Rail transit vibration reduction reliability evaluation method, device, equipment and medium

Abstract

The invention relates to the field of rail transit, and particularly discloses a rail transit vibration reduction reliability evaluation method and device, equipment and a medium, and the reliability evaluation method comprises the steps: obtaining a random factor and a time-varying factor which influence environment vibration; building a'parameter-response 'basic database based on the dynamic simulation model; based on an SE-U-Net neural network model, constructing an expanded database of a'parameter-response 'basic database sample; and reliability prediction and evaluation of rail transit vibration reduction are carried out based on a probability density evolution algorithm. According to the method, the comprehensive effect of random factors and time-varying factors which influence the rail transit vibration damping performance can be reflected, the influence of rail operation condition evolution on the vibration damping reliability is evaluated without depending on field testing, the influence degree of various factors on the reliability can be determined, the reliability evaluation time is shortened, the economic cost is reduced, and meanwhile the reliability evaluation efficiency is improved. The applicability of reliability evaluation is expanded, the accuracy of reliability evaluation is improved, and the vibration reduction design of rail transit can be accurately guided.

Description

Technical Field

[0001] This application relates to the field of rail transit, and in particular to a method, apparatus, equipment and medium for evaluating the reliability of vibration reduction in rail transit. Background Technology

[0002] With the rapid development of my country's rail transit industry, environmental vibration problems generated by urban rail transit have become a significant factor affecting residents' normal production and lives. To reduce environmental vibration and promote green development of the urban ecological environment, my country currently primarily uses Z-level vibration and its insertion loss to evaluate the vibration reduction effect of track vibration reduction measures during the vibration reduction design stage. In the current vibration reduction design, prediction, and evaluation system, the predicted results of Z-level vibration and its insertion loss are considered constant values. However, in actual evaluation of rail transit vibration reduction, operating conditions exhibit short-term randomness and long-term time variability. Specifically, these include random and time-varying factors such as train operating speed and axle load, wheel-rail irregularities and their development, and ground dynamic parameters. This results in the actual Z-level vibration and its insertion loss having a certain random distribution characteristic and evolving over time, rather than being a fixed value. Therefore, using constant values ​​to evaluate vibration reduction effectiveness is not practical and often leads to problems such as design compliance but engineering non-compliance, or design compliance but excessive vibration reduction in engineering, resulting in mismatches between design and engineering. This makes accurate prediction and design of vibration reduction impossible, leading to substandard vibration reduction designs or significant waste of engineering investment. Therefore, establishing a vibration reduction reliability prediction and evaluation method that can consider the probability distribution characteristics of multiple random factors and the evolution characteristics of multiple time-varying factors has become an urgent problem to be solved.

[0003] The published invention patent, "A Reliability Design Method for Track Vibration Reduction Measures" (CN...

[0004] Patent 117807672A proposes a track vibration reduction reliability design framework, which, based on the total probability formula, can calculate the probability density function under various factors affecting track vibration reduction effectiveness. Patent 116245278A, titled "A Vibration Control Performance Evaluation Method for the Whole Life Cycle of Metro Vibration-Reduced Tracks," considers long-term track service conditions such as track irregularity evolution and proposes a monitoring and evaluation system for the long-term service performance of vibration reduction measures based on the maximum Z-level vibration and its insertion loss. However, existing technical knowledge only preliminarily proposes a track vibration reduction reliability design framework or establishes a monitoring and evaluation system for the performance of vibration reduction measures under certain long-term service conditions, and still has the following shortcomings:

[0005] First, the existing track vibration reduction reliability design framework can only realize the reliability assessment of track vibration reduction effect under the influence of a single random factor. It cannot realize the combined effect of multiple random factors, nor does it consider the influence of time-varying factors, resulting in low reliability design accuracy and failing to accurately reflect the reliability of track transit on-site operation.

[0006] Second, while the existing continuous monitoring system for track vibration reduction measures can take into account the effects of time-varying factors such as the evolution of random irregularities and changes in train operating conditions, it is only applicable to the reliability assessment of a certain vibration reduction measure under long-term service conditions in a certain section. It cannot take into account the impact of the randomness of ground dynamic parameters or other random factors on reliability, and it relies on long-term real-time monitoring of the track, which consumes a lot of manpower and material resources and is difficult to promote.

[0007] Meanwhile, existing patents related to rail transit reliability only propose a reliability design approach based on the full probability formula, which can only achieve reliability design under the influence of a single factor. It cannot take into account the combined effects of various random and time-varying factors in the short-term and long-term operation phases of rail transit, and cannot obtain an accurate reliability design range for the actual operation of the line, nor can it accurately meet the actual requirements of vibration reduction on site.

[0008] Furthermore, existing technologies rely on long-term, continuous monitoring of track vibration response to determine the impact of time-varying factors on track vibration reduction reliability. This not only consumes significant human and material resources but also makes the method difficult to promote. Moreover, due to the complex testing environment on-site, it is difficult to determine the individual impacts of various random and time-varying factors on the reliability of vibration reduction, which is not conducive to guiding track vibration reduction design from a practical, multi-factor perspective.

[0009] Traditional dynamic calculations rely on solving complex dynamic equations, which is time-consuming and computationally expensive when dealing with a large number of combinations of random and time-varying factors required by the "parameter-response" database.

[0010] Based on this, the present invention develops an intelligent evaluation method for vibration reduction reliability in rail transit. Summary of the Invention

[0011] To reduce the time and economic cost of evaluating the reliability of vibration reduction in rail transit and improve the accuracy of predicting the reliability of vibration reduction in rail transit, this application provides a method, apparatus, equipment, and medium for evaluating the reliability of vibration reduction in rail transit. This method significantly reduces the time and economic cost of evaluating the reliability of vibration reduction in rail transit and improves the accuracy of predicting the reliability of vibration reduction in rail transit.

[0012] Firstly, this application provides a reliability evaluation method for vibration reduction in rail transit, which adopts the following technical solution:

[0013] A reliability evaluation method for vibration reduction in rail transit includes the following steps:

[0014] S1. Obtain the random and time-varying factors affecting environmental vibration;

[0015] S2. Build a basic database of "parameter-response" based on the dynamic simulation model;

[0016] S3. Construct an expanded database of "parameter-response" basic database samples based on the SE-U-Net neural network model;

[0017] S4. Reliability prediction and evaluation of vibration reduction in rail transit based on probability density evolution algorithm.

[0018] Preferably, step S1 includes the following steps:

[0019] S11, Obtain random factors;

[0020] Random factors include geological conditions, wheel-rail irregularities, and train operating conditions;

[0021] S12. Obtain time-varying factors;

[0022] Among them, time-varying factors include track irregularities that evolve over time, train operating conditions that have a time periodicity, and the deterioration of track elastic materials over long-term operation.

[0023] Preferably, the geological conditions include the dynamic modulus of the soil layer and the damping coefficient of the soil layer;

[0024] The wheel-rail irregularities include long-wave track irregularities, rail roughness, and wheel non-roundness.

[0025] The train operating conditions include train speed and axle load.

[0026] Preferably, step S2 includes the following steps:

[0027] S21. Select a dynamic simulation model;

[0028] S22. Input random and time-varying factors into the dynamic simulation model to calculate the vibration response parameters under different vibration reduction key parameters and different combinations of random and time-varying factors under different characteristic parameter conditions.

[0029] S23. Build a basic database of "parameter-response" based on vibration response parameters.

[0030] Preferably, step S3 includes the following steps:

[0031] S31. Select the SE-U-Net neural network model;

[0032] S32. Input the sample parameters from the “parameter-response” basic database built in step S2 into the SE-U-Net neural network model, output the supplementary samples, and store them in the “parameter-response” basic database to obtain the supplementary database, thereby enriching the samples in the “parameter-response” basic database.

[0033] Preferably, step S4 includes the following steps:

[0034] The formula for evaluating the reliability of track vibration reduction is:

[0035]

[0036] The probability density function of X(t) is:

[0037]

[0038] In the formula, X is the target physical quantity. These are the random parameters involved. It is x and The joint probability density function, equation (1) is a connection The joint probability density function and the corresponding velocity The one-dimensional convection probability density function;

[0039] The initial and boundary conditions for the probability density evolution problem are as follows:

[0040]

[0041] In the formula, δ is the Dirac function; P l X0 is the initial assigned probability, and X0 is the structural response at the initial moment.

[0042] The steps to solve the generalized probability density evolution equation are as follows:

[0043] ① In the random parameter space Ω Θ Select representative point set Where s is the number of multiple random variables, N sel The total number of points in the set;

[0044] ② For each representative point Solving for speed

[0045] ③ Obtained in step ② Substituting the discrete scheme into equation (3), and then solving it using the bilateral difference method under the initial conditions, the discrete space should be divided into a grid, and the grid nodes are denoted as (x i ,t k ), i = 0, ±1, ±2, ..., k = 0, 1, 2, ..., where x i =iΔx,t k =kΔt, k = 0, 1, 2, ..., Δx is the spatial step size in the x-direction, and Δt is the time step size. Thus, equation (1) is transformed into a system of algebraic equations, which can be solved to give the probability density values ​​at the nodes, i.e., p. XΘ (x,t);

[0046] ④. Combining the results of step ③, obtain the probability density:

[0047]

[0048] Secondly, the rail transit vibration reduction reliability evaluation device provided in this application adopts the following technical solution:

[0049] A reliability evaluation device for vibration reduction in rail transit includes:

[0050] The acquisition module is used to acquire random and time-varying factors that affect environmental vibration;

[0051] The database module is used to build a basic "parameter-response" database based on the dynamic simulation model;

[0052] An expanded database module is used to construct an expanded database of "parameter-response" basic database samples based on the SE-U-Net neural network model;

[0053] The track vibration reduction reliability evaluation module is used for reliability prediction and assessment of track traffic vibration reduction based on the probability density evolution algorithm.

[0054] Thirdly, this application provides an electronic device that adopts the following technical solution:

[0055] An electronic device comprising:

[0056] One or more processors;

[0057] Memory;

[0058] One or more applications, wherein the applications are stored in memory and configured to be executed by one or more processors, the applications being configured to: execute a rail transit vibration reduction reliability evaluation method as shown in any possible implementation of the first aspect.

[0059] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0060] A computer-readable storage medium includes: a computer program that can be loaded by a processor and executed to implement a rail transit vibration reduction reliability evaluation method as shown in any possible implementation of the first aspect.

[0061] In summary, this application includes the following beneficial technical effects:

[0062] 1. This application can reflect the combined effects of random and time-varying factors affecting the vibration reduction performance of rail transit, and does not rely on field tests to evaluate the impact of the evolution of rail operating conditions on vibration reduction reliability. It can clarify the degree of influence of various factors on reliability, while reducing the time and economic cost of reliability evaluation, expanding the applicability of reliability evaluation, improving the accuracy of reliability evaluation, and helping to accurately guide the vibration reduction design of rail transit.

[0063] 2. This application expands the "parameter-response" basic database by utilizing the SE-U-Net neural network model, thereby increasing the richness of the samples and improving the accuracy of reliability prediction and evaluation.

[0064] 3. This application enriches the various factors affecting track vibration reduction by providing a variety of random and time-varying factors, which is conducive to the subsequent formulation of better vibration reduction schemes for rail transit, so as to reduce the impact of rail transit vibration reduction on traffic. Attached Figure Description

[0065] Figure 1 This is a system diagram of the reliability evaluation method in the embodiments of this application.

[0066] Figure 2 This is the probability density function curve of the dynamic parameters of the soil layer beneath a tunnel in step S2 of this embodiment.

[0067] Figure 3 This is a schematic diagram of the vehicle-track-tunnel coupled vibration physical model in step S2 of this application embodiment.

[0068] Figure 4 This is a schematic diagram of the SE-U-Net model structure in step S3 of the embodiments of this application.

[0069] Figure 5 'a' is a graph of the training loss function in the embodiments of this application. Figure 5 b is a scatter plot of the predicted values ​​in the embodiments of this application. Figure 5 c is a comparison of the source strength time-domain results of the dynamic model and the SE-U-Net model in the embodiments of this application.

[0070] Figure 6 a is a schematic diagram of the probability density distribution of the maximum Z-level vibration of the tunnel source strength in an embodiment of this application. Figure 6 b is the reliability evolution curve of the maximum Z vibration level of the source strength 85 / 90dB over time in the embodiment of this application. Detailed Implementation

[0071] The present application will be further described in detail below with reference to the accompanying drawings.

[0072] This application discloses a method for evaluating the reliability of vibration reduction in rail transit.

[0073] Reference Figure 1 The reliability evaluation method for vibration reduction in rail transit includes the following steps:

[0074] S1. Obtain the random and time-varying factors affecting environmental vibration;

[0075] S11, Obtain random factors;

[0076] Specifically, the key random factors that commonly affect environmental vibration in rail transit operations usually include geological conditions, wheel-rail irregularities, and train operating conditions. Among these, geological conditions include the dynamic modulus of the soil and the damping coefficient of the soil; wheel-rail irregularities include long-wave irregularities of the track, rail roughness, and wheel non-roundness; and train operating conditions include train speed and axle load.

[0077] When evaluating the reliability of vibration reduction in rail transit, it is necessary to select random factors that affect the vibration reduction effect of rail transit based on the actual vibration reduction evaluation requirements of the project, and determine the influence mode and probability distribution characteristics of each factor.

[0078] The following example illustrates the random variation in axle load of a train. Axle load is the maximum total mass of a train vehicle borne by each axle, and its magnitude depends on the vehicle's tare weight and passenger capacity, with the passenger capacity being random. Variations in axle load not only change the total mass of the train but also affect the track stiffness, which is controlled by the load variation effect of the track's nonlinear elastic elements, thus introducing uncertainty into the vibration reduction effect.

[0079] Axle load is selected as a random factor, and its random distribution range is between the empty load and the maximum load. The accurate distribution characteristics within this range need to be determined by combining the relevant design specifications of rail transit and the passenger volume statistics of the target line.

[0080] S12. Obtain time-varying factors;

[0081] Among them, time-varying factors include track irregularities that evolve over time, train operating conditions that have a time periodicity, and the deterioration of track elastic materials over long-term operation.

[0082] Specifically, based on the actual needs of evaluating the reliability of rail transit vibration reduction, time-varying factors affecting the vibration reduction effect of rail transit should be selected, and the influence mode and time evolution characteristics of time-varying factors should be determined.

[0083] The following example illustrates the evolution of track irregularities over time. Under the long-term effects of wheel and rail wear, track irregularities will undergo significant changes over operating time, thereby altering wheel and rail excitation and causing the vibration reduction effect to evolve over time.

[0084] Track irregularities are selected as a time-varying factor. Their variation characteristics are determined by time factors such as daily train operation and rail grinding cycle. Their evolution conditions at each time point should be determined by combining the relevant design specifications of rail transit and the track irregularity measurement values ​​of the target line at the corresponding period.

[0085] Each random factor and time-varying factor has a corresponding probability and statistical function for fitting, and the control parameters of the fitting function are the characteristic parameters of each factor. When determining the characteristic parameters, it is necessary to first clarify the object of reliability evaluation and understand the randomness characteristics of the random samples of that object and the development trend of the time-varying factors.

[0086] If we consider the random factors of the soil layer beneath a tunnel, the probability distributions of its dynamic modulus and damping ratio conform to normal and log-normal distributions, respectively. The probability density curves for both are shown in the diagram below. Figure 2 As shown in the figure. The mean of the soil dynamic modulus is 279 MPa, and the root mean is 22 MPa; the mean of the damping ratio is 0.036, and the root mean is 0.002.

[0087] S2. Build a basic database of "parameter-response" based on the dynamic simulation model;

[0088] S21. Select a dynamic simulation model;

[0089] The selection of a dynamic simulation model needs to meet the following basic conditions:

[0090] a. It can reflect the actual operating status of the line;

[0091] b. It can reflect the changes in structural response to the characteristics of random and time-varying factors;

[0092] c. It can calculate the index results that can evaluate the track vibration reduction effect, such as the vertical vibration acceleration Z-level of the tunnel wall and the vertical vibration acceleration Z-level of the ground surface.

[0093] The maximum Z-level of the vertical vibration acceleration of the tunnel wall, i.e., the maximum Z-level of the source strength, is used as an indicator to evaluate the track vibration control effect. Taking the vehicle-track-tunnel coupled dynamic vertical model as an example, the track structure adopts a trapezoidal sleeper track, and the vehicle-track-tunnel coupled vibration physical model is as follows: Figure 3 As shown in the figure, from top to bottom, the components are: vehicle, rail, fastener, track slab (trapezoidal sleeper), elastic support layer (sleeper pad), tunnel, and underlying soil layer. The vehicle is simplified as a multi-rigid-body model with 10 degrees of freedom, the rail is simplified as an infinitely long Euler beam, the track slab and tunnel layer are simplified as elastic thin plates, and the fastener, sleeper pad, and underlying soil layer are simplified as spring dampers. The vibration differential equations of the vehicle and track system can be found in "Vehicle-Track Coupled Dynamics".

[0094] S22. Input random and time-varying factors into the dynamic simulation model to calculate the vibration response parameters under different vibration reduction key parameters and different combinations of random and time-varying factors under different characteristic parameter conditions.

[0095] Specifically, using dynamic simulation models, different key vibration reduction parameters are calculated, such as the high-frequency tangential dynamic stiffness of the track elastic element, the high-frequency loss factor of the track elastic element, and vibration response parameters under various calculation conditions combining random factor characteristic parameters and time-varying factor characteristic parameters.

[0096] S23. Build a basic database of "parameter-response" based on vibration response parameters.

[0097] Specifically, the vibration response parameters calculated in step S22 are used to build the "parameter-response" basic database required for reliability analysis.

[0098] S3. Construct an expanded database of "parameter-response" basic database samples based on the SE-U-Net neural network model;

[0099] S31. Select the SE-U-Net neural network model;

[0100] Since probability density evolution analysis requires a sufficient amount of data, while traditional dynamic simulation models require complex dynamic numerical / analytical analysis processes, they cannot generate a large number of working condition samples in a short period of time.

[0101] This embodiment uses the SE-U-Net neural network model, and the results of this model are as follows: Figure 4 As shown, its encoder-decoder structure, which is based on channel attention mechanism, residual connection and U-Net neural network structure, can support the input of time sampling sequence under various working conditions, identify the temporal characteristics of each single parameter and random parameter combination working condition, and thus realize the fast and accurate output of vibration response that meets the dynamic control conditions.

[0102] The SE-U-Net neural network model employs three layers of average pooling with varying lengths to initially calculate the changing trends of the input data, providing multi-scale and multi-level information for feature encoding. Then, multiple encoding blocks are used for feature encoding. These blocks primarily consist of multiple layers of Convolution-Batch Normalization-ReLU (CBR) and residual connections, and a channel attention (Squeeze-and-Excitation) mechanism is introduced to enhance feature extraction capabilities. The decoding layer comprises CBR and upsampling layers, with skip connections between encoding and decoding. This ensures that details in the multi-layer encoded features are preserved, reducing the risk of gradient vanishing and network degradation.

[0103] S32. Input the sample parameters from the “parameter-response” basic database built in step S2 into the SE-U-Net neural network model, output the supplementary samples, and store them in the “parameter-response” basic database to obtain the supplementary database, thereby enriching the samples in the “parameter-response” basic database.

[0104] Specifically, the sample parameters from the "parameter-response" basic database in step S2 are input into the SE-U-Net neural network model. The SE-U-Net neural network model then outputs supplementary samples, which are stored in the "parameter-response" basic database to enrich the sample set. This process enables rapid expansion of the "parameter-response" basic database, ensuring sufficient sample data for probability density evolution analysis.

[0105] S4. Reliability prediction and evaluation of vibration reduction in rail transit based on probability density evolution algorithm.

[0106] The formula for evaluating the reliability of track vibration reduction is:

[0107]

[0108] The probability density function of X(t) is:

[0109]

[0110] In the formula, X is the target physical quantity. These are the random parameters involved. It is x and The joint probability density function, equation (1) is a connection The joint probability density function and the corresponding velocity The one-dimensional convection probability density function;

[0111] The initial and boundary conditions for the probability density evolution problem are as follows:

[0112]

[0113] In the formula, δ is the Dirac function; P l X0 is the initial assigned probability, and X0 is the structural response at the initial moment.

[0114] The steps to solve the generalized probability density evolution equation are as follows:

[0115] ① In the random parameter space Ω Θ Select representative point set Where s is the number of multiple random variables, N sel The total number of points in the set;

[0116] ② For each representative point Solving for speed

[0117] ③ Obtained in step ② Substituting the discrete scheme into equation (3), and then solving it using the bilateral difference method under the initial conditions, the discrete space should be divided into a grid, and the grid nodes are denoted as (x i ,t k ), i = 0, ±1, ±2, ..., k = 0, 1, 2, ..., where x i =iΔx,t k =kΔt, k = 0, 1, 2, ..., Δx is the spatial step size in the x-direction, and Δt is the time step size. Thus, equation (1) is transformed into a system of algebraic equations, which can be solved to give the probability density values ​​at the nodes, i.e., p. XΘ (x,t);

[0118] ④. Combining the results of step ③, obtain the probability density:

[0119]

[0120] Specifically, the following is a corresponding example, taking the evaluation of the reliability of the source strength vibration reduction in a subway line tunnel as an example. The evaluation assesses the reliability of the tunnel source strength under the combined effects of the evolution of the rail roughness power spectrum, the underlying soil layer (soil dynamic modulus, damping ratio), and the random conditions of train speed within a 9-month grinding cycle.

[0121] (1) Select the random and time-varying factors that affect the reliability of track vibration reduction and determine the relevant characteristic parameters;

[0122] This project mainly studies the influence of train speed and random soil factors on vibration reduction reliability during the rail surface roughness grinding cycle. The characteristic parameters of each random factor and time-varying factor are shown in Tables 1 and 2. The grinding cycle is 9 months, and the power spectrum formula for rail roughness is shown in Equation (11).

[0123] Table 1 Random factors and their characteristic parameters

[0124]

[0125] Table 2 Time-varying factors and their characteristic parameters

[0126]

[0127]

[0128] In the formula, S(φ) is the power spectral density function of the short-wavelength irregularities on the rail surface (mm). 2 / m-1 ); φ is the spatial frequency (m -1 b represents the fitting parameters for the power spectrum of rail roughness at each level, and the specific values ​​are shown in Table 2.

[0129] (2) Build a "parameter-response" basic database based on dynamic simulation model and working conditions (taking the vehicle-track-tunnel coupled dynamic vertical model as an example).

[0130] Based on the actual operating conditions of this line, the dynamic simulation parameters were selected. The vehicle used was a Type B metro car, the rails were 60 gauge, and the vibration-damping track bed was a trapezoidal sleeper with a cross-sectional dimension of (0.58×2)×0.185. Specific parameters are shown in Tables 3, 4, and 5. The detailed calculation process for the vehicle-track-tunnel coupled dynamic vertical model is detailed in step S2.

[0131] Table 3 Vehicle Parameters

[0132]

[0133]

[0134] Table 4 Track Parameters

[0135]

[0136] Table 5 Tunnel structural dimensions

[0137] Tunnel parameters Parameter value unit tunnel inner diameter 6 m Tunnel lining thickness 0.3 m

[0138] Based on the above vehicle, track, and tunnel parameters, and combined with the cross-conditions of different random and time-varying factors shown in Table 6, the dynamic simulation model simulates and obtains the "parameter-response" basic database of tunnel source strength.

[0139] Table 6. Operating conditions with random and time-varying factors

[0140]

[0141] (3) Expand the number of samples based on the SE-U-Net neural network model;

[0142] The "parameter-response" dynamic results calculated by the dynamic simulation model are input into the SE-U-Net neural network model for training. The dynamic response feature parameters of the "parameter-response" basic database are extracted, and the sample size of the "parameter-response" basic database is expanded to obtain a supplementary database.

[0143] To determine the accuracy of the results generated by the SE-U-Net neural network model, the source strength time-domain results of the dynamic simulation model and the SE-U-Net model were compared, for example... Figure 5As shown in Figures (5a) and (5b), the SE-U-Net neural network model exhibits good convergence during training. After 150 epochs, the loss function converges to less than or equal to 0.1 and tends to stabilize. In Figure (5c), under the four working conditions, the simulation results of the dynamic simulation model have a high degree of fit with the results generated by the SE-U-Net model, which can meet the accuracy requirements of the expanded samples.

[0144] (4) Evaluate source strength reliability based on probability density evolution algorithm (PDEM);

[0145] The probability density evolution algorithm was used to analyze the "parameter-response" database generated in step S3. This yielded the probability density distribution evolution characteristics and reliability evolution characteristics of the maximum Z-level vibration of the tunnel source intensity within a 9-month grinding cycle, under the combined effects of rail roughness power spectrum evolution, underlying soil layers (soil dynamic modulus, damping ratio), and random vehicle speed conditions. Figure 6 As shown, the maximum Z-level vibration of the tunnel source strength increases with the number of months, and the reliability required to meet the limit of 85 dB decreases from nearly 100% in the first stage (January to March) to 70-80% in the second and third stages (April to September). If the reliability limit cannot be met according to national standards or engineering needs, the results of this reliability study can guide the improvement design of vibration reduction measures.

[0146] This application embodiment also provides a rail transit vibration reduction reliability evaluation device, which includes the following modules:

[0147] The acquisition module is used to acquire random and time-varying factors that affect environmental vibration;

[0148] The random factor acquisition unit is used to acquire and determine random factors;

[0149] The time-varying factor acquisition unit is used to acquire and determine time-varying factors;

[0150] The database module is used to build a basic "parameter-response" database based on the dynamic simulation model;

[0151] The dynamic simulation model selection unit is used to select a dynamic simulation model based on actual needs.

[0152] The first data input unit is used to input random and time-varying factors into the dynamic simulation model;

[0153] The dynamic simulation model calculation unit is used to calculate different key vibration reduction parameters and vibration response parameters under different combinations of characteristic parameters based on input random and time-varying factors.

[0154] The database unit is used to build a basic "parameter-response" database based on vibration response parameters.

[0155] An expanded database module is used to construct an expanded database of "parameter-response" basic database samples based on the SE-U-Net neural network model;

[0156] The model selection unit is used to select the SE-U-Net neural network model;

[0157] The second data input unit is used to input the sample parameters from the "parameter-response" basic database into the SE-U-Net neural network model;

[0158] SE-U-Net neural network model units are used to obtain supplementary samples based on the sample parameters output from the "parameter-response" basic database;

[0159] The supplementary database unit is used to store supplementary samples into the "parameter-response" basic database to obtain the supplementary database;

[0160] The track vibration reduction reliability evaluation module is used for reliability prediction and assessment of track traffic vibration reduction based on the probability density evolution algorithm.

[0161] This application also provides an electronic device, which includes a processor and a memory. The processor and memory are connected, for example, via a bus. Optionally, the electronic device may also include a transceiver. It should be noted that in practical applications, the transceiver is not limited to one unit, and the structure of this electronic device does not constitute a limitation on the embodiments of this application.

[0162] The processor can be a central processing unit (CPU), a general-purpose processor, a data signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0163] A bus can include a pathway for transmitting information between the aforementioned components. The bus can be a peripheral component interconnect standard PCI bus or an extended industry standard structure EISA bus, etc. Buses can be categorized as address buses, data buses, control buses, etc.

[0164] The memory may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a read-only optical disc (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0165] The memory stores the application code that executes the scheme of this application, and the processor controls its execution. The processor executes the application code stored in the memory to implement the content shown in the embodiment of a method for evaluating the reliability of vibration reduction in rail transit.

[0166] Electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Servers can also be included.

[0167] This application also provides a computer-readable storage medium storing a computer program that, when run on a computer, enables the computer to execute the corresponding content in an embodiment of a method for evaluating the reliability of vibration reduction in rail transit.

[0168] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0169] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A rail transit vibration damping reliability evaluation method, characterized by comprising the steps of: It comprises the following steps: S1, acquiring random factors and time-varying factors affecting environmental vibration; S2, building a "parameter-response" basic database based on a dynamic simulation model; S3, constructing an expansion database of "parameter-response" basic database samples based on an SE-U-Net neural network model; S4, predicting and evaluating the reliability of rail transit vibration reduction based on a probability density evolution algorithm.

2. The method according to claim 1, wherein The S1 comprises the following steps: S11, acquiring random factors; Wherein, the random factors include geological conditions, wheel-rail irregularities and train operation conditions; S12, acquiring time-varying factors; Wherein, the time-varying factors include track irregularities evolving over time, train operation conditions with time periodicity, and deterioration of track elastic materials over long-term operation.

3. The method according to claim 2, wherein The geological conditions include soil dynamic modulus and soil damping coefficient; The wheel-rail irregularities include track long-wave irregularities, rail roughness and wheel irregularity; The train operation conditions include speed and axle load.

4. The rail transit vibration-reducing reliability evaluation method of claim 1, wherein The S2 comprises the following steps: S21, selecting a dynamic simulation model; S22, inputting random factors and time-varying factors into the dynamic simulation model to calculate different vibration reduction key parameters and vibration response parameters of random factors and time-varying factors under different characteristic parameter combinations; S23, building a "parameter-response" basic database based on vibration response parameters.

5. The rail transit vibration-reducing reliability evaluation method of claim 1, wherein The S3 comprises the following steps: S31, selecting an SE-U-Net neural network model; S32, inputting sample parameters in the "parameter-response" basic database built in step S2 into the SE-U-Net neural network model, outputting supplementary samples and storing them into the "parameter-response" basic database to obtain a supplementary database, so as to enrich the samples of the "parameter-response" basic database.

6. The rail transit vibration-reducing reliability evaluation method of claim 1, wherein The S4 comprises the following steps: The evaluation formula of rail vibration reduction reliability is: The probability density function of X(t) is: where X is the target physical quantity, θ q is a random parameter involved, p XΘ (x, θ q , t) is the joint probability density function of x and θ q , and equation (1) is a one-dimensional convection probability density function connecting the joint probability density function of (X, θ q ) and the corresponding velocity . The initial conditions and boundary conditions of the probability density evolution problem are respectively: where δ is the Dirac function; P l is the initial assigned probability, and X0is the structural response quantity at the initial time. The steps for solving the generalized probability density evolution equation are as follows: Θ q 1,q s,q sel sel ​​​​​​​ ii. For each representative point θ q , the velocity ③ Obtained in step ② Substituting the discrete scheme into equation (3), and then solving it using the bilateral difference method under the initial conditions, the discrete space should be divided into a grid, and the grid nodes are denoted as (x i ,t k ), i = 0, ±1, ±2, ..., k = 0, 1, 2, ..., where x i =iΔx,t k =kΔt, k = 0, 1, 2, ..., Δx is the spatial step size in the x-direction, and Δt is the time step size. Thus, equation (1) is transformed into a system of algebraic equations, which can be solved to give the probability density values ​​at the nodes, i.e., p. XΘ (x,t); (4) Obtain the probability density by synthesizing the results of step (3):

7. A reliability evaluation device for vibration reduction in rail transit, characterized in that, It comprises: An acquisition module for acquiring random factors and time-varying factors affecting environmental vibration; A database module for building a "parameter-response" basic database based on a dynamic simulation model; An expansion database module for constructing an expansion database of "parameter-response" basic database samples based on an SE-U-Net neural network model; A rail vibration reduction reliability evaluation module for predicting and evaluating the reliability of rail transit vibration reduction based on a probability density evolution algorithm.

8. An electronic device, comprising: It comprises: One or more processors; Memory; One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to perform a rail transit vibration reduction reliability evaluation method according to any one of claims 1 to 7.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to realize the rail transit vibration reduction reliability evaluation method of any one of claims 1 to 7.

Images