Method for structural state perception and post-earthquake function loss evaluation of large-span transportation junction in high-intensity earthquake area

Abstract

The invention discloses a structural state perception and post-earthquake function loss evaluation method for a large-span transportation junction in a high-intensity earthquake area, and relates to the technical field of health monitoring and damage identification of important building structures. Comprising the steps of obtaining high-fidelity dynamic response data and low-fidelity dynamic response data of a large-span transportation junction structure under the action of an earthquake, and constructing a multi-fidelity structure response data set; establishing a data-physical driven large-span transportation junction structure earthquake response prediction model to obtain physical damage of the structure under the earthquake action; establishing a correlation mechanism between physical damage and functional loss of a main structure and a non-structural member of the large-span transportation hub under the action of an earthquake; and determining a weight coefficient of the function damage evaluation index, and establishing a post-earthquake function loss evaluation model of the large-span transportation junction structure. According to the method, a multi-source data fusion method is adopted, a large-span transportation junction structural state cross-modal cooperative sensing method is established, and rapid and accurate sensing of the damage state of the large-span transportation junction structural state is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of health monitoring and damage identification of important building structures, and more specifically to a method for sensing the structural state and assessing post-earthquake functional losses of large-span transportation hubs in high-intensity earthquake zones. Background Art

[0002] my country lies between the Pacific Rim and the Eurasian Seismic Belt, placing urban agglomerations at significant risk from earthquake disasters. Large-span transportation hubs, as core nodes of urban lifelines, have a significant impact on emergency response efficiency and socioeconomic stability through their post-earthquake resilience. Therefore, improving the seismic resilience of transportation hub structures remains a research hotspot and a cutting-edge trend in engineering structural disaster prevention and mitigation.

[0003] The existing monitoring system focuses on the key components of the main structure, with incomplete perception dimensions and a lack of coordinated monitoring of non-structural components such as curtain walls, pipes, and ceilings. In addition, cross-modal data fusion technology is immature, making it difficult to fully capture the state of structural damage. Traditional seismic response analysis methods rely on a large amount of computing resources. With the development of data-driven technologies such as neural networks and deep learning, the application of machine learning methods in structural response analysis is becoming increasingly mature. However, existing fusion algorithms mostly rely on experimental data experience and have not yet fully integrated the constitutive relationship of structural mechanics, resulting in deviations between theoretical models and actual engineering scenarios. In addition, the existing performance level classification criteria for large-span transportation hub structures are mostly focused on the physical response of the structure. A quantitative correlation mechanism between physical damage and functional loss (such as traffic interruption and economic impact) has not yet been established. At the same time, the influence of other factors such as personnel and economy has not been considered, and the performance evaluation index system is single.

[0004] Therefore, in order to improve the seismic resistance and disaster reduction capabilities of large-span transportation hub structures, the present invention proposes a method for structural state perception and post-earthquake functional loss assessment of large-span transportation hubs in high-intensity earthquake zones to solve the difficulties existing in the existing technology, which is a problem that technical personnel in this field urgently need to solve. Summary of the Invention

[0005] In view of this, the present invention provides a method for structural state perception and post-earthquake functional loss assessment of large-span transportation hubs in high-intensity earthquake zones. Through multi-source heterogeneous data fusion and edge computing technology, real-time and high-precision perception of structural damage status is achieved; a physical-data-driven earthquake response intelligent analysis model is constructed to improve the efficiency and reliability of earthquake response analysis; and a post-earthquake functional loss assessment model for large-span transportation hub structures is established to provide a theoretical basis for post-earthquake emergency rescue and recovery decisions.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] A method for sensing the structural state and assessing post-earthquake functional loss of a large-span transportation hub in a high-intensity earthquake zone includes the following steps:

[0008] S1. Obtain high-fidelity and low-fidelity dynamic response data of large-span transportation hub structures under earthquake action through shaking table tests and numerical simulations, respectively, and construct a multi-fidelity structural response dataset.

[0009] S2. Build a data-physics-driven seismic response prediction model for large-span transportation hub structures by combining graph neural networks and long short-term memory network algorithms to determine the physical damage to the structure under earthquake action.

[0010] S3. Use the fault tree analysis method to establish the correlation mechanism between physical damage and functional loss of the main structure and non-structural components of large-span transportation hubs under earthquake action;

[0011] S4. Based on the Delphi method, the weight coefficients of functional damage assessment indicators are determined. Combined with incremental dynamic time history analysis and Monte Carlo simulation, a post-earthquake functional loss assessment model for large-span transportation hub structures is established.

[0012] Optionally, S1 includes the following specific contents:

[0013] S11. Conduct shaking table tests on large-span transportation hub structures. Use accelerometers, displacement sensors, a DIC three-dimensional strain measurement system, and strain gauges to measure the acceleration, displacement, and strain responses of the test model. Install several GoPro sports cameras and video cameras with different viewing angles on the shaking table surface and main structure to capture images and videos of the overall and local responses of the test model.

[0014] S12. Perform multi-source data fusion on the sensory data collected during the shaking table test. By performing data cleaning, spatiotemporal registration, feature extraction, and feature fusion on images, videos, and sensor data, high-fidelity data on the dynamic response of large-span transportation hub structures is ultimately generated.

[0015] S13. Use finite element software to establish a numerical model of the main structure and non-structural components of a large-span transportation hub. Consider the non-structural components as the additional mass of the main structure and perform finite element analysis on the main structure. Consider the main structure response as the boundary condition of the non-structural components and perform finite element analysis on the non-structural components to generate low-fidelity data on the dynamic response of the large-span transportation hub structure.

[0016] Optionally, the specific content of multi-source data fusion in S12 is:

[0017] Data cleaning: Adaptive median filtering algorithm is used to eliminate noise, and cubic spline interpolation method is used to fill missing data;

[0018] Spatiotemporal registration: By adding timestamps to the collected data points, the temporal alignment of data with different sampling frequencies is achieved, and spatial interpolation technology is used to align the data spatially;

[0019] Feature extraction: extract features from preprocessed multi-source data;

[0020] Feature fusion: The principal component analysis method is used to fuse the extracted features to form high-fidelity data of the dynamic response of large-span transportation hub structures.

[0021] Optionally, feature extraction includes the following specific contents:

[0022] Extracting displacement features from displacement sensor data;

[0023] Extract the natural frequencies of the main structure and non-structural components from the acceleration sensor based on Fourier transform;

[0024] The appearance damage features of main structures and non-structural components are extracted from images and videos based on the improved edge detection algorithm.

[0025] Optionally, S2 includes the following specific contents:

[0026] S21. Use the earthquake motion sequence, structural nodes, and unit-related characteristic parameters as input data, and the time-history dynamic response at the corresponding position as output data. Assemble the input and output as learning samples, and divide the learning samples into training set, validation set, and test set, which are used for training, parameter adjustment, and performance evaluation of the prediction model respectively.

[0027] S22. Represent the structure as a graph, with the connection points of the structural components as nodes V of the graph, and the topological spatial relationships between the nodes as edges E of the graph. The discretized structure is a graph G = (V, E), and features related to the structural nodes and features related to the structural units are stored as nodes V and edges E of the graph, respectively.

[0028] S23. The training data is input into multiple parallel graph convolutional layers, and the node representation is updated through the message passing mechanism. The long short-term memory layer controls the flow of temporal information through the input gate, forget gate, and output gate. The Adam optimizer is used to adjust the learning rate, and the graph convolutional layer and the long short-term memory layer are optimized using Dropout regularization. The model is trained, verified, and tested based on the learning samples, and finally a seismic response prediction model for large-span transportation hub structures is obtained.

[0029] Optionally, the structural node and unit-related characteristic parameters in S21 include the following specific contents:

[0030] Structural nodes include: node coordinates, node mass, and frequency, mode, damping and combination of the structure;

[0031] Unit-related features include: unit length, unit section parameters, unit material parameters and combinations.

[0032] Optionally, S3 includes the following specific contents:

[0033] Using the fault tree analysis method, the loss of transportation hub function is taken as the top-level event of the fault tree, and then decomposed layer by layer into intermediate events such as damage to the main structure and destruction of non-structural components. Combined with expert consultation and historical earthquake damage data, the basic events of the fault tree are identified, and the logical relationship between the basic events and intermediate events is represented by logical OR gates and logical AND gates.

[0034] Alternatively, the formula for the logical OR gate is:

[0035]

[0036] Among them, D1 is an intermediate event, which is connected to the basic events X1 and X2 through a logic OR gate;

[0037] When only one of the basic events X1 and X2 occurs, the corrected formula of the logic OR gate is:

[0038]

[0039] The formula for the logic AND gate is:

[0040]

[0041] Among them, D2 is an intermediate event, which is connected to the basic events X3 and X4 through a logic AND gate;

[0042] When only one of the basic events X3 and X4 occurs, the corrected formula of the logic AND gate is:

[0043]

[0044] Optionally, S4 includes the following specific contents:

[0045] S41. Based on the results of the fault tree analysis, considering physical, economic and social factors, determine the multi-dimensional functional damage assessment indicators of large-span transportation hub structures;

[0046] S42. The Delphi method is used to weight the evaluation indicators. The opinions of experts on the questionnaire questions are collected anonymously multiple times, and feedback is provided. The opinions of experts are summarized, revised, and consulted several times. The consensus opinions of the experts are analyzed and summarized, and the weight coefficients of each evaluation indicator are finally determined.

[0047] S43. Use Monte Carlo simulation to generate earthquake intensity samples, conduct incremental dynamic time history analysis, record the structural response parameters under different earthquake scenarios, and form an engineering demand parameter matrix; judge the structural damage state based on the structural response under different earthquake scenarios, further calculate the functional loss index under each earthquake scenario, and use statistical analysis methods to establish the relationship between the functional loss index and the earthquake intensity and structural damage state.

[0048] Optionally, the multi-dimensional functional damage assessment indicators for large-span transportation hub structures in S41 include: primary indicators and secondary indicators; among them,

[0049] The first-level indicators include: physical damage, functional loss, and economic impact;

[0050] Secondary indicators include: structural damage, non-structural damage, economic losses, repair costs, number of people affected, and interruption time.

[0051] It can be seen from the above technical solutions that, compared with the prior art, the present invention provides a method for sensing the structural state and assessing post-earthquake functional loss of large-span transportation hubs in high-intensity earthquake zones, which has the following beneficial effects:

[0052] 1) To address the problem of imperfect performance perception of large-span transportation hub structures, the present invention adopts a multi-source data fusion method to establish a cross-modal collaborative performance perception method for large-span transportation hub structures, thereby achieving rapid and accurate perception of the damage status of large-span transportation hub structures.

[0053] 2) To address the problems of traditional numerical methods relying on large amounts of computing resources and data-driven machine learning models being highly data-dependent and having poor generalization capabilities, the present invention constructs a physical-data-driven seismic response analysis model for large-span transportation hub structures, enabling intelligent analysis of the seismic response of large-span transportation hub structures.

[0054] 3) In response to the problem of insufficient quantification of functional loss in large-span transportation hub structures, the present invention establishes a correlation mechanism between physical damage and functional loss in large-span transportation hub structures and non-structural components, and proposes functional loss assessment criteria for the main structures and non-structural components of large-span transportation hubs, thereby solving the problem of quantitative assessment of functional loss in large-span transportation hub structures after earthquakes. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0056] Figure 1A flow chart of a method for structural state perception and post-earthquake functional loss assessment of a large-span transportation hub in a high-intensity earthquake zone provided by the present invention;

[0057] Figure 2 Flowchart of S1 provided by the present invention;

[0058] Figure 3 This is a flow chart of S2 provided by the present invention. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0060] See also Figure 1 As shown, the present invention discloses a method for sensing the structural state of a large-span transportation hub in a high-intensity earthquake zone and assessing its post-earthquake functional loss, comprising the following steps:

[0061] S1. Obtain high-fidelity and low-fidelity dynamic response data of large-span transportation hub structures under earthquake action through shaking table tests and numerical simulations, respectively, and construct a multi-fidelity structural response dataset.

[0062] S2. Build a data-physics-driven seismic response prediction model for large-span transportation hub structures by combining graph neural networks and long short-term memory network algorithms to determine the physical damage to the structures under earthquakes.

[0063] S3. Use the fault tree analysis method to establish the correlation mechanism between physical damage and functional loss of the main structure and non-structural components of a large-span transportation hub under earthquake action;

[0064] S4. Based on the Delphi method, the weight coefficients of functional damage assessment indicators are determined. Combined with incremental dynamic time history analysis and Monte Carlo simulation, a post-earthquake functional loss assessment model for large-span transportation hub structures is established.

[0065] For further information, see Figure 2 As shown, the specific contents of S1 are:

[0066] S11. Conduct shaking table tests on large-span transportation hub structures. Use accelerometers, displacement sensors, a DIC three-dimensional strain measurement system, and strain gauges to measure the acceleration, displacement, and strain responses of the test model. Install several GoPro sports cameras and video cameras with different viewing angles on the shaking table surface and main structure to capture images and videos of the overall and local responses of the test model.

[0067] Specifically, the specific measurement contents of different sensors are as follows: (1) The acceleration sensor has a range of ±1.5g and is used to measure the absolute acceleration of the test object; (2) The displacement sensor uses a non-contact displacement meter, which measures the absolute displacement of the test model by arranging irregular speckle measurement points; (3) The strain gauge is a resistance strain gauge, which is used to measure the local strain of the test model; (4) The DIC three-dimensional strain measurement system is used to measure the three-dimensional strain and displacement of the glass in the glass curtain wall system.

[0068] S12. Perform multi-source data fusion on the sensory data collected during the shaking table test. By performing data cleaning, spatiotemporal registration, feature extraction, and feature fusion on images, videos, and sensor data, high-fidelity data on the dynamic response of large-span transportation hub structures is ultimately generated.

[0069] S13. Use finite element software to establish a numerical model of the main structure and non-structural components of a large-span transportation hub. Consider the non-structural components as the additional mass of the main structure and perform finite element analysis on the main structure. Consider the main structure response as the boundary condition of the non-structural components and perform finite element analysis on the non-structural components to generate low-fidelity data on the dynamic response of the large-span transportation hub structure.

[0070] Furthermore, the specific contents of multi-source data fusion in S12 are as follows:

[0071] Data cleaning: Adaptive median filtering algorithm is used to eliminate noise, and cubic spline interpolation method is used to fill missing data;

[0072] Spatiotemporal registration: By adding timestamps to the collected data points, the temporal alignment of data with different sampling frequencies is achieved, and spatial interpolation technology is used to align the data spatially;

[0073] Feature extraction: extract features from preprocessed multi-source data;

[0074] Feature fusion: The principal component analysis method is used to fuse the extracted features to form high-fidelity data of the dynamic response of large-span transportation hub structures.

[0075] Furthermore, the specific contents of feature extraction include:

[0076] Extracting displacement features from displacement sensor data;

[0077] Extract the natural frequencies of the main structure and non-structural components from the acceleration sensor based on Fourier transform;

[0078] The appearance damage features of main structures and non-structural components are extracted from images and videos based on the improved edge detection algorithm.

[0079] Specifically, for example, the overall color of the glass curtain wall structural glue joints, the falling of ceiling panels, and the breakage of pipe connections.

[0080] For further information, see Figure 3 As shown, the specific contents of S2 are:

[0081] S21. Use the earthquake motion sequence, structural nodes, and unit-related characteristic parameters as input data, and the time-history dynamic response at the corresponding position as output data. Assemble the input and output as learning samples, and divide the learning samples into training set, validation set, and test set, which are used for training, parameter adjustment, and performance evaluation of the prediction model respectively.

[0082] Specifically, the learning samples are divided into training set, validation set and test set in a ratio of 7:2:1.

[0083] S22. Represent the structure as a graph, with the connection points of the structural components as nodes V of the graph, and the topological spatial relationships between the nodes as edges E of the graph. The discretized structure is a graph G = (V, E), and features related to the structural nodes and features related to the structural units are stored as nodes V and edges E of the graph, respectively.

[0084] S23. The training data is input into multiple parallel graph convolutional layers, and the node representation is updated through the message passing mechanism. The long short-term memory layer controls the flow of temporal information through the input gate, forget gate, and output gate. The Adam optimizer is used to adjust the learning rate, and the graph convolutional layer and the long short-term memory layer are optimized using Dropout regularization. The model is trained, verified, and tested based on the learning samples, and finally a seismic response prediction model for large-span transportation hub structures is obtained.

[0085] Furthermore, the structural nodes and unit-related characteristic parameters in S21 include the following specific contents:

[0086] Structural nodes include: node coordinates, node mass, and frequency, mode, damping and combination of the structure;

[0087] Unit-related features include: unit length, unit section parameters, unit material parameters and combinations.

[0088] Furthermore, the specific contents of S3 are as follows:

[0089] Using the fault tree analysis method, the loss of transportation hub function is taken as the top-level event of the fault tree, and then decomposed layer by layer into intermediate events such as damage to the main structure and destruction of non-structural components. Combined with expert consultation and historical earthquake damage data, the basic events of the fault tree are identified, and the logical relationship between the basic events and intermediate events is represented by logical OR gates and logical AND gates.

[0090] Specifically, based on the above intermediate events, combined with expert consultation and historical earthquake damage data, all possible basic events are identified, such as excessive column end bending moments, curtain wall detachment, pipe rupture, and ceiling panel fall.

[0091] Furthermore, the formula for the logical OR gate is:

[0092]

[0093] Among them, D1 is an intermediate event, which is connected to the basic events X1 and X2 through a logic OR gate;

[0094] When only one of the basic events X1 and X2 occurs, the corrected formula of the logic OR gate is:

[0095]

[0096] The formula for the logic AND gate is:

[0097]

[0098] Among them, D2 is an intermediate event, which is connected to the basic events X3 and X4 through a logic AND gate;

[0099] When only one of the basic events X3 and X4 occurs, the corrected formula of the logic AND gate is:

[0100]

[0101] Furthermore, the specific contents of S4 are as follows:

[0102] S41. Based on the results of the fault tree analysis, considering physical, economic and social factors, determine the multi-dimensional functional damage assessment indicators of large-span transportation hub structures;

[0103] S42. The Delphi method is used to weight the evaluation indicators. The opinions of experts on the questionnaire questions are collected anonymously multiple times, and feedback is provided. The opinions of experts are summarized, revised, and consulted several times. The consensus opinions of the experts are analyzed and summarized, and the weight coefficients of each evaluation indicator are finally determined.

[0104] S43. Use Monte Carlo simulation to generate earthquake intensity samples, conduct incremental dynamic time history analysis, record the structural response parameters under different earthquake scenarios, and form an engineering demand parameter matrix; judge the structural damage state based on the structural response under different earthquake scenarios, further calculate the functional loss index under each earthquake scenario, and use statistical analysis methods to establish the relationship between the functional loss index and the earthquake intensity and structural damage state.

[0105] Furthermore, the multi-dimensional functional damage assessment indicators of the large-span transportation hub structure in S41 include: primary indicators and secondary indicators; among them,

[0106] The first-level indicators include: physical damage, functional loss, and economic impact;

[0107] Secondary indicators include: structural damage, non-structural damage, economic losses, repair costs, number of people affected, and interruption time.

[0108] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0109] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for sensing the structural state of a large-span transportation hub in a high-intensity earthquake zone and assessing its post-earthquake functional loss, characterized by: The following steps are involved: S1. Obtain high-fidelity and low-fidelity dynamic response data of large-span transportation hub structures under earthquake action through shaking table tests and numerical simulations, respectively, and construct a multi-fidelity structural response dataset. S2. Build a data-physics-driven seismic response prediction model for large-span transportation hub structures by combining graph neural networks and long short-term memory network algorithms to determine the physical damage to the structures under earthquakes. S3. Use the fault tree analysis method to establish the correlation mechanism between physical damage and functional loss of the main structure and non-structural components of a large-span transportation hub under earthquake action; S4. Based on the Delphi method, the weight coefficients of functional damage assessment indicators are determined. Combined with incremental dynamic time history analysis and Monte Carlo simulation, a post-earthquake functional loss assessment model for large-span transportation hub structures is established.

2. The method for structural state perception and post-earthquake function loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 1 is characterized in that: The specific contents of S1 are: S11. Conduct shaking table tests on large-span transportation hub structures. Use accelerometers, displacement sensors, a DIC three-dimensional strain measurement system, and strain gauges to measure the acceleration, displacement, and strain responses of the test model. Install several GoPro sports cameras and video cameras with different viewing angles on the shaking table surface and main structure to capture images and videos of the overall and local responses of the test model. S12. Perform multi-source data fusion on the sensory data collected during the shaking table test. By performing data cleaning, spatiotemporal registration, feature extraction, and feature fusion on images, videos, and sensor data, high-fidelity data on the dynamic response of large-span transportation hub structures is ultimately generated. S13. Use finite element software to establish a numerical model of the main structure and non-structural components of a large-span transportation hub. Consider the non-structural components as additional mass of the main structure and perform finite element analysis on the main structure. The main structural response is regarded as the boundary condition of non-structural components, and finite element analysis is performed on non-structural components to generate low-fidelity data of the dynamic response of large-span transportation hub structures.

3. The method for structural state perception and post-earthquake function loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 2 is characterized in that: The specific contents of multi-source data fusion in S12 are as follows: Data cleaning: Adaptive median filtering algorithm is used to eliminate noise, and cubic spline interpolation method is used to fill missing data; Spatiotemporal registration: By adding timestamps to the collected data points, the temporal alignment of data with different sampling frequencies is achieved, and spatial interpolation technology is used to align the data spatially; Feature extraction: extract features from preprocessed multi-source data; Feature fusion: The principal component analysis method is used to fuse the extracted features to form high-fidelity data of the dynamic response of large-span transportation hub structures.

4. The method for structural state perception and post-earthquake function loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 3 is characterized in that: The specific contents of feature extraction include: Extracting displacement features from displacement sensor data; Extract the natural frequencies of the main structure and non-structural components from the acceleration sensor based on Fourier transform; The appearance damage features of main structures and non-structural components are extracted from images and videos based on the improved edge detection algorithm.

5. The method for structural state perception and post-earthquake function loss assessment of large-span transportation hubs in high-intensity earthquake zones according to claim 3 is characterized in that: The specific contents of S2 are: S21. Use the earthquake motion sequence, structural nodes, and unit-related characteristic parameters as input data, and the time-history dynamic response at the corresponding position as output data. Assemble the input and output as learning samples, and divide the learning samples into training set, validation set, and test set, which are used for training, parameter adjustment, and performance evaluation of the prediction model respectively. S22. Represent the structure as a graph, with the connection points of the structural components as nodes V of the graph, and the topological spatial relationships between the nodes as edges E of the graph. The discretized structure is a graph G = (V, E), and features related to the structural nodes and features related to the structural units are stored as nodes V and edges E of the graph, respectively. S23. The training data is input into multiple parallel graph convolutional layers, and the node representation is updated through the message passing mechanism. The long short-term memory layer controls the flow of temporal information through the input gate, forget gate, and output gate. The Adam optimizer is used to adjust the learning rate, and the graph convolutional layer and the long short-term memory layer are optimized using Dropout regularization. The model is trained, verified, and tested based on the learning samples, and finally a seismic response prediction model for large-span transportation hub structures is obtained.

6. The method for structural state perception and post-earthquake functional loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 5 is characterized in that: The specific contents of the structural nodes and unit-related characteristic parameters in S21 include: Structural nodes include: node coordinates, node mass, and frequency, mode, damping and combination of the structure; Unit-related features include: unit length, unit section parameters, unit material parameters and combinations.

7. The method for structural state perception and post-earthquake functional loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 1 is characterized in that: The specific contents of S3 are: Using the fault tree analysis method, the loss of transportation hub function is taken as the top-level event of the fault tree, and then decomposed layer by layer into intermediate events such as damage to the main structure and destruction of non-structural components. Combined with expert consultation and historical earthquake damage data, the basic events of the fault tree are identified, and the logical relationship between the basic events and intermediate events is represented by logical OR gates and logical AND gates.

8. The method for structural state perception and post-earthquake functional loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 7 is characterized in that: The formula for the logical OR gate is: Among them, D1 is an intermediate event, which is connected to the basic events X1 and X2 through a logic OR gate; When only one of the basic events X1 and X2 occurs, the corrected formula of the logic OR gate is: The formula for the logic AND gate is: Among them, D2 is an intermediate event, which is connected to the basic events X3 and X4 through a logic AND gate; When only one of the basic events X3 and X4 occurs, the corrected formula of the logic AND gate is:

9. The method for structural state perception and post-earthquake functional loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 1 is characterized in that: The specific contents of S4 are: S41. Based on the results of the fault tree analysis, considering physical, economic and social factors, determine the multi-dimensional functional damage assessment indicators of large-span transportation hub structures; S42. The Delphi method is used to weight the evaluation indicators. The opinions of experts on the questionnaire questions are collected anonymously multiple times, and feedback is provided. The opinions of experts are summarized, revised, and consulted several times. The consensus opinions of the experts are analyzed and summarized, and the weight coefficients of each evaluation indicator are finally determined. S43. Use Monte Carlo simulation to generate earthquake intensity samples, conduct incremental dynamic time history analysis, record structural response parameters under different earthquake scenarios, and form an engineering demand parameter matrix; The structural damage state is judged based on the structural response under different earthquake scenarios, and the functional loss index under each earthquake scenario is further calculated. The statistical analysis method is used to establish the relationship between the functional loss index and the earthquake intensity and structural damage state.

10. The method for structural state perception and post-earthquake functional loss assessment of a large-span transportation hub in a high-intensity earthquake zone according to claim 9 is characterized in that: The multi-dimensional functional damage assessment indicators of medium and large span transportation hub structures in S41 include: primary indicators and secondary indicators; among them, The first-level indicators include: physical damage, functional loss, and economic impact; Secondary indicators include: structural damage, non-structural damage, economic losses, repair costs, number of people affected, and interruption time.

Images