Genetic algorithm-based rapid evaluation method for traffic safety of railway bridge after earthquake

Abstract

The invention discloses a rapid evaluation method for traffic safety of a railway bridge after an earthquake based on a genetic algorithm. The method comprises the following steps: creating a track-bridge nonlinear finite element model; simulating high-frequency acceleration seismic oscillation, simulating low-frequency velocity pulses, converting high-frequency acceleration into high-frequency velocity, and performing superposition of high-frequency and low-frequency components to generate a seismic oscillation sample; performing bridge response calculation on the track-bridge nonlinear finite element models under different earthquake magnitudes, and extracting post-earthquake track residual deformation; taking post-earthquake track residual deformation as compensation of an initial irregularity spectrum, and adding the initial irregularity spectrum into a trigonometric series superposition method to obtain a post-earthquake track irregularity spectrum; calculating to obtain a post-earthquake driving condition, and storing the earthquake magnitude and the corresponding driving state; and building a post-earthquake traffic safety rapid evaluation platform based on a genetic algorithm, and evaluating whether the railway bridge can safely run after the earthquake. The technical effect of the invention is that technical support is provided for rapidly evaluating the traffic safety of the railway bridge after the earthquake and stopping inspection.

Description

Technical Field

[0001] This invention relates to the field of bridge engineering safety assessment technology, specifically a rapid assessment method for post-earthquake railway bridge traffic safety based on genetic algorithms. Background Technology

[0002] In high-speed train operation, comfort and safety are closely related to track smoothness. To ensure train safety, railway staff immediately halt all trains and conduct comprehensive track inspections during earthquakes. Trains can only resume operation after the inspection is completed. However, earthquakes cause widespread vibrations, making it difficult to determine the inspection scope and key areas. This results in lengthy manual inspections, increasing the difficulty of restoring railway operations and causing incalculable losses. To address this issue and improve the efficiency of post-earthquake track and bridge traffic assessment, this invention discloses a rapid post-earthquake railway bridge traffic safety assessment method based on a genetic algorithm. Summary of the Invention

[0003] The purpose of this invention is to provide a rapid assessment method for post-earthquake railway bridge traffic safety based on genetic algorithms, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the invention provides the following technical solution: a rapid assessment method for post-earthquake railway bridge traffic safety based on genetic algorithms, specifically including the following steps:

[0005] S1: Collect existing railway bridge data, group representative railway bridges along the entire line according to site category, bridge type, and seismic isolation measures, and create a track-bridge nonlinear finite element model.

[0006] S2: The generalized evolution spectrum model is used to simulate high-frequency acceleration ground motion, and Gabor wavelets are used to simulate low-frequency velocity pulses. The high-frequency acceleration is converted into high-frequency velocity, and the high and low frequency components are superimposed to generate one thousand ground motion samples of different magnitudes.

[0007] S3: Calculate the bridge response of the track-bridge nonlinear finite element model under different earthquake magnitudes; if the components in the bridge structure enter the plastic state, the track deformation problem will no longer be discussed; the residual track deformation after the earthquake will only be extracted when all components remain in the elastic state after the earthquake.

[0008] S4: Treat the post-earthquake track residual deformation as compensation for the initial irregularity spectrum and add it to the trigonometric series superposition method to obtain the post-earthquake track irregularity spectrum.

[0009] S5: Import the post-earthquake track irregularity spectrum into the train-track-bridge coupled model in Simpack general finite element method, calculate the post-earthquake traffic conditions, record the magnitude and corresponding traffic conditions, and establish a database for storage.

[0010] S6: Build a rapid post-earthquake traffic safety assessment platform based on genetic algorithms. Take the measured ground motion as input, define the ground motion error range, find the optimal ground motion sample that meets the requirements through genetic algorithms, and retrieve the corresponding traffic status in the database. Repeat this process multiple times to find the traffic status corresponding to multiple optimal ground motion samples, and then calculate the average traffic status of these optimal ground motion samples to assess whether the railway bridge can be safely operated after the earthquake.

[0011] Preferably, the existing railway bridge data specifically includes the foundation structure, pier construction and reinforcement, seismic isolation measures, track slab type, train type, design speed, site category, and design intensity information;

[0012] The railway bridges along the entire line are grouped into representative bridge groups, specifically by grouping bridge sections with the same site soil type and bridge type into the same group.

[0013] The nonlinear finite element model of the track-bridge is established using ANSYS or Abaqus.

[0014] Preferably, the formula for constructing the generalized evolutionary spectrum model is as follows:

[0015] (1)

[0016] In the formula: It is a completely non-stationary process; The power spectrum of the two-sided evolution of non-stationary ground motion acceleration; The frequency is far from the walking distance; The number of discrete points in the frequency domain; ,in , These are the lower cutoff frequency and the upper cutoff frequency, respectively. For in the interval The basic random variable follows a uniform distribution; where , Indicates by Another number obtained from the mapping is used to construct phase relationships or random phase distributions; To and The corresponding calculation frequency, The duration of the earthquake. For in the interval The constant in.

[0017] Preferably, the low-frequency velocity pulse adopts an empirical formula based on Gabor wavelets:

[0018] (2)

[0019] In the formula, pulse velocity time history Peak pulse velocity (PGV), pulse period Pulse cycle number Pulse peak time Pulse phase angle , This represents the duration of the earthquake.

[0020] Preferably, formula (2) in step S2 is used when calculating low-frequency velocity pulses:

[0021] If abundant on-site measured ground motion data is available, linear regression can be used to fit the data and obtain PGV that meets the site requirements. , The fitting formula;

[0022] If measured ground motion data is insufficient, the measured data is combined with empirical fitting regression formulas, and the data is considered as random parameters that satisfy a normal distribution to obtain pulse velocity time histories that meet the actual site conditions.

[0023] Number of pulse cycles Treating it as following a log-normal distribution, the pulse phase angle Assuming a normal distribution, the accuracy of Gabor wavelets is verified using the mean and standard deviation.

[0024] Preferably, the step of converting high-frequency acceleration into high-frequency velocity, superimposing high- and low-frequency components to generate one thousand ground motion samples of different magnitudes specifically includes the following steps:

[0025] First, assuming that arrive During the time interval, the near-fault ground motion acceleration is ;

[0026] Then, the high-frequency velocity is obtained by integrating the time history of the high-frequency acceleration. ;

[0027] Next, the residual velocity peak value is utilized. right Amplitude modulation is performed to obtain the standardized high-frequency velocity time history:

[0028] (3)

[0029] In the formula: It is the residual velocity peak value after pulse identification and extraction from measured ground motion records;

[0030] Standardized high-frequency speed time history With low-frequency velocity pulse time history The velocity time histories of near-fault pulse-type ground motions are obtained by superimposing the data to obtain the dimensionality-reduced simulation:

[0031] (4);

[0032] Finally, the high and low frequency superposition calculation was performed using the MATLAB toolbox, and the number of samples was set to 1000, so that 1000 ground motion time histories under different magnitudes could be obtained quickly.

[0033] Preferably, in step S3:

[0034] If the structure of a railway bridge suffers component damage under the action of a seismic motion sample of a certain magnitude, it is directly determined that the bridge is not suitable for traffic.

[0035] The impact of track deformation on train safety and comfort is only considered when all components are operating within their elastic range.

[0036] Preferably, the specific operation of step S4 is as follows:

[0037] Assuming the initial track irregularity is a stationary Gaussian process, this process can be generated by the power spectral density (PSD) of the high-speed railway, and can be expressed as:

[0038] (5)

[0039] In the formula: Spatial frequency, unit: , where parameters A and k are constant terms.

[0040] The rail irregularity sample generated using the trigonometric series superposition method can be represented as:

[0041] (6)

[0042] In the formula: Stationary random process, Power spectral density function, Spatial frequency step size, Spatial frequency, distance, Indicates in A random constant uniformly distributed within a range;

[0043] For residual track deformation, the lateral and vertical deformation of the rails are key factors affecting train operation. Therefore, residual track deformation is used as an additional direction of track irregularity for compensation, and only the lateral and vertical deformations are considered. Since only the components in the post-earthquake bridge structure need to operate within the elastic range, only the lateral track deformation needs to be compensated for, thus obtaining the post-earthquake track irregularity spectrum.

[0044] Preferably, in step S5, the post-earthquake track irregularity spectrum is imported into the train-track-bridge coupled model in Simpack general finite element method to calculate the post-earthquake traffic conditions, specifically:

[0045] The derailment coefficient, wheel unloading rate, and wheel-rail force are used as indicators of train operation safety, while acceleration and sperling are used as indicators of train ride comfort.

[0046] The critical threshold for determining the train derailment coefficient on a bridge is calculated using the following formula:

[0047] (7)

[0048] The critical threshold for determining the rate of wheel removal on a bridge is given by the following formula:

[0049] (8)

[0050] In the formula: This represents the lateral force between the wheel and rail. This represents the vertical force between the wheel and the rail. Wheel load reduction rate, Unloading rate;

[0051] The critical threshold for judging the wheel-rail force of a train on a bridge is 170 kN.

[0052] According to the "Design Specifications for High-Speed ​​Railways" (TB10621-2014), the critical value for vertical acceleration is 0.13g, and the critical value for lateral acceleration is 0.10g.

[0053] According to the "Design Specifications for High-Speed ​​Railways" (TB10621-2014), the critical threshold value for Sperling is 2.5.

[0054] Based on the above parameter thresholds, the results output by Simpack are compared to obtain the driving conditions under different magnitudes; the magnitude and driving conditions are stored together in the database, and the data under the magnitude where the components exhibit plastic deformation are also stored in the database.

[0055] Preferably, the step S6 of building a rapid post-earthquake driving safety assessment platform based on genetic algorithms specifically includes:

[0056] First, the input is determined to be measured ground motion data, and then this data is encoded. Next, the maximum number of generations T, population size, crossover probability, and mutation probability are set. Then, M individuals are randomly generated as the initial population P0, and the fitness function is determined and the fitness value is calculated. The ground motion samples obtained by the genetic algorithm meet the requirements of measured ground motion. Finally, based on this, the genetic operator is determined to obtain multiple optimal ground motion samples that meet the requirements of measured ground motion.

[0057] Compared with existing technologies, the beneficial effects of the invention are:

[0058] 1. This method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithms uses a high-low frequency superposition method to simulate earthquake motion. The advantage of this approach is that when high frequencies dominate, the characteristics of the artificially synthesized earthquake motion are closer to far-field motion; while when both high and low frequencies contribute, it is closer to near-field motion. This method ensures the diversity of earthquake motion samples, including both far-field and near-field motion samples, which better reflects actual earthquake conditions.

[0059] 2. This method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm generates a large number of seismic motion samples, ensuring the richness of the database. As a result, the mean value obtained is more in line with reality, which can effectively improve the accuracy of the assessment.

[0060] 3. This method for rapid post-earthquake railway bridge traffic safety assessment based on genetic algorithms involves screening and classifying the entire railway line, then calculating the track deformation of the bridge under different earthquake magnitudes in various seismic samples. A residual track irregularity spectrum is generated using the residual track deformation, and this spectrum is superimposed with the initial track irregularity spectrum to obtain the post-earthquake track irregularity spectrum. The post-earthquake track irregularity spectrum is imported into the general-purpose finite element software Simpack to obtain the safety index for train operation at the design speed after the earthquake, and the drivability status is correspondingly stored in a database with the number of seismic samples at different earthquake magnitudes. Finally, a rapid assessment platform is established using a genetic algorithm. First, the measured seismic motion is used as the input value; second, the seismic motion error range is determined; and finally, the genetic algorithm finds the post-earthquake train operation status of the seismic motion samples in the database that meet the requirements. The average post-earthquake traffic status of these seismic motion samples is used to assess the safety of railway bridge traffic under the measured seismic motion. This method combines random parameters with seismic motion and uses high- and low-frequency superposition to improve the diversity of seismic motion samples. Using the sample mean to assess post-earthquake traffic is more consistent with reality. This method provides technical support for rapidly assessing the safety of railway bridge traffic after an earthquake and for stopping and inspecting railway bridges, and solves the problems existing in the current technology. Attached Figure Description

[0061] Figure 1This is a schematic diagram illustrating the process of creating a nonlinear finite element model of a track-bridge according to the present invention;

[0062] Figure 2 This is a schematic diagram of the process for establishing seismic motion samples according to the present invention;

[0063] Figure 3 This is a schematic diagram of the process for establishing a post-earthquake traffic condition database according to the present invention;

[0064] Figure 4 This is a schematic diagram of the fast evaluation process based on genetic algorithms of the present invention;

[0065] Figure 5 This is a schematic diagram of the grouping process for the entire track bridge of the present invention;

[0066] Figure 6 This is a schematic diagram of the mean and standard deviation of U-SRM and SRM of the present invention;

[0067] Figure 7 This is a graph showing the target value and simulated value of the present invention;

[0068] Figure 8 This is a schematic diagram of the optimal solution for measured ground motion based on genetic algorithm according to the present invention. Detailed Implementation

[0069] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] like Figures 1 to 4 As shown, this invention provides a technical solution: a method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithms, specifically including the following steps:

[0071] S1: Collect existing railway bridge data, group representative railway bridges along the entire line according to site category, bridge type, and seismic isolation measures, and create a track-bridge nonlinear finite element model.

[0072] This embodiment aims to analyze the entire railway line. Therefore, it is necessary to collect basic data on existing railway bridges along the entire line. This data includes information on foundation structure, pier construction and reinforcement, seismic isolation measures, track slab type, train type, design speed, site category, and design intensity. Since a single design scheme cannot fully reflect the actual bridge information, it is necessary to combine construction data from each section with the actual operation of the railway bridges to collect basic data on existing railway bridges. This allows for a more comprehensive and accurate acquisition of bridge information, further improving the reliability of the analysis.

[0073] The entire railway bridge network was grouped into representative bridge groups. Specifically, existing railway bridge data was collected and organized, grouping bridge sections with similar site soil types and bridge types into the same group. This method allows for the division of bridges into multiple groups, facilitating the evaluation of bridge performance using empirical and analogical methods, thereby effectively reducing the complexity and workload of the evaluation. Detailed data collection and grouping processes are as follows: Figure 5 The diagram shows the grouping process for the entire track bridge.

[0074] Based on the collected information on existing railway bridges, a nonlinear finite element model of the track-bridge interface is created. Various finite element software programs can be selected during modeling, with the appropriate level of precision chosen based on requirements.

[0075] The nonlinear finite element model of the track-bridge is built using ANSYS or Abaqus. This is because the subsequent interaction with Simpack needs to be considered during the modeling process.

[0076] S2: The generalized evolution spectrum model is used to simulate high-frequency acceleration ground motion, and Gabor wavelets are used to simulate low-frequency velocity pulses. The high-frequency acceleration is converted into high-frequency velocity, and the high and low frequency components are superimposed to generate one thousand ground motion samples of different magnitudes.

[0077] In this embodiment, the formula for constructing the generalized evolutionary spectrum model (U-SRM) is as follows:

[0078] (1)

[0079] In the formula: It is a completely non-stationary process; The power spectrum of the two-sided evolution of non-stationary ground motion acceleration; The frequency is far from the walking distance; The number of discrete points in the frequency domain; ,in , These are the lower cutoff frequency and the upper cutoff frequency, respectively. For in the interval The basic random variable follows a uniform distribution; where , Indicates by Another number obtained from the mapping is used to construct phase relationships or random phase distributions; To and The corresponding calculation frequency, The duration of the earthquake. For in the interval The constant in.

[0080] The seismic motion simulation method shown in formula (1) can effectively improve the ability to obtain representative samples of seismic motion, thereby helping to reduce the number of sampling times. Figure 6 The mean and standard deviation plots of U-SRM and SRM shown below demonstrate that this method can quickly obtain representative samples.

[0081] In this embodiment, the low-frequency velocity pulse adopts an empirical formula based on Gabor wavelets:

[0082] (2)

[0083] In the formula, pulse velocity time history Peak pulse velocity (PGV), pulse period Pulse cycle number Pulse peak time Pulse phase angle , This represents the duration of the earthquake.

[0084] In this embodiment, when calculating low-frequency velocity pulses, it is important to note that if abundant on-site measured ground motion data is available, linear regression can be used to fit and obtain PGV values ​​that meet site requirements. , If the measured ground motion data is insufficient, a combination of limited measured data and an empirical fitting regression formula can be used, treating the data as random parameters that satisfy a normal distribution. This approach can still yield pulse velocity time histories that meet actual site conditions. (Pulse cycle number) and pulse phase angle Consider both log-normal and normal distributions respectively. Figure 7 The figure shows the target value and the simulated value. The accuracy of the Gabor wavelet was verified using the mean and standard deviation.

[0085] Since it is necessary to superimpose the high-frequency components and low-frequency components of the model, but the calculation result of the high-frequency components is the acceleration time history while the calculation result of the low-frequency components is the velocity time history, it is necessary to unify the high-frequency and low-frequency components.

[0086] In this embodiment, high-frequency acceleration is converted into high-frequency velocity, and the high and low frequency components are superimposed to generate one thousand ground motion samples of different magnitudes. The specific steps include:

[0087] First, assuming that arrive During the time interval, the near-fault ground motion acceleration is ;

[0088] Then, the high-frequency velocity is obtained by integrating the time history of the high-frequency acceleration. ;

[0089] Next, the residual velocity peak value is utilized. right Amplitude modulation is performed to obtain the standardized high-frequency velocity time history:

[0090] (3)

[0091] In the formula: It is the residual velocity peak value after pulse identification and extraction from measured ground motion records;

[0092] Standardized high-frequency speed time history With low-frequency velocity pulse time history The velocity time histories of near-fault pulse-type ground motions are obtained by superimposing the data to obtain the dimensionality-reduced simulation:

[0093] (4);

[0094] Finally, the high and low frequency superposition calculation was performed using the MATLAB toolbox, and the number of samples was set to 1000, so that 1000 ground motion time histories under different magnitudes could be obtained quickly.

[0095] S3: Calculate the bridge response of the track-bridge nonlinear finite element model under different earthquake magnitudes; if the components in the bridge structure enter the plastic state, the track deformation problem will no longer be discussed; the residual track deformation after the earthquake will only be extracted when all components remain in the elastic state after the earthquake.

[0096] In this embodiment, seismic motion samples are input into a nonlinear finite element model of the track-bridge system to perform seismic response analysis under earthquake loading, thereby obtaining the response of the bridge structure. It is worth noting that if the track-bridge structure experiences component damage under a seismic motion sample of a certain magnitude, it is directly determined that the train cannot run. Only when all components are operating within their elastic range are the effects of track deformation on train safety and comfort considered.

[0097] S4: Treat the post-earthquake track residual deformation as compensation for the initial irregularity spectrum and add it to the trigonometric series superposition method to obtain the post-earthquake track irregularity spectrum.

[0098] The specific operation is as follows: Assuming the initial track irregularity is a stationary Gaussian process, this process can be generated by the power spectral density (PSD) of the high-speed railway, which can be expressed as:

[0099] (5)

[0100] In the formula: Spatial frequency, unit: , where parameters A and k are constant terms.

[0101] The rail irregularity sample generated using the trigonometric series superposition method can be represented as:

[0102] (6)

[0103] In the formula: Stationary random process, Power spectral density function, Spatial frequency step size, Spatial frequency, distance, Indicates in A random constant uniformly distributed within a range;

[0104] For residual track deformation, the lateral and vertical deformation of the rails are key factors affecting train operation. Therefore, residual track deformation is used as an additional direction of track irregularity for compensation, and only the lateral and vertical deformations are considered. Since only the components in the post-earthquake bridge structure need to operate within the elastic range, only the lateral track deformation needs to be compensated for, thus obtaining the post-earthquake track irregularity spectrum.

[0105] S5: Import the post-earthquake track irregularity spectrum into the train-track-bridge coupled model in Simpack general finite element method, calculate the post-earthquake traffic conditions, record the magnitude and corresponding traffic conditions, and establish a database for storage.

[0106] In this embodiment, the post-earthquake track irregularity spectrum is imported into the train-track-bridge coupled model in Simpack general finite element method to calculate the post-earthquake traffic conditions, specifically:

[0107] The derailment coefficient, wheel unloading rate, and wheel-rail force are used as indicators of train operation safety, while acceleration and sperling are used as indicators of train ride comfort.

[0108] The critical threshold for determining the train derailment coefficient on a bridge is calculated using the following formula:

[0109] (7)

[0110] The critical threshold for determining the rate of wheel removal on a bridge is given by the following formula:

[0111] (8)

[0112] In the formula: This represents the lateral force between the wheel and rail. This represents the vertical force between the wheel and the rail. Wheel load reduction rate, Unloading rate;

[0113] The critical threshold for judging the wheel-rail force of a train on a bridge is 170 kN.

[0114] According to the "Design Specifications for High-Speed ​​Railways" (TB10621-2014), the critical value for vertical acceleration is 0.13g, and the critical value for lateral acceleration is 0.10g.

[0115] According to the "Design Specifications for High-Speed ​​Railways" (TB10621-2014), the critical threshold value for Sperling is 2.5.

[0116] Based on the above parameter thresholds, the results output by Simpack are compared to obtain the driving conditions under different magnitudes; the magnitude and driving conditions are stored together in the database, and the data under the magnitude where the components exhibit plastic deformation are also stored in the database.

[0117] S6: Build a rapid post-earthquake traffic safety assessment platform based on genetic algorithms. Take the measured ground motion as input, define the ground motion error range, find the optimal ground motion sample that meets the requirements through genetic algorithms, and retrieve the corresponding traffic status in the database. Repeat this process multiple times to find the traffic status corresponding to multiple optimal ground motion samples, and then calculate the average traffic status of these optimal ground motion samples to assess whether the railway bridge can be safely operated after the earthquake.

[0118] like Figure 8 As shown, in this embodiment, a rapid post-earthquake driving safety assessment platform based on genetic algorithms is built, specifically including:

[0119] First, the input is determined to be measured ground motion data, and then this data is encoded. Next, the maximum number of generations T, population size, crossover probability, and mutation probability are set. Then, M individuals are randomly generated as the initial population P0, and the fitness function is determined and the fitness value is calculated. The ground motion samples obtained by the genetic algorithm meet the requirements of measured ground motion. Finally, based on this, the genetic operator is determined to obtain multiple optimal ground motion samples that meet the requirements of measured ground motion.

[0120] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A rapid assessment method for post-earthquake railway bridge traffic safety based on genetic algorithms, characterized in that: Specifically, the following steps are included: S1: Collect existing railway bridge data, group representative railway bridges along the entire line according to site category, bridge type, and seismic isolation measures, and create a track-bridge nonlinear finite element model. S2: The generalized evolution spectrum model is used to simulate high-frequency acceleration ground motion, and Gabor wavelets are used to simulate low-frequency velocity pulses. The high-frequency acceleration is converted into high-frequency velocity, and the high and low frequency components are superimposed to generate one thousand ground motion samples of different magnitudes. S3: Calculate the bridge response using nonlinear finite element models of the track-bridge system under different earthquake magnitudes; if components in the bridge structure enter a plastic state, track deformation will no longer be discussed. Only when all components remain in an elastic state after the earthquake can the residual deformation of the track after the earthquake be extracted; S4: Treat the post-earthquake track residual deformation as compensation for the initial irregularity spectrum and add it to the trigonometric series superposition method to obtain the post-earthquake track irregularity spectrum. S5: Import the post-earthquake track irregularity spectrum into the train-track-bridge coupled model in Simpack general finite element method, calculate the post-earthquake traffic conditions, record the magnitude and corresponding traffic conditions, and establish a database for storage. S6: Build a rapid post-earthquake traffic safety assessment platform based on genetic algorithms. Take the measured ground motion as input, define the ground motion error range, find the optimal ground motion sample that meets the requirements through genetic algorithms, and retrieve the corresponding traffic status in the database. Repeat this process multiple times to find the traffic status corresponding to multiple optimal ground motion samples, and then calculate the average traffic status of these optimal ground motion samples to assess whether the railway bridge can be safely operated after the earthquake.

2. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: The existing railway bridge data specifically includes the foundation structure, pier construction and reinforcement, seismic isolation measures, track slab type, train type, design speed, site category, and design intensity information. The railway bridges along the entire line are grouped into representative bridge groups, specifically including grouping bridge sections with the same site soil type and bridge type into the same group; The nonlinear finite element model of the track-bridge is established using ANSYS or Abaqus.

3. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: The formula for constructing the generalized evolutionary spectrum model is as follows: （1） In the formula: It is a completely non-stationary process; The power spectrum of the two-sided evolution of non-stationary ground motion acceleration; The frequency is far from the walking distance; The number of discrete points in the frequency domain; ,in , These are the lower cutoff frequency and the upper cutoff frequency, respectively. In the interval The basic random variable follows a uniform distribution; where , Indicates by Another number obtained from the mapping is used to construct phase relationships or random phase distributions; To and The corresponding calculation frequency, The duration of the earthquake. In the interval The constant in.

4. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: The low-frequency velocity pulse is derived using an empirical formula based on Gabor wavelets: （2） Where: pulse velocity time history Peak pulse velocity (PGV), pulse period Pulse cycle number Pulse peak time Pulse phase angle , This represents the duration of the earthquake.

5. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 4, characterized in that: When calculating low-frequency velocity pulses, the formula (2) is as follows: If abundant on-site measured ground motion data is available, linear regression can be used to fit the data and obtain PGV that meets the site requirements. , The fitting formula; If measured ground motion data is insufficient, the measured data is combined with empirical fitting regression formulas, and the data is considered as random parameters that satisfy a normal distribution to obtain pulse velocity time histories that meet the actual site conditions. Number of pulse cycles Treating it as following a log-normal distribution, the pulse phase angle Assuming a normal distribution, the accuracy of Gabor wavelets is verified using the mean and standard deviation.

6. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: The process of converting high-frequency acceleration into high-frequency velocity, superimposing high- and low-frequency components to generate one thousand ground motion samples of different magnitudes specifically includes the following steps: First, assuming that arrive During the time interval, the near-fault ground motion acceleration is ; Then, the high-frequency velocity is obtained by integrating the time history of the high-frequency acceleration. ; Next, the residual velocity peak value is utilized. right Amplitude modulation is performed to obtain the standardized high-frequency velocity time history: （3） In the formula: It is the residual velocity peak value after pulse identification and extraction from measured ground motion records; Standardized high-frequency speed time history With low-frequency velocity pulse time history The velocity time histories of near-fault pulse-type ground motions are obtained by superimposing the data to obtain the dimensionality-reduced simulation: （4）； Finally, the high and low frequency superposition calculation was performed using the MATLAB toolbox, and the number of samples was set to 1000, so that 1000 ground motion time histories under different magnitudes could be obtained quickly.

7. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: In step S3: If the structure of a railway bridge suffers component damage under the action of a seismic motion sample of a certain magnitude, it is directly determined that the bridge is not suitable for traffic. The impact of track deformation on train safety and comfort is only considered when all components are operating within their elastic range.

8. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: The specific operation of step S4 is as follows: Assuming the initial track irregularity is a stationary Gaussian process, this process can be generated by the power spectral density (PSD) of the high-speed railway, and can be expressed as: （5） In the formula: Spatial frequency, unit: Parameters A and k are constant terms; The rail irregularity sample generated using the trigonometric series superposition method can be represented as: （6） In the formula: Stationary random process, Power spectral density function, Spatial frequency step size, Spatial frequency, distance, Indicates in A random constant uniformly distributed within a range; For residual track deformation, the lateral and vertical deformation of the rails are key factors affecting train operation. Therefore, residual track deformation is used as an additional direction of track irregularity for compensation, and only the lateral and vertical deformations are considered. Since only the components in the post-earthquake bridge structure need to operate within the elastic range, only the lateral track deformation needs to be compensated for, thus obtaining the post-earthquake track irregularity spectrum.

9. A rapid assessment method for post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 8, characterized in that: The post-earthquake track irregularity spectrum was imported into the train-track-bridge coupled model in Simpack general finite element method to calculate the post-earthquake traffic conditions, specifically: The derailment coefficient, wheel unloading rate, and wheel-rail force are used as indicators of train operation safety, while acceleration and sperling are used as indicators of train ride comfort. The critical threshold for determining the train derailment coefficient on a bridge is calculated using the following formula: （7） The critical threshold for determining the rate of wheel removal on a bridge is given by the following formula: （8） In the formula: This represents the lateral force between the wheel and rail. This represents the vertical force between the wheel and the rail. Wheel load reduction rate, Unloading rate; The critical threshold for judging the wheel-rail force of a train on a bridge is 170 kN. According to the "Design Specifications for High-Speed ​​Railways" (TB10621-2014), the critical value for vertical acceleration is 0.13g, and the critical value for lateral acceleration is 0.10g. According to the "Design Specifications for High-Speed ​​Railways" (TB10621-2014), the Sperling critical threshold is 2.

5. Based on the above parameter thresholds, the results output by Simpack are compared to obtain the driving conditions under different magnitudes; the magnitude and driving conditions are stored together in the database, and the data under the magnitude where the components exhibit plastic deformation are also stored in the database.

10. The method for rapid assessment of post-earthquake railway bridge traffic safety based on genetic algorithm as described in claim 1, characterized in that: Step S6, which describes building a rapid post-earthquake driving safety assessment platform based on genetic algorithms, specifically includes: First, the input is determined to be measured ground motion data, and then this data is encoded. Next, the maximum number of generations T, population size, crossover probability, and mutation probability are set. Then, M individuals are randomly generated as the initial population P0, and the fitness function is determined and the fitness value is calculated. The ground motion samples obtained by the genetic algorithm meet the requirements of measured ground motion. Finally, based on this, the genetic operator is determined to obtain multiple optimal ground motion samples that meet the requirements of measured ground motion.

Images