Method for predicting long-term rigidity of hogging moment area of steel-concrete composite beam bridge

Abstract

The invention relates to the technical field of hogging moment region calculation, in particular to a hogging moment region long-term rigidity prediction method for a steel-concrete composite beam bridge. The invention discloses a long-term rigidity prediction method for a hogging moment area of a steel-concrete composite beam bridge. Comprising the steps that a welding stud connecting piece degradation time-varying model and a concrete bridge deck degradation full-time-domain model under the corrosion-fatigue coupling effect are constructed according to test data, and a steel-concrete composite beam bridge hogging moment area rigidity degradation model, steel-concrete composite beam bridge hogging moment area long-term rigidity prediction and the like are constructed. In the process of predicting the rigidity of the hogging moment area of the steel-concrete composite bridge, the deterioration mechanism of a welding stud connector under the corrosion-fatigue coupling effect is considered, the pitting corrosion condition under the environment influence and the welding stud crack condition under the vehicle load influence are considered, the influence of bridge deck slab cracks is considered in the pitting corrosion condition, and the rigidity of the hogging moment area of the steel-concrete composite bridge is predicted. In addition, the mutual relation among the welding stud pitting corrosion influence, the welding stud crack influence and the bridge deck crack influence is considered, and the rigidity of the hogging moment area of the steel-concrete combined bridge is accurately analyzed.

Description

Technical Field

[0001] The invention relates to the technical field of negative moment zone calculation, and in particular to a method for predicting the long-term stiffness of a negative moment zone of a steel-concrete composite beam bridge. Background Art

[0002] Steel-concrete composite beam bridges are widely used in bridge engineering because they combine the advantages of steel and concrete. However, during service, the long-term stiffness degradation problem in the negative moment zone has become increasingly prominent, affecting the durability and safety of the bridge. Existing literature shows that traditional RC beams only consider defects such as steel corrosion, fatigue fracture, and steel-concrete bond degradation under the action of corrosion-fatigue coupling. The relevant defects can be better simulated by reducing the relevant parameters in the beam model. However, in steel-concrete composite beams, the stiffness in the negative moment zone is affected by the degradation response of multiple components such as the concrete bridge deck and steel-concrete connectors. Therefore, it is urgent to explore the relationship between the degradation degree of each component and the stiffness degradation of the composite beam to better predict the stiffness of the negative moment zone of the steel-concrete composite beam bridge. Summary of the invention

[0003] In the process of predicting the stiffness of the steel-concrete composite bridge in the negative bending zone, the present invention takes into account the degradation mechanism of the welded rivet connector under the action of corrosion-fatigue coupling, the pitting corrosion under the influence of the environment and the welded rivet cracks under the influence of vehicle loads, and the influence of the bridge deck cracks in the pitting corrosion, and considers the relationship between the influence of the welded rivet pitting, the influence of the welded rivet cracks and the influence of the bridge deck cracks, and accurately analyzes the stiffness of the steel-concrete composite bridge in the negative bending zone; the uncertainty of the key parameters of the steel-concrete composite bridge in the prediction process is also taken into account, and data-driven is performed through actual monitoring data, and the future stiffness change trend range is generated through Monte Carlo simulation, providing a more robust and accurate prediction result.

[0004] The present invention provides a method for predicting long-term stiffness in a negative bending moment zone of a steel-concrete composite beam bridge, comprising: Based on the test data, a time-varying model of welded stud connector degradation under the action of corrosion-fatigue coupling and a full-time domain model of concrete bridge deck degradation were constructed. The time-varying degradation model of welded nail connectors and the full-time domain degradation model of concrete bridge deck are coupled to obtain the stiffness degradation model of the negative moment zone of steel-concrete composite beam bridge. Based on the actual bridge monitoring data and the simulated bridge monitoring data extended by the Bayesian method, the upper and lower bounds of the key parameters of the steel-concrete composite bridge are determined through the maximum density interval, and Monte Carlo simulation is performed according to the upper and lower bounds of the key parameters of the steel-concrete composite bridge. Several key parameter analysis groups of the steel-concrete composite bridge are sampled and all the key parameter analysis groups of the steel-concrete composite bridge are sent to the stiffness degradation model of the negative moment zone of the steel-concrete composite beam bridge for calculation, and the stiffness change trend of the negative moment zone of the steel-concrete composite beam bridge is statistically analyzed to realize the long-term stiffness prediction of the negative moment zone of the steel-concrete composite beam bridge.

[0005] Preferably, the method comprises: constructing a time-varying model of degradation of a welding stud connector under the action of corrosion-fatigue coupling according to test data, which specifically comprises the following steps: Electrochemical corrosion test method and fatigue loading test are used to simulate the corrosion-fatigue coupling effect, obtain test data, and extract welding nail size, welding nail connection stiffness, pitting corrosion rate, welding nail crack growth rate, concrete cover thickness, chloride ion diffusion rate, concrete fatigue crack width, concrete fatigue crack width depth and load force and load cycle number from the test data; The mapping relationship between pitting corrosion rate and welding nail size, concrete cover thickness, chloride ion diffusion rate, welding nail crack width and concrete fatigue crack width and depth is numerically simulated to construct the first welding nail connection degradation time-varying model; The mapping relationship between the crack growth rate of the weld stud and the size of the weld stud, the thickness of the concrete cover, the load force and the number of load cycles is numerically simulated to construct a time-varying degradation model of the second weld stud connection. The first weld rivet connector degradation time-varying model and the second weld rivet connector degradation time-varying model are coupled and calculated to construct the weld rivet connector degradation time-varying model, and the variable parameters of the weld rivet connector degradation time-varying model are adjusted according to the mapping relationship between the weld rivet connector stiffness and the pitting corrosion rate and the weld rivet crack growth rate.

[0006] Preferably, a full-time domain model of concrete bridge deck degradation under the action of corrosion-fatigue coupling is constructed based on the test data, which specifically includes the following steps: Extract concrete bridge deck stiffness, steel bar corrosion degree, internal void distribution, bridge deck crack length and bridge deck crack width from test data; Before the welding nails break, the mapping relationship between the concrete bridge deck stiffness and the degree of steel corrosion, the internal void distribution, the bridge deck crack length and the bridge deck crack width is digitized to construct the first concrete bridge deck degradation full time domain model; After the welding nails did not break, the interface slip reduction coefficient was introduced to correct the first concrete bridge deck degradation full time domain model, and the second concrete bridge deck degradation full time domain model was constructed; The first concrete bridge deck degradation full time domain model and the second concrete bridge deck degradation full time domain model are coupled and calculated to construct a concrete bridge deck degradation full time domain model.

[0007] Preferably, based on the actual bridge monitoring data and the simulated bridge monitoring data extended by the Bayesian method, the upper and lower bounds of the key parameters of the steel-concrete composite bridge are determined by the maximum density interval, which specifically includes the following steps: For each key parameter of the steel-concrete composite bridge, the corresponding time series of the key parameters of the steel-concrete composite bridge is collected, and the Bayesian inference of the key parameters of the steel-concrete composite bridge is performed through the Bayesian method to construct the probability distribution corresponding to the key parameters of the steel-concrete composite bridge. Then, based on the probability distribution corresponding to the key parameters of the steel-concrete composite bridge, the maximum density interval with the confidence level μ of the key parameters of the steel-concrete composite bridge is calculated as the upper and lower bounds of the key parameters of the steel-concrete composite bridge.

[0008] Preferably, Monte Carlo simulation is performed according to the upper and lower bounds of the key parameters of the steel-concrete composite bridge to sample and obtain a number of key parameter analysis groups of the steel-concrete composite bridge, which specifically includes the following steps: The key parameters of steel-concrete composite bridges are divided into epistemic uncertain variables and random uncertain variables. For each epistemic uncertain variable, a bounded cumulative distribution function is constructed according to the upper and lower bounds corresponding to the epistemic uncertain variable and the probability distribution corresponding to the epistemic uncertain variable. Traverse the key parameters of the steel-concrete composite bridge. For each key parameter of the steel-concrete composite bridge, if the key parameter of the steel-concrete composite bridge is a cognitive uncertain variable, extract the key parameter value of the steel-concrete composite bridge from the bounded cumulative distribution function; if the key parameter of the steel-concrete composite bridge is a random uncertain variable, extract the key parameter value of the steel-concrete composite bridge from the probability distribution function; until the traversal of the key parameters of the steel-concrete composite bridge is completed, all the key parameter values ​​of the steel-concrete composite bridge are combined into a steel-concrete composite bridge key parameter analysis group; repeat the operation of traversing the key parameters of the steel-concrete composite bridge several times to construct several steel-concrete composite bridge key parameter analysis groups.

[0009] Preferably, the coupling calculation refers to loading a model for coupling calculation through a finite element analysis model and solving it through a finite element method.

[0010] Preferably, for the missing data in the time series of key parameters of steel-concrete composite bridges, a deep learning network is designed through the AGT platform to predict and supplement the missing time series detection data. The present invention has the following advantages: In the process of predicting the stiffness of the steel-concrete composite bridge in the negative bending zone, the present invention takes into account the degradation mechanism of the welded rivet connector under the action of corrosion-fatigue coupling, the pitting corrosion under the influence of the environment and the welded rivet cracks under the influence of vehicle loads, and the influence of the bridge deck cracks in the pitting corrosion, and considers the relationship between the influence of the welded rivet pitting, the influence of the welded rivet cracks and the influence of the bridge deck cracks, and accurately analyzes the stiffness of the steel-concrete composite bridge in the negative bending zone; the uncertainty of the key parameters of the steel-concrete composite bridge in the prediction process is also taken into account, and data-driven is performed through actual monitoring data, and the future stiffness change trend range is generated through Monte Carlo simulation, providing a more robust and accurate prediction result. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a schematic flow chart of the method for predicting the long-term stiffness in the negative moment zone of a steel-concrete composite beam bridge adopted in an embodiment of the present invention.

[0012] Figure 2 It is a schematic flow chart of the degradation failure process of the welding stud connector under the action of corrosion-fatigue coupling in an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0014] like Figure 1 As shown, a method for predicting the long-term stiffness of a steel-concrete composite beam bridge in the negative moment zone includes: Based on the test data, a time-varying model of welded nail connection degradation under the action of corrosion-fatigue coupling and a full-time domain model of concrete bridge deck degradation are constructed, which specifically includes the following steps: The time-varying model of welded nail connection degradation and the full-time domain model of concrete bridge deck degradation are coupled and calculated to obtain the stiffness degradation model of the negative moment zone of the steel-concrete composite beam bridge; the coupling calculation refers to the model loaded by the finite element analysis model for coupling calculation and solved by the finite element method; Based on the actual bridge monitoring data and the simulated bridge monitoring data extended by the Bayesian method, the upper and lower bounds of the key parameters of the steel-concrete composite bridge are determined through the maximum density interval, and the Monte Carlo simulation is performed according to the upper and lower bounds of the key parameters of the steel-concrete composite bridge. Several key parameter analysis groups of the steel-concrete composite bridge are sampled, and all the key parameter analysis groups of the steel-concrete composite bridge are sent to the stiffness degradation model of the negative moment zone of the steel-concrete composite beam bridge for calculation, and the stiffness change trend of the negative moment zone of the steel-concrete composite beam bridge is statistically analyzed to achieve the long-term stiffness prediction of the negative moment zone of the steel-concrete composite beam bridge; Based on the test data, a time-varying model of the degradation of welded stud connectors under the action of corrosion-fatigue coupling is constructed, which specifically includes the following steps: Electrochemical corrosion test method and fatigue loading test are used to simulate the corrosion-fatigue coupling effect, obtain test data, and extract welding nail size, welding nail connector stiffness, pitting corrosion rate, welding nail crack growth rate, concrete cover thickness, chloride ion diffusion rate, concrete fatigue crack width, concrete fatigue crack width depth and load force and load cycle number from the test data; concrete fatigue crack width and concrete fatigue crack width depth environment affect the degree of environmental pitting corrosion of welding nail connector; The mapping relationship between pitting corrosion rate and welding nail size, concrete cover thickness, chloride ion diffusion rate, welding nail crack width and concrete fatigue crack width and depth is numerically simulated to construct the first welding nail connection degradation time-varying model; The mapping relationship between the crack growth rate of the weld stud and the size of the weld stud, the thickness of the concrete cover, the load force and the number of load cycles is numerically simulated to construct a time-varying degradation model of the second weld stud connection. The first time-varying degradation model of the welding nail connector and the second time-varying degradation model of the welding nail connector are coupled and calculated to construct the time-varying degradation model of the welding nail connector. The variable parameters of the time-varying degradation model of the welding nail connector are adjusted according to the mapping relationship between the stiffness of the welding nail connector and the pitting corrosion rate and the welding nail crack growth rate. One form of the time-varying degradation model of the welding nail connector is: ,in is the stiffness of the welding stud connection corresponding to time t, is the initial stiffness of the welded rivet connector, α is the effect of pitting corrosion rate on the stiffness of the welded rivet connector, and β is the effect of welded rivet crack growth rate on the stiffness of the welded rivet connector; This application adopts an electrochemical accelerated corrosion test method to explore the entire process of corrosion degradation of welded nail connectors under service conditions, and uses an X-ray tomography test method to quantitatively observe the corrosion state, crack extension morphology and steel-concrete bonding state of the welded nails during the test to obtain test data. Considering that the stiffness of the negative bending moment zone of the steel-concrete composite beam bridge is determined by the stiffness of the welded nail connector and the stiffness of the bridge deck, the degradation process of the welded nail connector under the coupling of corrosion and fatigue is first explored, and further considering that the welded nail connector will experience pitting due to environmental influences during use, as well as cracks under vehicle cyclic loads, both of which will affect the stiffness of the welded nail connector, and the two influences are mutual. Reference Figure 2For degradation stage ①, the effect of fatigue load is manifested as cracks in the bridge deck. At this time, the weld nails will be affected by external factors and pitting will occur. The pitting situation can be analyzed based on the first weld nail connector degradation time-varying model; for degradation stage ②, considering the stress concentration effect after the formation of pitting pits, the generation and expansion of weld nail fatigue cracks are caused. The growth rate of pitting pits is calculated based on the first weld nail connector degradation time-varying model. At the same time, the expansion rate of weld nail fatigue cracks is calculated based on the second weld nail connector degradation time-varying model. The pitting pit depth reaches the critical value of crack nucleation as the threshold for the transition from pitting pit growth to weld nail fatigue crack expansion, and is used as the lower time boundary of this degradation stage in the failure model. For degradation stage ③, the failure time boundary of steel-concrete connection is determined based on the weld nail connector degradation time-varying model, that is, the weld nail fracture; there are two judgment criteria for the failure of weld nail connectors, one is the depth of pitting pits to the failure boundary, and the other is the weld nail crack reaching the failure boundary; Based on the test data, a full time domain model of concrete bridge deck degradation under the action of corrosion-fatigue coupling is constructed, which includes the following steps: Extract concrete bridge deck stiffness, steel bar corrosion degree, internal void distribution, bridge deck crack length and bridge deck crack width from test data; Before the welding nails break, the mapping relationship between the concrete bridge deck stiffness and the degree of steel corrosion, the internal void distribution, the bridge deck crack length and the bridge deck crack width is digitized to construct the first concrete bridge deck degradation full time domain model; After the welding nails did not break, the interface slip reduction coefficient was introduced to correct the first concrete bridge deck degradation full time domain model, and the second concrete bridge deck degradation full time domain model was constructed; The first concrete bridge deck degradation full time domain model and the second concrete bridge deck degradation full time domain model are coupled and calculated to construct a concrete bridge deck degradation full time domain model.

[0015] When analyzing the degradation of concrete bridge decks, it is considered that after cracking, the stiffness of the welded rivet connector drops to a critical value, which will lead to a sharp increase in the slip at the welded rivet-concrete interface. On this basis, the correlation coefficient between the stiffness drop in the negative bending moment zone of the steel-concrete composite beam bridge and the interface slip is calculated by the finite element method, and is used as the interface slip reduction coefficient to correct the full time domain model of the first concrete bridge deck degradation.

[0016] Based on the actual bridge monitoring data and the simulated bridge monitoring data extended by the Bayesian method, the upper and lower bounds of the key parameters of the steel-concrete composite bridge are determined through the maximum density interval, which specifically includes the following steps: For each key parameter of the steel-concrete composite bridge, the key parameters of the steel-concrete composite bridge here include the initial stiffness of the bridge deck, the initial stiffness of the welded nail connector, the vehicle load, the chloride ion diffusion rate, etc., collect the corresponding time series of the key parameters of the steel-concrete composite bridge, and use the Bayesian method to perform Bayesian inference on the time series of the key parameters of the steel-concrete composite bridge to construct the probability distribution corresponding to the key parameters of the steel-concrete composite bridge. Then, based on the probability distribution corresponding to the key parameters of the steel-concrete composite bridge, the maximum density interval of the confidence level of the key parameters of the steel-concrete composite bridge is calculated. Here, μ is generally 95%, which serves as the upper and lower bounds of the key parameters of the steel-concrete composite bridge. It should be noted that the time series of the key parameters of the steel-concrete composite bridge here is obtained in combination with the actual design requirements of the bridge project and the actual bridge site. For the missing data in the time series of the key parameters of the steel-concrete composite bridge, a deep learning network can be designed through the AGT platform to predict and supplement the missing time series detection data. Monte Carlo simulation is performed according to the upper and lower bounds of the key parameters of the steel-concrete composite bridge, and several key parameter analysis groups of the steel-concrete composite bridge are obtained by sampling, which specifically includes the following steps: The key parameters of the steel-concrete composite bridge are divided into epistemic uncertain variables and random uncertain variables. For each epistemic uncertain variable, a bounded cumulative distribution function is constructed according to the upper and lower bounds corresponding to the epistemic uncertain variable and the probability distribution corresponding to the epistemic uncertain variable. The bounded cumulative distribution function is generally F(x)=P(X≤x), which represents the probability of X appearing before a certain value x. It should be noted that the epistemic uncertain variables here are limited understanding of the system, resulting in uncertain value ranges of some parameters. For example, the initial stiffness of the bridge deck and the initial stiffness of the welded nail connector have an error of ±5% under different construction quality. Random uncertain variables are random fluctuations that exist in nature or during use. For example, vehicle loads are random and different every day. Traverse the key parameters of the steel-concrete composite bridge. For each key parameter of the steel-concrete composite bridge, if the key parameter of the steel-concrete composite bridge is a cognitive uncertain variable, extract the key parameter value of the steel-concrete composite bridge from the bounded cumulative distribution function; if the key parameter of the steel-concrete composite bridge is a random uncertain variable, extract the key parameter value of the steel-concrete composite bridge from the probability distribution function. Here, the probability distribution function can be a normal distribution or a log-normal distribution, and the vehicle load generally adopts a normal distribution; until the key parameters of the steel-concrete composite bridge are traversed, all the key parameter values ​​of the steel-concrete composite bridge are combined into a steel-concrete composite bridge key parameter analysis group; repeat if The key parameter operations of steel-concrete composite bridges are traversed several times to construct several key parameter analysis groups of steel-concrete composite bridges; the key parameter analysis groups of steel-concrete composite bridges generated here are sent one by one to the stiffness degradation model of the negative moment zone of the steel-concrete composite beam bridge for calculation. For example, the change of stiffness in the negative moment zone after 20 years is calculated, and several groups of stiffness changes in the negative moment zone of the steel-concrete composite beam bridges can be obtained, such as a 30% decrease in stiffness and a 25% decrease in stiffness. The probability distribution statistics of the stiffness changes in the negative moment zone of all steel-concrete composite beam bridges are performed to obtain a 95% confidence range: a stiffness decrease of 30%-40%, which can provide more robust and accurate prediction results.

[0017] In the process of predicting the stiffness of the steel-concrete composite bridge in the negative bending zone, the present application takes into account the degradation mechanism of the weld connector under the action of corrosion-fatigue coupling, the pitting corrosion under the influence of the environment and the weld cracks under the influence of vehicle loads, and the influence of bridge deck cracks in the pitting corrosion, and considers the relationship between the influence of weld pitting corrosion, weld cracks and bridge deck cracks, and accurately analyzes the stiffness of the steel-concrete composite bridge in the negative bending zone; it also takes into account the uncertainty of key parameters of the steel-concrete composite bridge in the prediction process, drives the data through actual monitoring data, and generates the future stiffness change trend range through Monte Carlo simulation, providing more robust and accurate prediction results.

[0018] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. A method for predicting the long-term stiffness of a steel-concrete composite beam bridge in the negative bending moment zone, characterized in that: include: Based on the test data, a time-varying model of welded connection degradation under the action of corrosion-fatigue coupling and a full-time domain model of concrete bridge deck degradation were constructed; The time-varying degradation model of welded nail connectors and the full-time domain degradation model of concrete bridge deck are coupled to obtain the stiffness degradation model of the negative moment zone of steel-concrete composite beam bridge. Based on the actual bridge monitoring data and the simulated bridge monitoring data extended by the Bayesian method, the upper and lower bounds of the key parameters of the steel-concrete composite bridge are determined through the maximum density interval, and Monte Carlo simulation is performed according to the upper and lower bounds of the key parameters of the steel-concrete composite bridge. Several key parameter analysis groups of the steel-concrete composite bridge are sampled and all the key parameter analysis groups of the steel-concrete composite bridge are sent to the stiffness degradation model of the negative moment zone of the steel-concrete composite beam bridge for calculation, and the stiffness change trend of the negative moment zone of the steel-concrete composite beam bridge is statistically analyzed to realize the long-term stiffness prediction of the negative moment zone of the steel-concrete composite beam bridge.

2. The method for predicting long-term stiffness in negative moment zone of a steel-concrete composite beam bridge according to claim 1 is characterized in that: include: Based on the test data, a time-varying model of the degradation of welded stud connectors under the action of corrosion-fatigue coupling is constructed, which specifically includes the following steps: Electrochemical corrosion test method and fatigue loading test are used to simulate the corrosion-fatigue coupling effect, obtain test data, and extract welding nail size, welding nail connection stiffness, pitting corrosion rate, welding nail crack growth rate, concrete cover thickness, chloride ion diffusion rate, concrete fatigue crack width, concrete fatigue crack width depth and load force and load cycle number from the test data; The mapping relationship between pitting corrosion rate and welding nail size, concrete cover thickness, chloride ion diffusion rate, welding nail crack width and concrete fatigue crack width and depth is numerically simulated to construct the first welding nail connection degradation time-varying model; The mapping relationship between the crack growth rate of the weld stud and the size of the weld stud, the thickness of the concrete cover, the load force and the number of load cycles was numerically simulated to construct a time-varying degradation model of the second weld stud connection. The first weld rivet connector degradation time-varying model and the second weld rivet connector degradation time-varying model are coupled and calculated to construct the weld rivet connector degradation time-varying model, and the variable parameters of the weld rivet connector degradation time-varying model are adjusted according to the mapping relationship between the weld rivet connector stiffness and the pitting corrosion rate and the weld rivet crack growth rate.

3. The method for predicting long-term stiffness in negative moment zone of a steel-concrete composite beam bridge according to claim 2 is characterized in that: Based on the test data, a full time domain model of concrete bridge deck degradation under the action of corrosion-fatigue coupling is constructed, which includes the following steps: Extract concrete bridge deck stiffness, steel bar corrosion degree, internal void distribution, bridge deck crack length and bridge deck crack width from test data; Before the welding nails break, the mapping relationship between the concrete bridge deck stiffness and the degree of steel corrosion, the internal void distribution, the bridge deck crack length and the bridge deck crack width is digitized to construct the first concrete bridge deck degradation full time domain model; After the welding nails did not break, the interface slip reduction coefficient was introduced to correct the first concrete bridge deck degradation full time domain model, and the second concrete bridge deck degradation full time domain model was constructed; The first concrete bridge deck degradation full time domain model and the second concrete bridge deck degradation full time domain model are coupled and calculated to construct a concrete bridge deck degradation full time domain model.

4. The method for predicting long-term stiffness in negative moment zone of a steel-concrete composite beam bridge according to claim 3 is characterized in that: Based on the actual bridge monitoring data and the simulated bridge monitoring data extended by the Bayesian method, the upper and lower bounds of the key parameters of the steel-concrete composite bridge are determined through the maximum density interval, which specifically includes the following steps: For each key parameter of the steel-concrete composite bridge, the corresponding time series of the key parameters of the steel-concrete composite bridge is collected, and the Bayesian inference of the key parameters of the steel-concrete composite bridge is performed through the Bayesian method to construct the probability distribution corresponding to the key parameters of the steel-concrete composite bridge. Then, based on the probability distribution corresponding to the key parameters of the steel-concrete composite bridge, the maximum density interval with the confidence level μ of the key parameters of the steel-concrete composite bridge is calculated as the upper and lower bounds of the key parameters of the steel-concrete composite bridge.

5. The method for predicting long-term stiffness in negative moment zone of steel-concrete composite beam bridge according to claim 4 is characterized in that: Monte Carlo simulation is performed according to the upper and lower bounds of the key parameters of the steel-concrete composite bridge, and several key parameter analysis groups of the steel-concrete composite bridge are sampled, which specifically includes the following steps: The key parameters of steel-concrete composite bridges are divided into epistemic uncertain variables and random uncertain variables. For each epistemic uncertain variable, a bounded cumulative distribution function is constructed according to the upper and lower bounds corresponding to the epistemic uncertain variable and the probability distribution corresponding to the epistemic uncertain variable. Traverse the key parameters of the steel-concrete composite bridge. For each key parameter of the steel-concrete composite bridge, if the key parameter of the steel-concrete composite bridge is a cognitive uncertain variable, extract the key parameter value of the steel-concrete composite bridge from the bounded cumulative distribution function; if the key parameter of the steel-concrete composite bridge is a random uncertain variable, extract the key parameter value of the steel-concrete composite bridge from the probability distribution function; until the traversal of the key parameters of the steel-concrete composite bridge is completed, all the key parameter values ​​of the steel-concrete composite bridge are combined into a steel-concrete composite bridge key parameter analysis group; repeat the operation of traversing the key parameters of the steel-concrete composite bridge several times to construct several steel-concrete composite bridge key parameter analysis groups.

6. The method for predicting long-term stiffness in negative moment zone of steel-concrete composite beam bridge according to claim 5, characterized in that: Coupling calculation refers to the model that is loaded through the finite element analysis model for coupling calculation and solved by the finite element method.

7. The method for predicting long-term stiffness in negative moment zone of steel-concrete composite beam bridge according to claim 6 is characterized in that: Aiming at the missing data in the time series of key parameters of steel-concrete composite bridges, a deep learning network was designed through the AGT platform to predict and supplement the missing time series detection data.

Images