Large-span continuous rigid frame bridge bearing capacity data analysis method and system

Abstract

The invention provides a large-span continuous rigid frame bridge bearing capacity data analysis method which comprises the following steps: acquiring a long-term monitoring data sequence of a target bridge, performing load distribution analysis on the long-term monitoring data sequence, extracting load dynamic characteristics of the bridge, performing deformation correlation analysis on the long-term monitoring data sequence, and extracting deformation coupling characteristics of the bridge. The deformation coupling characteristics are used for representing the cooperative deformation relation between the bridge girder and the pier column; generating a comprehensive evaluation index based on the load dynamic characteristics and the deformation coupling characteristics, wherein the comprehensive evaluation index is used for quantifying the load performance degradation trend of the bridge structure in the full life cycle; and determining the bearing capacity state of the bridge in the current and future service stages according to a comparison result of the comprehensive evaluation index and a preset bearing capacity threshold value. According to the invention, the sensitivity of bearing performance degradation identification and the reliability of a prediction result can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and system for analyzing the bearing capacity data of a long-span continuous rigid frame bridge. Background Art

[0002] Bridge bearing capacity assessment is the core to ensure the safe operation of large-span continuous rigid frame bridges. The key lies in the accurate identification of bearing performance degradation trends through dynamic analysis of structural response data. In the existing technology, static monitoring data is usually used to independently analyze the load distribution characteristics or local deformation indicators, and the bearing capacity status is determined based on the preset safety factor threshold. However, this method has defects: single-point data cannot capture the dynamic evolution law of the bridge throughout its life cycle, resulting in a lag in the identification of degradation trends; the split analysis of load characteristics and deformation characteristics destroys the mechanical correlation of structural responses, making it difficult to accurately reflect the collaborative working status of the main beam and pier column; static assessment indicators lack a multi-dimensional feature fusion mechanism, and can neither quantify the spatiotemporal distribution characteristics of the residual bearing capacity nor establish a mathematical correlation model between the degradation process and the remaining life, resulting in a lack of foresight in maintenance decisions and difficulty in preventing the risk of systemic failure caused by the accumulation of local damage. Summary of the invention

[0003] The present application provides a method and system for analyzing the bearing capacity data of a long-span continuous rigid frame bridge.

[0004] According to one aspect of the present application, a method for analyzing the bearing capacity data of a large-span continuous rigid frame bridge is provided, comprising: obtaining a long-term monitoring data sequence of a target bridge, the long-term monitoring data sequence comprising a plurality of structural response data sets arranged according to monitoring time nodes during the service life of the bridge; performing a load distribution analysis on the long-term monitoring data sequence to extract the load dynamic characteristics of the bridge, the load dynamic characteristics being used to characterize the load transfer laws of each key section of the bridge at different monitoring time nodes; performing a deformation correlation analysis on the long-term monitoring data sequence to extract the deformation coupling characteristics of the bridge, the deformation coupling characteristics being used to characterize the coordinated deformation relationship between the main beam and the pier of the bridge; generating a comprehensive evaluation index based on the load dynamic characteristics and the deformation coupling characteristics, the comprehensive evaluation index being used to quantify the bearing performance degradation trend of the bridge structure during its entire life cycle; and determining the bearing capacity status of the bridge in the current and future service stages according to the comparison result between the comprehensive evaluation index and a preset bearing capacity threshold.

[0005] According to another aspect of the present application, a data analysis system is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor implements the method described above.

[0006] The bridge bearing capacity data analysis method provided by the present invention obtains a long-term monitoring data sequence arranged according to monitoring time nodes during the service life of the target bridge, combines the load dynamic characteristics extracted by load distribution analysis and the deformation coupling characteristics extracted by deformation correlation analysis, generates a comprehensive evaluation index and compares the preset threshold value, which can comprehensively quantify the bearing performance degradation trend of the bridge structure during the entire life cycle and accurately predict the future state. The long-term monitoring data sequence completely covers the dynamic behavior evolution law of the bridge during service through the correlation mapping of the time dimension and the structural response parameters; the load dynamic characteristics accurately capture the time-varying characteristics of the stress state of the key section through the spatial interpolation calculation of the load transfer law; the deformation coupling characteristics effectively reveal the degradation mechanism of the overall stiffness of the structure through the modeling of the coordinated deformation relationship between the main beam and the pier. The comprehensive evaluation index based on multi-source feature fusion incorporates the dynamic changes of structural stiffness, stress distribution and residual bearing capacity into a unified evaluation system, and combined with the trend comparison of historical benchmark values, it can break through the limitations of traditional static evaluation methods and significantly improve the sensitivity of bearing performance degradation identification and the reliability of prediction results. Through multi-level dynamic comparison of real-time monitoring data with preset thresholds, not only can the current bearing capacity safety level of the bridge be accurately determined, but also the risk evolution path of the future service stage can be prospectively identified based on the time series prediction model, providing a scientific basis for bridge maintenance decisions that is both timely and forward-looking, thereby effectively avoiding sudden structural failures caused by the accumulation of local damage and extending the safe service life of the bridge. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A schematic diagram of an application scenario of a method for analyzing the bearing capacity data of a long-span continuous rigid frame bridge according to an embodiment of the present application is shown.

[0008] Figure 2 A flow chart of a method for analyzing bearing capacity data of a long-span continuous rigid frame bridge according to an embodiment of the present application is shown.

[0009] Figure 3 A schematic diagram of the composition of a data analysis system according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0010] Figure 1 A schematic diagram of an application scenario provided according to an embodiment of the present application is shown. The application scenario includes one or more sensor monitoring devices 101, a data analysis system 120, and one or more networks 110 coupling the one or more sensor monitoring devices 101 to the data analysis system 120.

[0011] exist Figure 1In the configuration shown, the data analysis system 120 may include one or more components that implement the functions performed by the data analysis system 120. These components may include software components, hardware components, or a combination thereof that can be executed by one or more processors. A user operating the sensor monitoring device 101 may in turn utilize one or more applications to interact with the data analysis system 120 to utilize the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from application scenarios. Therefore, Figure 1 The sensor monitoring device 101 is an example of a system for implementing the various methods described herein and is not intended to be limiting. The sensor monitoring device 101 is used to monitor bridge data, such as pressure, and send the collected data to the data analysis system 120 .

[0012] The data analysis system 120 may include one or more general-purpose computers, dedicated server computers (e.g., PC (personal computer) servers, UNIX servers, mid-range servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and / or combination. In this application scenario, one or more databases 130 may also be included. In some embodiments, these databases may be used to store data and other information.

[0013] Please refer to Figure 2 The method for analyzing the bearing capacity data of a long-span continuous rigid frame bridge provided in the embodiment of the present application comprises the following steps: Step S100: Acquire a long-term monitoring data sequence of a target bridge, wherein the long-term monitoring data sequence includes a plurality of structural response data sets arranged according to monitoring time nodes during the service life of the bridge.

[0014] The long-term monitoring data sequence is the data sequence obtained by continuous monitoring of the target bridge throughout its entire service life. The structural response data set is a set of response data generated by the bridge due to various loads at each monitoring time node. These response data may include but are not limited to physical quantities such as stress, strain, displacement, acceleration, etc. at key parts of the bridge. For example, at a certain monitoring time node, the structural response data set may contain the stress value of the mid-span section of the bridge, the displacement value at the top of the pier, and the acceleration value of the main beam. The way to obtain the long-term monitoring data sequence can be achieved by installing various sensors on the bridge, such as stress sensors, displacement sensors, acceleration sensors, etc. These sensors will collect data at preset time intervals and transmit the collected data to the data storage center for storage and management.

[0015] Step S200: Perform load distribution analysis on the long-term monitoring data sequence to extract the load dynamic characteristics of the bridge. The load dynamic characteristics are used to characterize the load transfer law of each key section of the bridge at different monitoring time nodes.

[0016] The dynamic characteristics of loads are characteristic quantities that reflect the load transfer laws of bridges at different time nodes. In actual analysis, the load transfer laws of key sections of bridges at different monitoring time nodes will be affected by many factors, such as changes in vehicle loads and the influence of environmental factors. For example, during periods of high traffic volume, the loads borne by key sections of bridges will increase accordingly, and the load transfer laws will also change. By performing load distribution analysis on long-term monitoring data sequences, these dynamic characteristics of loads can be extracted to provide a basis for subsequent evaluation of the bearing capacity of bridges.

[0017] As an implementation mode, step S200, performing load distribution analysis on the long-term monitoring data sequence to extract the load dynamic characteristics of the bridge, may specifically include: Step S210: Separate a load distribution data subset from the long-term monitoring data sequence, where the load distribution data subset includes the moment monitoring values ​​and shear force monitoring values ​​of the mid-span and support sections of the bridge at each monitoring time node.

[0018] The load distribution data subset is a data subset related to load distribution that is screened out from the long-term monitoring data sequence. The bending moment monitoring value refers to the monitoring data of the bending moment borne by the mid-span and support sections of the bridge at each monitoring time node. The shear force monitoring value refers to the monitoring data of the shear force borne by the corresponding section. In actual operation, the load distribution data subset can be separated from the long-term monitoring data sequence through a data screening algorithm. For example, data related to the bending moment and shear force of the mid-span and support sections of the bridge can be screened out based on the label information of the data. These bending moment monitoring values ​​and shear force monitoring values ​​can intuitively reflect the stress conditions of each key section of the bridge at different monitoring time nodes.

[0019] Step S220: Decomposing the load distribution data subset into a constant load component sequence and a live load component sequence according to a preset load decomposition level, wherein the constant load component sequence is subjected to low-pass filtering to extract long-term trend items, and the live load component sequence is subjected to high-pass filtering to extract short-term fluctuation items.

[0020] The preset load decomposition level is a pre-set hierarchical structure for decomposing loads. The dead load component sequence is a component sequence related to the dead load in the load distribution data subset. The dead load usually refers to relatively stable loads such as the structural weight of the bridge itself and the weight of auxiliary facilities. The live load component sequence is a component sequence related to the live load. The live load includes vehicle loads, pedestrian loads, wind loads and other loads that vary greatly over time. For example, in practical applications, digital filters can be used to implement low-pass filtering and high-pass filtering to decompose the load distribution data subset into a dead load component sequence and a live load component sequence.

[0021] Step S230: dynamically superimpose the dead load component sequence and the live load component sequence using a time-varying weight distribution algorithm to generate an equivalent load spectrum corresponding to each monitoring time node.

[0022] The time-varying weight allocation algorithm is an algorithm that can dynamically adjust the weight according to time changes. The equivalent load spectrum is obtained by dynamically superimposing the constant load component sequence and the live load component sequence. It can comprehensively reflect the spectrum of the load conditions borne by the bridge at each monitoring time node. At different monitoring time nodes, the degree of effect of the constant load and live load on the bridge is different. Therefore, it is necessary to use a time-varying weight allocation algorithm to dynamically adjust the weights of the constant load component sequence and the live load component sequence, and then superimpose them to generate an equivalent load spectrum. For example, in time periods with large traffic flow, the effect of live load is relatively large. At this time, the weight of the live load component sequence can be appropriately increased; in time periods with small traffic flow, the effect of constant load is relatively prominent, and the weight of the constant load component sequence can be increased accordingly.

[0023] As an implementation mode, step S230, using a time-varying weight distribution algorithm to dynamically superimpose the dead load component sequence and the live load component sequence to generate an equivalent load spectrum corresponding to each monitoring time node, may specifically include: Step S231: Initialize the dead load weight coefficient of each monitoring time node based on the long-term trend slope of each monitoring time node in the dead load component sequence.

[0024] The long-term trend slope is the slope of the long-term change trend of the constant load component sequence at each monitoring time node. The constant load weight coefficient is a coefficient used to measure the proportion of the constant load in the equivalent load spectrum. At each monitoring time node, the long-term trend slope of the constant load component sequence reflects the changing trend of the constant load. For example, if the long-term trend slope of the constant load component sequence at a certain monitoring time node is large, it means that the constant load changes more dramatically near this time node. At this time, the constant load weight coefficient of this time node can be appropriately increased to more accurately reflect the effect of the constant load on the bridge. By analyzing the long-term trend slope of each monitoring time node in the constant load component sequence, the constant load weight coefficient of the corresponding time node can be initialized.

[0025] Step S232: Initialize the live load weight coefficient of the corresponding time node based on the short-term fluctuation amplitude of each monitoring time node in the live load component sequence.

[0026] The short-term fluctuation amplitude is the amplitude of the short-term fluctuation of the live load component sequence at each monitoring time node. The live load weight coefficient is a coefficient used to measure the proportion of the live load in the equivalent load spectrum. The short-term fluctuation amplitude of the live load reflects the severity of the change of the live load at that time node. For example, during peak traffic hours, the short-term fluctuation amplitude of the live load may be large, indicating that the live load changes more frequently and violently. At this time, the live load weight coefficient of the time node can be appropriately increased to better reflect the impact of the live load on the bridge. According to the short-term fluctuation amplitude of each monitoring time node in the live load component sequence, the live load weight coefficient of the corresponding time node can be initialized.

[0027] Step S233: Based on the real-time monitoring data set of the environment in which the bridge is located, the temperature gradient change rate, humidity cumulative effect factor and traffic flow impact index are extracted to generate an environmental dynamic impact set.

[0028] The real-time monitoring data set is a data set obtained by real-time monitoring of the environment in which the bridge is located. The temperature gradient change rate refers to the temperature change rate caused by temperature differences in different parts of the bridge structure. The humidity cumulative effect factor is a factor that reflects the cumulative impact of humidity on the bridge structure within a certain period of time. The traffic flow impact index is an index that measures the impact of traffic flow on the bridge structure. The environmental dynamic impact set is a set composed of factors such as the temperature gradient change rate, humidity cumulative effect factor and traffic flow impact index. For example, the real-time monitoring data set can be obtained by installing temperature sensors, humidity sensors and traffic flow monitoring equipment around the bridge, and then analyzing and processing these data to extract the temperature gradient change rate, humidity cumulative effect factor and traffic flow impact index to generate the environmental dynamic impact set.

[0029] Step S234: input the environmental dynamic impact set into the weight correction model, perform nonlinear coupling adjustment on the dead load weight coefficient and the live load weight coefficient, and obtain the corrected dead load weight and corrected live load weight at each monitoring time node.

[0030] The weight correction model is a model that can correct the constant load weight coefficient and the live load weight coefficient according to the input environmental dynamic impact set. Nonlinear coupling adjustment refers to considering the nonlinear interaction relationship between the constant load weight coefficient and the live load weight coefficient when adjusting them. The corrected constant load weight and the corrected live load weight are obtained after adjustment by the weight correction model, and can more accurately reflect the weights of the actual situation. For example, the weight correction model can use a neural network model, take the environmental dynamic impact set as input, and perform nonlinear coupling adjustment on the constant load weight coefficient and the live load weight coefficient through the learning and calculation of the neural network to obtain the corrected constant load weight and corrected live load weight for each monitoring time node.

[0031] Step S235: adopting the sliding time window mechanism, performing moving average processing on the modified constant load weights of the current monitoring time node and the previous preset number of nodes to generate a smoothed constant load weight sequence.

[0032] The sliding time window mechanism sets a fixed-length time window and slides the window on the time series for data processing. The preset number is the number of previous nodes that participate in the moving average processing. The moving average processing is the process of averaging the modified constant load weights of the current monitoring time node and the previous preset number of nodes. The smoothed constant load weight sequence is obtained after the moving average processing, which can make the constant load weight more smooth and stable.

[0033] Step S236: synchronously performing exponential decay processing on the modified live load weights of the current monitoring time node and the previous preset number of nodes to generate a decaying live load weight sequence.

[0034] Exponential decay processing is a method of calculating the decay of data, and its decay degree decreases exponentially over time. The decayed live load weight sequence is the live load weight sequence obtained after exponential decay processing. In practical applications, due to the large short-term impact of live loads, the impact of the live load weight of the previous node on the current node will gradually decrease over time. Therefore, the use of exponential decay processing can more reasonably consider the time decay characteristics of the live load weight. For example, the modified live load weight of the current monitoring time node and the previous preset number of nodes is calculated according to the exponential decay formula to obtain the decayed live load weight sequence.

[0035] Step S237: multiply the data of the corresponding time nodes in the smoothed constant load weight sequence and the constant load component sequence according to the weights to obtain the constant load weighted component.

[0036] The weighted component of the constant load is obtained by multiplying the smoothed constant load weight sequence with the data of the corresponding time node in the constant load component sequence according to the weight. At each monitoring time node, the smoothed constant load weight of the node is multiplied with the data of the corresponding node in the constant load component sequence to obtain the weighted component of the constant load of the node. For example, at the nth monitoring time node, the weight of the nth node in the smoothed constant load weight sequence is multiplied with the data of the nth node in the constant load component sequence to obtain the weighted component of the constant load of the nth node.

[0037] Step S238: multiply the data of the corresponding time nodes in the attenuated live load weight sequence and the live load component sequence according to the weights to obtain the live load weighted components.

[0038] The live load weighted component is the component obtained by multiplying the attenuated live load weight sequence with the data of the corresponding time node in the live load component sequence according to the weight. At each monitoring time node, the attenuated live load weight of the node is multiplied with the data of the corresponding node in the live load component sequence to obtain the live load weighted component of the node. For example, at the mth monitoring time node, the weight of the mth node in the attenuated live load weight sequence is multiplied with the data of the mth node in the live load component sequence to obtain the live load weighted component of the mth node.

[0039] Step S239: Perform vector superposition on the weighted components of the dead load and the weighted components of the live load at the same monitoring time node to generate an initial equivalent load spectrum.

[0040] Vector superposition is the operation of adding the weighted components of the dead load and the weighted components of the live load at the same monitoring time node. The initial equivalent load spectrum is obtained through vector superposition, which preliminarily reflects the spectrum of the load conditions borne by the bridge at each monitoring time node. At each monitoring time node, the weighted components of the dead load and the weighted components of the live load of the node are added to obtain the initial equivalent load value of the node, and so on to generate the initial equivalent load spectrum.

[0041] Step S2310: calling the load distribution verification model to perform a temporal and spatial alignment comparison between the initial equivalent load spectrum and the typical load pattern of the same type of bridge in the historical load distribution feature library.

[0042] The historical load distribution feature library is a database that stores the historical load distribution features of the same type of bridges. The typical load pattern refers to the representative load distribution pattern in the historical load distribution feature library. Spatiotemporal alignment comparison refers to the matching and comparison of the initial equivalent load spectrum and the typical load pattern in the time and space dimensions. For example, the load distribution verification model can use a pattern recognition algorithm to align the initial equivalent load spectrum with the typical load pattern of the same type of bridge in the historical load distribution feature library in time and space, and compare their similarities and differences in time and space. Through this comparison, it can be judged whether the initial equivalent load spectrum is reasonable and whether it conforms to the load distribution law of the same type of bridges.

[0043] Step S2311: According to the load peak position deviation and distribution uniformity difference in the comparison results, reversely adjust the coupling adjustment parameters of the weight correction model.

[0044] The load peak position deviation refers to the difference in the load peak position between the initial equivalent load spectrum and the typical load pattern. The distribution uniformity difference refers to the difference in the uniformity of the load distribution between the two. The coupling adjustment parameter is a parameter used to adjust the nonlinear coupling relationship between the dead load weight coefficient and the live load weight coefficient in the weight correction model. According to the comparison results, if the load peak position deviation is large or the distribution uniformity difference is obvious, it means that there is a large difference between the initial equivalent load spectrum and the typical load pattern, and the coupling adjustment parameters of the weight correction model need to be adjusted in the opposite direction. For example, if the load peak position of the initial equivalent load spectrum is earlier than the typical load pattern and the distribution uniformity is poor, the coupling adjustment parameters of the weight correction model can be appropriately adjusted to make the weight distribution of the dead load and the live load more reasonable, so that the initial equivalent load spectrum is closer to the typical load pattern.

[0045] Step S2312: re-execute the weight correction and subsequent superposition steps using the adjusted weight correction model until the similarity between the spatiotemporal distribution of the initial equivalent load spectrum and the historical typical load pattern reaches a preset threshold.

[0046] The preset threshold is a pre-set standard for measuring the similarity between the initial equivalent load spectrum and the historical typical load pattern. After the coupling adjustment parameters of the weight correction model are adjusted in reverse, it is necessary to re-execute the weight correction and subsequent superposition steps, that is, starting from step S234, recalculate the corrected constant load weight and the corrected live load weight, generate a smoothed constant load weight sequence and an attenuated live load weight sequence, calculate the constant load weighted component and the live load weighted component, and perform vector superposition to generate a new initial equivalent load spectrum. Then, the new initial equivalent load spectrum is aligned and compared with the historical typical load pattern in time and space again, and their similarity is calculated. Repeat this process until the similarity between the spatiotemporal distribution morphology of the initial equivalent load spectrum and the historical typical load pattern reaches the preset threshold.

[0047] Step S2313: Mark the verified initial equivalent load spectrum as the final equivalent load spectrum, and associate it with the structural response data set of the corresponding monitoring time node.

[0048] The final equivalent load spectrum is a verified spectrum that can accurately reflect the load conditions of the bridge at each monitoring time node. After marking the verified initial equivalent load spectrum as the final equivalent load spectrum, it needs to be associated with the structural response data set of the corresponding monitoring time node. In this way, the correspondence between load and structural response can be established, providing more comprehensive data support for subsequent analysis and evaluation. For example, at the i-th monitoring time node, the final equivalent load spectrum of the node is associated with the structural response data set of the node, so that when analyzing the structural response of the node, the load conditions borne by the node can be taken into account at the same time.

[0049] Step S240: Call the pre-trained load transfer model to perform spatial interpolation calculation on the equivalent load spectrum to obtain a load distribution cloud map within the full span of the bridge.

[0050] The pre-trained load transfer model is a pre-trained model that can transfer and calculate loads. Spatial interpolation calculation refers to the process of calculating the load data at other locations based on the load data of a finite number of known points. The load distribution cloud map is a graph that represents the load distribution within the full span of the bridge in the form of a cloud map. In practical applications, the equivalent load spectrum usually only contains the load data of some monitoring locations. In order to obtain the load distribution within the full span of the bridge, it is necessary to call the pre-trained load transfer model to perform spatial interpolation calculations on the equivalent load spectrum.

[0051] As an implementation mode, step S240, calling a pre-trained load transfer model to perform spatial interpolation calculation on the equivalent load spectrum to obtain a load distribution cloud map within the full span of the bridge, may specifically include: Step S241: Divide the load data of each monitoring time node in the equivalent load spectrum into a number of load action subdomains according to the longitudinal axis of the bridge, each subdomain including the starting position coordinates, the ending position coordinates and the average load intensity in the domain.

[0052] The load action subdomain is the sub-area obtained by dividing the load data in the equivalent load spectrum according to the longitudinal axis of the bridge. The starting position coordinates and the ending position coordinates respectively represent the starting and ending positions of each load action subdomain on the longitudinal axis of the bridge. The average load intensity within the domain refers to the average load size within each load action subdomain. For example, at a certain monitoring time node, the load data in the equivalent load spectrum is divided into multiple load action subdomains according to the longitudinal axis of the bridge. The starting position coordinates and the ending position coordinates of each subdomain can be obtained by measurement and calculation, and the average load intensity within the domain can be obtained by averaging the load data in the subdomain.

[0053] Step S242: According to the topological relationship of the bridge structure, the geometric connection characteristics of each load action subdomain and the adjacent unmonitored area are extracted to generate a load transfer path network.

[0054] The topological relationship of the bridge structure refers to the connection relationship and spatial layout between the various components of the bridge. The geometric connection characteristics refer to the geometric connection methods and characteristics between each load action subdomain and the adjacent unmonitored area. The load transfer path network is a network composed of load transfer paths between each load action subdomain and the adjacent unmonitored area. In practical applications, based on the topological relationship of the bridge structure, the geometric connection characteristics between each load action subdomain and the adjacent unmonitored area can be analyzed, for example, to determine whether they are connected by beam connection, node connection or other methods. Then, based on these geometric connection characteristics, a load transfer path network is generated to clarify the load transfer path in the bridge structure. For example, in a continuous rigid frame bridge, by analyzing the structural topological relationship of the bridge, the beam connection method between each span can be determined, thereby generating a load transfer path network, which provides a basis for subsequent load transfer calculations.

[0055] Step S243: The load intensity of the load action subdomain is used as the input node feature, and the geometric parameters of the load transfer path network are used as the edge feature to construct the load space propagation graph structure.

[0056] Input node features refer to parameters that serve as node features in the graph structure. Here, the load intensity of the load action subdomain is used as the input node feature. Edge features refer to parameters that serve as edge features in the graph structure. Here, the geometric parameters of the load transfer path network are used as edge features. The load space propagation graph structure is a structure that uses a graph structure to represent the propagation of loads in the bridge space. When constructing the load space propagation graph structure, each load action subdomain is taken as a node, and the node's feature is the load intensity of the subdomain; the load transfer path between each load action subdomain is taken as an edge, and the edge's feature is the geometric parameters of the load transfer path network. For example, in a simple bridge model, there are three load action subdomains A, B, and C. A, B, and C are taken as nodes, and the node's features are their load intensities, respectively. The load transfer paths between A and B, and between B and C are taken as edges, and the edge's features are the corresponding geometric parameters, thereby constructing the load space propagation graph structure.

[0057] Step S244: Call the graph convolution module in the pre-trained load transfer model to perform multi-scale feature aggregation on the load space propagation graph structure, and output the load intensity prediction value of each unmonitored area.

[0058] The graph convolution module can perform convolution operations on the graph structure and extract features from the graph structure. Multi-scale feature aggregation refers to the process of aggregating and fusing features in the graph structure at different scales. An unmonitored area refers to an area where load data is not directly monitored in the equivalent load spectrum. The load intensity prediction value is obtained by processing the load space propagation graph structure through the graph convolution module, which is a prediction result of the load intensity in the unmonitored area. For example, the pre-trained load transfer model can adopt a model based on a graph neural network, in which the graph convolution module can perform multi-scale feature aggregation on the load space propagation graph structure, and output the load intensity prediction value of each unmonitored area by learning the features of the nodes and edges in the graph structure.

[0059] Step S245: dynamically adjust the neighborhood sampling radius of the graph convolution module according to the gradient change direction between the measured load intensity and the predicted load intensity of the adjacent load action subdomain.

[0060] The gradient change direction refers to the change direction between the measured load intensity and the predicted load intensity of the adjacent load action subdomain. The neighborhood sampling radius is the range radius of the graph convolution module when performing the convolution operation. In practical applications, there may be a difference between the measured load intensity and the predicted load intensity of the adjacent load action subdomain. According to the gradient change direction of this difference, the neighborhood sampling radius of the graph convolution module can be dynamically adjusted. For example, if the gradient change between the measured load intensity and the predicted load intensity of the adjacent load action subdomain is large, it means that the load change is more drastic. At this time, the neighborhood sampling radius of the graph convolution module can be appropriately increased to obtain more adjacent node information and improve the accuracy of the prediction; conversely, if the gradient change is small, the neighborhood sampling radius can be appropriately reduced.

[0061] Step S246: splicing the predicted load intensity output by the adjusted graph convolution module with the measured load intensity according to the spatial position to form a preliminary load distribution surface.

[0062] The preliminary load distribution surface is obtained by splicing the predicted load intensity output by the adjusted graph convolution module with the measured load intensity according to the spatial position. At each spatial position, the predicted load intensity at that position is combined with the measured load intensity to form the load intensity value at that position. By analogy, a preliminary load distribution surface is formed over the entire bridge space. For example, at a certain cross-sectional position of the bridge, the predicted load intensity at that position is spliced ​​with the measured load intensity to obtain the final load intensity value at that position. The preliminary load distribution surface is formed by performing the same operation at each position of the bridge.

[0063] Step S247: Based on the stiffness distribution characteristics of the bridge main beam cross section, anisotropic smoothing correction is performed on the preliminary load distribution surface to eliminate inter-span mutation noise.

[0064] The stiffness distribution characteristics of the bridge girder section refer to the stiffness distribution of the bridge girder at different positions and directions. Anisotropic smoothing correction refers to the process of smoothing the preliminary load distribution surface while considering the stiffness distribution characteristics of the bridge girder section. Interspan mutation noise refers to the noise of sudden changes in the preliminary load distribution surface at the interspan position. In practical applications, the cross-sectional stiffness of the bridge girder may vary at different positions and directions, and this difference will affect the distribution of the load. Therefore, based on the stiffness distribution characteristics of the bridge girder section, anisotropic smoothing correction is performed on the preliminary load distribution surface, which can make the load distribution more reasonable and eliminate the interspan mutation noise. For example, a method based on finite element analysis can be used to smooth the preliminary load distribution surface according to the stiffness distribution characteristics of the bridge girder section, so that the load distribution at the interspan position is more continuous and smooth.

[0065] Step S248: Project the corrected load distribution surface into the three-dimensional coordinate system of the bridge, and generate a discretized load density matrix according to a preset resolution.

[0066] The three-dimensional coordinate system of the bridge is a coordinate system used to describe the spatial position of the bridge. The preset resolution is the preset accuracy for generating a discretized load density lattice. The discretized load density lattice is a lattice obtained by discretizing the corrected load distribution surface in the three-dimensional coordinate system of the bridge according to the preset resolution. After the corrected load distribution surface is projected into the three-dimensional coordinate system of the bridge, it is discretized according to the preset resolution, and the surface is converted into a series of discrete points, each of which corresponds to a load density value, forming a discretized load density lattice. For example, the preset resolution can be set to 1 meter. In the three-dimensional coordinate system of the bridge, a point is taken every 1 meter, and the load density value of the point is calculated to form a discretized load density lattice.

[0067] Step S249: Based on the spatial correlation of adjacent points in the load density lattice, an adaptive interpolation kernel function is used to fill the blank areas between the points to generate a continuous load distribution cloud map.

[0068] Spatial correlation refers to the degree of spatial association between the load density values ​​between adjacent points in the load density lattice. The adaptive interpolation kernel function is an interpolation function that can automatically adjust the interpolation parameters according to the spatial correlation of adjacent points. The blank area refers to the unfilled area between adjacent points in the discretized load density lattice. The continuous load distribution cloud map is obtained by filling the blank areas between points with the adaptive interpolation kernel function. It is a cloud map that can continuously represent the load distribution within the full span of the bridge. In practical applications, according to the spatial correlation of adjacent points in the load density lattice, a suitable adaptive interpolation kernel function is selected, the blank areas between adjacent points are interpolated, the blank areas are filled, and a continuous load distribution cloud map is generated. For example, the Gaussian interpolation kernel function can be used to automatically adjust the interpolation parameters according to the difference in the spatial distance between adjacent points and the load density values, fill the blank areas between points, and generate a continuous load distribution cloud map.

[0069] Step S2410: perform a morphological similarity comparison between the load distribution cloud map and the typical load patterns in the historical load database, and extract the peak area contour and valley area boundary in the cloud map.

[0070] The historical load database is a database that stores historical load data and typical load patterns. Morphological similarity comparison refers to the process of comparing and analyzing the morphology of the load distribution cloud map and the typical load pattern in the historical load database. The peak area contour refers to the contour of the area with higher load intensity in the load distribution cloud map. The valley area boundary refers to the boundary of the area with lower load intensity in the load distribution cloud map. In practical applications, the generated load distribution cloud map is compared with the typical load pattern in the historical load database for morphological similarity to determine whether the load distribution cloud map conforms to historical laws. Through comparison, the peak area contour and valley area boundary in the cloud map are extracted to provide important information for subsequent analysis and evaluation. For example, an image matching algorithm can be used to compare the load distribution cloud map with the typical load pattern in the historical load database to extract the peak area contour and valley area boundary.

[0071] Step S2411: According to the morphological differences in the comparison results, reversely optimize the feature aggregation weight coefficients of the graph convolution module.

[0072] Morphological difference refers to the difference in morphology between the load distribution cloud map and the typical load pattern in the historical load database. The feature aggregation weight coefficient is the weight coefficient used to aggregate features in the graph convolution module. According to the comparison results, if there are morphological differences between the load distribution cloud map and the typical load pattern, it means that the feature aggregation effect of the graph convolution module may not be ideal, and it is necessary to reversely optimize the feature aggregation weight coefficient of the graph convolution module. For example, if the peak area position of the load distribution cloud map is inconsistent with the typical load pattern, the feature aggregation weight coefficient of the graph convolution module can be adjusted so that the graph convolution module pays more attention to the features related to the peak area when aggregating features, thereby improving the accuracy of load prediction.

[0073] Step S2412: Re-execute the load intensity prediction and subsequent correction steps using the optimized load transfer model until the matching degree between the profile characteristics of the load distribution cloud map and the historical typical pattern exceeds a set threshold.

[0074] The set threshold is a pre-set standard for measuring the degree of match between the contour features of the load distribution cloud map and the historical typical pattern. After the feature aggregation weight coefficient of the reverse optimization graph convolution module, the optimized load transfer model is used to re-execute the load intensity prediction and subsequent correction steps, that is, starting from step S244, the multi-scale feature aggregation of the load space propagation graph structure is re-performed, and the load intensity prediction value of each unmonitored area is output, and then subsequent adjustment, splicing, correction and other steps are performed to generate a new load distribution cloud map. The new load distribution cloud map is again compared with the historical typical pattern for morphological similarity, and their matching degree is calculated. Repeat this process until the matching degree between the contour features of the load distribution cloud map and the historical typical pattern exceeds the set threshold. For example, the set threshold can be set to 90%. When the matching degree between the contour features of the new load distribution cloud map and the historical typical pattern reaches 90%, it is considered that the load distribution cloud map has met the requirements.

[0075] Step S2413: The load distribution cloud map that has finally passed the verification is stored in time and space alignment with the structural response data set of the corresponding monitoring time node, and the coordinate range and strength level of the key load concentration area are marked.

[0076] The load distribution cloud map that has passed the final verification is the load distribution cloud map that meets the requirements after multiple optimizations and verifications. Time-space alignment storage refers to the process of matching and storing the load distribution cloud map that has passed the final verification with the structural response data set of the corresponding monitoring time node in time and space. The key load concentration area refers to the area in the load distribution cloud map where the load intensity is high and has a greater impact on the bridge structure. The coordinate range refers to the position range of the key load concentration area in the three-dimensional coordinate system of the bridge. The strength grade refers to the load intensity grade of the key load concentration area. When the load distribution cloud map that has passed the final verification is stored in time and space with the structural response data set of the corresponding monitoring time node, it is necessary to ensure the consistency of the two in time and space for subsequent analysis and research. At the same time, marking the coordinate range and strength grade of the key load concentration area can provide an important reference for the maintenance and management of the bridge. For example, when storing, the load distribution cloud map and the structural response data set can be associated according to the monitoring time node, and the coordinate range and strength grade of the key load concentration area can be marked in the load distribution cloud map.

[0077] As an implementation manner, the method provided in the embodiment of the present invention further includes a verification process of the load transfer model, which may specifically include: Step S2401: Obtain the three-dimensional displacement field measured data of the target bridge under the standard load test condition, wherein the three-dimensional displacement field measured data includes the spatial coordinate changes of the main beam mid-span and pier top section.

[0078] The standard load test condition refers to the condition set when the bridge is subjected to a load test. The three-dimensional displacement field measured data is obtained through actual measurement, and is the three-dimensional displacement field data of the target bridge under the standard load test condition. The mid-span of the main beam refers to the middle position of the main beam of the bridge. The pier top section refers to the section at the top of the pier. The change in spatial coordinates refers to the change in the spatial coordinates of the mid-span of the main beam and the pier top section relative to the initial state under the standard load test condition. In actual operation, high-precision measuring equipment, such as total stations, laser rangefinders, etc., can be used to measure the three-dimensional displacement field of the target bridge under the standard load test condition, obtain the change in the spatial coordinates of the mid-span of the main beam and the pier top section, and obtain the three-dimensional displacement field measured data.

[0079] Step S2402: input the load parameters of the standard load test condition into the pre-established parameterized finite element model, and generate theoretical displacement field distribution data after performing multi-scale meshing.

[0080] A parametric finite element model is a finite element model established in a parametric way, which can easily simulate and analyze different structural parameters and load conditions. Multi-scale meshing refers to the process of using meshes of different scales to divide the finite element model according to different parts and characteristics of the structure. The theoretical displacement field distribution data is obtained by simulating and calculating the parameterized finite element model, which is the theoretical displacement field distribution of the target bridge under the standard load test condition. In practical applications, the load parameters of the standard load test condition, such as load size, load distribution, etc., are input into the pre-established parametric finite element model, and then the model is multi-scale meshed to improve the calculation accuracy and efficiency. Finally, the model is calculated by the finite element analysis software to generate the theoretical displacement field distribution data.

[0081] Step S2403: extract the absolute error vector between the measured value of the vertical displacement at the mid-span of the main beam in the three-dimensional displacement field measured data and the theoretical value at the corresponding position in the theoretical displacement field distribution data, and simultaneously calculate the relative error distribution pattern between the measured value and the theoretical value of the horizontal displacement at the pier top.

[0082] The absolute error vector refers to the difference vector between the measured value of the vertical displacement at the mid-span of the main beam in the measured data of the three-dimensional displacement field and the theoretical value of the corresponding position in the distribution data of the theoretical displacement field. The relative error distribution pattern refers to the spatial distribution of the relative error between the measured value and the theoretical value of the horizontal displacement at the pier top. In actual operation, the measured value of the vertical displacement at the mid-span of the main beam is extracted from the measured data of the three-dimensional displacement field, and the theoretical value of the corresponding position is extracted from the distribution data of the theoretical displacement field. The difference between them is calculated to obtain the absolute error vector. At the same time, the relative error between the measured value and the theoretical value of the horizontal displacement at the pier top is calculated, and its spatial distribution is analyzed to obtain the relative error distribution pattern.

[0083] Step S2404: Identify the correction priority of the main beam stiffness parameters according to the direction angle of the absolute error vector, and give priority to adjusting the elastic modulus adjustment factor orthogonal to the direction of the maximum absolute error.

[0084] The direction angle refers to the direction angle of the absolute error vector in space. The main beam stiffness parameter is a parameter that describes the main beam stiffness characteristics, such as elastic modulus. The elastic modulus adjustment factor is a factor used to adjust the elastic modulus of the main beam. In practical applications, the stiffness error of the main beam in different directions can be determined based on the direction angle of the absolute error vector. Prioritizing the adjustment of the elastic modulus adjustment factor that is orthogonal to the direction of the maximum absolute error can more effectively reduce the error. For example, if the direction angle of the absolute error vector indicates that the error is the largest in a certain direction, then the elastic modulus adjustment factor orthogonal to that direction is adjusted first so that the stiffness of the main beam in that direction is more in line with the actual situation.

[0085] Step S2405: Based on the spatial correlation of the relative error distribution pattern, the boundary stiffness proportionality coefficient of the pier foundation constraint condition is synchronously corrected.

[0086] Spatial correlation refers to the degree of spatial correlation of the relative error distribution pattern. The pier foundation constraint condition refers to the constraint relationship between the pier foundation and the foundation. The boundary stiffness proportional coefficient is a coefficient that describes the boundary stiffness of the pier foundation constraint condition. In practical applications, the spatial correlation of the relative error distribution pattern reflects the stress conditions and deformation differences of the pier foundation at different positions. Based on this spatial correlation, the boundary stiffness proportional coefficient of the pier foundation constraint condition is synchronously corrected, which can make the pier foundation constraint condition more reasonable and improve the accuracy of the model. For example, if the relative error distribution pattern shows that the relative error in a certain area is large, it means that the pier foundation constraint condition in this area may need to be adjusted. At this time, the boundary stiffness proportional coefficient of this area can be corrected accordingly.

[0087] Step S2406: re-import the corrected elastic modulus adjustment factor and boundary stiffness proportional coefficient into the parameterized finite element model, and iteratively perform the load-displacement response calculation until the absolute error vector modulus and relative error rate of all measuring points are lower than a preset convergence threshold.

[0088] The preset convergence threshold is a pre-set standard for measuring the convergence degree of the calculation results. After re-importing the corrected elastic modulus adjustment factor and boundary stiffness proportional coefficient into the parametric finite element model, the load-displacement response calculation is performed again to obtain new theoretical displacement field distribution data. Then, the absolute error vector and relative error rate between the measured value of each measuring point in the three-dimensional displacement field measured data and the new theoretical value are extracted again. Repeat this process, continuously correcting the elastic modulus adjustment factor and boundary stiffness proportional coefficient until the absolute error vector modulus and relative error rate of all measuring points are lower than the preset convergence threshold.

[0089] Step S2407: extract the mid-span bending moment distribution cloud diagram and the pier bottom shear gradient curve from the final converged finite element model output to generate a load transfer feature verification fingerprint.

[0090] The mid-span bending moment distribution cloud map is a cloud map that shows the distribution of bending moment at the mid-span position of the bridge main beam. The shear gradient curve at the bottom of the pier is a curve that describes the change of shear gradient at the bottom of the pier. The load transfer feature verification fingerprint is composed of the mid-span bending moment distribution cloud map and the shear gradient curve at the bottom of the pier, and is used to verify the characteristic fingerprint of the load transfer feature. In the final converged finite element model output, the mid-span bending moment distribution cloud map and the shear gradient curve at the bottom of the pier are extracted and combined to generate the load transfer feature verification fingerprint. This verification fingerprint can reflect the transfer characteristics of the bridge under load and is used for subsequent verification and comparison.

[0091] Step S2408: dynamically spatially register the load transfer feature verification fingerprint with the load distribution cloud map monitored in real time, and calculate the morphological similarity index of the two in the gradient change area.

[0092] Dynamic spatial registration refers to the process of spatially matching and aligning the load transfer feature verification fingerprint and the real-time monitored load distribution cloud map. The gradient change area refers to the area with large gradient changes in the load transfer feature verification fingerprint and the real-time monitored load distribution cloud map. The morphological similarity index is an indicator to measure the degree of morphological similarity between the two in the gradient change area. In practical applications, a spatial registration algorithm is used to dynamically spatially register the load transfer feature verification fingerprint and the real-time monitored load distribution cloud map so that they are aligned in space. Then, the morphological features of the two in the gradient change area are analyzed, and the morphological similarity index is calculated. For example, an image matching algorithm can be used to calculate the morphological similarity index of the two in the gradient change area to determine their similarity.

[0093] Step S2409: When the morphological similarity index is lower than the historical baseline level for three consecutive monitoring cycles, the parameter recalibration instruction of the load transfer model is triggered.

[0094] The historical baseline level refers to the average level or reference level of the morphological similarity index in the historical monitoring data. The parameter recalibration instruction is an instruction used to trigger the recalibration of the load transfer model parameters. In the actual monitoring process, the morphological similarity index of the load transfer feature verification fingerprint and the real-time monitored load distribution cloud map in the gradient change area is continuously calculated. If the indicator is lower than the historical baseline level for three consecutive monitoring cycles, it means that the parameters of the load transfer model may have deviated and need to be recalibrated. At this time, the parameter recalibration instruction of the load transfer model is triggered to start the parameter recalibration process.

[0095] Step S24010: Call the latest three-dimensional displacement field measured data according to the recalibration instruction, re-execute the correction process of the elastic modulus adjustment factor and the boundary stiffness proportional coefficient, and update the material constitutive relationship of the finite element model.

[0096] The latest three-dimensional displacement field measured data refers to the three-dimensional displacement field measured data of the target bridge under the standard load test condition obtained by the latest monitoring. The material constitutive relationship is the relationship that describes the mechanical properties of the material. After receiving the parameter recalibration instruction, the latest three-dimensional displacement field measured data is called, and the correction process of the elastic modulus adjustment factor and the boundary stiffness proportional coefficient is re-executed according to the process of steps S2403-S2406. By correcting these parameters, the material constitutive relationship of the finite element model is updated, so that the finite element model can more accurately reflect the actual mechanical properties of the bridge. For example, according to the latest three-dimensional displacement field measured data, the absolute error vector and the relative error distribution pattern are recalculated, the elastic modulus adjustment factor and the boundary stiffness proportional coefficient are adjusted, and the constitutive relationships such as the elastic modulus and boundary constraints of the material in the finite element model are updated.

[0097] Step S24011: reversely inject the theoretical displacement field distribution data output by the updated finite element model into the training data set of the load transfer model to replace the outdated response features in the historical samples.

[0098] The theoretical displacement field distribution data is obtained by calculating the updated finite element model, and is the theoretical displacement field distribution of the target bridge under the standard load test condition. The training data set is a data set used to train the load transfer model. Outdated response characteristics refer to response characteristics in historical samples that can no longer accurately reflect the actual situation of the bridge. After updating the material constitutive relationship of the finite element model, the updated theoretical displacement field distribution data is calculated by the finite element analysis software. These data are reversely injected into the training data set of the load transfer model to replace the outdated response characteristics in the historical samples, so that the training data set more accurately reflects the actual situation of the bridge. For example, the displacement values, stress values ​​and other data in the updated theoretical displacement field distribution data replace the corresponding outdated data in the training data set to improve the quality of the training data set.

[0099] Step S24012: Perform online retraining of the load transfer model based on the injected training data set, optimize the weight distribution strategy in the spatial interpolation calculation, and form an adaptive verification closed loop of the load distribution analysis module.

[0100] Online retraining refers to the process of real-time training of the model based on new data during the operation of the model. The weight allocation strategy refers to the strategy of assigning weights to different data points in the spatial interpolation calculation. The adaptive verification closed loop refers to a closed-loop system that enables the load distribution analysis module to automatically adapt to the actual changes in the bridge through continuous verification and adjustment. After the updated theoretical displacement field distribution data is injected into the training data set, the load transfer model is retrained online based on the injected training data set. Through retraining, the weight allocation strategy in the spatial interpolation calculation is optimized, so that the load transfer model can perform spatial interpolation calculation more accurately. At the same time, by continuously repeating the above verification and adjustment process, an adaptive verification closed loop of the load distribution analysis module is formed to ensure the accuracy and reliability of the load distribution analysis. For example, in the retraining process, a deep learning algorithm is used to train the load transfer model, adjust the parameters of the model, optimize the weight allocation strategy, and improve the performance of the model.

[0101] Step S300: performing deformation correlation analysis on the long-term monitoring data sequence to extract the deformation coupling characteristics of the bridge. The deformation coupling characteristics are used to characterize the coordinated deformation relationship between the bridge main beam and the pier.

[0102] Deformation correlation analysis refers to an in-depth analysis of data related to bridge deformation in a long-term monitoring data sequence to understand the correlation between the deformations of various parts of the bridge. The deformation coupling feature is a characteristic quantity that reflects the cooperative deformation relationship between the bridge main beam and the pier. In actual bridge structures, the main beam and the pier will deform under the action of force. The deformations between them are not independent of each other, but there is a cooperative deformation relationship. By performing deformation correlation analysis on the long-term monitoring data sequence, this deformation coupling feature can be extracted, providing an important basis for evaluating the structural performance and safety of the bridge.

[0103] As an implementation mode, step S300, performing deformation correlation analysis on the long-term monitoring data sequence to extract the deformation coupling characteristics of the bridge, may specifically include: Step S310: extracting the main beam deflection monitoring sequence and the pier column inclination monitoring sequence from the long-term monitoring data sequence, and establishing a time domain synchronous correspondence between the main beam deflection monitoring sequence and the pier column inclination monitoring sequence.

[0104] The main beam deflection monitoring sequence refers to the time series data extracted from the long-term monitoring data sequence, which reflects the changes in the deflection of the bridge main beam. The pier tilt monitoring sequence refers to the time series data that reflects the changes in the tilt of the bridge pier. The time domain synchronous correspondence refers to matching and corresponding the main beam deflection monitoring sequence and the pier tilt monitoring sequence in the time dimension, so that their data at the same time point can be correlated with each other. In actual operation, the main beam deflection monitoring sequence and the pier tilt monitoring sequence are extracted from the long-term monitoring data sequence through data screening and processing. Then, according to the monitoring time nodes, the time domain synchronous correspondence between them is established.

[0105] Step S320: extracting the local deformation waveform in the main beam deflection monitoring sequence through the time window sliding algorithm, and extracting the phase offset in the pier column tilt monitoring sequence corresponding to the time window.

[0106] The time window sliding algorithm is an algorithm that performs sliding window operations on time series data. By setting a time window of fixed length, the window is slid on the time series for data processing. The local deformation waveform refers to the deformation waveform in the local time period extracted by the time window sliding algorithm in the main beam deflection monitoring sequence. The phase offset refers to the phase offset relative to the main beam deflection monitoring sequence in the pier tilt monitoring sequence of the corresponding time window. In practical applications, the time window sliding algorithm is used to slide the time window on the main beam deflection monitoring sequence to extract the local deformation waveform in each time window. At the same time, in the pier tilt monitoring sequence of the corresponding time window, the phase relationship between it and the main beam deflection monitoring sequence is analyzed to extract the phase offset. For example, the time window length is set to 10 monitoring time nodes, and the window is slid on the main beam deflection monitoring sequence to extract the local deformation waveform in each window; at the same time, in the pier tilt monitoring sequence of the corresponding window, the phase difference between it and the main beam deflection monitoring sequence is calculated to obtain the phase offset.

[0107] Step S330: construct a main beam-pier column displacement transfer function, input the local deformation waveform into the input end of the transfer function, and use the phase offset as a feedback correction item.

[0108] The main beam-pier displacement transfer function is a function used to describe the transfer relationship between the main beam displacement and the pier displacement. When constructing the function, the local deformation waveform is used as input to reflect the displacement change of the main beam; the phase offset is used as a feedback correction term to correct the output of the transfer function so that the transfer function can more accurately reflect the coordinated deformation relationship between the main beam and the pier. For example, the main beam-pier displacement transfer function can be in the form of a linear function or a nonlinear function, which is selected and constructed according to actual conditions. The local deformation waveform is input into the input end of the transfer function, and the predicted displacement of the pier is obtained by calculating the function. At the same time, the phase offset is used as a feedback correction term to correct the predicted displacement.

[0109] Step S340: using an iterative approximation algorithm to optimize the parameter combination of the main beam-pier column displacement transfer function until the error between the pier column displacement curve predicted at the output end and the measured data is less than a set tolerance.

[0110] The iterative approximation algorithm is an algorithm that gradually approaches the optimal solution through continuous iterative calculations. The parameter combination refers to the value combination of each parameter in the main beam-pier displacement transfer function. The set tolerance is a pre-set allowable range for measuring the error between the predicted pier displacement curve and the measured data. In practical applications, the iterative approximation algorithm is used to continuously adjust the parameter combination of the main beam-pier displacement transfer function and calculate the predicted pier displacement curve at the output end. Then, the predicted pier displacement curve is compared with the measured data to calculate the error. If the error is higher than the set tolerance, continue to adjust the parameter combination and repeat the above process until the error is lower than the set tolerance. For example, the iterative approximation algorithm can use Newton iteration method or gradient descent method, etc., to optimize the parameter combination and improve the accuracy of the transfer function through continuous iterative calculation.

[0111] As an implementation mode, in step S340, the optimization process of the main beam-pier column displacement transfer function includes: Step S341: extracting stiffness coupling term coefficients and damping correction term weights from the initial parameter combination of the transfer function.

[0112] The stiffness coupling term coefficient is a coefficient used to describe the stiffness coupling relationship between the main beam and the pier in the main beam-pier displacement transfer function. The damping correction term weight is the weight used to correct the damping term in the transfer function. In the initial parameter combination of the transfer function, multiple parameters are included, from which the stiffness coupling term coefficient and the damping correction term weight are extracted for subsequent optimization and adjustment. For example, in a simple main beam-pier displacement transfer function, the parameter combination may include the stiffness coupling term coefficient, the damping correction term weight, the displacement amplification coefficient, etc. The stiffness coupling term coefficient and the damping correction term weight are extracted from this initial parameter combination.

[0113] Step S342: calling the main beam deflection monitoring sequence and the pier column inclination monitoring sequence in the historical displacement data set to generate a displacement transfer verification sample set.

[0114] The historical displacement data set is a data set that stores the main beam deflection monitoring sequence and pier column tilt monitoring sequence in the historical monitoring data. The displacement transfer verification sample set is generated by calling the main beam deflection monitoring sequence and pier column tilt monitoring sequence in the historical displacement data set, and is used to verify the sample set of the main beam-pier column displacement transfer function. In practical applications, the main beam deflection monitoring sequence and pier column tilt monitoring sequence of different time periods are extracted from the historical displacement data set, and they are combined together to generate the displacement transfer verification sample set.

[0115] Step S343: input the verification sample set into the transfer function for forward calculation, and output the predicted pier column displacement curve.

[0116] Forward calculation refers to the process of inputting the main beam deflection monitoring sequence in the verification sample set into the main beam-pier displacement transfer function, calculating according to the calculation rules of the function, and obtaining the predicted pier displacement curve. After the verification sample set is input into the transfer function, the transfer function calculates and outputs the predicted pier displacement curve based on the input main beam deflection monitoring sequence and the current parameter combination. For example, the main beam deflection monitoring sequence of a certain time period in the verification sample set is input into the transfer function, and the predicted pier displacement curve of the time period is obtained through the calculation of the function.

[0117] Step S344: extracting the gradient difference distribution characteristics between the predicted curve and the measured curve, and generating a parameter correction direction vector.

[0118] The gradient difference distribution feature refers to the difference in gradient change between the predicted pier displacement curve and the measured pier displacement curve. The parameter correction direction vector is generated based on the gradient difference distribution feature and is a vector used to indicate the direction of parameter adjustment. In practical applications, the predicted pier displacement curve is compared with the measured pier displacement curve, and the difference in gradient change is analyzed to extract the gradient difference distribution feature. Then, based on the gradient difference distribution feature, the parameter correction direction vector is generated to determine the adjustment direction of the stiffness coupling term coefficient and the damping correction term weight. For example, if the gradient change of the predicted curve is significantly different from the measured curve in some areas, it means that the parameters of the transfer function in these areas may need to be adjusted, and the parameter correction direction vector is generated based on the difference.

[0119] Step S345: adjusting the spatial weight distribution of the stiffness coupling term coefficient and the time attenuation factor of the damping correction term weight according to the correction direction vector.

[0120] The spatial weight distribution refers to the weight distribution of the stiffness coupling term coefficient in space. The time attenuation factor refers to the attenuation coefficient of the damping correction term weight over time. According to the parameter correction direction vector, the spatial weight distribution of the stiffness coupling term coefficient and the time attenuation factor of the damping correction term weight are adjusted. For example, if the parameter correction direction vector indicates that the stiffness coupling term coefficient of a certain area needs to be increased, the spatial weight distribution of the area is adjusted accordingly; if it indicates that the time attenuation characteristics of the damping correction term weight need to be adjusted, the time attenuation factor is adjusted. Through this adjustment, the main beam-pier displacement transfer function can more accurately reflect the cooperative deformation relationship between the main beam and the pier.

[0121] Step S346: Input the adjusted parameter combination into the digital twin verification interface to match the time-frequency characteristics with the real-time monitored pier column dynamic response data.

[0122] The digital twin verification interface is an interface for verifying the parameter combination of the main beam-pier column displacement transfer function. The real-time monitored pier column dynamic response data refers to the response data of the pier column under dynamic action obtained by real-time monitoring, such as acceleration, displacement, etc. Time-frequency characteristic matching refers to the process of inputting the adjusted parameter combination into the digital twin verification interface, and matching and comparing the output of the transfer function with the real-time monitored pier column dynamic response data in the time and frequency domains. In practical applications, the adjusted parameter combination is input into the digital twin verification interface, and the transfer function is called through the interface for calculation to obtain the predicted dynamic response data of the pier column. Then, the predicted dynamic response data is matched with the real-time monitored pier column dynamic response data for time-frequency characteristics, and their similarities and differences in time and frequency are analyzed. For example, a time-frequency analysis method, such as wavelet transform, is used to perform time-frequency analysis on the predicted dynamic response data and the real-time monitoring data, and their time-frequency characteristics are compared.

[0123] Step S347: When the deviation of the low-frequency component energy ratio in the matching result exceeds a preset threshold, the anisotropic compensation mechanism of the stiffness coupling term coefficient is triggered.

[0124] The deviation of low-frequency component energy ratio refers to the difference in low-frequency component energy ratio between the predicted dynamic response data and the real-time monitoring data. The preset threshold is a pre-set allowable range for measuring the deviation of low-frequency component energy ratio. The anisotropic compensation mechanism of the stiffness coupling term coefficient is a mechanism for compensating for the difference of stiffness coupling term coefficients in different directions. In the process of time-frequency characteristic matching, if the deviation of low-frequency component energy ratio in the matching result exceeds the preset threshold, it means that the transfer function may have a large error in low-frequency response, and the anisotropic compensation mechanism of the stiffness coupling term coefficient needs to be triggered.

[0125] Step S348: re-execute the displacement transfer calculation using the compensated parameter combination until the phase lag characteristics of the predicted curve meet the time domain waveform of the measured data.

[0126] The phase lag characteristic refers to the phase lag of the predicted pier displacement curve relative to the measured pier displacement curve. The time domain waveform fit refers to the degree of waveform similarity between the predicted curve and the measured data in the time domain. After triggering the anisotropic compensation mechanism of the stiffness coupling term coefficient, the displacement transfer calculation is re-executed using the compensated parameter combination to obtain a new predicted pier displacement curve. Then, the new predicted curve is compared with the measured data to analyze their phase lag characteristics and time domain waveform fit. If the fit does not meet the standard, continue to adjust the parameter combination and repeat the above process until the phase lag characteristics of the predicted curve meet the time domain waveform fit of the measured data.

[0127] Step S349: Associating the optimal parameter combination that has finally passed the verification to the quantitative expression of the deformation coupling feature, and generating a transfer function feature update instruction.

[0128] The optimal parameter combination is the parameter combination that minimizes the error between the predicted curve output by the main beam-pier displacement transfer function and the measured data after multiple optimizations and verifications. The quantitative expression of the deformation coupling feature is to express the deformation coupling feature between the bridge main beam and the pier using a numerical value or mathematical expression. The transfer function feature update instruction is an instruction for updating the main beam-pier displacement transfer function feature. After obtaining the optimal parameter combination that has passed the final verification, it is associated with the quantitative expression of the deformation coupling feature so that the deformation coupling feature can more accurately reflect the cooperative deformation relationship between the main beam and the pier. At the same time, a transfer function feature update instruction is generated to update the characteristic parameters of the transfer function to ensure the accuracy and effectiveness of the transfer function. For example, the parameters such as the stiffness coupling term coefficient and the damping correction term weight in the optimal parameter combination are associated with the quantitative expression of the deformation coupling feature to generate a transfer function feature update instruction.

[0129] Step S3410: Synchronously adjust the reference threshold for determining displacement anomaly in the bridge health monitoring system based on the update instruction.

[0130] The bridge health monitoring system is a system used to monitor the health status of bridges in real time. The baseline threshold for displacement anomaly judgment is a standard threshold for judging whether the displacement of a bridge is abnormal. Based on the transfer function characteristic update instruction, the baseline threshold for displacement anomaly judgment in the bridge health monitoring system is adjusted synchronously. Because the displacement response of the bridge may change after the parameters of the main beam-pier displacement transfer function are updated, the original baseline threshold for displacement anomaly judgment may no longer apply. Therefore, it is necessary to adjust the baseline threshold according to the updated transfer function characteristics so that the bridge health monitoring system can more accurately judge the displacement anomaly of the bridge.

[0131] Step S400: Generate a comprehensive evaluation index based on the dynamic characteristics of the load and the deformation coupling characteristics, where the comprehensive evaluation index is used to quantify the degradation trend of the bearing performance of the bridge structure over its entire life cycle.

[0132] The dynamic characteristics of loads are extracted by load distribution analysis of long-term monitoring data series, and are characteristic quantities that reflect the load transfer law of bridges at different monitoring time nodes. The deformation coupling characteristics are extracted by deformation correlation analysis of long-term monitoring data series, and are characteristic quantities that characterize the coordinated deformation relationship between the bridge main beam and the pier. The comprehensive evaluation index is generated after comprehensive analysis of the dynamic characteristics of loads and the deformation coupling characteristics, and is used to quantify the degradation trend of the bearing performance of bridge structures throughout their life cycle. In practical applications, the bearing performance of bridges is affected by both loads and deformations, so it is necessary to combine the dynamic characteristics of loads and the deformation coupling characteristics for analysis. By generating comprehensive evaluation indicators, the degradation trend of the bearing performance of bridge structures throughout their life cycle can be evaluated more comprehensively and accurately, providing a scientific basis for the maintenance and management of bridges.

[0133] As an implementation mode, step S400 generates a comprehensive evaluation index based on the dynamic characteristics of the load and the deformation coupling characteristics, which may specifically include: Step S410: Establish a load-deformation joint analysis model, where the input end of the model receives equivalent load spectrum parameters in the load dynamic characteristics and main beam-pier column displacement transfer function parameters in the deformation coupling characteristics.

[0134] The load-deformation joint analysis model is a model used to comprehensively analyze the dynamic characteristics of loads and the coupling characteristics of deformations. The equivalent load spectrum parameters are important parameters in the dynamic characteristics of loads, reflecting the equivalent load conditions borne by the bridge at each monitoring time node. The main beam-pier displacement transfer function parameters are key parameters in the deformation coupling characteristics, describing the displacement transfer relationship between the main beam and the pier. When establishing the load-deformation joint analysis model, the equivalent load spectrum parameters and the main beam-pier displacement transfer function parameters are used as inputs of the model, so that the model can simultaneously consider the effects of load and deformation on the bridge structure. For example, the load-deformation joint analysis model can adopt a neural network model, with the equivalent load spectrum parameters and the main beam-pier displacement transfer function parameters as input nodes of the input layer, and analyze the combined effects of load and deformation through learning and calculation of the neural network.

[0135] Step S420: setting a multi-level evaluation node in the joint analysis model, wherein the first-level node is used to calculate the structural stiffness degradation rate, the second-level node is used to calculate the stress redistribution coefficient, and the third-level node is used to calculate the residual bearing reserve.

[0136] Multi-level evaluation nodes are evaluation nodes at different levels in the load-deformation joint analysis model, and each node has different evaluation functions. The structural stiffness degradation rate refers to the degree of stiffness degradation of the bridge structure over time during use. The first-level node calculates the structural stiffness degradation rate by analyzing the input equivalent load spectrum parameters and the main beam-pier displacement transfer function parameters. The stress redistribution coefficient is a coefficient that reflects the stress redistribution of the bridge structure during the force process. The second-level node calculates the stress redistribution coefficient based on the calculation results of the first-level node and the input parameters. The residual bearing reserve refers to the additional load that the bridge structure can still withstand in the current state. The third-level node calculates the residual bearing reserve by combining the calculation results of the first two levels of nodes. For example, in the joint analysis model, the first-level node can calculate the structural stiffness degradation rate by comparing the stiffness values ​​of the bridge in the initial state and the current state; the second-level node can calculate the stress redistribution coefficient based on the deformation of the structure and the mechanical properties of the material; the third-level node can calculate the residual bearing reserve based on the current bearing capacity and design bearing capacity of the structure.

[0137] Step S430: Input the output values ​​of the three-level evaluation nodes into the weighted fusion module to obtain the fused comprehensive evaluation value. The weight coefficient of the weighted fusion module is dynamically adjusted according to the bridge design parameters.

[0138] The weighted fusion module is a module used to fuse the output values ​​of the three-level evaluation nodes. The comprehensive evaluation value is the value obtained by fusing the output values ​​of the three-level evaluation nodes through the weighted fusion module, which comprehensively reflects the bearing capacity of the bridge structure. The weight coefficient is the coefficient used to adjust the weight of the output value of each evaluation node in the weighted fusion module. These weight coefficients are dynamically adjusted according to the bridge design parameters. Different bridge design parameters, such as the type, span, and material of the bridge, will affect the importance of the output value of each evaluation node to the comprehensive evaluation value. For example, for a long-span bridge, the structural stiffness degradation rate may have a greater impact on the comprehensive evaluation value. At this time, the weight coefficient of the output value of the first-level node can be appropriately increased. In practical applications, the weight coefficient of the weighted fusion module is determined according to the design parameters of the bridge, and the output values ​​of the three-level evaluation nodes are weighted and summed according to the weights to obtain the fused comprehensive evaluation value.

[0139] Step S440: converting the fused comprehensive evaluation value into a standardized bearing capacity index through a nonlinear mapping function.

[0140] The standardized bearing capacity index is obtained after conversion by a nonlinear mapping function, and is used to uniformly measure the bearing performance of bridges. The integrated evaluation value after fusion may have different dimensions and value ranges. In order to facilitate comparison and evaluation, it is necessary to convert it into a standardized bearing capacity index through a nonlinear mapping function. For example, the nonlinear mapping function can be in the form of a logarithmic function, an exponential function, etc., to map the integrated evaluation value after fusion to a value range, such as between 0 and 100, to obtain a standardized bearing capacity index.

[0141] Step S450: compare the bearing capacity index with the historical benchmark value to generate a comprehensive evaluation index including the current state level and the remaining life prediction.

[0142] The historical benchmark value refers to the reference value of the bearing capacity index of the bridge at different service stages in the past, which can reflect the normal change trend of the bridge's bearing performance. Trend comparison is the process of comparing the current bearing capacity index with the historical benchmark value and analyzing its change trend. The current state level is the level of the current bearing performance state of the bridge based on the comparison result of the bearing capacity index and the historical benchmark value. The remaining life prediction is the prediction of the remaining service life of the bridge based on the change trend of the bearing capacity index. By comparing the bearing capacity index with the historical benchmark value, it can be judged whether the bearing capacity of the bridge is in a normal state, a declining state or an abnormal state, thereby determining the current state level. At the same time, according to the decline rate and trend of the bearing capacity index, the remaining life of the bridge can be predicted. For example, if the current bearing capacity index is lower than the historical benchmark value and the downward trend is obvious, it means that the bearing capacity of the bridge is declining and may be in a risk warning state, and its remaining life needs to be further analyzed. Finally, the current state level and the remaining life prediction information are integrated to generate a comprehensive evaluation index containing this information.

[0143] As an implementation method, step S450 compares the trend of the bearing capacity index with the historical benchmark value to generate a comprehensive evaluation index including the current state level and the remaining life prediction, which may specifically include: Step S451: extracting a historical benchmark value set of the bearing capacity index from the bridge service history database, wherein the historical benchmark value set includes the mean value, fluctuation range and degradation rate characteristic parameters of the bearing capacity index at different service stages.

[0144] The bridge service history database is a database that stores various data of the bridge throughout its service process, including historical data of the bearing capacity index. The historical benchmark value set is extracted from the bridge service history database, and contains a set of mean values, fluctuation ranges, and degradation rate characteristic parameters of the bearing capacity index at different service stages. The mean value of the bearing capacity index reflects the average bearing performance of the bridge at a certain service stage, the fluctuation range indicates the amplitude of change of the bearing capacity index at this stage, and the degradation rate characteristic parameter describes the degradation rate of the bearing capacity index over time. For example, by querying and analyzing the bridge service history database, the bearing capacity index of the bridge in different years is extracted, and the mean value, fluctuation range, and degradation rate characteristic parameters of each year are calculated to form a historical benchmark value set.

[0145] Step S452: Arrange the bearing capacity index sequence continuously acquired in the current monitoring period in chronological order, and use the trend decomposition algorithm to separate the long-term degradation trend term and the short-term environmental disturbance term.

[0146] The current monitoring period refers to the time period during which monitoring is currently being performed. The bearing capacity index sequence is a sequence of bearing capacity indices continuously obtained during the current monitoring period. The trend decomposition algorithm is an algorithm used to decompose a time series into different components. The long-term degradation trend term refers to the long-term downward trend of the bearing capacity index over time, which reflects the impact of factors such as aging and damage of the bridge structure on the bearing performance. The short-term environmental disturbance term refers to the fluctuation effect on the bearing capacity index caused by short-term environmental factors such as climate change and temporary loads. In practical applications, trend decomposition algorithms such as seasonal decomposition and wavelet decomposition are used to decompose the bearing capacity index sequence in the current monitoring period into long-term degradation trend terms and short-term environmental disturbance terms. For example, the seasonal decomposition algorithm is used to decompose the bearing capacity index sequence into trend terms, seasonal terms and residual terms, where the trend term is the long-term degradation trend term, and the seasonal term and the residual term can be combined into the short-term environmental disturbance term.

[0147] Step S453: Dynamically time-warp the long-term degradation trend item and the benchmark trend curve of the corresponding service stage in the historical benchmark value set, and calculate the trend offset and curvature difference.

[0148] Dynamic time warping matching is an algorithm used to compare the similarity of two time series. It can dynamically adjust the two series on the time axis to make them match the best. The benchmark trend curve is the trend curve of the bearing capacity index corresponding to the service stage in the historical benchmark value set. The trend offset refers to the vertical offset between the long-term degradation trend item and the benchmark trend curve, which reflects the difference between the current bridge bearing performance degradation and the historical situation. The curvature difference refers to the difference in the curvature of the curves between the two, which can reflect the change in the speed of bearing performance degradation. In practical applications, the dynamic time warping matching algorithm is used to match the long-term degradation trend item with the benchmark trend curve, and the trend offset and curvature difference are calculated. For example, through the dynamic time warping algorithm, the best match between the long-term degradation trend item and the benchmark trend curve on the time axis is found, and their vertical distance at each time point is calculated to obtain the trend offset; at the same time, the curvature changes of the two curves are analyzed and the curvature difference is calculated.

[0149] Step S454: Based on the direction and magnitude of the trend offset, determine whether the stiffness degradation mode of the bridge structure in the current monitoring period belongs to the uniform settlement type, the local damage type or the overall fatigue type.

[0150] The direction and amplitude of the trend offset reflect the relative position and degree of difference between the long-term degradation trend item and the benchmark trend curve. The uniform settlement type stiffness degradation mode refers to the uniform stiffness degradation of the bridge structure as a whole, which is similar to the uniform settlement situation. The local damage type stiffness degradation mode refers to the damage of a local part of the bridge structure, resulting in a significant decrease in the stiffness of that part. The overall fatigue type stiffness degradation mode refers to the fatigue damage of the bridge structure as a whole due to long-term load bearing, resulting in a gradual decrease in stiffness. According to the direction and amplitude of the trend offset, the stiffness degradation mode of the bridge structure can be judged. For example, if the direction of the trend offset is negative and the amplitude is large, and the curve is relatively smooth, it may belong to the overall fatigue type stiffness degradation mode; if the trend offset suddenly increases in a local area, it may belong to the local damage type stiffness degradation mode; if the trend offset is relatively uniform over the entire time series, it may belong to the uniform settlement type stiffness degradation mode.

[0151] Step S455: According to the stiffness degradation pattern matching, a preset remaining life prediction rule library is selected, and a corresponding life prediction model is selected and its time window length adjustment parameter is loaded.

[0152] The preset remaining life prediction rule library is a pre-established database containing remaining life prediction rules corresponding to different stiffness degradation modes. The life prediction model is a model used to predict the remaining life of a bridge. Different stiffness degradation modes may require different life prediction models. The time window length adjustment parameter is a parameter used to adjust the time window length of the life prediction model. It can adjust the time range of the prediction model according to actual conditions. After determining the stiffness degradation mode of the bridge structure, match it in the preset remaining life prediction rule library according to the mode and select the corresponding life prediction model. At the same time, load the time window length adjustment parameter of the model to adapt to different prediction needs.

[0153] Step S456: Input the terminal slope of the long-term degradation trend term, the curvature difference and the average degradation rate in the historical benchmark value set into the life prediction model, and iteratively deduce the exponential attenuation trajectory of the bearing capacity for multiple future service cycles.

[0154] The terminal slope of the long-term degradation trend term reflects the degradation rate of the current bearing capacity index, the curvature difference reflects the change in the degradation rate, and the average degradation rate in the historical benchmark value set provides a historical degradation reference. The life prediction model receives these parameters and iteratively deduces the attenuation trajectory of the bearing capacity index for multiple future service cycles. In practical applications, the terminal slope, curvature difference and average degradation rate of the long-term degradation trend term are used as inputs and substituted into the calculation formula of the life prediction model for iterative calculations. Each iterative calculation obtains a new bearing capacity index, and so on, to obtain the attenuation trajectory of the bearing capacity index for multiple future service cycles. For example, the life prediction model can use a linear regression model or a nonlinear regression model to calculate the bearing capacity index of each future service cycle based on the input parameters to form an attenuation trajectory.

[0155] Step S457: During the simulation process, the attenuation trajectory is monitored in real time for similarity to the trajectory of similar bridge failure cases in the historical benchmark value set, and the time window length is dynamically adjusted to optimize the prediction step length.

[0156] Similar bridge failure cases refer to the trajectory of changes in the bearing capacity index when a bridge with similar structure, use environment and other conditions as the target bridge fails. Trajectory similarity refers to the degree of similarity between the attenuation trajectory and the trajectory of similar bridge failure cases. The time window length is the time range used for prediction in the life prediction model, and the prediction step refers to the time interval between each prediction. In the process of deducing the bearing capacity index attenuation trajectory for multiple future service cycles, the attenuation trajectory is monitored in real time to determine the similarity of the trajectory of similar bridge failure cases in the historical benchmark value set. If the similarity is high, it means that the target bridge may have a high risk of failure. At this time, the time window length can be dynamically adjusted to shorten the prediction step in order to more accurately predict the remaining life of the bridge.

[0157] Step S458: When the attenuation trajectory reaches a preset critical failure threshold for the first time, the corresponding number of service cycles is recorded as the remaining life prediction value.

[0158] The preset critical failure threshold is a pre-set value used to determine whether the bridge has reached the bearing capacity index threshold of the failure state. When the deduced bearing capacity index decay trajectory first touches the preset critical failure threshold, it means that the bridge may fail at that moment. The corresponding number of service cycles at this time is recorded and used as the remaining life prediction value. For example, the preset critical failure threshold is 20. During the deduction process, when the bearing capacity index decay trajectory drops to 20 for the first time, the corresponding number of service cycles recorded is 5 years, and the remaining life prediction value is 5 years.

[0159] Step S459: According to the coverage ratio between the fluctuation range of the carrying capacity index in the current monitoring period and the fluctuation range of the historical benchmark value, the status level is divided into normal operation level, risk warning level or emergency intervention level.

[0160] The fluctuation range of the bearing capacity index in the current monitoring period refers to the range between the maximum and minimum values ​​of the bearing capacity index in the current monitoring period. The fluctuation range of the historical benchmark value refers to the fluctuation range of the bearing capacity index corresponding to the service stage in the historical benchmark value set. The coverage ratio refers to the degree of overlap between the fluctuation range of the bearing capacity index in the current monitoring period and the fluctuation range of the historical benchmark value. According to the size of the coverage ratio, the status level of the bridge can be divided. If the coverage ratio is high, it means that the fluctuation of the current bearing capacity index is similar to the historical situation, and the bridge is in the normal operation level; if the coverage ratio is moderate, but the fluctuation range tends to exceed the historical range, it means that the bridge may have certain risks and is at the risk warning level; if the coverage ratio is low and the fluctuation range greatly exceeds the historical range, it means that the bridge may face an emergency and is at the emergency intervention level.

[0161] Step S4510: Encode the remaining life prediction value, state level and stiffness degradation mode into a structured evaluation vector, embedding a timestamp and a spatial location identifier.

[0162] The structured assessment vector is a vector formed by encoding information such as the remaining life prediction value, state level and stiffness degradation pattern. The timestamp is used to record the time of the assessment, and the spatial location identifier is used to indicate the specific location of the bridge. After obtaining the remaining life prediction value, state level and stiffness degradation pattern, they are encoded to form a structured assessment vector. At the same time, the timestamp and spatial location identifier are embedded so that the assessment result has time and space information. For example, the remaining life prediction value is represented by a numerical value, the state level is represented by a classification code, and the stiffness degradation pattern is represented by another classification code. These codes are combined into a vector, and then the timestamp and spatial location identifier are added to form a structured assessment vector.

[0163] Step S4511: Correct the short-term evaluation deviation caused by environmental disturbance items by searching the similarity between the evaluation vector and the historical evaluation case library.

[0164] The historical evaluation case library is a database that stores case information of bridge evaluations in history. Similarity retrieval refers to the process of searching for cases similar to the current evaluation vector in the historical evaluation case library. The environmental disturbance term is separated in step S452. Due to the fluctuation of the bearing capacity index caused by short-term environmental factors, it may cause deviations in short-term evaluations. By performing similarity retrieval on the current evaluation vector and the historical evaluation case library, similar historical cases are found, and the impact and correction methods of environmental disturbance terms in these cases are analyzed, and the current evaluation results are corrected to eliminate the short-term evaluation deviations caused by environmental disturbance terms. For example, a case with a high degree of similarity to the current evaluation vector is found in the historical evaluation case library, and it is found that the environmental disturbance term in this case causes the evaluation result to be high. By analyzing its correction method, the current evaluation result is adjusted accordingly.

[0165] Step S4512: Convert the corrected evaluation vector into a comprehensive evaluation index including a visualized degradation path map, maintenance priority recommendations, and a list of key monitoring areas.

[0166] The visual degradation path diagram is a chart that graphically displays the degradation process of the bridge's bearing performance, which can intuitively reflect the changing trend of the bridge's bearing performance in the future. The maintenance priority recommendation is a recommendation to prioritize bridge maintenance work based on the evaluation results. The list of key monitoring areas is a list of areas that need to be monitored on the bridge. These areas may be areas that are prone to problems or have already been damaged. The corrected evaluation vector is converted to generate a comprehensive evaluation index that includes a visual degradation path diagram, maintenance priority recommendations, and a list of key monitoring areas. For example, a visual degradation path diagram is drawn based on the remaining life prediction value and bearing capacity exponential decay trajectory in the evaluation vector; maintenance priority recommendations are formulated based on the state level and stiffness degradation mode; and a list of key monitoring areas is determined based on the damage found during the evaluation process.

[0167] Step S4513: Dynamically associate the comprehensive evaluation indicators with the real-time monitoring data in the bridge digital twin model, triggering the model self-update mechanism to synchronously reflect the latest structural status.

[0168] The digital twin model of a bridge is a virtual model obtained by digitally modeling the physical entity of the bridge. It can reflect the structural status of the bridge in real time. The comprehensive evaluation indicators are dynamically associated with the real-time monitoring data in the digital twin model of the bridge, so that the comprehensive evaluation indicators can be combined with the real-time monitoring data. When the comprehensive evaluation indicators change, the model self-update mechanism is triggered to update the digital twin model of the bridge to synchronously reflect the latest structural status of the bridge. For example, comprehensive evaluation indicators such as visual degradation path diagrams, maintenance priority recommendations, and lists of key monitoring areas are associated with real-time displacement, stress and other monitoring data in the digital twin model of the bridge. When the comprehensive evaluation indicators are updated, the relevant parameters and visual displays in the digital twin model of the bridge are automatically updated, so that the model can accurately reflect the latest situation of the bridge.

[0169] Step S500: Determine the bearing capacity status of the bridge in the current and future service stages according to the comparison result between the comprehensive evaluation index and the preset bearing capacity threshold.

[0170] The preset bearing capacity threshold is a pre-set standard threshold used to classify the bearing capacity status of the bridge. By comparing the comprehensive evaluation index with the preset bearing capacity threshold, the bearing capacity status of the bridge in the current and future service stages can be determined. The comprehensive evaluation index comprehensively reflects the bearing performance of the bridge, including information such as the current status level and the remaining life prediction. The preset bearing capacity threshold can be set according to factors such as the design requirements and safety standards of the bridge. For example, based on the comparison results, it can be determined whether the bridge is in a safe service state, a state that requires restricted access, or a state that requires emergency closure. At the same time, combined with the remaining life prediction and the future bearing capacity exponential decay trajectory in the comprehensive evaluation index, the changing trend of the bearing capacity status of the bridge in the future service stage can also be predicted.

[0171] As an implementation mode, step S500, based on the comparison result between the comprehensive evaluation index and the preset bearing capacity threshold, determines the bearing capacity status of the bridge in the current and future service stages, which may specifically include: Step S510: Compare the carrying capacity index in the comprehensive evaluation index with the preset carrying capacity thresholds step by step. The preset carrying capacity thresholds include a normal use state threshold, a restricted traffic state threshold, and an emergency closure state threshold.

[0172] The bearing capacity index is an important component of the comprehensive evaluation index, which reflects the current bearing performance of the bridge. The preset bearing capacity thresholds include the normal use state threshold, the restricted traffic state threshold and the emergency closure state threshold. These thresholds are set according to factors such as the design standards and safety requirements of the bridge. The normal use state threshold is the lower limit of the bearing capacity index that the bridge can pass and use normally. When the bearing capacity index is higher than this threshold, the bridge is in normal use. The restricted traffic state threshold is when the bearing capacity index is lower than this threshold, it is necessary to restrict the passage of the bridge, such as limiting the vehicle load and speed. The emergency closure state threshold is when the bearing capacity index is lower than this threshold, the bridge needs to be closed urgently, prohibiting vehicles and pedestrians from passing, to ensure safety. In actual operation, the bearing capacity index in the comprehensive evaluation index is compared step by step with these preset bearing capacity thresholds to determine the current bearing state of the bridge. For example, the bearing capacity index is first compared with the normal use state threshold. If it is higher than this threshold, the bridge is in normal use; if it is lower than the normal use state threshold, it is compared with the restricted traffic state threshold, and so on.

[0173] Step S520: Determine the current carrying capacity state according to the interval affiliation between the carrying capacity index and the preset carrying capacity threshold value, where the current carrying capacity state includes a safe service state, a restricted traffic preparation state, and an emergency closure trigger state.

[0174] The interval affiliation refers to the relationship between the bearing capacity index falling within different intervals divided by the preset bearing capacity threshold. According to the interval affiliation between the bearing capacity index and the preset bearing capacity threshold, the current bearing capacity status of the bridge can be determined. If the bearing capacity index is higher than the normal use status threshold, the bridge is in a safe service state, and the bridge can be used and used normally. If the bearing capacity index is lower than the normal use status threshold but higher than the restricted access status threshold, the bridge is in a restricted access preparation state, and it is necessary to pay close attention to the status of the bridge, and some preventive restricted access measures may be taken. If the bearing capacity index is lower than the restricted access status threshold, the bridge is in an emergency closure trigger state, and the bridge needs to be closed immediately to prohibit vehicles and pedestrians from passing.

[0175] Step S530: input the historical bearing capacity index sequence of the comprehensive evaluation index into the time series prediction model, and output the future bearing capacity index prediction sequence for the next three monitoring periods.

[0176] The historical bearing capacity index sequence is a sequence of the bearing capacity index of the bridge recorded in the comprehensive evaluation index over a period of time in the past. The time series prediction model is a model used to predict time series data. It can predict future data values ​​based on the changing trend of historical data. The historical bearing capacity index sequence is input into the time series prediction model. The model outputs the future bearing capacity index prediction sequence for the next three monitoring periods by analyzing and learning the historical data. For example, the time series prediction model can use the autoregressive integrated moving average model (ARIMA), long short-term memory network (LSTM), etc., and take the historical bearing capacity index sequence as input. After the model is calculated, the bearing capacity index prediction values ​​for the next three monitoring periods are obtained to form the future bearing capacity index prediction sequence.

[0177] Step S540: Compare the future bearing capacity index prediction sequence with the preset bearing capacity threshold value period by period, and identify the future monitoring period in which the future bearing capacity index prediction sequence is lower than the threshold level corresponding to the current monitoring period for the first time.

[0178] Cycle-by-cycle comparison refers to the process of comparing each predicted value in the future bearing capacity index prediction sequence with the preset bearing capacity threshold one by one. The threshold level corresponding to the current monitoring cycle refers to the preset bearing capacity threshold level corresponding to the bridge status in the current monitoring cycle, such as the normal use status threshold, the restricted traffic status threshold or the emergency closure status threshold. By comparing the future bearing capacity index prediction sequence with the preset bearing capacity thresholds cycle by cycle, find the future monitoring cycle in which the future bearing capacity index prediction sequence is lower than the threshold level corresponding to the current monitoring cycle for the first time. For example, in the current monitoring cycle, the bridge is in a safe service state, corresponding to the normal use status threshold. Each predicted value in the future bearing capacity index prediction sequence is compared with the normal use status threshold to find the future monitoring cycle corresponding to the first predicted value lower than the threshold.

[0179] Step S550: Determine the degradation trend of the bearing capacity state in the future service stage according to the time interval of the future monitoring cycle that is first lower than the current threshold level and the exponential decline gradient, and generate a bearing capacity state decision report including the current bearing capacity state classification and the future bearing capacity state warning level.

[0180] The time interval refers to the time difference between the future monitoring period when the level is first lower than the current threshold level and the current monitoring period. The exponential decline gradient refers to the rate of decline of the bearing capacity index in the process of the future bearing capacity index prediction sequence approaching the current threshold level. According to the time interval and the exponential decline gradient, the degradation trend of the bearing capacity state in the future service stage can be determined. If the time interval is short and the exponential decline gradient is large, it means that the bearing capacity state of the bridge will degrade rapidly and face a higher risk in the future; if the time interval is long and the exponential decline gradient is small, it means that the bearing capacity state of the bridge will degrade slowly. According to the current bearing capacity state classification (such as safe service state, restricted traffic preparation state, emergency closure trigger state) and the future bearing capacity state degradation trend, the future bearing capacity state warning level is generated, such as low risk warning, medium risk warning, high risk warning, etc. Finally, the current bearing capacity state classification and the future bearing capacity state warning level are integrated to generate a bearing capacity state decision report, which provides a decision basis for the management and maintenance of the bridge.

[0181] As an implementation method, the training process of the time series prediction model includes: Step S5301: extract the bearing capacity index time series from the historical monitoring data, and intercept the continuous degradation characteristic segments according to the preset cycle length.

[0182] Historical monitoring data is the data accumulated from long-term monitoring of bridges, which contains the historical records of bearing capacity index. The bearing capacity index time series is a sequence of bearing capacity indexes arranged in chronological order, extracted from the historical monitoring data. The preset cycle length is the pre-set time length for intercepting the continuous degradation characteristic segment. The continuous degradation characteristic segment refers to the time period in which the bearing capacity index shows a continuous degradation trend over a period of time. In actual operation, the bearing capacity index data is screened out from the historical monitoring data to form a bearing capacity index time series. Then, according to the preset cycle length, the continuous degradation characteristic segment is intercepted in the time series.

[0183] Step S5302: Adopt an adaptive decomposition algorithm to decompose each characteristic segment into multiple intrinsic modal components and trend residuals.

[0184] The adaptive decomposition algorithm is an algorithm that can automatically decompose data according to its own characteristics. The intrinsic mode component refers to the component with different frequencies and amplitudes obtained by decomposing the characteristic segment through the adaptive decomposition algorithm. Each intrinsic mode component represents a change pattern in the data. The trend remainder is the part remaining after decomposition that represents the long-term trend of the data. In practical applications, adaptive decomposition algorithms such as empirical mode decomposition (EMD) and ensemble empirical mode decomposition (EEMD) are used to decompose each continuous degradation characteristic segment into multiple intrinsic mode components and trend remainders. For example, the empirical mode decomposition algorithm is used to decompose a continuous degradation characteristic segment to obtain multiple intrinsic mode components, each with different frequencies and amplitudes, reflecting the changes in the bearing capacity index on different time scales, and a trend remainder is obtained, representing the long-term degradation trend of the characteristic segment.

[0185] Step S5303: extracting the time-frequency domain energy distribution characteristics for each intrinsic modal component, and establishing an associated mapping relationship with the external environmental variables.

[0186] The energy distribution characteristics in the time-frequency domain refer to the energy distribution of the intrinsic modal components in the time and frequency domains, which can reflect the energy concentration and change law of the intrinsic modal components. External environmental variables refer to external factors that may affect the bridge bearing capacity index, such as temperature, humidity, traffic flow, etc. For each intrinsic modal component, time-frequency analysis methods such as wavelet transform and short-time Fourier transform are used to extract its energy distribution characteristics in the time-frequency domain. Then, the relationship between these characteristics and external environmental variables is analyzed to establish an associated mapping relationship. For example, by performing a wavelet transform on a certain intrinsic modal component, its energy distribution characteristics in the time-frequency domain are obtained, and then the correlation between the characteristics and external environmental variables such as temperature and traffic flow is analyzed to establish an associated mapping relationship so that the influence of external environmental factors can be considered in the prediction.

[0187] Step S5304: configure an independent autoregressive predictor for each modal component according to the association mapping relationship, and load the covariance matrix of the corresponding environmental variables.

[0188] An autoregressive predictor is a predictor based on an autoregressive model, which can predict future data based on historical data. A covariance matrix is ​​a matrix used to describe the correlation between external environmental variables. According to the association mapping relationship established in step S5303, an independent autoregressive predictor is configured for each intrinsic modal component. Each autoregressive predictor can be predicted based on the historical data of the modal component and the external environmental variables associated therewith. At the same time, the covariance matrix of the corresponding environmental variables is loaded so as to consider the correlation between the external environmental variables in the prediction process. For example, for an intrinsic modal component, an autoregressive predictor is configured according to its association mapping relationship with temperature and traffic flow, and the covariance matrix of temperature and traffic flow is loaded. The predictor can predict the future value of the modal component based on the historical data of the modal component, the current values ​​of temperature and traffic flow, and the correlation between them.

[0189] Step S5305: Input the trend residual into the nonlinear expansion predictor and generate a long-term degradation baseline in combination with the bridge material aging parameters.

[0190] The nonlinear extended predictor is a predictor that can handle nonlinear relationships. It can further predict the trend residual. The bridge material aging parameters are parameters that describe the degree of aging of bridge materials during use, such as the carbonization depth of concrete, the corrosion rate of steel, etc. The trend residual is input into the nonlinear extended predictor, and combined with the bridge material aging parameters to generate a long-term degradation baseline. The long-term degradation baseline reflects the degradation trend of the bridge bearing capacity index in the long term, and it takes into account the influence of factors such as bridge material aging. For example, the nonlinear extended predictor can use a neural network model, take the trend residual and the bridge material aging parameters as input, and generate a long-term degradation baseline through model learning and calculation. This baseline can be used to predict the change trend of the bridge bearing capacity index over a long period of time.

[0191] Step S5306: calling the parameter coordinator to perform time domain alignment and energy ratio calibration on the outputs of each predictor to generate a fusion prediction result.

[0192] The parameter coordinator is a module used to coordinate the outputs of each predictor. It can perform time domain alignment and energy ratio calibration on the outputs of each predictor. Time domain alignment refers to aligning the outputs of each predictor in the time dimension so that they have the same time starting point and time interval. Energy ratio calibration refers to weighted summing of the outputs of each predictor according to the energy ratio of each intrinsic modal component and trend residual term to obtain a more accurate fusion prediction result. In actual applications, the parameter coordinator is called to perform time domain alignment on the prediction outputs of the intrinsic modal components of each regression predictor and the prediction outputs of the trend residual term of the nonlinear expansion predictor, and then perform weighted summing according to the energy ratio of each component to generate a fusion prediction result.

[0193] Step S5307: perform a morphological similarity comparison between the fusion result and the forward-looking window of the measured data to extract the spatial distribution hotspots of the prediction error.

[0194] The forward-looking window refers to a data window of a future period of time selected from the measured data for comparison with the fusion prediction results. Morphological similarity comparison refers to the comparison of the shape and trend of the forward-looking windows of the fusion prediction results and the measured data to analyze their similarity. The spatial distribution hotspot of the prediction error refers to the area or time period with large errors in the prediction error. The forward-looking windows of the fusion prediction results and the measured data are compared for morphological similarity, and the accuracy of the prediction is evaluated by calculating the similarity indicators between the two, such as the correlation coefficient and the mean square error. At the same time, the distribution of the prediction error in time and space is analyzed to extract the spatial distribution hotspots of the prediction error. For example, the data of the next 10 monitoring periods in the measured data are selected as the forward-looking window, the fusion prediction results are compared with the forward-looking window, and the correlation coefficient is calculated. If the correlation coefficient is low, it means that the prediction accuracy is poor. Further analyze the distribution of the prediction error in these 10 monitoring periods to find the time period with large errors as the spatial distribution hotspot of the prediction error.

[0195] Step S5308: Adjust the weight allocation strategy of each predictor according to the hotspot distribution, and focus on strengthening the prediction logic of the modal components associated with the hotspot area.

[0196] The weight allocation strategy refers to the weight allocation method used when weighted summing the outputs of each predictor in the parameter coordinator. The hotspot area refers to the area or time period corresponding to the hotspot of the spatial distribution of the prediction error. According to the hotspot of the spatial distribution of the prediction error, the weight allocation strategy of each predictor is adjusted to increase the weight of the prediction logic of the modal component associated with the hotspot area. Because the prediction error in the hotspot area is large, it means that the prediction of the modal components corresponding to these areas may be inaccurate, and it is necessary to focus on strengthening its prediction logic. For example, if the hotspot of the spatial distribution of the prediction error is concentrated in a certain time period, and this time period is mainly related to a certain intrinsic modal component, then the weight of the autoregressive predictor corresponding to the intrinsic modal component is increased, and the parameters and model structure of the predictor are optimized to improve its prediction accuracy.

[0197] Step S5309: Connect the optimized predictor combination to the real-time data stream interface to establish a dynamic parameter exchange channel between the prediction model and the monitoring system.

[0198] The real-time data stream interface is an interface for receiving real-time monitoring data. The dynamic parameter exchange channel refers to a channel through which parameters can be exchanged in real time between the prediction model and the monitoring system. The optimized predictor combination is connected to the real-time data stream interface so that the prediction model can obtain the latest data of the monitoring system in real time. At the same time, a dynamic parameter exchange channel between the prediction model and the monitoring system is established. During the prediction process, if the parameters of the monitoring system change, such as changes in external environmental variables, updates to bridge structure parameters, etc., these changes can be transmitted to the prediction model in real time through the dynamic parameter exchange channel, and the prediction model can adjust the prediction results in time according to these changes. For example, when the monitoring system detects a sudden increase in temperature, the temperature change information is transmitted to the prediction model through the dynamic parameter exchange channel, and the prediction model can adjust the prediction results according to the correlation mapping relationship between temperature and each modal component.

[0199] Step S5310: Based on the real-time feedback data in the exchange channel, perform online fine-tuning of the predictor parameters to maintain the model prediction accuracy.

[0200] Real-time feedback data refers to data obtained from the monitoring system in real time through the dynamic parameter exchange channel. Online fine-tuning refers to the process of adjusting the parameters of the predictor in real time according to the real-time feedback data during the operation of the model. Based on the real-time feedback data in the exchange channel, the parameters of the predictor are fine-tuned online to maintain the prediction accuracy of the model. Because the actual situation of the bridge is constantly changing, the real-time feedback data can reflect these changes. By fine-tuning the predictor parameters online, the prediction model can better adapt to these changes and improve the prediction accuracy. For example, when the real-time feedback data shows that the change trend of a certain inherent modal component deviates from the prediction result, the parameters of the autoregressive predictor corresponding to the modal component are fine-tuned according to the deviation, such as adjusting the autoregressive coefficient, etc., to improve the accuracy of the prediction.

[0201] Please refer to Figure 3, is a structural block diagram of the data analysis system 120 of the present application, and the data analysis system 120 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in a ROM (i.e., read-only memory) 1002 or a computer program loaded from a storage unit 1008 to a RAM (i.e., random access memory) 1003. In RAM 1003, various programs and data required for the operation of the data analysis system 120 can also be stored. The computing unit 1001, ROM 1002, and RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004. Multiple components in the data analysis system 120 are connected to the I / O interface 1005, including: an input unit 1006, an output unit 1007, a storage unit 1008, and a communication unit 1009. The input unit 1006 can be any type of device that can input information to the data analysis system 120. The output unit 1007 can be any type of device that can present information. The storage unit 1008 may include, but is not limited to, a magnetic disk and an optical disk.

Claims

1. A method for analyzing the bearing capacity data of a long-span continuous rigid frame bridge, characterized in that: include: Acquire a long-term monitoring data sequence of a target bridge, wherein the long-term monitoring data sequence includes a plurality of structural response data sets arranged according to monitoring time nodes during the service life of the bridge; Performing load distribution analysis on the long-term monitoring data sequence to extract the load dynamic characteristics of the bridge, wherein the load dynamic characteristics are used to characterize the load transfer law of each key section of the bridge at different monitoring time nodes; Performing deformation correlation analysis on the long-term monitoring data sequence to extract deformation coupling characteristics of the bridge, wherein the deformation coupling characteristics are used to characterize the cooperative deformation relationship between the bridge main beam and the pier column; generating a comprehensive evaluation index based on the load dynamic characteristics and the deformation coupling characteristics, wherein the comprehensive evaluation index is used to quantify the load-bearing performance degradation trend of the bridge structure over its entire life cycle; The bearing capacity status of the bridge in the current and future service stages is determined based on the comparison result of the comprehensive evaluation index with the preset bearing capacity threshold.

2. The method according to claim 1, characterized in that The performing load distribution analysis on the long-term monitoring data sequence to extract the load dynamic characteristics of the bridge includes: Separating a load distribution data subset from the long-term monitoring data sequence, wherein the load distribution data subset includes the moment monitoring value and the shear force monitoring value of the mid-span and support section of the bridge at each monitoring time node; Decomposing the load distribution data subset into a dead load component sequence and a live load component sequence according to a preset load decomposition level, wherein the dead load component sequence is subjected to low-pass filtering to extract long-term trend items, and the live load component sequence is subjected to high-pass filtering to extract short-term fluctuation items; The dead load component sequence and the live load component sequence are dynamically superimposed by using a time-varying weight distribution algorithm to generate an equivalent load spectrum corresponding to each monitoring time node; Calling the pre-trained load transfer model to perform spatial interpolation calculation on the equivalent load spectrum to obtain a load distribution cloud map within the full span of the bridge; The geometric parameters and mechanical parameters of the maximum gradient change area in the load distribution cloud map are used as the load dynamic characteristics.

3. The method according to claim 2, characterized in that The step of performing deformation correlation analysis on the long-term monitoring data sequence to extract deformation coupling characteristics of the bridge includes: Extracting a main beam deflection monitoring sequence and a pier column tilt monitoring sequence from the long-term monitoring data sequence, and establishing a time domain synchronous correspondence between the main beam deflection monitoring sequence and the pier column tilt monitoring sequence; Extracting the local deformation waveform in the main beam deflection monitoring sequence by using a time window sliding algorithm, and extracting the phase offset in the pier column tilt monitoring sequence corresponding to the time window; Constructing a main beam-pier displacement transfer function, inputting the local deformation waveform into the input end of the transfer function, and using the phase offset as a feedback correction term; An iterative approximation algorithm is used to optimize the parameter combination of the main beam-pier column displacement transfer function until the error between the pier column displacement curve predicted at the output end and the measured data is less than a set tolerance; The optimized characteristic parameter set of the main beam-pier column displacement transfer function is used as the deformation coupling feature.

4. The method according to claim 3, characterized in that The generating of a comprehensive evaluation index based on the load dynamic characteristics and the deformation coupling characteristics comprises: Establishing a load-deformation joint analysis model, wherein the input end of the model receives the equivalent load spectrum parameters in the load dynamic characteristics and the main beam-pier column displacement transfer function parameters in the deformation coupling characteristics; In the joint analysis model, a multi-level evaluation node is set, wherein the first-level node is used to calculate the structural stiffness degradation rate, the second-level node is used to calculate the stress redistribution coefficient, and the third-level node is used to calculate the residual bearing reserve; The output values ​​of the three-level evaluation nodes are input into a weighted fusion module to obtain a fused comprehensive evaluation value, wherein the weight coefficient of the weighted fusion module is dynamically adjusted according to the bridge design parameters; The fused comprehensive evaluation value is converted into a standardized bearing capacity index through a nonlinear mapping function; The bearing capacity index is trend-compared with historical benchmark values ​​to generate a comprehensive evaluation index including current status level and remaining life prediction.

5. The method according to claim 4, characterized in that Determining the bearing capacity status of the bridge in the current and future service stages according to the comparison result between the comprehensive evaluation index and the preset bearing capacity threshold includes: Compare the carrying capacity index in the comprehensive evaluation index with the preset carrying capacity thresholds step by step, wherein the preset carrying capacity thresholds include a normal use state threshold, a restricted traffic state threshold, and an emergency closure state threshold; Determine the current carrying capacity state according to the interval affiliation between the carrying capacity index and the preset carrying capacity threshold, wherein the current carrying capacity state includes a safe service state, a restricted traffic preparation state, and an emergency closure trigger state; Inputting the historical bearing capacity index sequence of the comprehensive evaluation index into the time series prediction model, and outputting the future bearing capacity index prediction sequence for the next three monitoring periods; Comparing the future bearing capacity index prediction sequence with the preset bearing capacity threshold value period by period, and identifying the future monitoring period in which the future bearing capacity index prediction sequence is lower than the threshold level corresponding to the current monitoring period for the first time; Based on the time interval and exponential decline gradient of the future monitoring period when the level falls below the current threshold for the first time, the degradation trend of the bearing capacity state in the future service stage is determined, and a bearing capacity state decision report including the current bearing capacity state classification and the future bearing capacity state warning level is generated.

6. The method according to claim 3, characterized in that The optimization process of the main beam-pier column displacement transfer function includes: Extracting stiffness coupling term coefficients and damping correction term weights from the initial parameter combination of the transfer function; The main beam deflection monitoring sequence and pier column tilt monitoring sequence in the historical displacement data set are called to generate a displacement transfer verification sample set; Inputting the verification sample set into the transfer function for forward calculation, and outputting a predicted pier column displacement curve; Extract the gradient difference distribution characteristics between the predicted curve and the measured curve, and generate the parameter correction direction vector; Adjusting the spatial weight distribution of the stiffness coupling term coefficient and the time attenuation factor of the damping correction term weight according to the correction direction vector; The adjusted parameter combination is input into the digital twin verification interface to match the time-frequency characteristics with the real-time monitored pier column dynamic response data; When the deviation of the energy proportion of the low-frequency component in the matching result exceeds a preset threshold, the anisotropic compensation mechanism of the stiffness coupling term coefficient is triggered; The displacement transfer calculation is re-executed using the compensated parameter combination until the phase lag characteristics of the predicted curve and the time domain waveform of the measured data meet the standard; Associating the optimal parameter combination that has finally passed the verification to the quantitative expression of the deformation coupling feature to generate a transfer function feature update instruction; The benchmark threshold for displacement anomaly determination in the bridge health monitoring system is synchronously adjusted based on the update instructions.

7. The method according to claim 5, characterized in that The training method of the time series prediction model includes: Extract the bearing capacity index time series from the historical monitoring data and intercept the continuous degradation characteristic segments according to the preset period length; Adopting adaptive decomposition algorithm to decompose each characteristic segment into multiple intrinsic modal components and trend residuals; Extract the energy distribution characteristics in the time-frequency domain for each intrinsic modal component and establish a correlation mapping relationship with external environmental variables; According to the association mapping relationship, an independent autoregressive predictor is configured for each modal component, and a covariance matrix of the corresponding environmental variables is loaded; Inputting the trend residual into a nonlinear expansion predictor and combining it with bridge material aging parameters to generate a long-term degradation baseline; Call the parameter coordinator to perform time domain alignment and energy ratio calibration on the outputs of each predictor to generate a fusion prediction result; The fusion result is compared with the forward-looking window of the measured data for morphological similarity, and the spatial distribution hot spots of the prediction error are extracted; Adjust the weight allocation strategy of each predictor according to the hotspot distribution, and focus on strengthening the prediction logic of the modal components associated with the hotspot area; Connect the optimized predictor combination to the real-time data stream interface to establish a dynamic parameter exchange channel between the prediction model and the monitoring system; Based on the real-time feedback data in the exchange channel, online fine-tuning of the predictor parameters is performed to maintain the model prediction accuracy.

8. The method according to claim 2, characterized in that The verification method of the load transfer model includes: Obtaining measured three-dimensional displacement field data of the target bridge under standard load test conditions, wherein the measured three-dimensional displacement field data includes spatial coordinate changes of the main beam mid-span and pier top sections; Inputting the load parameters of the standard load test condition into a pre-established parameterized finite element model, and performing multi-scale meshing to generate theoretical displacement field distribution data; Extracting the absolute error vector between the measured value of the vertical displacement at the mid-span of the main beam in the three-dimensional displacement field measured data and the theoretical value at the corresponding position in the theoretical displacement field distribution data, and calculating the relative error distribution mode between the measured value and the theoretical value of the horizontal displacement at the pier top; Identifying the correction priority of the main beam stiffness parameter according to the direction angle of the absolute error vector, and giving priority to adjusting the elastic modulus adjustment factor orthogonal to the direction of the maximum absolute error; Based on the spatial correlation of the relative error distribution pattern, the boundary stiffness proportionality coefficient of the pier foundation constraint condition is synchronously corrected; The modified elastic modulus adjustment factor and boundary stiffness proportional coefficient are re-imported into the parameterized finite element model, and the load-displacement response calculation is iteratively performed until the absolute error vector modulus and relative error rate of all measuring points are lower than the preset convergence threshold; Extract the mid-span bending moment distribution cloud diagram and pier bottom shear gradient curve from the final converged finite element model output to generate the load transfer feature verification fingerprint; Dynamically registering the load transfer feature verification fingerprint with the load distribution cloud map monitored in real time, and calculating the morphological similarity index of the two in the gradient change area; When the morphological similarity index is lower than the historical baseline level for three consecutive monitoring cycles, a parameter recalibration instruction of the load transfer model is triggered; According to the recalibration instruction, the latest measured data of the three-dimensional displacement field is called, the correction process of the elastic modulus adjustment factor and the boundary stiffness proportional coefficient is re-executed, and the material constitutive relationship of the finite element model is updated; The theoretical displacement field distribution data output by the updated finite element model is back-injected into the training data set of the load transfer model to replace the outdated response features in the historical samples; Based on the injected training data set, the load transfer model is retrained online to optimize the weight distribution strategy in the spatial interpolation calculation, thus forming an adaptive verification closed loop of the load distribution analysis module.

9. The method according to claim 2, characterized in that The time-varying weight distribution algorithm is used to dynamically superimpose the dead load component sequence and the live load component sequence to generate an equivalent load spectrum corresponding to each monitoring time node, including: Initializing the dead load weight coefficient of the corresponding time node based on the long-term trend slope of each monitoring time node in the dead load component sequence; Initializing the live load weight coefficient of the corresponding time node based on the short-term fluctuation amplitude of each monitoring time node in the live load component sequence; Based on the real-time monitoring data set of the bridge environment, the temperature gradient change rate, humidity cumulative effect factor and traffic flow impact index are extracted to generate the environmental dynamic impact set; The environmental dynamic impact set is input into the weight correction model, and the dead load weight coefficient and the live load weight coefficient are adjusted nonlinearly to obtain the corrected dead load weight and the corrected live load weight at each monitoring time node; The sliding time window mechanism is adopted to perform moving average processing on the modified constant load weights of the current monitoring time node and the previous preset number of nodes to generate a smoothed constant load weight sequence; Synchronously perform exponential decay processing on the modified live load weights of the current monitoring time node and the previous preset number of nodes to generate a decaying live load weight sequence; Multiplying the data of the corresponding time nodes in the smoothed dead load weight sequence and the dead load component sequence by weight to obtain the dead load weighted component; Multiplying the data of the corresponding time nodes in the attenuated live load weight sequence and the live load component sequence according to the weights to obtain the live load weighted components; Perform vector superposition on the weighted components of the dead load and the live load at the same monitoring time node to generate the initial equivalent load spectrum; Calling a load distribution verification model to perform a temporal and spatial alignment comparison between the initial equivalent load spectrum and a typical load pattern of the same type of bridge in a historical load distribution feature library; According to the load peak position deviation and distribution uniformity difference in the comparison results, reversely adjust the coupling adjustment parameters of the weight correction model; Re-execute the weight correction and subsequent superposition steps using the adjusted weight correction model until the similarity between the spatiotemporal distribution form of the initial equivalent load spectrum and the historical typical load pattern reaches a preset threshold; The verified initial equivalent load spectrum is marked as the final equivalent load spectrum and associated with the structural response data set of the corresponding monitoring time node.

10. A data analysis system, characterized in that: include: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method described in any one of claims 1 to 9.

Images