Bridge structure risk prediction method and system based on digital twinning

Abstract

The invention provides a bridge structure risk prediction method and system based on digital twinning, and relates to the technical field of bridge structure risk prediction, and the method comprises the steps: obtaining bridge stress node displacement data, environment humidity data, and bridge visual and acoustic monitoring data through the system; the system constructs a digital twinborn model, simulates stress and deformation of a bridge under different load conditions, and calculates a stress deformation potential risk coefficient according to a maximum stress value and an average deformation value. The monitoring risk coefficient is calculated by extracting the total crack length in the visual monitoring data and the frequency characteristics of the acoustic data. And calculating a final risk coefficient by combining the stress deformation potential risk coefficient, the monitoring risk coefficient and the humidity data so as to judge the safety condition of the bridge. According to the method, the accuracy and reliability of bridge safety assessment in a complex environment are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of bridge structure risk prediction, and specifically to a bridge structure risk prediction method and system based on digital twins. Background Art

[0002] Over the past few decades, bridges, as vital infrastructure, have played a key role in urban transportation and logistics. The safety hazards of bridge structures have long been a focus of concern within the engineering community and society. Due to the complexities of the natural environment and usage, bridge accidents frequently occur, resulting in significant economic losses and safety risks. This has prompted governments and relevant agencies worldwide to continuously explore and adopt new technologies to enhance bridge safety monitoring and management.

[0003] In the prior art, publication number CN118364555B discloses a bridge structure risk prediction method and system combined with deep learning. The method obtains labeled data containing sample bridge data and risk features, introduces attention units, and further processes these focused features through a deep learning network to generate a description of bridge abnormality features. The risk feature labeled data and abnormal features are fused to generate fused data, which is then used to derive new bridge structure data. The data is used to construct a risk learning dataset, and the risk prediction model is optimized by training error parameters. However, current technologies mostly rely on a single technical means. For example, using only machine learning to judge the status of a bridge may miss other potential risk factors, resulting in insufficient accuracy and real-time performance of the prediction results.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a bridge structure risk prediction method and system based on digital twins to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A bridge structure risk prediction method based on digital twins includes the following steps:

[0008] S1: Obtain bridge stress node displacement data, humidity data of the bridge's surrounding environment, and visual monitoring data and acoustic monitoring data of the bridge;

[0009] S2: Build a digital twin model to simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk factor of stress and deformation based on the maximum stress value and average deformation value;

[0010] S3: Based on the visual monitoring data and acoustic monitoring data of the bridge, the total length of the cracks and the frequency characteristic data of the acoustic data are extracted to calculate the monitoring risk coefficient;

[0011] S4: Calculate the final risk factor based on the stress-deformation potential risk factor, monitoring risk factor, and humidity data to determine the safety of the bridge.

[0012] The supports, mid-span, and connections in the bridge are set as stress nodes. A spatial coordinate system is established based on the environmental area where the bridge is located. A point is randomly selected in the bridge area as the origin. At the selected origin, the length direction of the bridge is the x-axis, the direction perpendicular to the x-axis and parallel to the bridge width is the y-axis, and the height direction of the bridge perpendicular to the x-axis and y-axis is the z-axis. According to the established spatial coordinate system, the position coordinates of each stress node are determined and recorded as (x j ,y j ,z j ), where j is the index number of the stress node, j = 1, 2, …, m, where m is the total number of stress nodes in the environmental area where the bridge is located;

[0013] The formula for calculating the displacement of the j-th stress node is:

[0014]

[0015] Among them, δ j is the displacement of the jth stress node, x now is the x-axis coordinate of the j-th stress node after movement, y now is the y-axis coordinate of the stress node after movement, z now is the z-axis coordinate of the stress node after movement;

[0016] The visual monitoring data is used to identify the total length of cracks based on the acquired bridge surface images. The acoustic monitoring data is used to identify the frequency characteristic data obtained by the acoustic sensor. The data is detected by the sensor through the acoustic signal changes inside the bridge. The data is extracted through signal processing technology to analyze the frequency and amplitude characteristics, thereby identifying potential internal damage.

[0017] Furthermore, the logic of building the digital twin technology is:

[0018] Through the CAD model, the geometric structure and material properties of the actual bridge are fully reproduced in the digital space; the collected real-time stress node displacement and humidity data are combined with the dynamic response of the digital model. By inputting the collected real-time stress node displacement, humidity data, visual monitoring data and acoustic monitoring data into the digital twin model, synchronous updates with the actual bridge status are achieved. Through FEA software, these real-time data are applied to the model to simulate the stress and deformation of the bridge under different load conditions.

[0019] Furthermore, the maximum stress value is obtained from the stress node distribution value, and the stress node distribution value formula is:

[0020] σ j =K j ·δ j

[0021] Among them, σ j is the stress of the jth stress node, L j is the stress conversion matrix, δ j is the stress node displacement;

[0022] The maximum stress value formula is:

[0023] σ max =max(σ1,σ2,…,σ m )

[0024] Among them, σ max is the maximum stress value, and m is the total number of stress nodes.

[0025] Furthermore, the calculation formula of the average deformation value is:

[0026]

[0027] Among them, δ avg is the average deformation value, δ j is the displacement of the jth stress node.

[0028] Furthermore, the calculation formula of the stress deformation potential risk coefficient is:

[0029] G potential =α·σ max +β·δ avg

[0030] Among them, G potential is the potential risk coefficient of stress and deformation, α and β are the weight coefficients of the maximum stress value and the average deformation value, respectively, where α and β are both greater than 0, and α<β.

[0031] Furthermore, the calculation formula of the monitoring risk coefficient is:

[0032] F=w1·L crack +w2·A acoustic

[0033] Among them, F is the monitoring risk coefficient, L crack is the total length of the crack, A acoustic is the acoustic characteristic amplitude, w1 and w2 are the weight coefficients of the total crack length and acoustic characteristic amplitude respectively, where w1 and w2 are both greater than 0, and w1 <w2。

[0034] Furthermore, the calculation formula of the final risk coefficient is:

[0035]

[0036] Among them, R final is the final risk coefficient, S is the humidity change, and the humidity change is the difference between the current humidity value and the preset humidity threshold g. γ1, γ2, and γ3 are the weight coefficients of the monitoring risk coefficient, the potential risk coefficient of stress and deformation, and the humidity change, respectively. γ1, γ2, and γ3 are all greater than 0, and γ3<γ1<γ2. ρ, μ, and θ are adjustment coefficients.

[0037] Furthermore, the logic for judging the safety of a bridge is as follows:

[0038] When F < 0.5 and G potential <0.3, R final <2.0, low risk state, the bridge is in normal condition and no immediate maintenance is required; when 0.5≤F<1.0 or 0.3 <G potential <0.7, 2.0≤R final <4.0, medium risk state, requires regular monitoring, and may require small-scale maintenance; when 1.0≤F<1.5 or 0.7 <G potential ≤1.0, 4.0≤R final <6.0, high risk state, immediate measures need to be taken for preventive maintenance; when F≥1.5 or G potential When ≥1.0, R final ≥6.0, extremely high risk status, must be immediately inspected and repaired.

[0039] The present invention also provides a bridge structure risk prediction system based on digital twins. The prediction system is obtained by executing the above-mentioned bridge structure risk prediction method based on digital twins. The specific modules include:

[0040] Acquisition module: acquires bridge stress node displacement data, humidity data of the bridge's surrounding environment, as well as visual monitoring data and acoustic monitoring data of the bridge;

[0041] Stress and deformation risk module: Build a digital twin model to simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk coefficient of stress and deformation based on the maximum stress value and average deformation value;

[0042] Monitoring risk module: Based on the visual monitoring data and acoustic monitoring data of the bridge, the total length of the cracks and the frequency characteristic data of the acoustic data are extracted to calculate the monitoring risk coefficient;

[0043] Judgment module: Calculate the final risk factor based on the potential risk factor of stress and deformation, monitoring risk factor and humidity data to judge the safety of the bridge.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] The present invention significantly improves the accuracy and real-time performance of bridge safety monitoring by introducing a bridge structure risk prediction system based on digital twins. Specifically, the system comprehensively constructs a digital twin model of the bridge by acquiring bridge stress node displacement, environmental humidity, visual and acoustic monitoring data. The model can simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk coefficient of stress and deformation by combining the maximum stress value and the average deformation value. In addition, the system further uses visual and acoustic data to calculate the monitoring risk coefficient, making the risk assessment more comprehensive and accurate. By combining humidity data, the system calculates the final risk coefficient to determine the safety of the bridge. This multi-dimensional data integration and analysis method enables the status of the bridge to be evaluated more accurately. Through this intelligent monitoring and prediction mechanism, the present invention not only improves the safety management level of the bridge, but also provides strong support for preventive maintenance, greatly enhancing the adaptability and responsiveness in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0047] Figure 2 This is a schematic diagram of the overall system design of the present invention. DETAILED DESCRIPTION

[0048] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0049] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0050] Example:

[0051] See also Figure 1 , the present invention provides a technical solution:

[0052] A bridge structure risk prediction method and system based on digital twins, specifically comprising the following steps:

[0053] S1: Obtain bridge stress node displacement data, humidity data of the bridge's surrounding environment, and visual monitoring data and acoustic monitoring data of the bridge;

[0054] The supports, mid-span, and connections of the bridge are set as stress nodes. The supports of the bridge are structural components that bear the weight of the bridge body and traffic loads, and transfer the loads to the piers or abutments. The mid-span is the central part of the bridge span, and the connection is the part where the various components of the bridge are connected, such as the connection between beams and beams, and beams and columns. A spatial coordinate system is established based on the environmental area where the bridge is located. A point in the bridge area is randomly selected as the origin. At the selected origin, the length direction of the bridge is the x-axis direction, the direction perpendicular to the x-axis and parallel to the width of the bridge is the y-axis, and the height direction of the bridge perpendicular to the x-axis and y-axis is the z-axis. According to the established spatial coordinate system, the position coordinates of each stress node are determined, which are recorded as (x j ,y j ,z j ), where j is the index number of the stress node, j = 1, 2, …, m, where m is the total number of stress nodes in the environmental area where the bridge is located;

[0055] The formula for calculating the displacement of the j-th stress node is:

[0056]

[0057] Among them, δ j is the displacement of the jth stress node, x now is the x-axis coordinate of the j-th stress node after movement, y now is the y-axis coordinate of the stress node after movement, z now The z-axis coordinate of the stress node after it moves

[0058] The visual monitoring data is obtained based on the surface image of the bridge to identify the total length of the cracks. The visual sensor uses a laser scanner to obtain an image of the structure surface, uses a filter to remove noise in the image, adjusts the contrast and brightness to highlight the crack characteristics, applies an edge detection algorithm to identify the edge of the crack, uses expansion and corrosion operations to highlight the crack, and measures the total length of the crack through a geometric algorithm. The acoustic monitoring data is frequency characteristic data obtained by the acoustic sensor. The data detects changes in acoustic signals inside the bridge through the sensor. The data is extracted through signal processing technology, using Fourier transform to convert the signal into the frequency domain, identify characteristic frequencies, and analyze frequency and amplitude characteristics to identify potential internal damage.

[0059] S2: Build a digital twin model to simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk factor of stress and deformation based on the maximum stress value and average deformation value;

[0060] The CAD model fully reproduces the actual bridge's geometry and material properties in digital space. Real-time data on stress node displacements and humidity levels is collected and integrated with the dynamic response of the digital model. By inputting these data, along with visual and acoustic monitoring data, into the digital twin model, it is synchronized with the actual bridge's state. This real-time data is then applied to the model using FEA software to simulate the bridge's stress and deformation under varying load conditions.

[0061] The maximum stress value is obtained from the stress node distribution value, and the stress node distribution value formula is:

[0062] σ j =K j ·δ j

[0063] Among them, σ j is the stress of the jth stress node, K j To obtain the stress conversion matrix, the basic characteristics of the material are determined according to the material properties at the stress node position. The structure is discretized using the finite element method to determine the influence area of ​​each stress node. The conversion matrix is ​​calculated based on the geometric relationship between the material properties and the stress nodes.

[0064] The maximum stress value formula is:

[0065] σ max =max(σ1,σ2,…,σ m )

[0066] Among them, σ max is the maximum stress value, and m is the total number of stress nodes.

[0067] The calculation formula of the average deformation value is:

[0068]

[0069] Among them, δ avg is the average deformation value, δ j is the displacement of the jth stress node.

[0070] The calculation formula of the stress deformation potential risk coefficient is:

[0071] G potential =α·σ max +β·δ avg

[0072] Among them, G potential is the potential risk coefficient of stress deformation, which reflects the potential danger of the bridge caused by stress load. When G potential is larger, it indicates that the stress node load of the bridge is too large and the potential risk of the bridge is high. α and β are the weight coefficients of the maximum stress value and the average deformation value respectively. Among them, both α and β are greater than 0, and α < β. According to expert experience, the value of α can be set to 0.1 - 0.5. This setting is based on the impact of the maximum stress on the bridge, and this range can effectively ensure that the impact of the maximum stress on the potential risk coefficient of stress deformation is moderate. The value of β can be set to 0.5 - 1.0. The average deformation value is usually a direct reflection of the stress deformation of the bridge, and the impact of deformation in various materials and structures may be more significant. Although the maximum stress is important, its impact is more indirect, and the impact of deformation in various materials and structures may be more significant.

[0073] S3: According to the visual monitoring data and acoustic monitoring data of the bridge, extract the total crack length and the frequency characteristic data of the acoustic data, and calculate the monitoring risk coefficient;

[0074] The calculation formula of the monitoring risk coefficient is:

[0075] F = w1·L crack + w2·A acoustic

[0076] Among them, F is the monitoring risk coefficient, which reflects the cracking degree of the internal and external materials of the bridge. When F is larger, it indicates that the cracking condition of the bridge materials is serious. The total crack length and the frequency characteristic data of the acoustic data are the external cracking condition and the internal cracking condition of the bridge materials respectively. L crack is the total crack length, A acoustic is the acoustic characteristic amplitude. w1 and w2 are the weight coefficients of the total crack length and the acoustic characteristic amplitude respectively. Among them, both w1 and w2 are greater than 0, and w1 < w2. According to expert experience, the value of w1 can be set to 0.1 - 0.4. The setting is based on the impact of the total crack length on the bridge risk, and this range can effectively ensure that the impact of the total crack length on the monitoring risk coefficient is moderate. The value of w2 can be set to 0.4 - 1.0. The acoustic characteristic amplitude is usually a direct reflection of the bridge risk. The acoustic signal can detect tiny and early material changes before the cracks are obvious. Although the crack length directly reflects physical damage, the acoustic signal can provide more comprehensive structural health information. Therefore, w1 < w2.

[0077] S4: According to the potential risk coefficient of stress deformation, the monitoring risk coefficient and the humidity data, calculate the final risk coefficient and judge the safety condition of the bridge.

[0078] The calculation formula of the final risk coefficient is:

[0079]

[0080] Among them, R final is the final risk factor, R final The larger it is, the greater the risk to the bridge structure. S is the humidity change, and the humidity change is the difference between the current humidity value and the preset humidity threshold g. γ1, γ2 and γ3 are the monitoring risk coefficient, stress deformation potential risk coefficient and humidity change weight coefficient, respectively. Among them, γ1, γ2 and γ3 are all greater than 0, and γ3<γ1<γ2. The range of γ2 is 0.5-1.0, which is set to the maximum because stress deformation has the most direct and serious impact on structural safety and may cause structural failure. The range of γ1 is 0.3-0.4. The monitoring risk coefficient combines cracks and acoustic characteristics and can reflect potential problems of the structure. The range of γ3 is 0.1-0.2, which is set to the minimum because although humidity change affects the structure, it usually does not directly cause damage. ρ, μ and θ are preset adjustment coefficients. According to expert experience, they range from 0 to 10.

[0081] The logic for judging the safety of a bridge is as follows:

[0082] When F < 0.5 and G potential <0.3, R final <2.0, low risk state, the bridge is in normal condition and no immediate maintenance is required; when 0.5≤F<1.0 or 0.3 <G potential <0.7, 2.0≤R final <4.0, medium risk state, requires regular monitoring, and may require small-scale maintenance; when 1.0≤F<1.5 or 0.7 <G potential ≤1.0, 4.0≤R final <6.0, high risk state, immediate measures need to be taken for preventive maintenance; when F≥1.5 or G potential When ≥1.0, R final ≥6.0, extremely high risk status, must be immediately inspected and repaired.

[0083] See also Figure 2 The present invention also provides a bridge structure risk prediction system based on digital twins. The prediction system is obtained by executing the above-mentioned bridge structure risk prediction method. The specific modules include:

[0084] Acquisition module: acquires bridge stress node displacement data, humidity data of the bridge's surrounding environment, as well as visual monitoring data and acoustic monitoring data of the bridge;

[0085] Stress and deformation risk module: Build a digital twin model to simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk coefficient of stress and deformation based on the maximum stress value and average deformation value;

[0086] Monitoring risk module: Based on the visual monitoring data and acoustic monitoring data of the bridge, the total length of the cracks and the frequency characteristic data of the acoustic data are extracted to calculate the monitoring risk coefficient;

[0087] Judgment module: Calculate the final risk factor based on the potential risk factor of stress and deformation, monitoring risk factor and humidity data to judge the safety of the bridge.

[0088] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0089] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0090] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0091] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A bridge structure risk prediction method based on digital twins, characterized in that: The specific steps include: S1: Obtain the displacement data of the bridge stress nodes, collect the humidity data of the bridge surrounding environment, and the visual monitoring data and acoustic monitoring data of the bridge; S2: Build a digital twin model to simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk coefficient of stress and deformation based on the maximum stress value and average deformation value; S3: Based on the visual monitoring data and acoustic monitoring data of the bridge, the total length of the cracks and the frequency characteristic data of the acoustic data are extracted to calculate the monitoring risk coefficient; S4: Calculate the final risk factor based on the potential risk factor of stress deformation, monitoring risk factor and humidity data to determine the safety of the bridge.

2. A bridge structure risk prediction method based on digital twin according to claim 1, characterized in that: The specific method for obtaining the displacement data of bridge stress nodes is: The supports, mid-span, and joints of the bridge are set as stress nodes. A spatial coordinate system is established based on the environment in which the bridge is located. A point in the bridge area is randomly selected as the origin. At the selected origin, the length of the bridge is the x-axis, the direction perpendicular to the x-axis and parallel to the bridge width is the y-axis, and the height of the bridge perpendicular to the x-axis and y-axis is the z-axis. According to the established spatial coordinate system, the position coordinates of each stress node are determined, recorded as (x j ,y j ,z j ), where j is the index number of the stress node, j = 1, 2, ..., m, where m is the total number of stress nodes in the environmental area where the bridge is located; The formula for calculating the displacement of the jth stress node is: Among them, δ j is the displacement of the jth stress node, x now is the x-axis coordinate of the jth stress node after it moves, and y now is the y-axis coordinate of the stress node after it moves, z now is the z-axis coordinate of the stress node after movement; The visual monitoring data is used to identify the total length of cracks based on the acquired bridge surface image. The acoustic monitoring data is used to identify the frequency characteristic data obtained by the acoustic sensor. The data is detected by the sensor through the acoustic signal changes inside the bridge. The data is extracted through signal processing technology to analyze the frequency and amplitude characteristics, thereby identifying potential internal damage.

3. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The logic of building digital twin technology is: Through the CAD model, the geometric structure and material properties of the actual bridge are fully reproduced in the digital space. The collected real-time stress node displacement and humidity data are combined with the dynamic response of the digital model. By inputting the collected real-time stress node displacement, humidity data, visual monitoring data and acoustic monitoring data into the digital twin model, synchronous updates with the actual bridge status are achieved. Through FEA software, these real-time data are applied to the model to simulate the stress and deformation of the bridge under different load conditions.

4. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The maximum stress value is obtained from the stress node distribution value, and the stress node distribution value formula is: s j =K j ·d j Among them, σ j is the stress of the jth stress node, K j is the stress transformation matrix of the jth stress node, δ j is the stress node displacement; The maximum stress value formula is: s max =max(σ1,σ2,…,σ m ) Among them, σ max is the maximum stress value, and m is the total number of stress nodes.

5. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The calculation formula of the average deformation value is: Among them, δ avg is the average deformation value, δ j is the displacement of the jth stress node.

6. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The calculation formula of the stress deformation potential risk coefficient is: G potential =a·s max +b·d avg Among them, G potential is the potential risk coefficient of stress deformation, α and β are the weight coefficients of the maximum stress value and the average deformation value, respectively, where α and β are both greater than 0, and α<β.

7. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The calculation formula of the monitoring risk coefficient is: F=w1·L crack +w2·A acoustic Among them, F is the monitoring risk factor, L crack is the total length of the crack, A acoustic is the acoustic characteristic amplitude, w1 and w2 are the weight coefficients of the total crack length and the acoustic characteristic amplitude, respectively, where w1 and w2 are both greater than 0, and w1 <w2。 8. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The calculation formula of the final risk coefficient is: Among them, R final is the final risk coefficient, S is the humidity change, the humidity change is the difference between the current humidity value and the preset humidity threshold g, γ1, γ2 and γ3 are the weight coefficients of monitoring risk coefficient, stress deformation potential risk coefficient and humidity change, respectively, where γ1, γ2 and γ3 are all greater than 0, and γ3<γ1<γ2, ρ, μ and θ are adjustment coefficients.

9. The bridge structure risk prediction method based on digital twin according to claim 1 is characterized by: The logic for judging the safety of a bridge is as follows: When F < 0.5 and G potential <0.3, R final <2.0, low risk state, the bridge is in normal condition and no immediate maintenance is required; when 0.5≤F<1.0 or 0.3 <G potential <0.7, 2.0≤R final <4.0, medium risk state, requiring regular monitoring and small-scale maintenance; when 1.0≤F<1.5 or 0.7 <G potential ≤1.0, 4.0≤R final <6.0, high risk status, immediate measures need to be taken for preventive maintenance; When F ≥ 1.5 or G potential ≥1.0, R final ≥6.0, extremely high risk status, must be immediately inspected and repaired.

10. A bridge structure risk prediction system based on digital twins, characterized by: The system is obtained by executing the bridge structure risk prediction method of the digital twin according to any one of claims 1 to 9, and the specific modules include: Acquisition module: obtains the displacement data of bridge stress nodes, humidity data of the bridge's surrounding environment, and visual monitoring data and acoustic monitoring data of the bridge; Stress deformation risk module: Build a digital twin model to simulate the stress and deformation of the bridge under different load conditions, and calculate the potential risk coefficient of stress deformation based on the maximum stress value and average deformation value; Monitoring risk module: Based on the visual monitoring data and acoustic monitoring data of the bridge, the total length of the cracks and the frequency characteristic data of the acoustic data are extracted to calculate the monitoring risk coefficient; Judgment module: Calculate the final risk factor based on the potential risk factor of stress deformation, monitoring risk factor and humidity data to judge the safety of the bridge.

Images