Safety assessment method for cable-stayed bridges based on Bayesian decision network

Abstract

The present invention aims to provide a cable-stayed bridge safety assessment method incorporating a Bayesian decision network for operational assessment and prediction. This method utilizes the physical and material properties of cable-stayed bridge components to construct a component-level degradation model. A force transmission network is established based on the load transmission path (load → girder → cable → pylon) of the cable-stayed bridge. The CPT is obtained by learning from a sample library of monitoring data (or numerical simulations). The UT can be calculated by referring to the relevant table data in the "Highway Bridge Technical Condition Assessment Standard" (JTG / T H21-2011). During decision assessment, monitoring data is combined to determine the path by comparing the utility values ​​of the paths' BDNs. A global cable-stayed bridge degradation model is then constructed based on the utility values ​​and decisions output by each path, enabling safety assessment and prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge health monitoring, and in particular to a cable-stayed bridge safety assessment method combined with artificial intelligence decision-making, and specifically to a cable-stayed bridge safety assessment method combined with a Bayesian decision network. Background Art

[0002] Bridge technical condition assessment is a complex and systematic task. It combines data from bridge inspections and monitoring to assess the bridge's structural load-bearing capacity, degradation level, and other aspects of its service life. This assessment serves as a crucial basis for bridge operations, management, maintenance, reinforcement, and renovation. Regular or real-time safety assessments are crucial for ensuring bridge operational safety. Furthermore, developing intelligent analysis and processing methods for massive, redundant, and complex inspection and monitoring data, and subsequently establishing an intelligent safety early warning system for bridge structures, has significant engineering significance and will be a path to achieving smart bridges and intelligent operations and maintenance.

[0003] Current bridge safety assessment methods are broadly categorized into appearance survey and evaluation methods, the Analytic Hierarchy Process (AHP), load testing, reliability theory, and artificial intelligence methods such as neural networks, each with its own advantages, disadvantages, and applicability. The technical condition of bridges is assessed through conventional inspections and apparent damage surveys, with expert scoring combined with AHP methods. However, these methods suffer from poor timeliness and subjectivity. On-site load testing provides an intuitive understanding of the bridge's bearing capacity, but static load testing is costly and often requires blocking traffic on the bridge. Safety assessment methods based on reliability theory require identifying the primary failure mode of the bridge, and nonlinear functional functions are often difficult to define, significantly limiting their application in the field. The "black box" operation of neural networks results in a lack of mechanical interpretability of the variable inference process within the bridge structure, and the mapping relationship between input and output parameters lacks physical meaning.

[0004] Therefore, in order to overcome the shortcomings of traditional civil structure safety assessment methods such as poor timeliness, strong subjectivity, and high implementation costs, and at the same time solve the problems of artificial neural network assessment methods lacking interpretability and physical meaning, the present invention proposes to use Bayesian Decision Network (BDN) for analysis, which can be regarded as an extension of the Bayesian network. Decision nodes and utility nodes are added to the Bayesian network and expressed as a finite, directed acyclic graph. BDN constructs a visual decision-making system with causal logic reasoning capabilities in a graphical mode, which can directly and clearly understand the specific influencing factors of the decision-making process and the proportion of each influencing factor. Combining probabilistic reasoning with utility at the same time is a method of intuitively illustrating and visualizing knowledge, which has been applied in planning problem solving, search, diagnosis and decision-making. Summary of the Invention

[0005] The technical solution specifically adopted by the present invention to solve the technical problem is:

[0006] A cable-stayed bridge safety assessment method combined with a Bayesian decision network includes the following steps.

[0007] Step S1: Based on the external load conditions, component geometry and material properties of the cable-stayed bridge, a component-level degradation model based on the degradation mechanism is established for each component level;

[0008] Step S2: Establishing a Bayesian decision network for a cable-stayed bridge, including: defining network node variables, establishing a network topology including each force transmission path of the cable-stayed bridge, calculating the conditional probability table CPT between each uncertainty node and the utility table UT of the utility node, and assembling the established network topology and the calculated CPT and UT;

[0009] Step S3: Safety assessment of cable-stayed bridge: Based on the utility value calculated by UT, a Bayesian decision network is used to perform decision evaluation on each force transmission path: first, the local path of the cable-stayed bridge is evaluated, and then the utility value of each output of the local path is weighted averaged to complete the safety assessment of the cable-stayed bridge system.

[0010] Furthermore, after step S3, the overall utility value output under the component-level degradation model is fitted into a system-level degradation model.

[0011] Furthermore, step S2 specifically includes the following steps:

[0012] Step S2-1: Define the cable-stayed bridge BDN node variables:

[0013] The load node is defined as an uncertain node. Considering that the actual cable-stayed bridge is often subjected to the combined effects of constant and live loads, the live load is the main factor causing the stress changes of each component during the operation of the cable-stayed bridge. It is set as F i (i=1,2,...);

[0014] Define the cable node as an uncertain node, one cable as a cable node, and the maximum cable stress as the variable of the cable node, set as C i (i=1,2,...);

[0015] The main beam node is defined as an uncertain node. The anchor point between the cable and the main beam is used as the boundary to divide the main beam into different beam segments. The maximum deflection of the main beam segment is the variable of the main beam node, set as B i (i=1,2,...);

[0016] The tower node is defined as an uncertain node. The anchor point between the cable and the tower is used as the boundary to divide the tower into different tower segments. The maximum stress value of the tower segment is the variable, set as T i (i=1,2,...);

[0017] Define path nodes as decision nodes. Establish a force transmission network based on the force transmission path of the cable-stayed bridge: "load → main beam → cable → bridge tower". The nodes contain two states: "failure" and "safe". There are paths 1, 2, ...;

[0018] Define utility nodes as the basis for decision nodes to make decisions: Each utility node has a corresponding UT, which contains the utility values ​​of each decision node, set as U i (i=1,2,...);

[0019] Step S2-2: Establish the Bayesian decision network topology of each force transmission path of the cable-stayed bridge. Use conditional arcs to connect uncertainty nodes in sequence. Then add function arcs from cable nodes and decision nodes to utility nodes. Finally, add information arcs from tower nodes to decision nodes.

[0020] Step S2-3: Calculate the CPT of each uncertainty node and the UT of the utility node;

[0021] Step S2-4: Assemble the established network topology together with CPT and UT to form the Bayesian decision network of the cable-stayed bridge.

[0022] Furthermore, in step S2-3, CPT is obtained by learning from monitoring data or numerical analysis samples; UT is calculated with reference to the relevant table data in the "Highway Bridge Technical Condition Assessment Standard" (JTG / T H21-2011); wherein, CPT is a function of the bridge structure-level degradation model, and the utility node in UT is a specific numerical value that does not change with the degradation of bridge performance.

[0023] And, a cable-stayed bridge safety assessment system combined with a Bayesian decision network, based on a computer system, comprising:

[0024] The degradation model generation module is used to establish a component-level degradation model based on the degradation mechanism of each component according to the external load conditions, component geometry and material properties of the cable-stayed bridge;

[0025] The Bayesian decision network module for cable-stayed bridges is used to establish a Bayesian decision network for cable-stayed bridges, including: defining network node variables, establishing a network topology that includes the force transmission paths of the cable-stayed bridge, calculating the conditional probability table (CPT) between each uncertainty node and the utility table (UT) of the utility node, and assembling the established network topology and the calculated CPT and UT;

[0026] In addition, the cable-stayed bridge safety assessment module is used to make decision evaluations on each force transmission path through a Bayesian decision network based on the utility value calculated by UT: first, the local path of the cable-stayed bridge is evaluated, and then the utility value of each output of the local path is weighted averaged to complete the safety assessment of the cable-stayed bridge system.

[0027] Furthermore, the degradation model generation module is used to fit the overall utility value output under the component-level degradation model into a system-level degradation model.

[0028] Compared with the existing technology, the present invention and its preferred embodiment propose a cable-stayed bridge safety assessment method combined with a Bayesian decision network. This method can combine basic bridge information and inspection and monitoring data to understand the performance degradation of the bridge, infer the current status of the cable-stayed bridge based on the Bayesian decision network, and predict the time for maintenance. The advantages are: (1) The use of the Bayesian decision network avoids the shortcomings of traditional safety assessment methods such as poor timeliness, strong subjectivity, and high implementation costs; (2) The proposed method can not only complete the assessment of the local path of the cable-stayed bridge, but also achieve the assessment of the system in combination with the local assessment; (3) The future mechanical properties of the cable-stayed bridge can be predicted, thereby planning the maintenance time of the bridge and reducing the maintenance cost throughout its life cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0030] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention;

[0031] Figure 2 Schematic diagram of the definition of the main beam nodes and cable nodes of a cable-stayed bridge according to an embodiment of the present invention;

[0032] Figure 3 Schematic diagram of the definition of a cable-stayed bridge tower node according to an embodiment of the present invention;

[0033] Figure 4 1 is a schematic diagram of a cable-stayed bridge decision network topology definition according to an embodiment of the present invention.

[0034] Figure 5 Schematic diagram of a local decision network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] To make the features and advantages of this patent more clearly understood, the following embodiments are specifically described in detail as follows:

[0036] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in this specification have the same meaning as commonly understood by those skilled in the art to which this application belongs.

[0037] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or combinations thereof.

[0038] The present invention proposes a cable-stayed bridge safety assessment method combined with a Bayesian decision network, which can combine basic bridge information and inspection and monitoring data to understand the performance degradation of the bridge, infer the current status of the cable-stayed bridge based on the Bayesian decision method, and predict the time for maintenance. Figure 1 The technical solution shown includes the following steps.

[0039] Step S1: Based on the external load conditions, component geometry, and material properties of the cable-stayed bridge, a component-level degradation model based on the degradation mechanism is established.

[0040] Step S2: Establishing the cable-stayed bridge BDN, including defining network node variables, establishing a network topology containing the force transmission paths of the cable-stayed bridge, calculating the conditional probability table (CPT) between each uncertainty node and the utility table (UT) of the utility node, and assembling the established network topology and the calculated CPT and UT.

[0041] Step S2-1: Define the BDN node variables of the cable-stayed bridge.

[0042] The load node is defined as an uncertainty node, which is represented by an elliptical box. Considering that the actual cable-stayed bridge is often subjected to the combined effects of constant and live loads, the live load is the main factor causing the stress changes of each component during the operation of the cable-stayed bridge, and is set as F i(i=1,2,...);

[0043] Define the cable node as an uncertainty node, represented by an elliptical box. A cable is a cable node, and the maximum cable stress is the variable of the cable node, set as C i (i=1,2,...);

[0044] Define the main beam node as an uncertain node, represented by an elliptical box. Use the anchor point between the cable and the main beam as the boundary to divide the main beam into different beam segments. The maximum deflection of the main beam segment is the variable of the main beam node, set as B i (i=1,2,...);

[0045] The tower node is defined as an uncertain node, represented by an elliptical box. The tower is divided into different tower sections with the anchor point between the cable and the tower as the boundary. The maximum stress value of the tower section is the variable, set as T i (i=1,2,...);

[0046] Define path nodes as decision nodes, represented by rectangular boxes. Establish a force transmission network based on the force transmission path of a cable-stayed bridge: "load → girder → cable → tower." Nodes can have two states: "failed" and "safe." Path 1, Path 2, ..., are defined.

[0047] Define the utility node, which is represented by a diamond box and is the basis for the decision node to make decisions. Each utility node has a corresponding UT, which contains the utility values ​​of each decision node, set as U i (i=1,2,...);

[0048] Step S2-2: Establish the BDN topology of each force transmission path of the cable-stayed bridge. Use conditional arcs to connect the uncertainty nodes in sequence. Then add function arcs from the cable nodes and decision nodes to the utility nodes. Finally, add information arcs from the tower nodes to the decision nodes.

[0049] Step S2-3: Calculate the CPT and UT of each uncertainty node. The CPT can be obtained from monitoring data or numerical analysis samples, while the UT can be calculated using the relevant table data in the "Highway Bridge Technical Condition Assessment Standard" (JTG / TH21-2011). Here, the CPT is also a function of the bridge structure-level degradation model, while the utility node is a specific value that does not change with bridge performance degradation.

[0050] Step S2-4: Assemble the established network topology together with the CPT and UT to form a cable-stayed bridge BDN.

[0051] Step S3: Cable-stayed bridge safety assessment. Based on the utility values ​​calculated by the UT, the BDN performs a decision-making assessment of each force transmission path. First, the local paths of the cable-stayed bridge (e.g., main girder → cable → pylon) are assessed. The utility values ​​of each local path are then weighted averaged to complete the safety assessment of the cable-stayed bridge system.

[0052] The overall utility value output under the component-level degradation model can be further fitted into a system-level degradation model. The generated system-level degradation model can predict the future performance of the bridge.

[0053] The present invention is further described below with reference to a specific example.

[0054] like Figure 1 As shown, this embodiment combines a Bayesian decision network with a cable-stayed bridge safety assessment method for evaluating and predicting cable-stayed bridges during their operation. It utilizes the physical and material properties of cable-stayed bridge components to construct a cable-stayed bridge component-level degradation model, and establishes a force transmission network based on the cable-stayed bridge's force transmission path (load → main beam → cable → bridge tower). The CPT is obtained by learning from a sample library composed of monitoring data (or numerical simulation). The UT can be calculated by referring to the relevant table data in the "Highway Bridge Technical Condition Assessment Standard" (JTG / TH21-2011). During decision evaluation, monitoring data is combined to make path decisions by comparing the utility values ​​of the path BDNs. Then, the utility values ​​and decisions output by each path are used to establish an overall cable-stayed bridge degradation model to achieve safety assessment and prediction.

[0055] Figure 1 The evaluation method shown includes the following steps:

[0056] Step S1: Based on the external load conditions, component geometry, and material properties of the cable-stayed bridge, a component-level degradation model based on the degradation mechanism is established.

[0057] Step S2: Establishing the cable-stayed bridge BDN, including defining network node variables, establishing a network topology containing the force transmission paths of the cable-stayed bridge, calculating the CPT between each uncertainty node and the UT of the utility node, and assembling the established network topology, CPT, and UT;

[0058] Step S2-1: Define the BDN node variables of the cable-stayed bridge.

[0059] As Figure 2 As an example, the cable-stayed bridge shown in the figure is combined with Figure 3 , each node variable in step S2-1 is defined as:

[0060] The load node is defined as an uncertainty node, which is represented by an elliptical box. Considering that the actual cable-stayed bridge is often subjected to the combined effects of constant and live loads, the live load is the main factor causing the stress changes of each component during the operation of the cable-stayed bridge, and is set as Fi (i=1,2,...);

[0061] Define the cable node as an uncertainty node, which is represented by an elliptical box. Figure 2 As shown, a cable is a cable node. Consider the maximum cable stress as the variable of the cable node, set as C i (i=1,2,...);

[0062] Define the main beam node as an uncertain node, which is represented by an elliptical box. Figure 2 As shown in the figure, the main beam is divided into different beam segments with the anchor point between the cable and the main beam as the boundary. The maximum deflection of the main beam segment is the variable of the main beam node, which is set as B i (i=1,2,...);

[0063] The tower node is defined as an uncertainty node and is represented by an elliptical frame. Figure 3 As shown in the figure, the bridge tower is divided into different tower sections with the anchor point between the cable and the tower as the boundary. The maximum stress value of the tower section is a variable, set as T i (i=1,2,...);

[0064] Define path nodes as decision nodes, represented by rectangular boxes. Establish a force transmission network based on the force transmission path of a cable-stayed bridge: "load → girder → cable → tower." Nodes can have two states: "failed" and "safe." Path 1, Path 2, ..., are defined.

[0065] Define the utility node, which is represented by a diamond box and is the basis for the decision node to make decisions. Each utility node has UT, which contains the utility values ​​of each decision node, set as U i (i=1,2,...);

[0066] Step S2-2: Establish the BDN topology of each force transmission path of the cable-stayed bridge. Use conditional arcs to connect uncertainty nodes in sequence, then add function arcs from cable nodes and decision nodes to utility nodes, and finally use information arcs to connect tower nodes and decision nodes.

[0067] like Figure 4 、 Figure 5 As shown in step S2-2, the BDN topology of each force transmission path of the cable-stayed bridge is:

[0068] (1) The force transmission path "load→main beam" is the topology between the vehicle load node and the main beam node. During one pass, the load will pass through each main beam segment in sequence and transmit the force to each main beam. Therefore, the load node in the BDN points to each main beam node.

[0069] (2) The force transmission path "main beam → cable" is the topology between the main beam node and the cable node. The lower end of the cable is anchored on both sides of the main beam, providing vertical elastic support for the main beam. The vehicle load is transmitted to the cable through the main beam, so the main beam node is the parent node of the cable node, that is, the child node of a main beam segment is all the cable nodes connected to this main beam segment;

[0070] (3) The force transmission path "cable → bridge tower" is the topology between the cable node and the tower node. The load borne by the tower section of a cable-stayed bridge comes from the deadweight of the upper tower section and the reaction force of the cable tension. Considering that the stress of the tower is difficult to calculate, the tower node can be used as an observation node to influence the decision-making node.

[0071] (4) “C_ and C`_ local decision network” is the local decision network of the force transmission paths X and Y. The network runs from the load node to the tower node, the tower nodes and path decision nodes are connected by information arcs, and the utility nodes are pointed to by the cable nodes and decision nodes. Considering Figure 2 The cable-stayed bridge has a total of 20 cables, so the decision network constructed has a total of 20 local decision networks.

[0072] Step S2-3: Calculate the CPT and UT of each uncertainty node. The CPT can be obtained from monitoring data or numerical analysis samples, while the UT can be calculated based on the relevant table data in the "Highway Bridge Technical Condition Assessment Standard" (JTG / T H21-2011). In this embodiment, the CPT is also a function of the bridge structure-level degradation model, while the utility node is a specific numerical value that does not change with bridge performance degradation.

[0073] Step S2-4: Assemble the established network topology together with the CPT and UT to form a cable-stayed bridge BDN.

[0074] Step S3: Cable-stayed bridge safety assessment. Based on the utility values ​​calculated by the UT, a decision-making evaluation is performed on each force transmission path using the BDN. The local paths of the cable-stayed bridge (e.g., girder → cable → pylon) are first evaluated. The utility values ​​output from each local path are then weighted averaged to complete the safety assessment of the cable-stayed bridge system. The overall utility values ​​output from the component-level degradation model are then fitted into a system-level degradation model. This generated system-level degradation model can predict the future performance of the bridge.

[0075] When the failure value of a utility node on a cable-stayed bridge path is high, the decision node for that path declares a "failure" decision; otherwise, it declares a "safe" decision. The utility of the entire cable-stayed bridge is calculated from the 20 local utilities. When the overall utility failure value is high, the overall decision node declares a "failure" decision; otherwise, it declares a "safe" decision.

[0076] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0077] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0078] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0079] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0080] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other manner. Any person skilled in the art may utilize the above-disclosed technical content to modify or modify the present invention into equivalent embodiments. However, any simple modifications, equivalent variations, and modifications to the above embodiments that do not depart from the technical content of the present invention and are based on the technical essence of the present invention remain within the scope of protection of the present invention.

[0081] This patent is not limited to the above-mentioned optimal implementation mode. Anyone can derive various other forms of cable-stayed bridge safety assessment methods combined with Bayesian decision networks under the inspiration of this patent. All equal changes and modifications made according to the scope of the patent application of this invention should fall within the scope of this patent.

Claims

1. A cable-stayed bridge safety assessment method based on Bayesian decision network, characterized in that: The following steps are involved: Step S1: Based on the external load conditions, component geometry and material properties of the cable-stayed bridge, a component-level degradation model based on the degradation mechanism is established for each component level; Step S2: Establishing a Bayesian decision network for a cable-stayed bridge, including: defining network node variables, establishing a network topology including each force transmission path of the cable-stayed bridge, calculating the conditional probability table CPT between each uncertainty node and the utility table UT of the utility node, and assembling the established network topology and the calculated CPT and UT; Specifically include: Step S2-1: Define the cable-stayed bridge BDN node variables: The load node is defined as an uncertain node. Considering that actual cable-stayed bridges often bear the combined effects of constant and live loads, where live load is the main factor causing stress changes in various components during the operation of cable-stayed bridges, it is set as ; Define the cable node as an uncertain node, one cable as a cable node, and the maximum cable stress as the variable of the cable node, set as ; The main beam node is defined as an uncertain node. The anchor point between the cable and the main beam is used as the boundary to divide the main beam into different beam segments. The maximum deflection of the main beam segment is the variable of the main beam node, which is set as ; The tower node is defined as an uncertain node. The anchor point between the cable and the tower is used as the boundary to divide the tower into different tower segments. The maximum stress value of the tower segment is the variable, set as ; Define path nodes as decision nodes. Establish a force transmission network based on the force transmission path of the cable-stayed bridge: "load → main beam → cable → bridge tower". The nodes contain two states: "failure" and "safe". There are paths 1, 2, ...; Define utility nodes as the basis for decision nodes to make decisions: Each utility node has a corresponding UT, which contains the utility values ​​of each decision node, set as ; Step S2-2: Establish the Bayesian decision network topology of each force transmission path of the cable-stayed bridge. Use conditional arcs to connect uncertainty nodes in sequence. Then add function arcs from cable nodes and decision nodes to utility nodes. Finally, add information arcs from tower nodes to decision nodes. Step S2-3: Calculate the CPT of each uncertainty node and the UT of the utility node; Step S2-4: Assemble the established network topology together with CPT and UT to form the Bayesian decision network of the cable-stayed bridge; Step S3: Safety assessment of cable-stayed bridge: Based on the utility value calculated by UT, a Bayesian decision network is used to perform decision evaluation on each force transmission path: first, the local path of the cable-stayed bridge is evaluated, and then the utility value of each output of the local path is weighted averaged to complete the safety assessment of the cable-stayed bridge system.

2. The cable-stayed bridge safety assessment method combined with a Bayesian decision network according to claim 1 is characterized by: After step S3, the overall utility value output under the component-level degradation model is fitted into a system-level degradation model.

3. The cable-stayed bridge safety assessment method in combination with a Bayesian decision network according to claim 1 is characterized in that: In step S2-3, CPT is obtained by learning from monitoring data or numerical analysis samples; UT is calculated with reference to the relevant table data in the "Highway Bridge Technical Condition Assessment Standard" JTG / T H21-2011; among them, CPT is a function of the bridge structure-level degradation model, and the utility node in UT is a specific numerical value that does not change with the degradation of bridge performance.

4. A cable-stayed bridge safety assessment system combined with Bayesian decision network, characterized in that: A method for implementing the method according to any one of claims 1 to 3, based on a computer system, comprising: The degradation model generation module is used to establish a component-level degradation model based on the degradation mechanism of each component according to the external load conditions, component geometry and material properties of the cable-stayed bridge; The Bayesian decision network module for cable-stayed bridges is used to establish a Bayesian decision network for cable-stayed bridges, including: defining network node variables, establishing a network topology that includes the force transmission paths of the cable-stayed bridge, calculating the conditional probability table (CPT) between each uncertainty node and the utility table (UT) of the utility node, and assembling the established network topology and the calculated CPT and UT; In addition, the cable-stayed bridge safety assessment module is used to make decision evaluations on each force transmission path through a Bayesian decision network based on the utility value calculated by UT: first, the local path of the cable-stayed bridge is evaluated, and then the utility value of each output of the local path is weighted averaged to complete the safety assessment of the cable-stayed bridge system.

5. The cable-stayed bridge safety assessment system combined with Bayesian decision network according to claim 4 is characterized in that: Also includes: The degradation model generation module is used to fit the overall utility value output under the component-level degradation model into a system-level degradation model.

Images