Bridge structure service state space-time correlation diagnosis method based on graph network and state space model

Abstract

The invention provides a bridge structure service state space-time correlation diagnosis method based on a graph network and a state space model. According to the method, a space-time diagram structure is adopted, a space-time diagram fusion learning module is constructed, deep learning is carried out in combination with a diagram neural network and time series data, meanwhile, data characteristics of the service state of a bridge structural component in two dimensions of space and time are considered, and modeling and dynamic space-time evolution of a single bridge complex component system are more accurately predicted. According to the method, a state space model is adopted, a structured state space sequence is constructed, a selectivity mechanism is introduced, model behaviors are dynamically adjusted, and optimal utilization of computing resources is achieved. According to the method, spatial-temporal characteristic decomposition and fusion are carried out on service state monitoring data of multiple types of bridge components with different complexities, so that processing requirements of service state spatial-temporal data of the bridge structural components with multiple types, multiple scales and multiple granularities are better met, and the universality and prediction efficiency of the method are improved.

Description

Technical Field

[0001] The present invention relates to the fields of bridge engineering, structural health monitoring, spatiotemporal data analysis, and deep learning technology, and in particular to a spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model. The method can be directly applied to the fields of intelligent bridges, intelligent bridge operation and maintenance, intelligent infrastructure, and intelligent disaster prevention and mitigation. Background Art

[0002] As a key component of the urban transportation system, the stability of the service status of bridges is directly related to the smoothness and safety of urban transportation. As a complex system composed of multiple types of structural components, bridges will inevitably be affected by the coupling of multiple factors such as the natural environment, traffic loads, and material aging during long-term service. The structural performance will continue to degrade, and even extreme situations may occur where the entire bridge collapses due to the failure of local components. Therefore, accurate and efficient prediction of the service status of bridge structures is of great significance to ensuring urban traffic safety and improving the efficiency of bridge operation and maintenance management.

[0003] Traditional bridge service status diagnosis methods mainly rely on the prediction of the service status of local components of single bridges. The research objects of interest are often limited to local components such as bridge towers, main beams, and cables. They fail to integrate and utilize the shared characteristics of multiple types of responses on bridge structures, and ignore the similarities between different bridge components and the spatiotemporal correlation in terms of damage evolution. As a result, the associated synergistic effects between the components of single bridges have not been fully explored.

[0004] At present, deep learning technology has achieved rapid development, and domestic and foreign research has carried out research on using deep learning methods to predict the service status of bridges. However, current research mainly focuses on:

[0005] (1) Deep learning-based diagnosis of the condition of single bridge components: For example, the damage location and severity assessment of cable-stayed bridges based on message passing neural networks uses the cable force data of multiple cables as input and outputs the location and cross-sectional area of ​​the damaged cables, which only considers the spatial correlation of different cables on a single bridge.

[0006] (2) For the diagnosis of the structural status of key components, since the structural responses of the components of a single bridge have mechanical correlation and deformation coordination characteristics, the spatiotemporal correlation between the multi-type responses of a single bridge is very important for status diagnosis, which has not been addressed in current research;

[0007] (3) In addition, existing studies often only focus on a single type of data, such as acceleration, cable force, vertical displacement, etc., and have not yet established a general deep learning method for spatiotemporal correlation prediction of multi-type, multi-scale, and multi-granular spatiotemporal data on the service status of bridges (including structural dynamic response, static response, quasi-static response, etc.).

[0008] In recent years, spatiotemporal graph structures and state-space models have provided new ideas and advanced algorithms for studying spatiotemporal correlation diagnosis. Spatiotemporal graph neural networks have shown broad application prospects in the intelligent diagnosis of the service status of bridge structures. However, current research still faces huge challenges. The problems that need to be solved urgently include: how to construct a suitable spatiotemporal graph structure to accurately reflect the complex spatiotemporal correlation relationship between different components of a single bridge, how to design an effective spatiotemporal graph neural network model to capture the spatiotemporal dependency of the evolution of the service status of bridge structure components, and how to make full use of the spatiotemporal data of multi-type bridge component status monitoring to improve prediction accuracy and efficiency. Summary of the invention

[0009] The purpose of the present invention is to solve the problems in the prior art and propose a spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model.

[0010] The present invention is implemented by the following technical scheme. The present invention proposes a spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model. The method comprises the following steps:

[0011] Step 1: Determine the bridge structure and multi-type component objects to be studied, and establish a single bridge graph structure model that considers the spatiotemporal correlation of the responses of different components;

[0012] Step 2: Design the spatiotemporal graph prediction task based on multi-type perception data of the service status of bridge structural components and construct the corresponding input and output data sets;

[0013] Step 3: Establish a spatiotemporal correlation diagnosis model for the service status of a bridge structure based on a graph network and a state space model. The graph selective state space module of the spatiotemporal correlation diagnosis model for the service status of a bridge structure includes a normalization layer, a dynamic filtering graph convolution layer, a splicing layer, a linear layer, a segmentation layer, a one-dimensional convolution layer, a SiLU activation layer, a graph state space selection mechanism, an element-by-element multiplication layer, and a residual connection layer. The dynamic filtering graph convolution layer learns the complex spatial dependency of the service status sequence data of multiple types of components of the bridge structure, and the graph state space selection mechanism learns the evolution and spatiotemporal correlation of the service status of multiple types of components of the bridge.

[0014] Step 4: Training and prediction of the spatiotemporal correlation diagnosis model for the service status of bridge structures; Based on the gradient descent error back propagation, the spatiotemporal correlation diagnosis model for the service status of bridge structures is trained, and the trained model is used to perform spatiotemporal correlation prediction and diagnosis on the service status of key components including quasi-static response and dynamic response.

[0015] Furthermore, the step 1 specifically includes:

[0016] Step 1: Collection and preprocessing of multi-source perception data on the service status of multi-type bridge components;

[0017] (1) Multi-source perception data collection of the service status of multiple types of bridge components: Collect historical service status data of key components with pseudo-static response and dynamic response;

[0018] (2) Preprocessing of multi-source perception data on the service status of multi-type bridge components: Clean the collected data, remove outliers and missing values, and normalize the multi-type data to ensure data consistency and accuracy;

[0019] Step 1 and 2: Establish a single bridge graph structural model that considers the temporal and spatial correlation of the responses of different components;

[0020] (1) Definition of structural nodes in the single-body bridge graph: Each key component structure on the single-body bridge is regarded as a node in the graph, and the node attributes are pseudo-static response and dynamic response;

[0021] (2) Definition of structural edges in the single bridge graph: Based on the spatial connection relationship between the multi-type components of the bridge structure, the edges between nodes are preliminarily defined, and the edges are undirected edges;

[0022] (3) Generation of monomer bridge graph structure: The monomer bridge graph structure is represented by G = (V; E), where V is the set of component nodes and E is the set of edges; the monomer bridge graph structure is generated according to the definitions of nodes and edges respectively; the adjacency matrix A of the monomer bridge graph structure is determined according to the definitions of nodes and edges.

[0023] Furthermore, the step 2 specifically includes:

[0024] Step 21: For the bridge structure service status diagnosis task, the historical monitoring data of pseudo-static response and dynamic response are used as input, and the pseudo-static response and dynamic response time history of multiple types of bridge components in the future are predicted as output;

[0025] Step 22: Embed the input dataset as Where N is the number of key components of a single bridge, T is the time step of the input data, and D is the feature dimension, including the pseudo-static response and dynamic response of the structure; the corresponding output data set embedding is expressed as Where T p is the predicted output data time step;

[0026] Step 2 and 3: For the spatiotemporal data of the service status of multi-type, multi-scale and multi-granular bridge structures and multi-type components, including the pseudo-static response and dynamic response of the structure, three different time granularities are used as input data, including recent data X r , Periodic data X c and trend data X q; Recent data is a historical time series directly adjacent to the forecast period. Periodic data consists of segments whose past days in the same time period as the forecast period are integer multiples of α, and trend data consists of segments whose past days in the same time period as the forecast period are integer multiples of β.

[0027] Furthermore, the establishment of the spatiotemporal correlation diagnosis model of the bridge structure service status specifically includes:

[0028] Step 31: Convert the three time-granularity data X of the service status of multiple types of bridge structure components into r , X c and X q The outputs of N spatiotemporal graph convolution blocks are weighted fused;

[0029] Step 32: Enter X r The outputs passed to the M graph-selective state-space modules are weighted fused again;

[0030] Step 33: Get the final output after the fully connected layer

[0031] Furthermore, the dynamic filtering graph convolution layer specifically refers to the dynamic changes of multi-hop connection relationships and graph structure information in the learning graph; its calculation method is to convert the recent data X of the service status of multiple types of components of the bridge structure into r After being transformed by layer normalization, it is passed to the dynamic filter graph convolution layer:

[0032]

[0033] Where A is the original adjacency matrix of the monomer bridge graph structure, D is the degree matrix, and A norm is the normalized adjacency matrix, represents the power series of the adjacency matrix, B is the basic filter used to dynamically adjust the adjacency matrix of the monomer bridge graph structure, T(B) represents the linear transformation of the basic filter B matrix, λ is the weight coefficient between the manually assigned original adjacency matrix and the transformed filter, is the dynamically adjusted adjacency matrix, are weights and biases respectively, and B are the parameters to be optimized. is the result of the i-th dynamic filter graph convolution layer, H cat is the splicing result; Each element in B is initialized from a mean of 0 to a range of Random values ​​are taken from uniform distribution; m is a hyperparameter, and N is the number of components contained in the single bridge structure.

[0034] Furthermore, the graph state space selection mechanism is specifically: after the dynamic filtering graph convolution layer is performed, it is used to learn the service state evolution and spatiotemporal correlation of multiple types of bridge components. The specific method is: the result H of the dynamic filtering graph convolution layer splicing cat The output H after the linear layer core_sec Core sequence H core and the minor sequence H sec , the two sequences have the same dimension, and the core sequence is passed to the graph state space selection mechanism GSSSM after one-dimensional convolution and SiLU activation, and the output H is obtained gsssm , the calculation formula is as follows:

[0035]

[0036] The graph state space selection mechanism GSSSM includes two parts: the state space selection mechanism parameter calculation update module and the graph selective scanning module. gsssm It is element-wise multiplied with the secondary sequence activated by SiLU, and then passes through the linear layer to obtain H out :

[0037] H out =W out (H gsssm ⊙SiLU(H sec ))+b out (3)

[0038] Where W out and b out Respectively represent the weight and bias parameters to be optimized; H out and layer normalized X r Residual connection is performed to obtain the output of the graph selective state space module of the spatiotemporal correlation diagnosis model of the bridge structure service status.

[0039] Furthermore, the state space selection mechanism parameter calculation update module of the graph state space selection mechanism GSSSM comprises the following specific steps:

[0040] (1) Calculate the input-independent parameters A and D, A log Each row of the matrix is ​​initialized to the logarithm of a continuous integer sequence from 1 to the number of columns, and all elements in D are initialized to 1. A is the state transfer matrix, and D is the direct gain;

[0041] (2) Use a linear layer to process the input H c ' ore , we get H c ″ ore ;

[0042] (3) Split H c ″ oreThe parameters Δ, B, and C related to the input of service condition monitoring data of multi-type components of bridge structures are obtained. The dimensions of B and C are the same, B and C are the input matrix and output matrix respectively, and Δ represents the conversion parameter from the continuous state space model to the discrete state space model.

[0043] (4) Use the linear layer and SoftPlus activation function to adjust Δ to obtain Δ′;

[0044] (5) Call the graph selective scanning module GraphSelectiveScan and pass in H c ' ore ,Δ′,A,B,C,D are updated and calculated to get the output H gsssm ;

[0045] The parameter calculation and update formula of the state space selection mechanism GSSSM is as follows:

[0046]

[0047] Furthermore, the graph selective scanning module GraphSelectiveScan of the graph state space selection mechanism GSSSM comprises the following specific steps:

[0048] (1) Obtaining the dynamically adjusted monomer bridge graph structure adjacency matrix from the dynamic filter graph convolution layer Initialize the filling matrix and convert the adjacency matrix Copy to the upper left corner of the filled matrix to get P pad_adj Filling matrix, P pad_adj Perform matrix multiplication matmul on the filled matrix and Δ′ to obtain the updated parameter Δ″;

[0049] (2) Use Δ″ to discretize and update A and B to obtain ΔA and ΔB, where einsum represents the Einstein summation convention;

[0050] (3) Initialize the state Z in the state space selection mechanism to 0, and perform state Z from 1 to T for time step i in the service status monitoring data of multi-type components of the bridge structure. i Update and output H i Calculation of

[0051] (4) Output H i Use stack operation to get H stack , and finally H c ' ore Add the weight coefficient and direct gain D to get H gsssm ;

[0052] The graph selective scanning calculation formula of the graph state space selection mechanism GSSSM is:

[0053]

[0054] The present invention also proposes an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model are implemented.

[0055] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of the spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] (1) By embedding data with perception at different time granularities (recent data, periodic data, and trend data), the model can capture the dynamic changes of service status data of multi-type, multi-scale, and multi-granular bridge structures and multi-type components at different time scales, thereby improving the versatility, accuracy, and robustness of the model.

[0058] (2) A dynamic filtering graph convolutional network module is used to improve the processing capability of spatiotemporal data on the service status of multiple types of components of bridge structures, including those with noise and uncertainty.

[0059] (3) By designing segmentation operations and residual connections, features can be separated and fused more effectively, capturing richer spatiotemporal information levels in the service status monitoring data of multi-type components of bridge structures, thereby improving the representation ability and prediction performance of the spatiotemporal correlation prediction model.

[0060] (4) A dynamic parameter adjustment mechanism and an efficient selectivity mechanism are designed in the spatiotemporal graph selective state space module. By dynamically adjusting the state space model parameters (including the state transfer matrix, the control input matrix, and the output matrix), important time series information can be automatically selected according to the input characteristics of the service status monitoring data of multiple types of bridge structure components, thereby realizing input-dependent dynamic feature selection and improving the flexibility and adaptability of the model.

[0061] (5) The model proposed in the present invention has linear time complexity, which significantly reduces the computational cost and improves the processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0063] Figure 1 It is a flow chart of the spatiotemporal correlation diagnosis method of the service status of bridge structures based on graph networks and state space models;

[0064] Figure 2 It is a schematic diagram of the structural model of a single bridge considering the temporal and spatial correlation of the responses of different components; (a) is a cable-stayed bridge, and (b) is a suspension bridge;

[0065] Figure 3 This is the architecture diagram of the spatiotemporal correlation diagnosis model for the service status of bridge structures based on graph networks and state space models;

[0066] Figure 4 It is a schematic diagram of the graph selective state space module architecture of the spatiotemporal correlation diagnosis model of the service status of the bridge structure;

[0067] Figure 5 It is a schematic diagram of the dynamic filtering graph convolutional layer architecture of the spatiotemporal correlation diagnosis model of the service status of bridge structures;

[0068] Figure 6 It is a schematic diagram of the parameter calculation and updating method of the state space selection mechanism of the spatiotemporal correlation diagnosis model of the service state of the bridge structure;

[0069] Figure 7 It is a schematic diagram of the graph selective scanning method of the spatiotemporal correlation diagnosis model of the service status of bridge structures. DETAILED DESCRIPTION

[0070] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] Combination Figure 1-Figure 7 The present invention proposes a spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model, and the method comprises the following steps:

[0072] Step 1: Determine the bridge structure and multi-type component objects to be studied, and establish a single bridge graph structure model that considers the spatiotemporal correlation of the responses of different components;

[0073] The step 1 specifically includes:

[0074] Step 1: Collection and preprocessing of multi-source perception data on the service status of multi-type bridge components;

[0075] (1) Multi-source sensing data collection of the service status of various types of bridge components: Collect historical service status data of key components including pseudo-static response (vertical displacement of main beam, strain of key cross-section of main beam, support reaction, cable force, etc.) and dynamic response (vertical, lateral, and longitudinal vibration acceleration of main beam, etc.);

[0076] (2) Preprocessing of multi-source perception data on the service status of multi-type bridge components: Clean the collected data, remove outliers and missing values, and normalize the multi-type data to ensure data consistency and accuracy;

[0077] Step 1 and 2: Establish a single bridge graph structural model that considers the temporal and spatial correlation of the responses of different components;

[0078] (1) Definition of structural nodes in the single-body bridge diagram: Each key structural component on the single-body bridge is regarded as a node in the diagram, and the node attributes are pseudo-static response (vertical displacement of the main beam, strain of the key section of the main beam, support reaction, cable force, etc.) and dynamic response (vertical, lateral, and longitudinal vibration acceleration of the main beam, etc.);

[0079] (2) Definition of structural edges in the single bridge graph: Based on the spatial connection relationship between the multi-type components of the bridge structure, the edges between nodes are preliminarily defined, and the edges are undirected edges;

[0080] (3) Generation of monomer bridge graph structure: The monomer bridge graph structure is represented by G = (V; E), where V is the set of component nodes and E is the set of edges; the monomer bridge graph structure is generated according to the definitions of nodes and edges respectively; the adjacency matrix A of the monomer bridge graph structure is determined according to the definitions of nodes and edges.

[0081] Taking a large bridge as an example, the graph structure model of a single bridge considering the temporal and spatial correlation of the responses of different components is as follows: Figure 2 shown.

[0082] The structure of a cable-stayed bridge is as follows Figure 2 As shown in (a), the number of bridge towers is 2, the whole bridge is equipped with 4×18 pairs of parallel steel cables, and the number of main beam sections is 36. Therefore, according to the above method, the whole bridge is modeled as 2 bridge tower nodes, 72 cable nodes, 36 main beam nodes and 2 foundation nodes.

[0083] A suspension bridge Figure 2As shown in (b), the symmetrically arranged half bridge has one tower, two pairs of main cables, 2×14 pairs of parallel steel wire slings on the main cables, and 14 main beam sections. Therefore, according to the above method, the symmetrically arranged half bridge is modeled as one tower node, two main cable nodes, 28 sling nodes, 14 main beam nodes, and one foundation node.

[0084] Step 2: Design the spatiotemporal graph prediction task based on multi-type perception data of the service status of bridge structural components and construct the corresponding input and output data sets;

[0085] The step 2 specifically includes:

[0086] Step 21: For the bridge structure service status diagnosis task, the historical monitoring data of pseudo-static response and dynamic response are used as input, and the pseudo-static response and dynamic response time history of multiple types of bridge components in the future are predicted as output; the degree of change between the predicted value and the true value is used as the basis for service status diagnosis. If the deviation between the predicted value and the true value is small, it means that the service status has not changed and is in a healthy state; otherwise, it means that the corresponding bridge component has deteriorated;

[0087] Step 22: Embed the input dataset as Where N is the number of key components of a single bridge, T is the time step of the input data, and D is the feature dimension, including the pseudo-static response and dynamic response of the structure; the corresponding output data set embedding is expressed as Where T p is the predicted output data time step;

[0088] Step 2 and 3: For the spatiotemporal data of the service status of multi-type, multi-scale and multi-granular bridge structures and multi-type components, including the pseudo-static response and dynamic response of the structure, three different time granularities are used as input data, including recent data X r , Periodic data X c and trend data X q ; Recent data is a historical time series directly adjacent to the forecast period, periodic data is composed of segments whose past days in the same time period as the forecast period are integer multiples of α, and trend data is composed of segments whose past days in the same time period as the forecast period are integer multiples of β. Among them, the period window is expressed by variables α and β, respectively, and the possible values ​​are: 1 for 1 day, 7 for 1 week, 30 for 1 month, 365 for 1 year, etc. The length of the three types of input data, recent data, periodic data, and trend data (i.e., the number of time steps T contained in each data segment) is set according to the actual forecast needs to adapt to different forecast targets and historical data characteristics.

[0089] Step 3: Establish a spatiotemporal correlation diagnosis model for the service status of a bridge structure based on a graph network and a state space model. The graph selective state space module of the spatiotemporal correlation diagnosis model for the service status of a bridge structure includes a normalization layer, a dynamic filtering graph convolution layer, a splicing layer, a linear layer, a segmentation layer, a one-dimensional convolution layer, a SiLU activation layer, a graph state space selection mechanism, an element-by-element multiplication layer, and a residual connection layer. The dynamic filtering graph convolution layer learns the complex spatial dependency of the service status sequence data of multiple types of components of the bridge structure, and the graph state space selection mechanism learns the evolution and spatiotemporal correlation of the service status of multiple types of components of the bridge.

[0090] The establishment of the spatiotemporal correlation diagnosis model of the bridge structure service status specifically includes:

[0091] Step 31: Convert the three time-granularity data X of the service status of multiple types of bridge structure components into r , X c and X q The outputs of N spatiotemporal graph convolution blocks are weighted fused;

[0092] Step 32: Enter X r The outputs passed to the M graph-selective state-space modules are weighted fused again;

[0093] Step 33: Get the final output after the fully connected layer

[0094] The architecture of the spatiotemporal correlation diagnosis model for bridge structure service status based on graph network and state space model is as follows: Figure 3 As shown in the figure, N and M are the number of spatiotemporal graph convolution modules and graph selective state space modules, which are preset hyperparameters. The output of the previous module is passed to the next module as input and then passed forward in sequence.

[0095] The graph-selective state space module architecture of the spatiotemporal correlation diagnosis model of the bridge structure service status in step 3.2 is as follows: Figure 4 As shown, it mainly includes layer normalization layer, dynamic filtering graph convolution layer for learning complex spatial dependencies of service status sequence data of multi-type components of bridge structure, splicing layer, linear layer, segmentation layer, one-dimensional convolution layer, SiLU activation layer, graph state space selection mechanism for learning service status evolution and spatiotemporal correlation of multi-type components of bridge, element-by-element multiplication layer, residual connection layer, etc.

[0096] The dynamic filtering graph convolution layer specifically refers to the dynamic changes of multi-hop connection relationships and graph structure information in the learning graph; its calculation method is to convert the recent data X of the service status of multiple types of components of the bridge structure into r After being transformed by layer normalization, it is passed to the dynamic filter graph convolution layer:

[0097]

[0098] Where A is the original adjacency matrix of the monomer bridge graph structure, D is the degree matrix, and A norm is the normalized adjacency matrix, represents the power series of the adjacency matrix, B is the basic filter used to dynamically adjust the adjacency matrix of the monomer bridge graph structure, T(B) represents the linear transformation of the basic filter B matrix, λ is the weight coefficient between the manually assigned original adjacency matrix and the transformed filter, is the dynamically adjusted adjacency matrix, are weights and biases respectively, and B are the parameters to be optimized. is the result of the i-th dynamic filter graph convolution layer, H cat is the splicing result; Each element in B is initialized from a mean of 0 to a range of Random values ​​are taken from uniform distribution; m is a hyperparameter, and N is the number of components contained in the single bridge structure.

[0099] The graph state space selection mechanism is specifically: after the dynamic filtering graph convolution layer is performed, it is used to learn the evolution of the service state and the spatiotemporal correlation of multiple types of bridge components. The specific method is: the result H of the dynamic filtering graph convolution layer splicing cat The output H after the linear layer core_sec Core sequence H core and the minor sequence H sec , the two sequences have the same dimension, and the core sequence is passed to the graph state space selection mechanism GSSSM after one-dimensional convolution and SiLU activation, and the output H is obtained gsssm , the calculation formula is as follows:

[0100]

[0101] The graph state space selection mechanism GSSSM includes two parts: the state space selection mechanism parameter calculation update module and the graph selective scanning module. gsssm It is element-wise multiplied with the secondary sequence activated by SiLU, and then passes through the linear layer to obtain H out :

[0102] H out =W out (H gsssm ⊙SiLU(H sec ))+b out (3)

[0103] Where W out and b out Respectively represent the weight and bias parameters to be optimized; H outand layer normalized X r Residual connection is performed to obtain the output of the graph selective state space module of the spatiotemporal correlation diagnosis model of the bridge structure service status.

[0104] The state space selection mechanism parameter calculation and update module of the graph state space selection mechanism GSSSM comprises the following specific steps:

[0105] (1) Calculate the input-independent parameters A and D, A log Each row of the matrix is ​​initialized to the logarithm of a continuous integer sequence from 1 to the number of columns, and all elements in D are initialized to 1. A is the state transfer matrix, and D is the direct gain;

[0106] (2) Use a linear layer to process the input H c ' ore , we get H c ″ ore ;

[0107] (3) Split H c ″ ore The parameters Δ, B, and C related to the input of service condition monitoring data of multi-type components of bridge structures are obtained. The dimensions of B and C are the same, B and C are the input matrix and output matrix respectively, and Δ represents the conversion parameter from the continuous state space model to the discrete state space model.

[0108] (4) Use the linear layer and SoftPlus activation function to adjust Δ to obtain Δ′;

[0109] (5) Call the graph selective scanning module GraphSelectiveScan and pass in H c ' ore ,Δ′,A,B,C,D are updated and calculated to get the output H gsssm ;

[0110] The parameter calculation and update formula of the state space selection mechanism GSSSM is as follows:

[0111]

[0112] The graph selective scanning module GraphSelectiveScan of the graph state space selection mechanism GSSSM has the following specific steps:

[0113] (1) Obtaining the dynamically adjusted monomer bridge graph structure adjacency matrix from the dynamic filter graph convolution layer Initialize the fill matrix (filled with 1), and convert the adjacency matrix Copy to the upper left corner of the filled matrix to get P pad_adj Filling matrix, P pad_adjPerform matrix multiplication matmul on the filled matrix and Δ′ to obtain the updated parameter Δ″;

[0114] (2) Use Δ″ to discretize and update A and B to obtain ΔA and ΔB, where einsum represents the Einstein summation convention;

[0115] (3) Initialize the state Z in the state space selection mechanism to 0, and perform state Z from 1 to T for time step i in the service status monitoring data of multi-type components of the bridge structure. i Update and output H i Calculation of

[0116] (4) Output H i Use stack operation to get H stack , and finally H c ' ore Add the weight coefficient and direct gain D to get H gsssm ;

[0117] The graph selective scanning calculation formula of the graph state space selection mechanism GSSSM is:

[0118]

[0119] Step 4: Training and prediction of the spatiotemporal correlation diagnosis model for the service status of bridge structures: Based on the gradient descent error back propagation, the spatiotemporal correlation diagnosis model for the service status of bridge structures is trained, and the trained model is used to predict and diagnose the service status of key components including pseudo-static response and dynamic response in spatiotemporal correlation. Thus, a spatiotemporal correlation diagnosis system for the service status of bridge structures is established.

[0120] The step 4 specifically includes:

[0121] Step 41: Based on the gradient descent error back propagation training graph selective state space model, the optimization algorithm includes optimizers such as AddmW and SGD, and the training objective function adopts the relative mean square error (RMSE) with regularization constraints, so that the generated pseudo-static response (vertical displacement of the main beam, strain of the key section of the main beam, support reaction, cable force, etc.) and dynamic response (vertical, lateral, longitudinal vibration acceleration of the main beam, etc.) and other key components The predicted values ​​of the service state are as close as possible to the actual values ​​of the actual monitoring data:

[0122]

[0123] In the formula, is the time history data of the service state (including pseudo-static response and dynamic response) of the key bridge components predicted by the model in the future k steps, X (t+1):(t+k)is the real time-history data of the service status (including quasi-static response and dynamic response) of key bridge components in the future k steps; N is the number of components contained in the bridge structure, k is the time step, and D is the characteristic dimension, including quasi-static response (vertical displacement of the main beam, strain of the key section of the main beam, support reaction, cable force, etc.) and dynamic response (vertical, lateral, longitudinal vibration acceleration of the main beam, etc.); B represents the batch size, λ is the regularization coefficient, and R(Θ) is the regularization term.

[0124] Step 42: For the task of diagnosing the service status of multiple types of components of a single bridge, the trained model is used to perform spatiotemporal correlation prediction of the service status of key components including quasi-static response (vertical displacement of the main beam, strain of key sections of the main beam, support reaction, cable force, etc.) and dynamic response (vertical, lateral, longitudinal vibration acceleration of the main beam, etc.).

[0125] Step 43: Use the degree of change between the predicted value and the true value as the basis for service status diagnosis. If the deviation between the predicted value and the true value is small, it means that the service status has not changed and is in a healthy state; otherwise, it means that the corresponding bridge component has deteriorated, thereby establishing a spatiotemporal correlation diagnosis system for the service status of the bridge structure.

[0126] The present invention proposes a spatiotemporal correlation diagnosis method for the service status of bridge structures based on graph networks and state space models. The method aims to improve the diagnostic capability, accuracy, efficiency and versatility of the service status of bridge structures by capturing the complex spatiotemporal correlation and spatiotemporal evolution laws inherent in the service status monitoring data of different components of a single bridge. The innovative points are:

[0127] (1) Using a spatiotemporal graph structure, we construct a spatiotemporal graph fusion learning module, combining graph neural networks with deep learning of time series data, and considering the data characteristics of the service status of bridge structural components in both spatial and temporal dimensions to more accurately model and predict the dynamic spatiotemporal evolution of complex component systems of single bridges;

[0128] (2) Adopting the state space model, constructing a structured state space sequence, and introducing a selective mechanism to dynamically adjust the model behavior and achieve optimal utilization of computing resources;

[0129] (3) By decomposing and fusing the spatiotemporal characteristics of the service status monitoring data of multi-type components of bridges of different complexity, the method can better adapt to the processing requirements of the spatiotemporal data of the service status of multi-type, multi-scale, and multi-granular bridge structural components, thereby improving the versatility and prediction efficiency of the method.

[0130] The present invention also proposes an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model are implemented.

[0131] The present invention also proposes a computer-readable storage medium for storing computer instructions, which, when executed by a processor, implement the steps of the spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model.

[0132] The memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus RAM (DRRAM). It should be noted that the memory of the method described in the present invention is intended to include, but is not limited to, these and any other suitable types of memory.

[0133] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disc (SSD)).

[0134] In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in a processor or an instruction in the form of software. The steps of the method disclosed in conjunction with the embodiment of the present application can be directly embodied as a hardware processor for execution, or a combination of hardware and software modules in a processor for execution. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in conjunction with its hardware. To avoid repetition, it is not described in detail here.

[0135] It should be noted that the processor in the embodiment of the present application can be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor can be combined and performed. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0136] The above is a detailed introduction to the spatiotemporal correlation diagnosis method for the service status of a bridge structure based on a graph network and a state space model proposed in the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A spatiotemporal correlation diagnosis method for bridge structure service status based on graph network and state space model, characterized in that: The method comprises the following steps: Step 1: Determine the bridge structure and multi-type component objects to be studied, and establish a single bridge graph structure model that considers the spatiotemporal correlation of the responses of different components; Step 2: Design the spatiotemporal graph prediction task based on multi-type perception data of the service status of bridge structural components and construct the corresponding input and output data sets; Step 3: Establish a spatiotemporal correlation diagnosis model for the service status of a bridge structure based on a graph network and a state space model. The graph selective state space module of the spatiotemporal correlation diagnosis model for the service status of a bridge structure includes a normalization layer, a dynamic filtering graph convolution layer, a splicing layer, a linear layer, a segmentation layer, a one-dimensional convolution layer, a SiLU activation layer, a graph state space selection mechanism, an element-by-element multiplication layer, and a residual connection layer. The dynamic filtering graph convolution layer learns the complex spatial dependency of the service status sequence data of multiple types of components of the bridge structure, and the graph state space selection mechanism learns the evolution and spatiotemporal correlation of the service status of multiple types of components of the bridge. Step 4: Training and prediction of the spatiotemporal correlation diagnosis model for the service status of bridge structures; Based on the gradient descent error back propagation, the spatiotemporal correlation diagnosis model for the service status of bridge structures is trained, and the trained model is used to perform spatiotemporal correlation prediction and diagnosis on the service status of key components including quasi-static response and dynamic response.

2. The method according to claim 1, characterized in that: The step 1 specifically includes: Step 1: Collection and preprocessing of multi-source perception data on the service status of multi-type bridge components; (1) Multi-source perception data collection of the service status of multiple types of bridge components: Collect historical service status data of key components with pseudo-static response and dynamic response; (2) Preprocessing of multi-source perception data on the service status of multi-type bridge components: Clean the collected data, remove outliers and missing values, and normalize the multi-type data to ensure data consistency and accuracy; Step 1 and 2: Establish a single bridge graph structural model that considers the temporal and spatial correlation of the responses of different components; (1) Definition of structural nodes in the single-body bridge graph: Each key component structure on the single-body bridge is regarded as a node in the graph, and the node attributes are pseudo-static response and dynamic response; (2) Definition of structural edges in the single bridge graph: Based on the spatial connection relationship between the multi-type components of the bridge structure, the edges between nodes are preliminarily defined, and the edges are undirected edges; (3) Generation of monomer bridge graph structure: The monomer bridge graph structure is represented by G = (V; E), where V is the set of component nodes and E is the set of edges; the monomer bridge graph structure is generated according to the definitions of nodes and edges respectively; according to the definitions of nodes and edges, the adjacency matrix A of the monomer bridge graph structure is determined.

3. The method according to claim 2, characterized in that The step 2 specifically includes: Step 21: For the bridge structure service status diagnosis task, the historical monitoring data of pseudo-static response and dynamic response are used as input, and the pseudo-static response and dynamic response time history of multiple types of bridge components in the future are predicted as output; Step 22: Embed the input dataset as Where N is the number of key components of a single bridge, T is the time step of the input data, and D is the feature dimension, including the pseudo-static response and dynamic response of the structure; the corresponding output data set embedding is expressed as Where T p is the predicted output data time step; Step 2 and 3: For the spatiotemporal data of the service status of multi-type, multi-scale and multi-granular bridge structures and multi-type components, including the pseudo-static response and dynamic response of the structure, three different time granularities are used as input data, including recent data X r , Periodic data X c and trend data X q ; Recent data is a historical time series directly adjacent to the forecast period. Periodic data consists of segments whose past days in the same time period as the forecast period are integer multiples of α, and trend data consists of segments whose past days in the same time period as the forecast period are integer multiples of β.

4. The method according to claim 3, characterized in that The establishment of the spatiotemporal correlation diagnosis model of the bridge structure service status specifically includes: Step 31: Convert the three time-granularity data X of the service status of multiple types of bridge structure components into r , X c and X q The outputs of N spatiotemporal graph convolution blocks are weighted fused; Step 32: Enter X r The outputs passed to the M graph-selective state-space modules are weighted fused again; Step 33: Get the final output after the fully connected layer 5. The method according to claim 4, characterized in that The dynamic filtering graph convolution layer specifically refers to the dynamic changes of multi-hop connection relationships and graph structure information in the learning graph; its calculation method is to convert the recent data X of the service status of multiple types of components of the bridge structure into r After being transformed by layer normalization, it is passed to the dynamic filter graph convolution layer: Where A is the original adjacency matrix of the monomer bridge graph structure, D is the degree matrix, and A norm is the normalized adjacency matrix, represents the power series of the adjacency matrix, B is the basic filter used to dynamically adjust the adjacency matrix of the monomer bridge graph structure, T(B) represents the linear transformation of the basic filter B matrix, λ is the weight coefficient between the manually assigned original adjacency matrix and the transformed filter, is the dynamically adjusted adjacency matrix, are weights and biases respectively, and B are the parameters to be optimized. is the result of the i-th dynamic filter graph convolution layer, H cat is the splicing result; Each element in B is initialized from a mean of 0 to a range of Random values ​​are taken from uniform distribution; m is a hyperparameter, and N is the number of components contained in the single bridge structure.

6. The method according to claim 5, characterized in that The graph state space selection mechanism is specifically: after the dynamic filtering graph convolution layer is performed, it is used to learn the evolution of the service state and the spatiotemporal correlation of multiple types of bridge components. The specific method is: the result H of the dynamic filtering graph convolution layer splicing cat The output H after the linear layer core_sec Core sequence H core and the minor sequence H sec , the two sequences have the same dimension, and the core sequence is passed to the graph state space selection mechanism GSSSM after one-dimensional convolution and SiLU activation, and the output H is obtained gsssm , the calculation formula is as follows: The graph state space selection mechanism GSSSM includes two parts: the state space selection mechanism parameter calculation update module and the graph selective scanning module. gsssm It is element-wise multiplied with the secondary sequence activated by SiLU, and then passes through the linear layer to obtain H out : H out =W out (H gsssm ⊙SiLU(H sec ))+b out (3) Where W out and b out Respectively represent the weight and bias parameters to be optimized; H out and layer normalized X r Residual connection is performed to obtain the output of the graph selective state space module of the spatiotemporal correlation diagnosis model of the bridge structure service status.

7. The method according to claim 6, characterized in that The state space selection mechanism parameter calculation and update module of the graph state space selection mechanism GSSSM comprises the following specific steps: (1) Calculate the input-independent parameters A and D, A log Each row of the matrix is ​​initialized to the logarithm of a continuous integer sequence from 1 to the number of columns, and all elements in D are initialized to 1. A is the state transfer matrix, and D is the direct gain; (2) Use a linear layer to process the input H c ' ore , we get H c ″ ore ; (3) Split H c ″ ore The parameters Δ, B, and C related to the input of service condition monitoring data of multi-type components of bridge structures are obtained. The dimensions of B and C are the same, B and C are the input matrix and output matrix respectively, and Δ represents the conversion parameter from the continuous state space model to the discrete state space model. (4) Use the linear layer and SoftPlus activation function to adjust Δ to obtain Δ′; (5) Call the graph selective scanning module GraphSelectiveScan and pass in H c ' ore ,Δ′,A,B,C,D are updated and calculated to get the output H gsssm ; The parameter calculation and update formula of the state space selection mechanism GSSSM is as follows:

8. The method according to claim 7, characterized in that The graph selective scanning module GraphSelectiveScan of the graph state space selection mechanism GSSSM has the following specific steps: (1) Obtaining the dynamically adjusted monomer bridge graph structure adjacency matrix from the dynamic filter graph convolution layer Initialize the filling matrix and convert the adjacency matrix Copy to the upper left corner of the filled matrix to get P pad_adj Filling matrix, P pad_adj Perform matrix multiplication matmul on the filled matrix and Δ′ to obtain the updated parameter Δ″; (2) Use Δ″ to discretize and update A and B to obtain ΔA and ΔB, where einsum represents the Einstein summation convention; (3) Initialize the state Z in the state space selection mechanism to 0, and perform state Z from 1 to T for time step i in the service status monitoring data of multi-type components of the bridge structure. i Update and output H i Calculation of (4) Output H i Use stack operation to get H stack , and finally H c ' ore Add the weight coefficient and direct gain D to get H gsssm ; The graph selective scanning calculation formula of the graph state space selection mechanism GSSSM is:

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium for storing computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images