and angle correction value. Conduct numerical interval statistics on these parameters to determine the value ranges and distribution characteristics of each type of parameter, and obtain the control parameter characteristic data. These characteristic data reflect the specific control requirements of the current lifting condition. When extracting parameter records of similar conditions from the historical case database, similarity criteria need to be defined. The similarity criteria include key characteristics such as structural type, dimensional ratio, weight distribution, and environmental conditions. Screen out historical cases with higher similarity through these criteria, and then extract the control parameter records in these cases. Standardize the extracted parameter records to unify the data formats and dimensions between different cases and establish the corresponding relationships between parameters. These corresponding relationships reflect the variation laws of parameters under different conditions and constitute the historical parameter mapping data.

[0056] When calculating the similarity between the control parameter characteristic data and the historical parameter mapping data, a multi-dimensional similarity measurement method is adopted. For each control parameter, calculate its similarity degree with the corresponding parameter in the historical data, comprehensively considering the numerical differences, variation trends, and control effects of the parameters. The historical control strategies with higher similarity are screened out, and these strategies include parameter configuration schemes and control process records. Systematically organize the screened control strategies to obtain the control strategy database. Conduct time series analysis on the data in the control strategy database, focusing on the variation characteristics of the control parameters over time. Use time series analysis methods to identify the periodic, trend, and mutation characteristics of parameter changes. Model the variation laws of each type of parameter to obtain mathematical models describing the parameter evolution process. These models reflect the variation characteristics of parameters in different stages and constitute the parameter evolution sequence. When performing prediction and inference based on the parameter evolution sequence, a progressive prediction method is adopted. According to the historical variation trend and current state of the parameters, predict the parameter values at the next moment. The prediction process considers the mutual influences and constraint relationships between parameters to ensure the rationality of the prediction results. The predicted parameter values constitute the control parameter update data, and these data reflect the latest variation trends of the control parameters.

[0057] Organize the control parameter update data according to the execution priority and time sequence relationship. Determine the execution order of the parameters. Usually, the parameters affecting the structural stability are placed first, followed by the precision control parameters, and finally the auxiliary control parameters. Organize the parameters into an instruction sequence in this order, and each instruction contains the specific execution time, target parameter value, and execution conditions.

[0058] For example: Extract the position parameters (coordinate values and allowable deviations), moment parameters (magnitude and direction), and compensation parameters (displacement and angle compensation values) of the four lifting points from the lifting control data packet. Retrieve the lifting records of the same type of roof truss from the historical database and find five similar cases. Establish a parameter mapping relationship by comparing the structural characteristics and construction conditions of the current working condition with those of the historical cases. Calculate the parameter similarity and screen out the two most matching historical cases as references. Analyze the variation law of the parameters in these two cases to obtain a mathematical model describing the parameter evolution process. Predict the next control parameters according to the model, such as the fine-tuning amount of the lifting point position and the compensation value of the moment. Organize these parameters into an execution instruction sequence in the order of "stability control - precision control - auxiliary control" to guide the precise implementation of the lifting process.

[0059] The above describes the real-time monitoring method for the lifting accuracy of large steel structures in the embodiments of the present application. Next, the real-time monitoring system for the lifting accuracy of large steel structures in the embodiments of the present application will be described. Please refer to Figure 3 , an embodiment of the real-time monitoring system for the lifting accuracy of large steel structures in the embodiments of the present application includes: An acquisition module 201, configured to scan and acquire the edge of the steel structure through a distributed sensing network, obtain structure angular velocity data, strain data, and load distribution data, and obtain an original monitoring data set; An extraction module 202, configured to extract and analyze the structural transient response characteristics according to the original monitoring data set, and generate a structural state feature matrix including local deformation data and overall torsion data; A coupling module 203, configured to calculate and analyze the coupling relationship between local deformation and overall torsion according to the structural state feature matrix, obtain torsion deformation evolution data and stress distribution prediction data, and integrate them to obtain a dynamic response prediction matrix; A fusion module 204, configured to perform spatio-temporal registration and data fusion on the real-time acquired monitoring data based on the dynamic response prediction matrix, and construct a state evaluation data set including torsion deformation indexes and stress distribution indexes; An optimization module 205, configured to perform dynamic optimization calculation on the lifting parameters according to the state evaluation data set, and generate a lifting control data packet including lifting point position parameters, lifting moment parameters, and compensation control parameters; A control module 206, configured to perform intelligent reasoning analysis by using the lifting control data packet in combination with historical case data, and output a real-time control instruction sequence to achieve accuracy monitoring and control of the lifting process.

[0060] Through the collaborative cooperation of the above-mentioned various components, the edge of the steel structure is scanned and collected through a distributed sensing network, realizing the multi-source heterogeneous data collection of structural angular velocity data, strain data, and load distribution data, and solving the problem that it is difficult for traditional single sensors to comprehensively monitor the structural state; through the extraction and analysis of transient response characteristics of the original monitoring data set, a structural state feature matrix containing local deformation data and overall torsion data is generated, realizing a comprehensive description of the structural deformation characteristics; the coupling relationship calculation and analysis method is adopted to deeply analyze the correlation between local deformation and overall torsion, obtain the torsional deformation evolution data and stress distribution prediction data, and integrate these data to form a dynamic response prediction matrix, providing a reliable basis for structural deformation prediction; based on the dynamic response prediction matrix, the real-time collected monitoring data is subjected to spatio-temporal registration and data fusion, and a state evaluation data set containing torsional deformation indicators and stress distribution indicators is constructed, improving the spatio-temporal consistency of the monitoring data; according to the state evaluation data set, a dynamic optimization algorithm is used to calculate the lifting parameters in real time, generating a lifting control data packet containing hoisting point position parameters, lifting moment parameters, and compensation control parameters, making the generation of control parameters more scientific; through intelligent reasoning and analysis in combination with historical case data, a real-time control instruction sequence is output, realizing the closed-loop optimization of the precision monitoring and control of the lifting process. In the application of artificial intelligence algorithms, the present invention innovatively combines deep learning with traditional mechanical models, establishes a prediction model that is more in line with engineering practice through the mining and learning of historical data, and at the same time uses an intelligent reasoning algorithm to dynamically optimize the control strategy, significantly improving the accuracy and real-time performance of control instructions, making the entire monitoring and control scheme more adaptable and robust.

[0061] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it; although the present application 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 on 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 embodiments of the present application.

Claims

1. A real-time monitoring method for the hoisting precision of large steel structures, characterized in that The real-time monitoring method for the hoisting accuracy of large steel structures includes: Scanning and collecting the edges of the steel structure through a distributed sensing network to obtain structural angular velocity data, strain data, and load distribution data, and obtaining an original monitoring data set; According to the original monitoring data set, extracting and analyzing the transient response characteristics of the structure, and generating a structural state feature matrix including local deformation data and overall torsion data; Based on the structural state feature matrix, calculating and analyzing the coupling relationship between local deformation and overall torsion, obtaining torsion deformation evolution data and stress distribution prediction data, and integrating them to obtain a dynamic response prediction matrix; Based on the dynamic response prediction matrix, performing spatio-temporal registration and data fusion on the real-time collected monitoring data, and constructing a state evaluation data set including torsion deformation indexes and stress distribution indexes; According to the state evaluation data set, performing dynamic optimization calculation on the hoisting parameters, and generating a hoisting control data packet including hoisting point position parameters, hoisting moment parameters, and compensation control parameters; Using the hoisting control data packet, combining historical case data for intelligent reasoning analysis, and outputting a real-time control instruction sequence to achieve the accuracy monitoring and control of the hoisting process.

2. The real-time monitoring method for the hoisting accuracy of large steel structures according to claim 1, characterized in that, The scanning and collecting the edges of the steel structure through a distributed sensing network to obtain structural angular velocity data, strain data, and load distribution data, and obtaining an original monitoring data set includes: Arranging fiber Bragg grating sensors along the edge of the steel structure in a spiral pattern at a spacing of 0.5 m, performing high-frequency sampling on the edge of the structure, and performing wavelength demodulation analysis at each sampling point to obtain the initial strain signal. Then, performing multi-scale decomposition on the initial strain signal through wavelet transform, performing threshold processing and reconstruction on the decomposed signals in each frequency band, and generating structural edge strain data; Continuously sampling the key nodes of the steel structure by a gyroscope array at a sampling frequency of 200 Hz, collecting the original triaxial angular velocity data, screening the frequency band of the original data through a band-pass filter, removing high-frequency noise and low-frequency drift, and then using the Kalman filter algorithm to perform state estimation and prediction on the processed data to obtain structural angular velocity data; Arranging dynamic strain sensors at the force application points of the hoisting equipment, collecting load time-series data, converting the time-domain data to the frequency domain through fast Fourier transform, identifying and eliminating interference frequency components, and then reconstructing the signal using inverse Fourier transform to obtain high-signal-to-noise load distribution data; Performing temperature compensation and linear calibration on the structural edge strain data, removing the influence of environmental factors, then calculating the strain change rate through the sliding window method, and extracting features in combination with the strain amplitude to generate edge strain feature data; Performing dimensionality reduction processing on the structural angular velocity data through principal component analysis, extracting the main change features, and then performing resampling and synchronization processing on the data through the spline interpolation algorithm to obtain angular velocity feature data; Performing maximum-minimum normalization on the load distribution data, performing standardization processing based on the statistical characteristics of historical data, and then eliminating short-term fluctuations through the moving average method to construct load feature data; Align the edge strain characteristic data, angular velocity characteristic data, and load characteristic data according to the time reference, establish data association through timestamp matching, perform interpolation and outlier processing on the aligned data, and integrate the three types of characteristic data into a structural state description matrix through a data fusion algorithm to generate the original monitoring dataset.

3. The real-time monitoring method for the hoisting precision of large steel structures according to claim 1, characterized in that, Extract and analyze the structural transient response characteristics based on the original monitoring dataset to generate a structural state characteristic matrix including local deformation data and overall torsion data, including: Intercept the data of the original monitoring dataset through time-domain segmentation, select the data segments within the key time window, and generate time-series characteristic data; Conduct spatial correlation analysis on the data of each measurement point in the time-series characteristic data, calculate the correlation coefficient matrix between the measurement points, and screen out the key measurement point combinations; Perform multi-dimensional decomposition on the data of the key measurement point combinations, extract the displacement component, velocity component, and acceleration component to obtain the local response characteristics; Perform spatial mapping on the local response characteristics through data reconstruction, establish a node displacement field, and obtain local deformation data; Conduct overall coordination analysis based on the local deformation data, calculate the deformation gradient and deformation rate of each region, and construct overall torsion data; Organize and arrange the local deformation data and overall torsion data according to the spatio-temporal correspondence relationship to generate the structural state characteristic matrix.

4. The real-time monitoring method for the hoisting precision of large steel structures according to claim 1, characterized in that, Calculate and analyze the coupling relationship between local deformation and overall torsion based on the structural state characteristic matrix, obtain the torsional deformation evolution data and stress distribution prediction data, and integrate them to obtain the dynamic response prediction matrix, including: Perform segmented processing on the structural state characteristic matrix according to the time series, divide the continuous data stream into multiple time windows, and extract the local deformation and overall torsion data segments within each window; Conduct spatial decoupling analysis on the local deformation data segments, calculate the relative displacement and deformation rate of each monitoring point, and generate local deformation feature vectors; Calculate the torsional angle distribution for the overall torsion data segments, establish the trajectory of the torsion center, and obtain the overall torsion feature vectors; Conduct coupling measurement on the local deformation feature vectors and overall torsion feature vectors through tensor operations, establish a stress-deformation mapping relationship, and obtain the torsional deformation evolution data; Perform correction calculation on the stress-deformation mapping relationship based on the local stress concentration coefficient, construct a stress distribution field, and generate the stress distribution prediction data; Combine and associate the torsional deformation evolution data and stress distribution prediction data according to the spatio-temporal correspondence principle, establish a time-varying feature mapping matrix, and form the dynamic response prediction matrix.

5. The real-time monitoring method for the hoisting accuracy of large steel structures according to claim 1, characterized in that Based on the dynamic response prediction matrix, perform spatio-temporal registration and data fusion on the real-time collected monitoring data, and construct a state evaluation dataset including torsional deformation indicators and stress distribution indicators, including: Align the data of the dynamic response prediction matrix through time-series sampling, extract the reference time information of each monitoring point, and generate a time reference sequence; Compare the timestamps of the real-time collected monitoring data, establish a mapping relationship with the time reference sequence, and obtain the time registration data; Perform spatial mapping on the time registration data using the monitoring point coordinate information, calculate the relative position deviations of each node, and obtain the spatial registration data; Perform correlation analysis on the time registration data and the spatial registration data with the dynamic response prediction matrix, calculate the deviation distribution, and generate the torsional deformation index; Perform grid division on the monitoring area according to the torsional deformation index, calculate the stress distribution characteristics at the grid nodes, and construct the stress distribution index; Organize and sort the torsional deformation index and the stress distribution index according to the topological relationship to obtain the multi-dimensional feature data table, which constitutes the state evaluation data set.

6. The real-time monitoring method for the hoisting accuracy of large steel structures according to claim 1, characterized in that According to the state evaluation data set, perform dynamic optimization calculation on the hoisting parameters to generate a hoisting control data packet including hoisting point position parameters, hoisting moment parameters, and compensation control parameters, including: Perform gradient analysis on the torsional deformation index in the state evaluation data set, calculate the spatial distribution of the structure deformation sensitive area, and obtain the deformation sensitivity distribution data; Perform force analysis on the structure stress points based on the deformation sensitivity distribution data, calculate the stress concentration degree of each node, and obtain the stress distribution weight data; Perform numerical superposition on the deformation sensitivity distribution data and the stress distribution weight data, evaluate and analyze the stress state of the hoisting points, and generate the hoisting point position parameters; Perform moment balance calculation for the hoisting point position parameters, establish the force balance relationship between the hoisting points, and construct the hoisting moment parameters; Predict the deformation trend during the hoisting process according to the hoisting moment parameters, calculate the displacement compensation amount of each key point, and obtain the compensation control parameters; Integrate the hoisting point position parameters, hoisting moment parameters, and compensation control parameters according to the control priority to generate the hoisting control data packet.

7. The real-time monitoring method for the hoisting accuracy of large steel structures according to claim 1, characterized in that Use the hoisting control data packet, combine with historical case data for intelligent reasoning analysis, and output a real-time control instruction sequence to realize the precision monitoring and control of the hoisting process, including: Perform feature analysis on the hoisting control data packet, extract the numerical intervals of the hoisting point position parameters, hoisting moment parameters, and compensation control parameters, and obtain the control parameter feature data; Extract the control parameter records under similar working conditions from the historical case data, establish the parameter correspondence relationship, and generate the historical parameter mapping data; Perform similarity calculation on the control parameter feature data and the historical parameter mapping data, screen out the historical control strategies with high matching degree, and construct the control strategy database; Perform time series analysis on the control parameters in the control strategy database, calculate the change trend of each control parameter, and obtain the parameter evolution sequence; Perform prediction and reasoning on the control parameters based on the parameter evolution sequence, calculate the control parameter values at the next moment, and generate the control parameter update data; Organize the control parameter update data in time series according to the control execution order to generate the real-time control instruction sequence.

8. A real-time monitoring system for the hoisting accuracy of large steel structures, which is used to implement the real-time monitoring method for the hoisting accuracy of large steel structures described in any one of claims 1-7, characterized in that, The large steel structure hoisting precision real-time monitoring system includes: An acquisition module for scanning and acquiring the steel structure edge through a distributed sensing network to obtain the structure angular velocity data, strain data, and load distribution data, and obtain the original monitoring data set; An extraction module, which is used to extract and analyze the structural transient response characteristics according to the original monitoring data set, and generate a structural state feature matrix including local deformation data and overall torsion data; A coupling module, which is used to calculate and analyze the coupling relationship between local deformation and overall torsion according to the structural state feature matrix, obtain the torsional deformation evolution data and stress distribution prediction data, and integrate them to obtain a dynamic response prediction matrix; A fusion module, which is used to perform spatio-temporal registration and data fusion on the real-time collected monitoring data based on the dynamic response prediction matrix, and construct a state evaluation data set including torsional deformation indexes and stress distribution indexes; An optimization module, which is used to perform dynamic optimization calculation on the hoisting parameters according to the state evaluation data set, and generate a hoisting control data packet including hoisting point position parameters, hoisting moment parameters and compensation control parameters; A control module, which is used to perform intelligent reasoning analysis by using the hoisting control data packet in combination with historical case data, output a real-time control instruction sequence, and realize the precision monitoring and control of the hoisting process.

Images