Temperature effect monitoring method and system for metal roof

Abstract

The invention discloses a temperature effect monitoring method and system for a metal roof, particularly relates to the technical field of building structure health monitoring, and is used for solving the problems of decoupling of temperature monitoring data and structural mechanical response, failure of a static early warning threshold and the like in the prior art. Acquiring three-dimensional temperature field data through a distributed temperature sensing network, and extracting a temperature gradient characteristic value from the three-dimensional temperature field data; a piezoelectric film sensor is used for collecting constraint point micro-vibration signals, and dynamic constraint reaction force spectrum characteristic values are extracted through signal transformation and manifold learning; establishing a temperature gradient and mechanical characteristic coupling model to generate a temperature stress evaluation value; dynamically adjusting an early warning threshold range according to the environmental parameters; when the evaluation value exceeds a threshold value, a structure safety early warning signal is output; multi-physical field coupling analysis of temperature field distribution and structural response is achieved, an environment-adaptive dynamic early warning mechanism is constructed, and the accuracy and reliability of metal roof temperature effect monitoring are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of building structure health monitoring. More specifically, the present invention relates to a method and system for monitoring the temperature effect of a metal roof. Background Art

[0002] Metal roofs are widely used in large public buildings such as airport terminals and stadiums due to their high structural strength and strong construction convenience. Existing metal roof temperature monitoring technologies mainly collect data through local point temperature sensors and make safety judgments by combining fixed thresholds. Such technical solutions have been proven to provide basic temperature information in engineering practice, but limited by the single-point data collection mode, it is difficult to completely characterize the overall temperature distribution characteristics of large-scale roofs.

[0003] The defects of the existing technology are as follows: the temperature monitoring data is decoupled from the structural mechanics response, specifically manifested as: single-point temperature measurement cannot establish a three-dimensional temperature gradient field, resulting in distorted evaluation of the temperature load action range; there is a lack of a real-time correlation mechanism between temperature data and structural strain and displacement, and it is impossible to quantify the actual impact of temperature changes on structural safety; the static warning threshold is disconnected from dynamic environmental parameters, and the failure risk is significant under seasonal alternation and extreme weather conditions. Summary of the Invention

[0004] In order to overcome the above defects of the existing technology, the present invention provides a method and system for monitoring the temperature effect of a metal roof to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions: A method for monitoring the temperature effect of a metal roof, comprising the following steps: S1: Obtain three-dimensional temperature field data of the metal roof through a distributed temperature sensing network; S2: Extract the temperature gradient eigenvalue of the metal roof from the three-dimensional temperature field data; S3: Collect the micro-vibration signal sequence of the metal roof constraint points through a piezoelectric film sensor array, process the micro-vibration signal sequence by using the Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extract the dynamic constraint reaction force spectrum eigenvalue from the instantaneous frequency-energy spectrum by using the Riemannian manifold learning method; S4: Establish a coupling relationship model between the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue to generate a temperature stress evaluation value of the metal roof; S5: Dynamically adjust the warning threshold range according to the solar radiation intensity and environmental wind speed parameters; S6: When the temperature stress evaluation value exceeds the warning threshold range, output a structural safety warning signal for the metal roof.

[0006] Further, three-dimensional temperature field data of the metal roof is obtained through a distributed temperature sensing network, including: The distributed temperature sensing network includes multiple temperature measurement optical fiber layout subsystems at different angles; the three-dimensional temperature field data collects discrete temperature point data on the surface of the metal roof through the distributed optical fiber temperature measurement technology of the temperature measurement optical fiber layout subsystem, performs spatial correlation analysis on the discrete temperature point data, and reconstructs and generates three-dimensional temperature field data through an improved interpolation algorithm.

[0007] Further, the temperature measurement optical fiber layout subsystem covers the entire metal roof area.

[0008] Further, temperature gradient eigenvalue of the metal roof is extracted from the three-dimensional temperature field data, including: Calculating the temperature difference values at different spatial positions on the metal roof, identifying the physical boundary constraint conditions of the metal roof based on the three-dimensional temperature field data, correcting the temperature difference values in combination with the physical boundary constraint conditions, and extracting eigenvectors from the corrected temperature difference values as temperature gradient eigenvalues through the principal component analysis algorithm.

[0009] Further, a micro-vibration signal sequence of the metal roof constraint points is collected through a piezoelectric film sensor array, the Hilbert-Huang transform is used to process the micro-vibration signal sequence to generate an instantaneous frequency-energy spectrum, and the dynamic constraint reaction force spectrum eigenvalue is extracted from the instantaneous frequency-energy spectrum through the Riemannian manifold learning method, including: Decomposing the micro-vibration signal sequence into intrinsic mode function components through the Hilbert-Huang transform; Performing Hilbert spectrum analysis on the intrinsic mode function components to generate an instantaneous frequency-energy spectrum; Mapping the instantaneous frequency-energy spectrum to the Riemannian manifold space to construct a tangent vector field; Calculating the geodesic distance matrix eigenvalue of the tangent vector field; Selecting the eigenvector corresponding to the maximum geodesic distance matrix eigenvalue as the dynamic constraint reaction force spectrum eigenvalue.

[0010] Further, the micro-vibration signal sequence is obtained by collecting the vibration response of the metal roof constraint points through a piezoelectric film sensor array.

[0011] Further, a coupling relationship model between the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue is established to generate a temperature stress evaluation value of the metal roof, including: Constructing a covariance tensor of the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue; Performing orthogonal decomposition on the covariance tensor to obtain an eigenmode matrix; Establishing a mapping relationship between the eigenmode matrix and the temperature stress value through the least squares regression algorithm; Input the real-time characteristic mode matrix into the mapping relationship to generate the temperature stress evaluation value of the metal roof.

[0012] Furthermore, dynamically adjust the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters, including: Call the benchmark threshold range in the historical working condition database; Construct a fuzzy logic rule base for solar radiation intensity and environmental wind speed parameters; Match the correction factor in the fuzzy logic rule base according to the real-time solar radiation intensity and environmental wind speed parameters; Multiply the benchmark threshold range by the correction factor to generate the dynamic early warning threshold range.

[0013] Furthermore, when the temperature stress evaluation value exceeds the early warning threshold range, output the structural safety early warning signal of the metal roof, including: Map the temperature stress evaluation value to the three-dimensional grid coordinates of the metal roof; Identify the grid coordinate area that exceeds the early warning threshold range to generate a risk area identifier; Construct a risk state vector based on the risk area identifier; Encode the risk state vector into a structural safety early warning signal through the orthogonal coding conversion algorithm.

[0014] On the other hand, the present invention provides a temperature effect monitoring system for a metal roof, including the following modules: A temperature sensing module for obtaining three-dimensional temperature field data of the metal roof through a distributed temperature sensing network; A gradient extraction module for extracting the temperature gradient characteristic value of the metal roof from the three-dimensional temperature field data; A mechanical characteristic module for collecting the micro-vibration signal sequence of the metal roof constraint points through a piezoelectric film sensor array, processing the micro-vibration signal sequence by the Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extracting the dynamic constraint reaction force spectrum characteristic value from the instantaneous frequency-energy spectrum through the Riemannian manifold learning method; A thermal-mechanical coupling module for establishing a coupling relationship model between the temperature gradient characteristic value and the dynamic constraint reaction force spectrum characteristic value to generate the temperature stress evaluation value of the metal roof; A threshold dynamic module for dynamically adjusting the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters; A safety early warning module for outputting the structural safety early warning signal of the metal roof when the temperature stress evaluation value exceeds the early warning threshold range.

[0015] Compared with the prior art, the present invention has the following beneficial effects: Through the deep collaboration between the distributed temperature sensing network and the extraction of mechanical characteristics, the technical solution of the present invention constructs a holographic monitoring system for the temperature effect of metal roofs. Compared with the traditional single-point monitoring mode, its core breakthrough lies in realizing the real-time coupling analysis of the three-dimensional temperature field distribution and the structural mechanical response. Specifically, the spatial refinement extraction of the temperature gradient eigenvalue solves the problem of distorted evaluation of the temperature load action range, and the dynamic constraint reaction force spectrum eigenvalue generated based on the spectral analysis of micro-vibration signals establishes a quantitative correlation mechanism between temperature changes and structural constraint forces in the field of engineering monitoring. This collaborative analysis mode of multi-physical field characteristics enables the temperature stress evaluation process to truly reflect the thermo-mechanical coupling effect of the metal roof. Through the early warning decision-making mechanism with environmental parameter adaptability, the dynamic optimization of the safety protection of the metal roof is realized. Based on the fuzzy rule base of solar radiation intensity and environmental wind speed parameters, the early warning threshold range is dynamically adjusted, effectively overcoming the failure risk of fixed thresholds during seasonal alternation and extreme weather. The real-time comparison mechanism between the temperature stress evaluation value and the dynamic threshold, combined with orthogonal coding conversion to generate anti-interference early warning signals, not only greatly improves the early warning accuracy rate, but also forms a closed-loop control chain from data acquisition to risk response. This paradigm shift from "passive monitoring" to "active protection" provides reliable technical support for the full-life cycle safety management of large building metal roofs. Brief Description of the Drawings

[0016] Figure 1 It is a flowchart of a method for monitoring the temperature effect of a metal roof according to the present invention; Figure 2 It is a structural schematic diagram of a system for monitoring the temperature effect of a metal roof according to the present invention. Detailed Embodiments

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] Embodiment 1: Figure 1 A method for monitoring the temperature effect of a metal roof according to the present invention is given, which includes the following steps: S1: Obtain the three-dimensional temperature field data of the metal roof through a distributed temperature sensing network; S2: Extract the temperature gradient eigenvalue of the metal roof from the three-dimensional temperature field data; S3: Collect the micro-vibration signal sequence of the metal roof constraint points through a piezoelectric thin film sensor array, process the micro-vibration signal sequence using the Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extract the dynamic constraint reaction force spectrum eigenvalue from the instantaneous frequency-energy spectrum through the Riemannian manifold learning method; S4: Establish a coupling relationship model between the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue to generate the temperature stress evaluation value of the metal roof; S5: Dynamically adjust the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters; S6: When the temperature stress evaluation value exceeds the early warning threshold range, output the structural safety early warning signal of the metal roof.

[0019] Deploy multiple temperature measurement optical fiber laying subsystems at angles to each other on the metal roof. The included angle range between the temperature measurement optical fiber laying subsystems is set between 45 degrees and 135 degrees to ensure that the laying paths of all temperature measurement optical fiber laying subsystems cross-cover the entire metal roof area without monitoring blind spots. Specifically, when implemented, the first temperature measurement optical fiber path is laid along the ridge direction of the metal roof, the second temperature measurement optical fiber path is laid along the eave direction at the edge of the metal roof, and the third temperature measurement optical fiber path is laid along the diagonal direction of the metal roof. The triangular grid layout formed by the three groups of temperature measurement optical fiber paths ensures that any roof area is covered by at least two temperature measurement optical fiber paths. The fiber sensing node density of the temperature measurement optical fiber laying subsystem is controlled to be no less than 4 measurement points per square meter. For example, in a large terminal project, a total of 48,600 measurement points are set in a 12,000-square-meter roof area.

[0020] Collect the discrete temperature point data on the surface of the metal roof through the distributed optical fiber temperature measurement technology. The implementation process of this technology includes: the laser pulse generator emits a laser pulse signal with a central wavelength of 1550 nanometers to the temperature measurement optical fiber laying subsystem, and the single pulse width is controlled at the order of 100 nanoseconds; the receiving end collects the backscattered light signal for time-domain analysis, and inversely calculates the absolute temperature value of each sensing node on the optical fiber path according to the intensity ratio of the Stokes light and anti-Stokes light of the Raman scattering effect. Each temperature value corresponds to the specific coordinate position in the three-dimensional space coordinate system on the surface of the metal roof, and the coordinate positioning error is less than 2 millimeters, finally forming a discrete temperature point data set containing spatial coordinates and temperature values. For example, in the implementation process, the discrete temperature point data format recorded by a single sensing node is (X coordinate value, Y coordinate value, Z coordinate value, temperature value).

[0021] Perform a spatial correlation analysis on discrete temperature point data. This analysis consists of two core processing stages: In the first stage, establish a three-dimensional spatial coordinate grid matrix for the metal roof, with a grid cell size of 0.25 meters by 0.25 meters, and map each discrete temperature point data to the central coordinate position of the corresponding grid cell; In the second stage, calculate the temperature autocorrelation function values of adjacent grid cells, and use the Moran's I index calculation formula in spatial statistics to calculate the correlation coefficient: The Moran's I index calculation requires constructing a spatial weight matrix. Taking the current grid cell as the center, calculate the weighted average temperature of its surrounding eight neighboring grid cells. The weight coefficient is determined according to the Euclidean distance between grid cells. For example, the weight coefficient for a grid with an interval of 1 meter is set to 0.85, and the weight coefficient for a grid with an interval of 2 meters is reduced to 0.45. The results output by the spatial correlation analysis identify the coordinates of the temperature abnormally fluctuating areas.

[0022] Reconstruct and generate three-dimensional temperature field data through an improved interpolation algorithm. The improved interpolation algorithm implements three specific optimizations under the framework of standard Kriging interpolation: The first is to introduce a heat conduction boundary constraint condition, taking the fixed connection part of the metal roof as an interpolation boundary that the isotherm cannot cross. For example, the ridge weld area is set as an isothermal boundary; The second is to increase the gradient direction correction factor. When the included angle between the interpolation direction and the measured temperature change gradient direction is greater than 45 degrees, the interpolation weight in this direction is automatically multiplied by the correction coefficient of 0.6; The third is to set the iteration termination condition as the overall temperature change amount between two consecutive interpolation results being less than 0.5 degrees Celsius. During the reconstruction process, abnormal temperature point data is detected and removed in real time. The criterion for determining abnormal points is data points that exceed the average temperature value of the surrounding eight neighboring areas by more than 10 degrees Celsius. The final generated three-dimensional temperature field data has a spatial resolution of 0.0625 square meters. For example, in the reconstruction results of a stadium project, there are 192,000 temperature data points.

[0023] The temperature measurement optical fiber laying subsystem uses an armored distributed temperature measurement optical cable with an IP68 protection level; The spatial correlation analysis uses an ellipsoidal model to calculate the spatial weight coefficient, and the major axis direction of the ellipsoid is consistent with the roof slope direction; The improved interpolation algorithm dynamically adjusts the interpolation step during the data processing. When the monitored temperature gradient is greater than 15 degrees Celsius per meter, the interpolation step is automatically reduced to 60% of the original value. The relative error of the three-dimensional temperature field data obtained through the above implementation process does not exceed 1.2%. For example, the deviation between the measured temperature value and the reconstructed value is less than 0.46 degrees Celsius in a 38-degree Celsius high-temperature environment. All processing processes are realized based on the geometric topology structure and material heat conduction characteristics of the metal roof, ensuring that the three-dimensional temperature field data truly reflects the thermodynamic state of the roof.

[0024] The three-dimensional temperature field data is sourced from the real-time acquisition results of a distributed temperature sensing network. This data contains the temperature measurement values at each spatial coordinate point on the metal roof surface. When calculating the temperature difference values at different spatial positions on the metal roof, the Euclidean distance between adjacent nodes in the three-dimensional grid coordinate system is selected as the calculation benchmark. The specific implementation process is as follows: First, determine that the distance threshold range between spatial positions is controlled between 1 meter and 3 meters. For example, on the roof plane, grid nodes with a 2-meter distance are selected as the calculation units. Then, compare the temperature values at multiple spatial position points within each calculation unit. The absolute temperature difference algorithm is used to calculate the temperature difference value, that is, the absolute value of the difference between the temperature value at the current spatial position point and the temperature value at the adjacent spatial position point. Taking the ridge position of the metal roof as an example, when the temperature at the ridge point is 52 degrees Celsius and the adjacent eaves point is 38 degrees Celsius, the temperature difference value in this area is 14 degrees Celsius. During the implementation process, a temperature difference value matrix is established to store the calculation results of all spatial positions. The distance threshold is set based on ensuring that the grid cells cover the temperature-sensitive areas and meet the three-dimensional reconstruction accuracy requirements.

[0025] Identify the physical boundary constraint conditions of the metal roof based on the three-dimensional temperature field data. The physical boundary constraint conditions are defined as the structural characteristics of the fixed connection parts of the metal roof. This identification process includes three stages: In the first stage, potential boundary coordinates are located by detecting the temperature mutation gradient. When the temperature change rate between adjacent nodes exceeds 10 degrees Celsius per meter, the boundary identification mechanism is triggered. In the second stage, verify the effectiveness of the identified physical boundary by comparing with the coordinate library of the structural joint positions in the building structure design drawing. In the third stage, establish a constraint attribute classification system, and classify the constraint types into three categories: rigid connection constraint type, sliding connection constraint type, and free boundary constraint type. For example, the weld seam of the metal roof panel is classified as the rigid connection constraint type, and the roof expansion joint is classified as the sliding connection constraint type. Finally, an eigen description table containing the boundary spatial position coordinates, the constraint action radius, and the direction of the constraint force action is output. The set range of the constraint action radius is from 30 centimeters to 150 centimeters, and the specific value depends on the size of the connecting piece.

[0026] The temperature difference value is corrected in combination with physical boundary constraint conditions, and a differential processing mechanism is implemented according to the constraint type during the correction process. For the rigid connection constraint area, the temperature difference value is multiplied by a heat transfer suppression coefficient, and the value range of the heat transfer suppression coefficient is set between 0.5 and 0.9, and the value is determined by the thermal conductivity characteristics of the fixed connection component material. For example, the coefficient in the stainless steel connection component area is taken as 0.7; for the sliding connection constraint area, a displacement compensation factor is added to the temperature difference value, and the value of the compensation factor is equal to the theoretical displacement prediction value divided by the coefficient of linear expansion of the material; for the free boundary area, the original calculation result of the temperature difference value is directly used without correction. In the implementation case of a large terminal building, after the metal roof expansion joint area is identified as the sliding connection constraint type, a displacement compensation factor of 0.28 is added to all temperature difference values within a radius of 80 cm in this area. When the correction mechanism is executed, the parameters in the physical boundary constraint condition database are automatically matched to establish a correspondence matrix between the temperature difference value and the constraint characteristics.

[0027] The eigenvector is extracted from the corrected temperature difference value as the temperature gradient eigenvalue through the principal component analysis algorithm. The implementation process of the principal component analysis algorithm includes four core operation steps. First, the covariance matrix of the corrected temperature difference value is constructed, and the matrix dimension is the same as the number of roof space grid cells; second, the eigenvalue decomposition is performed on the covariance matrix to solve the set of eigenvectors; then, the dominant eigenvectors are selected according to the eigenvalue size sorting, and the selection criterion is set as the principal component direction with a cumulative contribution rate of more than 80%; finally, the selected eigenvectors are normalized. The extracted temperature gradient eigenvalue includes two key parameters: the direction eigenvector component and the modulus eigenvalue component. The direction eigenvector component represents the projection angle of the dominant heat flow direction in the three-dimensional space coordinate system, and the modulus eigenvalue component represents the temperature gradient intensity. During the implementation process, the dimension of the eigenvector is consistent with the three-dimensional space coordinate system. For example, the direction eigenvector component of the temperature gradient eigenvalue extracted from a certain stadium roof area is (0.89, 0.44, 0.09), and the modulus eigenvalue component is 28.3 degrees Celsius per meter, indicating that the heat flow direction forms an angle of 26 degrees with the roof plane and the gradient intensity is 28.3 degrees Celsius per meter. The iteration termination condition of the principal component analysis algorithm is set as the change amount of the eigenvector direction is less than 0.15 degrees or the change amount of the modulus is less than 0.4 degrees Celsius per meter.

[0028] A special scenario processing mechanism is configured during the implementation process: when rainfall or snowfall weather is detected by the weather station, the surface temperature difference correction mode is automatically activated, and this mode excludes the temperature sudden change points caused by surface moisture evaporation when calculating the temperature difference value; when the deviation between the temperature gradient direction and the roof slope direction exceeds 30 degrees, the system triggers a direction recalibration process, and this process optimizes the result by weighted averaging of three groups of eigenvectors in the adjacent area. The physical property parameters of the metal roof material are stored in the system characteristic parameter database. For example, the value of the coefficient of linear expansion parameter of steel is 12×10-6 / °C, the linear expansion coefficient parameter of the aluminum alloy material is taken as 23×10 -6 / °C, providing an accurate calculation basis for the temperature difference value correction. All the extracted temperature gradient eigenvalue components are displayed in real time in the building model through a 3D visualization system. Different color depths are used to identify the modulus eigenvalue components, and arrow vectors are used to identify the direction eigenvector components, forming a complete roof heat flow distribution map. The final output result is automatically associated with the temperature effect evaluation process, providing thermodynamic characteristic inputs for subsequent temperature stress analysis.

[0029] Through the temperature gradient eigenvalue extraction process, the heat conduction characteristics of the metal roof are quantified into mathematically computable and analyzable features. For example, during the change of winter sunshine radiation intensity, the maximum modulus eigenvalue component recorded in the south area of the roof reaches 42.7 °C / m. This quantification result provides accurate input parameters for temperature stress evaluation. During the implementation process, a data quality monitoring mechanism is established. When it is detected that the modulus change of the temperature gradient eigenvalues extracted continuously five times in the same area exceeds 15%, a recalibration instruction for the distributed temperature sensing network is automatically triggered to ensure the continuity and reliability of the temperature effect monitoring data. The implementation case of the eigenvalue extraction algorithm shows that in a terminal project with a roof area of 30,000 square meters, the complete temperature gradient eigenvalue extraction cycle is controlled to be completed within 25 seconds, meeting the real-time requirements of structural health monitoring.

[0030] Piezoelectric film sensor arrays are installed at the positions of the metal roof restraint points, and the restraint points include fixed support nodes, weld joint parts, and expansion joint connector areas. The effective measurement range of each piezoelectric film sensor covers an area with a diameter of 12 mm, and the sensor spacing is set so that the maximum distance between adjacent restraint points does not exceed 1.8 m. For example, a sensor array is arranged in the key area of the roof ridge to form a 0.9 m equidistant grid. The sensor array is directly attached to the metal surface, and the mechanical vibration is converted into an electrical signal waveform through a charge conversion circuit. The sampling frequency is set to 20,000 Hz, and the signal resolution reaches 0.001 m / s². The collected micro-vibration signal sequence contains time series and spatial coordinate information. For example, the coordinate of a certain weld position is recorded as three-dimensional values, corresponding to a continuous vibration data point sequence collected within 5 seconds.

[0031] The piezoelectric film sensor array acquires the vibration response of the metal roof constraint point in real time to form a micro-vibration signal sequence, which is transmitted to the signal processing unit through the data acquisition system. After receiving the original signal, the signal processing unit first performs preprocessing operations: automatically detect abnormal pulse signals and use median filtering to eliminate interference points. The filter window width is set to 51 sampling points; then the signal is normalized so that all amplitude values are mapped to the standard range of -1.0 to 1.0. The processed micro-vibration signal sequence is stored in a time-space matrix format. The matrix row represents the time series index and the column represents the spatial sensor node number. For example, 128 sensor nodes form a 128-column data matrix.

[0032] The micro-vibration signal sequence is decomposed into intrinsic mode function components by Hilbert-Huang transform, which implements the empirical mode decomposition algorithm. The decomposition process starts with an iterative screening operation: first, the local maximum and minimum points of the input signal are identified, and then the cubic spline interpolation method is used to connect these extreme points to form the upper envelope and the lower envelope. Then the mean curve of the upper and lower envelopes is calculated, and finally the mean curve is subtracted from the original signal to obtain the preliminary modal components. After each iteration, check whether the component meets the intrinsic mode function conditions: the difference between the number of zero crossings and the number of extreme points does not exceed one; the local mean of all positions approaches zero. The iterative termination condition is set to the standard deviation of three consecutive screenings less than 0.3, and finally multiple intrinsic vibration components that meet the conditions are output. For example, a weld vibration signal is decomposed into five groups of intrinsic mode function components with a frequency distribution of 180 Hz to 1900 Hz.

[0033] The Hilbert spectrum analysis of the intrinsic mode function components is performed to generate the instantaneous frequency-energy spectrum, and the orthogonal demodulation technology is used in the analysis process. First, the Hilbert transform is performed on each intrinsic mode function component to generate the corresponding analytical signal; secondly, the instantaneous phase angle of the analytical signal is calculated, and the instantaneous frequency value is solved by phase angle differential operation, and the frequency calculation time step is set to 0.05 milliseconds; finally, the frequency-energy relationship matrix is constructed according to the instantaneous frequency value and the envelope amplitude value at the corresponding moment. The instantaneous frequency-energy spectrum eventually forms a two-dimensional distribution diagram, with the horizontal axis representing the frequency value, the vertical axis representing the energy density value, and the spectrum resolution set to 0.5 Hz. For example, the peak energy density recorded at 125 Hz at a certain expansion joint position reaches 1.8 volts per Hz.

[0034] Map the instantaneous frequency - energy spectrum to the Riemannian manifold space to construct a tangent vector field, and the mapping process is realized based on differential geometry theory. First, define an orthogonal tangent space coordinate system in the frequency - energy plane, and set the origin of the coordinate system at the lowest - energy point of the spectrum. Secondly, convert each data point in the instantaneous frequency - energy spectrum into a tangent vector. The length of the tangent vector is determined by the natural logarithm of the energy density value, and the direction angle is determined by the phase difference between the frequency value and the reference frequency value. After the tangent vector field is constructed, a set of spatial vectors is formed, and each vector is associated with a specific spatial coordinate position. For example, the tangent vector modulus of 0.43 and the direction angle of 27.5 degrees are generated at the point corresponding to a frequency of 325 Hz. The vector set is automatically associated with the three - dimensional coordinate system of the roof to establish a mapping relationship.

[0035] Calculate the eigenvalues of the geodesic distance matrix of the tangent vector field, and the calculation process implements an iterative optimization algorithm. First, construct a position connection graph of the tangent vectors, and set the connection condition as the frequency difference is less than 5 Hz and the spatial distance is less than 0.7 m. Secondly, solve the geodesic distance between each pair of connected points. This distance represents the shortest path length on the manifold surface, and the Floyd algorithm is used to iteratively calculate the shortest path, and the iterative convergence threshold is set as the maximum value of the difference between adjacent distances does not exceed 0.03. Finally, store the calculation result as a symmetric distance matrix, and the matrix dimension is equal to the total number of tangent vectors. In the implementation case, the distance matrix formed by 256 tangent vectors in a certain area of the roof is processed by eigenvalue decomposition using the Jacobi method.

[0036] Select the eigenvector corresponding to the largest eigenvalue of the geodesic distance matrix as the eigenvalue of the dynamic constraint reaction spectrum, and this process implements a standardization processing flow. First, obtain the set of all eigenvalues through the eigenvalue decomposition algorithm, and control the calculation accuracy of the eigenvalue decomposition within 0.001. Secondly, sort all the eigenvalues in descending order, and select the largest eigenvalue as the main eigen - component. Then, extract the eigenvector corresponding to this main eigenvalue and perform vector normalization processing. Finally, store the normalized eigenvector in association with the corresponding physical space coordinates. The final expression form of the eigenvalue of the dynamic constraint reaction spectrum includes two components: modulus and phase angle. For example, the modulus eigen - quantity of the eigenvector extracted from a certain fixed support node is 0.94, and the phase - angle eigen - quantity is 31.6 degrees, indicating the strength and direction characteristics of the constraint reaction at this position. The calculation time of the entire feature extraction algorithm is controlled to be completed within 350 milliseconds.

[0037] Implementation process configures special working condition handling rules: when the environmental vibration intensity exceeds the set threshold, the noise reduction processing mode is automatically activated; when the temperature exceeds 80 degrees Celsius, the temperature compensation algorithm is started to calibrate the sensor sensitivity. The mechanical parameters of the metal roof material are stored in the system physical property database. For example, the elastic modulus of steel takes a value of 206 GPa, and the Poisson's ratio takes a value of 0.28, providing a basis for physical property parameters for feature extraction. All generated dynamic constraint reaction force spectrum eigenvalue is real-time displayed in the 3D model through a color cloud map. The color depth of the cloud map corresponds to the magnitude of the eigenvector, and the vector direction identifies the reaction force direction.

[0038] The feature extraction system sets up a data quality monitoring mechanism: when the convergence time of the eigenvalue calculation exceeds 450 milliseconds for three consecutive times, the algorithm parameter configuration is automatically optimized; when the fluctuation amplitude of the eigenvector direction angle exceeds 15 degrees, the direction recalibration process is triggered. In the actual measurement of a stadium roof project, piezoelectric film sensors cover 32 key constraint points. The average number of eigenvalue convergence iterations calculated by the geodesic distance matrix is 17 times, and the entire process takes less than 1.9 seconds to complete. The correlation coefficient between the dynamic constraint reaction force spectrum eigenvalue and the temperature load experimental data reaches 0.93, verifying the technical reliability of the metal roof temperature effect monitoring.

[0039] The temperature gradient eigenvalue is derived from the thermodynamic state data of the metal roof, including two elements: the direction eigenvector component and the magnitude eigenvalue component. The direction eigenvector component identifies the angular projection of the main heat flow direction in the three-dimensional coordinate system, and the magnitude eigenvalue component reflects the intensity of the temperature change rate (in degrees Celsius per meter). The dynamic constraint reaction force spectrum eigenvalue characterizes the structural mechanical response, including two parameters: the magnitude eigenquantity and the phase angle eigenquantity. The magnitude eigenquantity represents the binding force intensity (dimensionless), and the phase angle eigenquantity represents the reaction force direction angle (in degrees). These two types of eigenvalues are synchronously input into the coupled modeling system through a data interface, and the input timestamp alignment accuracy is controlled within 20 milliseconds to ensure data consistency in the time dimension.

[0040] In the process of constructing the covariance tensor of the structural temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue, first define a three-dimensional tensor structure: the first dimension corresponds to the modulus eigenvalue component of the temperature gradient eigenvalue, the second dimension corresponds to the modulus eigenvalue quantity of the dynamic constraint reaction force spectrum eigenvalue, and the third dimension corresponds to the angular difference of the direction correlation between the two types of eigenvalues. The calculation of tensor elements adopts the covariance formula: the covariance value within each spatial grid cell is the product of the modulus eigenvalue component of the temperature gradient eigenvalue and the modulus eigenvalue quantity of the dynamic constraint reaction force spectrum eigenvalue within the cell minus the product of their respective means. Before calculation, data standardization processing is performed: the modulus eigenvalue component of the temperature gradient eigenvalue is divided by its historical maximum value for dimension normalization, and the modulus eigenvalue quantity of the dynamic constraint reaction force spectrum eigenvalue is divided by its historical mean for dimensionless processing. The dimension of the finally formed covariance tensor is consistent with the number of roof grid partitions. For example, a 32-partition system generates a 32×32×32 three-dimensional tensor matrix, and each tensor element represents the thermo-mechanical coupling strength of a specific spatial region.

[0041] Perform orthogonal decomposition on the covariance tensor to obtain the eigenmode matrix, and implement the high-order singular value decomposition algorithm. The decomposition process includes four stages: in the first stage, expand the three-dimensional tensor into a matrix sequence, and each slice matrix corresponds to a specific angular difference interval of direction correlation; in the second stage, perform singular value decomposition on each slice matrix, and retain the principal components with a singular value contribution rate exceeding 85%; in the third stage, reconstruct the core tensor and calculate the eigenvectors; in the fourth stage, sort the eigenvectors according to the variance contribution rate to form the eigenmode matrix. The termination condition for decomposition iteration is set that the change amount of eigenvectors between two adjacent times is less than 0.03, and the maximum number of iterations is limited to 100 times. Each column vector of the eigenmode matrix corresponds to an independent eigenmode, and the row number of the matrix is associated with a specific spatial grid coordinate. For example, the eigenvector corresponding to a certain roof panel area (X3,Y7) is [0.48, 0.32, 0.21, 0.07], indicating the weight coefficient distribution of the four main modes in this area.

[0042] The mapping relationship between the characteristic mode matrix and the temperature stress value is established through the least - squares regression algorithm, and a step - by - step training mechanism is adopted. First, collect the historical working condition data set: retrieve the characteristic mode matrix records and the corresponding measured temperature stress values (unit: megapascal) from the metal roof structure health monitoring system, and the data set size is not less than 500 groups of valid samples. Second, construct a multiple linear regression equation: set the dependent variable as the temperature stress value, and the independent variable as the weight coefficient vector of the characteristic mode matrix. Then, perform the regression coefficient solution: calculate the coefficient value through the algorithm of minimizing the sum of squared residuals, and the residual threshold is set to 0.25 megapascal. Finally, verify the model accuracy: use the ten - fold cross - validation method to check the correlation coefficient between the predicted value and the measured value. The final expression form of the mapping relationship is the weight coefficient vector. For example, the coefficient vector [0.87, 0.35, - 0.12, 0.09] in the regression equation represents the contribution weights of the four characteristic modes to the temperature stress. During the model training process, abnormal sample points are automatically excluded, and the criterion for determining abnormal points is the data points exceeding 3 times the standard deviation.

[0043] Input the real - time characteristic mode matrix into the mapping relationship to generate the temperature stress evaluation value of the metal roof, and dynamic verification is implemented during the generation process. First, read the current characteristic mode matrix data and match the weight coefficient vector trained historically according to the grid coordinate index. Second, calculate the temperature stress evaluation value: the temperature stress value of each grid cell is equal to the dot product of the row vector of the characteristic mode matrix of this cell and the regression coefficient vector. Finally, perform the physical range verification: when the calculation result exceeds the yield strength range of the metal material, secondary calculation is automatically triggered. The evaluation value is output as a spatial distribution matrix, and the matrix elements represent the temperature stress values (unit: megapascal) of each grid point. For example, the evaluation value of the ridge area (X5,Y9) is 152.7 megapascal. The system establishes an output quality monitoring mechanism: when it is detected that the stress value mutation between adjacent grids exceeds 25 megapascal, the quality of the input characteristic values is automatically verified; when the material temperature exceeds 120 degrees Celsius, the high - temperature stress correction algorithm is started to adjust the output value.

[0044] Implement a material property adaptive mechanism during the implementation process: automatically match the corresponding material parameter database according to the actual material type of the metal roof panel. For example, the aluminum alloy plate matches the elastic modulus parameter of 70000 megapascal and the thermal expansion coefficient of 0.000023 per degree Celsius, and the galvanized steel plate matches the elastic modulus parameter of 206000 megapascal and the thermal expansion coefficient of 0.000012 per degree Celsius. The special working condition handling rules include: start the dynamic load compensation mode during the typhoon warning period and adjust the stress evaluation value by increasing the wind speed correction factor; when the ice and snow load is detected, activate the freeze - thaw effect correction algorithm.

[0045] Evaluate the self-optimization function of the system settings parameters: When the cumulative monitored data exceeds 10,000 groups, automatically update the least squares regression coefficient; When the change rate of the characteristic mode exceeds 15% continuously for 10 times, trigger the orthogonal decomposition recalibration. In the application case of the terminal building, the roof is divided into 256 evaluation grids, the characteristic mode matrix contains 8 main mode components, and the spatial resolution of the temperature stress evaluation value output by the regression calculation reaches 0.5 meters. The evaluation results are displayed in real time through a three-dimensional cloud map, using a gradient color from blue to red to identify the stress range from 0 MPa to 250 MPa, and the warm color areas identify the high-risk parts. The entire temperature stress evaluation process takes an average of 1.3 seconds to complete on a standard hardware platform, meeting the real-time monitoring requirements of large building roofs.

[0046] The system maintenance unit configures a data traceability mechanism: Store snapshots of the characteristic mode matrix for the last 30 days, which can be traced back and analyzed when abnormal evaluation values are detected; Establish a material aging correction model to automatically adjust the elastic modulus parameters according to the service life of the metal roof. The finally output temperature stress evaluation value matrix is transmitted to the structural health monitoring platform through the safety warning interface, providing a quantitative basis for the maintenance decision of the metal roof.

[0047] When calling the reference threshold range in the historical working condition database, the historical working condition database system stores the complete temperature stress record data of the metal roof structure in the past five monitoring years. The reference threshold range is defined as the critical interval of the temperature stress for the safe bearing of the metal roof, and this interval is determined by statistical analysis of historical monitoring data: First, extract the statistical distribution characteristics of all temperature stress monitoring values, calculate the percentile value distribution, take the 95th percentile value as the initial value of the upper limit reference threshold, and take the 5th percentile value as the initial value of the lower limit reference threshold; Then screen out the extreme working condition data and exclude the abnormal records exceeding 3 times the standard deviation range; Finally, form a spatial distribution matrix of the reference threshold range, where the rows of the matrix correspond to the roof space grid partition numbers, and the columns represent the classification results of different season types. The spatial distribution matrix of the reference threshold range performs data update operations in March every year. When updating, the newly added complete annual monitoring data is incorporated to recalculate the percentile value distribution. For example, the reference threshold range record for the ridge area of a terminal building project in summer is from 112 MPa to 168 MPa, and this data is obtained by analyzing 12,500 groups of effective data from five summer monitoring cycles from 2019 to 2023.

[0048] In the process of constructing the fuzzy logic rule base for solar radiation intensity and ambient wind speed parameters, the fuzzy set definitions of input and output variables are clarified: the solar radiation intensity parameter is divided into three fuzzy subsets, including the weak radiation intensity level, the medium radiation intensity level, and the strong radiation intensity level; the ambient wind speed parameter is divided into three fuzzy subsets, including the gentle breeze level, the moderate wind level, and the strong wind level; the fuzzy set of the defined output correction factor is divided into three categories: the decrease level, the maintain level, and the increase level. The construction of the rule base includes two stages: rule generation and parameter setting. In the rule generation stage, the correlation between input and output is determined through historical data regression analysis, and 9 core fuzzy rules are established. In the parameter setting stage, the membership function parameters of each fuzzy set are determined. For example, the radiation parameter value boundary of the weak radiation intensity level is set to 0 watts per square meter to 400 watts per square meter, and the strong radiation intensity level is set to 800 watts per square meter to 1200 watts per square meter. All fuzzy rules are expressed using standard conditional statements. For example, rule 6 states that when the solar radiation intensity is at the medium radiation intensity level and the ambient wind speed is at the moderate wind level, the correction factor takes the median value of 1.0 at the maintain level. After the rule base is established, it is verified and tested with 300 sets of historical data, and the fitting error is controlled within 5.3%.

[0049] When matching the correction factor in the fuzzy logic rule base according to the real-time solar radiation intensity and ambient wind speed parameters, three consecutive processing stages are executed: in the fuzzification stage, the measured values of the real-time collected solar radiation intensity (unit: watts per square meter) and the ambient wind speed (unit: meters per second) are converted into membership degree values of each fuzzy subset through the triangular membership function. For example, when the radiation value is 650 watts per square meter, the membership degree in the medium radiation intensity level reaches 0.82. In the rule triggering stage, the applicability intensity of all 9 rules is calculated, and the applicability intensity value is the algebraic product result of the membership degrees of the input variables. In the defuzzification stage, the centroid calculation method is used to perform a weighted average on the output results of all rules to calculate the final correction factor value. The real-time matching process is executed once every 10 minutes and is automatically shortened to once per minute under extreme weather conditions. Matching example: when the real-time solar radiation intensity is 880 watts per square meter (membership degree of the strong radiation intensity level is 0.92) and the ambient wind speed is 3.8 meters per second (membership degree of the moderate wind level is 0.67), the output value of the correction factor is calculated to be 1.15.

[0050] When generating a dynamic warning threshold range by multiplying a reference threshold range by a correction factor, first extract the base value of the current spatial grid cell from the spatial distribution matrix of the reference threshold range; secondly, perform matrix multiplication operations: multiply the lower limit value of the reference threshold by the correction factor value, and multiply the upper limit value of the reference threshold by the correction factor value; then perform physical boundary verification: when the calculation result exceeds the allowable strength of the metal material, it is automatically truncated to 90% of the allowable strength of the material. The output result of the dynamic warning threshold range is in the form of a spatial matrix, and the matrix elements contain upper and lower limit value pairs. For example, the original reference threshold range for a certain metal roof panel area is from 128 MPa to 184 MPa, and when the correction factor is 1.15, the new dynamic warning threshold range generated is from 147 MPa to 211 MPa. The system establishes a threshold rationality monitoring mechanism: when it is detected that the lower limit of the dynamic threshold exceeds 85% of the upper limit, the manual review process is automatically started; when a certain spatial unit has not been updated for three consecutive calculations, a data source verification alarm is triggered.

[0051] Special scenario automatic processing rules are configured during the implementation process: when the rainfall weather sensor is activated, the rainy day correction mode is enabled, and this mode forces the solar radiation intensity parameter to zero; when the ambient wind speed sensor fails, the average wind speed value in the last three hours is used as an alternative parameter; when the metal roof temperature exceeds 150 degrees Celsius, the high-temperature material softening correction coefficient is activated. The material characteristic parameters of the metal roof are called in real time from the material database. For example, the yield strength of 304 stainless steel is 205 MPa, and the ultimate strength is 515 MPa, providing a basis for threshold boundary verification. The update frequency of the dynamic warning threshold range under standard working conditions is set to once every 10 minutes, and the update process is controlled to be completed within 0.7 seconds.

[0052] The system realizes an adaptive optimization function: incrementally update the reference threshold range automatically every quarter, and incorporate the newly added monitoring cycle data into the statistical calculation; when the matching error of the fuzzy rule base exceeds 10% for 15 consecutive days, the rule parameter re-learning program is automatically started. In the actual measurement project of an industrial factory building, the dynamic warning threshold range reduces the false alarm rate of the system from 28.7% of the traditional method to 7.3%. The finally generated dynamic warning threshold range data is transmitted to the warning decision-making system through a secure interface, providing a dynamic judgment benchmark for the safety status assessment of the metal roof.

[0053] The monitoring platform establishes a full-cycle traceability function: stores the dynamic threshold data sequence for the last 45 days, supports the playback analysis of the historical threshold trajectory; when a certain spatial partition exceeds the dynamic threshold three consecutive times, automatically mark this area as a key monitoring area and increase the monitoring frequency. The average execution time of the entire dynamic adjustment process on the edge computing device is 0.8 seconds, meeting the real-time response requirements for the temperature effect monitoring of the metal roof.

[0054] The temperature stress evaluation value is derived from the output data of the temperature effect evaluation system. This data is stored in the form of a spatial matrix, and the matrix row and column numbers are indexed corresponding to the three-dimensional grid coordinate system of the metal roof. When mapping the temperature stress evaluation value to the three-dimensional grid coordinates of the metal roof, a spatial coordinate index mapping table is established: the index table contains the three-dimensional coordinate values of the center point of each grid cell, and the coordinate value accuracy reaches 0.01 meters; rapid positioning is achieved through the hash mapping algorithm, and the hash key value is generated by combining the grid row and column numbers, and the hash value stores the corresponding coordinate (X, Y, Z) numerical pair. The mapping process ensures that each temperature stress evaluation value is accurately associated with the physical location of the roof. For example, the temperature stress evaluation value of 148 MPa is recorded at the coordinate position (12.5 meters, 24.8 meters, 8.7 meters) of the grid number 5 in the roof partition number.

[0055] When generating a risk area identifier for the grid coordinate area that exceeds the warning threshold range, the warning threshold range uses dynamically updated spatial matrix data. Three key operations are implemented in the identification process: First, perform a point-to-point threshold comparison operation, and compare the temperature stress evaluation value of each grid cell with the upper and lower limit values of the dynamic warning threshold range at the same position in real time; Second, mark the risk grid cells. When the temperature stress evaluation value is greater than the upper limit value or less than the lower limit value of the dynamic warning threshold range, it is marked as a risk state; Finally, perform regional clustering analysis, and use the breadth-first search algorithm to identify the spatially continuous risk grid set to form a closed polygon risk area. The output of the risk area identifier includes the polygon vertex coordinate sequence and the unique area identification code. For example, the vertex sequence of area number RISK-2023-109 includes 8 position points such as coordinate points (X7, Y8), (X7, Y9), (X8, Y9).

[0056] When constructing a risk state vector based on the risk area identifier, the risk state vector is defined as a multi-dimensional feature array. The vector construction includes four-dimensional parameters: The first-dimensional parameter calculates the projected area value of the risk area (in square meters), which is obtained by summing the areas of the triangular grids through the polygon triangulation method. For example, the projected area of a certain area is recorded as 3.8 square meters; The second-dimensional parameter extracts the maximum stress excess value (in MPa), which takes the maximum absolute value of the excess part of all grid cells in the area; The third-dimensional parameter calculates the risk intensity index, which is equal to the product of the projected area value and the maximum stress excess value; The fourth-dimensional parameter records the area centroid coordinate value, which is calculated by the average value of the polygon vertex coordinates. The vector construction implements normalization processing: the projected area value is divided by the roof total area normalization coefficient, and the maximum stress excess value is divided by the material yield strength normalization coefficient.

[0057] When encoding the risk status vector into a structural safety warning signal through the orthogonal coding conversion algorithm, the Hadamard matrix is used as the orthogonal basis in the orthogonal coding conversion algorithm. The encoding process includes five standard steps: First, perform vector data quantization, multiplying each dimension parameter of the risk status vector by 1000 and rounding to convert it into an integer sequence; second, generate the Hadamard matrix, and the matrix order is equal to the vector dimension number; then perform the orthogonal transformation operation, multiplying the risk status vector by the Hadamard matrix to generate an intermediate coding sequence; then add a cyclic redundancy check code, using the CRC-16 algorithm to generate a 2-byte check code; finally, convert it into a hexadecimal string format for output. For example, a risk status vector [38, 185, 7030, 12524875] is encoded to output the warning signal "8E3A-F5C2-9B01-D4F6". An error control mechanism is set in the encoding process: when the check code verification fails, the last three valid signal data packets are automatically retransmitted.

[0058] In the implementation process, an emergency handling rule is configured: when a super-large risk area (projection area exceeds 8 square meters) is detected, the fast encoding mode is started, and the encoding processing time is compressed to be completed within 50 milliseconds; when there are multiple risk areas at the same time, a priority scheduling strategy is adopted, and the priority order is sorted according to the risk intensity index. The metal roof material parameters are called from the material database in real time. For example, the yield strength parameter of 304 stainless steel material takes a value of 205 MPa, and the ultimate strength parameter takes a value of 515 MPa, providing a physical basis for the normalization calculation.

[0059] The system implements a signal optimization mechanism: when the same risk area is continuously warned more than three times, the warning level is automatically increased; when the risk area is lifted, a warning cancellation signal code is generated. In the terminal building project, the measured error rate of warning signal transmission is lower than one in one hundred thousand. The warning signal is transmitted to the monitoring center through the industrial wireless sensor network, and the communication protocol follows the IEEE802.15.4g standard specification.

[0060] The monitoring platform establishes a full-cycle traceability system: stores all warning signal data and associated risk area snapshots in the last 48 hours; when the signal check fails three times in a row, an artificial intervention request is automatically initiated. The implementation data shows that the average end-to-end delay from risk identification to signal output is 0.85 seconds, and the matching accuracy between the warning signal and the actual structural safety status reaches 94.6%. The entire processing flow runs stably on the industrial Internet of Things edge computing platform, and the response timeliness meets the real-time safety monitoring requirements of the metal roof. The finally output structural safety warning signal triggers the sound and light alarm device in the monitoring center and the mobile terminal push system, forming a complete closed-loop of monitoring-warning-response.

[0061] The system maintenance unit configures a signal self-check mechanism: performs an offline signal quality test every day at dawn, and the test data packets cover all encoding combinations; automatically switches to an anti-interference encoding mode when the signal-to-noise ratio of the communication environment is lower than 15 decibels. In the actual measurement of the stadium project, the early warning system accurately predicted the buckling risk of the ridge part 32 minutes in advance, avoiding structural safety accidents. All early warning signal records are synchronously stored in the cloud disaster recovery system, supporting the traceability analysis of historical data for five years.

[0062] The metal roof temperature effect monitoring method breaks through the data decoupling limitation of traditional monitoring technologies through the deep coupling of multi-source sensing and intelligent algorithms: Adopts a collaborative layout scheme of a distributed temperature sensing network and a piezoelectric film sensor array, establishes a non-linear correlation between the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue, and replaces the conventional direct measurement of strain / displacement. Among them, the dynamic constraint reaction force spectrum eigenvalue decomposes the signal eigenmode through Hilbert-Huang transform and introduces Riemannian manifold learning to construct a spectral geometric invariant. This interdisciplinary technology integration breaks through the analysis paradigm of mechanical response signals and significantly improves the perception accuracy of temperature-structure coupling characteristics.

[0063] Constructs a covariance tensor of the temperature gradient eigenvalue and the mechanical eigenvalue, extracts the thermo-mechanical coupling main mode through high-order orthogonal decomposition, and then establishes a temperature stress evaluation mapping model through least squares regression. This model transforms the complex interaction relationship between temperature load and structural response into a quantifiable feature matrix operation, solves the problem of the lack of a thermo-mechanical parameter correlation mechanism in traditional methods, and provides an accurate mathematical model basis for temperature stress evaluation.

[0064] The early warning threshold range is based on the solar radiation intensity and environmental wind speed parameters, and dynamically corrects the reference threshold through a fuzzy rule base. This mechanism adjusts the safety boundary in real time in combination with meteorological parameters, realizing a fundamental transformation from a static threshold to environmental adaptability. Especially, the risk state vector is transformed into an anti-interference early warning signal through an orthogonal coding conversion algorithm, ensuring communication reliability in complex industrial environments.

[0065] The above technical solutions form a complete technical closed-loop through three-level integration of cross-scale signal processing (micro-vibration sequence → manifold feature), cross-dimensional modeling (three-dimensional temperature field → thermo-mechanical tensor), and dynamic decision optimization (fuzzy rule → orthogonal coding). Its core breakthrough lies in upgrading the temperature effect monitoring from a passive mode of "single-point data acquisition - fixed threshold judgment" to an active protection system of "multi-source coupling analysis - environmental adaptive decision-making", overcoming the inherent defect that existing technologies cannot quantify the dynamic response of temperature-structure.

[0066] Example 2: Figure 2The structural schematic diagram of a temperature effect monitoring system for a metal roof according to the present invention is given. A temperature effect monitoring system for a metal roof includes the following modules: A temperature sensing module, configured to obtain three-dimensional temperature field data of the metal roof through a distributed temperature sensing network; A gradient extraction module, configured to extract temperature gradient characteristic values of the metal roof from the three-dimensional temperature field data; A mechanical characteristic module, configured to collect a micro-vibration signal sequence of the metal roof constraint points through a piezoelectric film sensor array, process the micro-vibration signal sequence by using Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extract dynamic constraint reaction force spectrum characteristic values from the instantaneous frequency-energy spectrum through a Riemannian manifold learning method; A thermal-mechanical coupling module, configured to establish a coupling relationship model between the temperature gradient characteristic values and the dynamic constraint reaction force spectrum characteristic values, and generate a temperature stress evaluation value of the metal roof; A threshold dynamic module, configured to dynamically adjust the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters; A safety early warning module, configured to output a structural safety early warning signal of the metal roof when the temperature stress evaluation value exceeds the early warning threshold range.

[0067] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and threshold selections in the calculations are set by those skilled in the art according to the actual situation.

[0068] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0069] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application of the technical solution and the invention constraints. Those skilled in the art can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0070] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0071] In several embodiments provided by the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or modules can be in electrical, mechanical, or other forms.

[0072] As described above, this is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0073] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for monitoring the temperature effect of a metal roof, characterized in that, It includes the following steps: S1: Obtain the three-dimensional temperature field data of the metal roof through a distributed temperature sensing network; S2: Extract the temperature gradient eigenvalue of the metal roof from the three-dimensional temperature field data; S3: Collect the micro-vibration signal sequence of the metal roof constraint points through a piezoelectric film sensor array, process the micro-vibration signal sequence using the Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extract the dynamic constraint reaction force spectrum eigenvalue from the instantaneous frequency-energy spectrum through the Riemannian manifold learning method; S4: Establish a coupling relationship model between the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue to generate the temperature stress evaluation value of the metal roof; S5: Dynamically adjust the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters; S6: When the temperature stress evaluation value exceeds the early warning threshold range, output the structural safety warning signal of the metal roof.

2. The temperature effect monitoring method for a metal roof according to claim 1, characterized in that Obtain the three-dimensional temperature field data of the metal roof through a distributed temperature sensing network, including: The distributed temperature sensing network includes multiple temperature measurement optical fiber layout subsystems at angles to each other; the three-dimensional temperature field data collects the discrete temperature point data on the surface of the metal roof through the distributed optical fiber temperature measurement technology of the temperature measurement optical fiber layout subsystem, performs spatial correlation analysis on the discrete temperature point data, and reconstructs and generates the three-dimensional temperature field data through an improved interpolation algorithm.

3. The temperature effect monitoring method for a metal roof according to claim 2, characterized in that, The temperature measurement optical fiber layout subsystem covers the entire metal roof area.

4. The temperature effect monitoring method for a metal roof according to claim 1, characterized in that, Extract the temperature gradient eigenvalue of the metal roof from the three-dimensional temperature field data, including: Calculate the temperature difference value at different spatial positions of the metal roof, identify the physical boundary constraint conditions of the metal roof based on the three-dimensional temperature field data, correct the temperature difference value in combination with the physical boundary constraint conditions, and extract the eigenvector from the corrected temperature difference value as the temperature gradient eigenvalue through the principal component analysis algorithm.

5. The temperature effect monitoring method for a metal roof according to claim 1, characterized in that Collect the micro-vibration signal sequence of the metal roof constraint points through a piezoelectric film sensor array, process the micro-vibration signal sequence using the Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extract the dynamic constraint reaction force spectrum eigenvalue from the instantaneous frequency-energy spectrum through the Riemannian manifold learning method, including: Decompose the micro-vibration signal sequence into intrinsic mode function components through the Hilbert-Huang transform; Perform Hilbert spectrum analysis on the intrinsic mode function components to generate an instantaneous frequency-energy spectrum; Map the instantaneous frequency-energy spectrum to the Riemannian manifold space to construct a tangent vector field; Calculate the geodesic distance matrix eigenvalue of the tangent vector field; Select the eigenvector corresponding to the maximum geodesic distance matrix eigenvalue as the dynamic constraint reaction force spectrum eigenvalue.

6. The temperature effect monitoring method for a metal roof according to claim 5, characterized in that, The micro-vibration signal sequence is obtained by collecting the vibration response of the metal roof constraint points through a piezoelectric film sensor array.

7. The temperature effect monitoring method for a metal roof according to claim 1, characterized in that Establish a coupling relationship model between the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue to generate the temperature stress evaluation value of the metal roof, including: Construct the covariance tensor of the temperature gradient eigenvalue and the dynamic constraint reaction force spectrum eigenvalue; Perform orthogonal decomposition on the covariance tensor to obtain the eigenmode matrix; Establish the mapping relationship between the eigenmode matrix and the temperature stress value through the least squares regression algorithm; Input the real-time eigenmode matrix into the mapping relationship to generate the temperature stress evaluation value of the metal roof.

8. A temperature effect monitoring method for a metal roof according to claim 1, characterized in that Dynamically adjust the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters, including: Call the reference threshold range in the historical working condition database; Construct a fuzzy logic rule base for solar radiation intensity and environmental wind speed parameters; Match the correction factor in the fuzzy logic rule base according to the real-time solar radiation intensity and environmental wind speed parameters; Multiply the reference threshold range by the correction factor to generate a dynamic early warning threshold range.

9. A method for monitoring the temperature effect of a metal roof according to claim 1, characterized in that, When the temperature stress evaluation value exceeds the early warning threshold range, output a structural safety early warning signal for the metal roof, including: Map the temperature stress evaluation value to the three-dimensional grid coordinates of the metal roof; Identify the grid coordinate area that exceeds the early warning threshold range to generate a risk area identifier; Construct a risk state vector based on the risk area identifier; Encode the risk state vector into a structural safety early warning signal through the orthogonal coding conversion algorithm.

10. A temperature effect monitoring system for a metal roof, which is used to implement the temperature effect monitoring method for a metal roof according to any one of claims 1-9, characterized in that, Include the following modules: Temperature sensing module, used to obtain the three-dimensional temperature field data of the metal roof through a distributed temperature sensing network; Gradient extraction module, used to extract the temperature gradient characteristic value of the metal roof from the three-dimensional temperature field data; Mechanical feature module, used to collect the micro-vibration signal sequence of the metal roof constraint points through a piezoelectric film sensor array, process the micro-vibration signal sequence using the Hilbert-Huang transform to generate an instantaneous frequency-energy spectrum, and extract the dynamic constraint reaction force spectrum characteristic value from the instantaneous frequency-energy spectrum through the Riemannian manifold learning method; Thermal-mechanical coupling module, used to establish a coupling relationship model between the temperature gradient characteristic value and the dynamic constraint reaction force spectrum characteristic value, and generate the temperature stress evaluation value of the metal roof; Threshold dynamic module, used to dynamically adjust the early warning threshold range according to the solar radiation intensity and environmental wind speed parameters; Safety early warning module, used to output a structural safety early warning signal for the metal roof when the temperature stress evaluation value exceeds the early warning threshold range.

Images