Long-span bridge soft flutter response prediction method and device
By constructing a multivariate polynomial regression model and optimizing the damping and stiffness coefficients using a digital twin, the uncertainty problem in predicting the soft flutter response of long-span bridges was solved, achieving higher accuracy in prediction.
Patent Information
- Application Number
- CN202411618686.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-13
- Publication Date
- 2026-01-23
AI Technical Summary
Existing technologies for predicting the soft flutter response of long-span bridges suffer from uncertainties and inaccuracies in wind tunnel testing and numerical simulation methods. It is difficult to accurately identify the damping coefficient and stiffness coefficient, leading to inaccurate prediction results.
By constructing a multivariate polynomial regression model to optimize parameters, and combining wind tunnel tests and numerical simulations, a digital twin is established to optimize the damping coefficient and stiffness coefficient, thereby improving prediction accuracy.
It improves the prediction accuracy of soft flutter response of long-span bridges, overcomes the uncertainties of wind tunnel tests and numerical simulations, breaks through the limitations of wind tunnel tests, and can predict soft flutter response at wind speeds that have not been tested in wind tunnels or exceed the measurement range of wind tunnel tests.
Smart Images

Figure CN121389222A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of bridge safety technology, and more specifically, to a method and apparatus for predicting the soft flutter response of long-span bridges. Background Technology
[0002] Bridge flutter is a self-excited vibration phenomenon caused by the interaction between wind and the bridge structure, posing a serious threat to the safety and performance of the bridge structure. Bridges will experience flutter when the wind speed exceeds the critical flutter wind speed. This phenomenon manifests as limit cycle flutter with a stable amplitude, and the amplitude increases with increasing wind speed. Once the amplitude reaches a certain level, it may lead to structural damage or failure. To avoid the harm caused by flutter to bridges, wind tunnel tests or numerical simulations are typically used during the bridge design phase to predict the bridge's flutter response.
[0003] However, both methods have limitations and uncertainties. In terms of wind tunnel testing, due to the nonlinear characteristics of the damping and stiffness of the bridge deck model structure as a function of amplitude, and the influence of non-wind-induced aerodynamic effects in the wind tunnel on the damping and stiffness of the bridge deck model structure, structural parameters such as the bridge deck damping coefficient and stiffness coefficient are uncertain and cannot be accurately identified.
[0004] The simplification of numerical simulation models may ignore certain structural details or use idealized material properties, which may affect the accuracy of the simulation. At the same time, the damping coefficient and stiffness coefficient are key parameters that affect the amplitude of bridge deck flutter, but due to their uncertainty, they cannot be accurately set in numerical simulation. Summary of the Invention
[0005] The purpose of this invention is to provide a method and apparatus for predicting the soft flutter response of long-span bridges, thereby improving the aforementioned problems. To achieve this objective, the technical solution adopted by this invention is as follows:
[0006] In a first aspect, this application provides a method for predicting the soft flutter response of long-span bridges, including:
[0007] Obtain parameter information for long-span bridges;
[0008] A physical model and an initial numerical model are constructed based on the parameter information of the long-span bridge. The optimization parameters of the initial numerical model include the damping coefficient and the stiffness coefficient.
[0009] Based on the physical model, soft flutter wind tunnel tests were conducted to obtain the measured vibration response under different test wind speeds. The measured vibration response includes the measured amplitude and frequency in the vertical direction, as well as the measured amplitude and frequency in the torsional direction.
[0010] The parameters of the multivariate polynomial regression model with preset optimization parameters are optimized to obtain the optimal numerical model applicable to different test wind speeds. The objective function of the parameter optimization is constructed based on the simulated vibration response and the measured vibration response. The simulated vibration response is calculated through the initial numerical model.
[0011] Interpolating the optimal numerical model yields a digital twin of the long-span bridge;
[0012] Predicting soft flutter response of long-span bridges based on the digital twin.
[0013] Secondly, this application also provides a device for predicting the soft flutter response of long-span bridges, comprising:
[0014] The information acquisition module is used to acquire parameter information of long-span bridges;
[0015] The construction module is used to construct a physical model and an initial numerical model based on the parameter information of the long-span bridge. The optimization parameters of the initial numerical model include the damping coefficient and the stiffness coefficient.
[0016] The test module is used to conduct soft flutter wind tunnel tests based on the physical model to obtain the measured vibration response under different test wind speeds. The measured vibration response includes the measured amplitude and frequency in the vertical direction and the measured amplitude and frequency in the torsional direction.
[0017] The optimization module is used to perform parameter optimization on a multivariate multinomial regression model with preset optimization parameters to obtain the optimal numerical model suitable for different test wind speeds. The objective function of the parameter optimization is constructed based on the simulated vibration response and the measured vibration response. The simulated vibration response is calculated through the initial numerical model.
[0018] An interpolation module is used to interpolate the optimal numerical model to obtain a digital twin of the long-span bridge;
[0019] The prediction module is used to predict the soft flutter response of long-span bridges based on the digital twin.
[0020] The beneficial effects of this invention are as follows:
[0021] This invention overcomes the uncertainties of wind tunnel testing and numerical simulation in simulating bridge flutter by establishing a multivariate polynomial regression model for parameter optimization. Simultaneously, it constructs a digital twin, improving the prediction accuracy of flutter response for long-span bridges, achieving higher reliability compared to separate wind tunnel testing and numerical simulation. Furthermore, the established digital twin is used to predict flutter response at wind speeds that have not been tested in a wind tunnel or exceed the measurement range of wind tunnel tests, overcoming the limitations of wind tunnel testing, eliminating the influence of high blockage ratios under large amplitudes, and improving the accurate prediction capability of flutter response for long-span bridges.
[0022] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. Attached Figure Description
[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0024] Figure 1 This is a schematic diagram of the flow chart of the method for predicting soft flutter response of long-span bridges as described in this embodiment of the invention;
[0025] Figure 2 This is a schematic diagram of the structure of the long-span bridge soft flutter response prediction device described in this embodiment of the invention.
[0026] The markings in the diagram are: 800, long-span bridge soft flutter response prediction device; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. Detailed Implementation
[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.
[0028] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.
[0029] Example 1:
[0030] This embodiment provides a method for predicting the soft flutter response of long-span bridges.
[0031] See Figure 1 The figure shows that the method includes steps S100, S200, S300, S400, S500 and S600.
[0032] Step S100: Obtain parameter information for long-span bridges;
[0033] Step S200: Construct a physical model and an initial numerical model based on the parameter information of the long-span bridge. The optimization parameters of the initial numerical model include the damping coefficient and the stiffness coefficient.
[0034] In this embodiment, a physical model of the long-span bridge is fabricated based on its original design. The physical model is then installed inside a wind tunnel using a spring suspension system, forming a test system for soft flutter wind tunnel testing. A laser displacement meter and an accelerometer are installed within the test system to collect the vibration response of the long-span bridge during the test. Additionally, a cobra anemometer is installed inside the wind tunnel to collect flow field information. All equipment is connected to a data acquisition system mounted on a computer via a charge amplifier.
[0035] Step S300: Based on the physical model, a soft flutter wind tunnel test is conducted to obtain the measured vibration response under different test wind speeds. The measured vibration response includes the measured vertical amplitude and measured vibration frequency, as well as the measured amplitude and measured vibration frequency in the torsional direction.
[0036] In this embodiment, for the constructed physical model, the mass, moment of inertia, damping coefficient, and stiffness coefficient of the bridge deck are the most important parameters affecting the vibration characteristics of long-span bridge decks. Among them, the mass and moment of inertia of the bridge deck are determined during the design stage of the physical model and can be measured by equipment such as electronic scales; therefore, the mass and moment of inertia can be accurately measured.
[0037] Due to the nonlinearity of the damping coefficient and stiffness coefficient, and the interference from non-wind-induced additional damping and non-wind-induced additional mass in the wind tunnel, these two parameters cannot be measured accurately. Therefore, the total damping coefficient and total stiffness coefficient at different amplitudes obtained from the free decay vibration time history of the bridge deck under zero wind speed (windless environment) are used as reference values for the damping coefficient and stiffness coefficient, i.e., preset damping coefficient and preset stiffness coefficient.
[0038] In this embodiment, when conducting wind tunnel tests based on the logistics model, the measured vibration response of the physical model under different wind speeds is obtained by controlling the wind speed. Simultaneously, the initial numerical model is used for numerical simulation of the wind tunnel tests.
[0039] Step S400: Perform parameter optimization on the multivariate polynomial regression model of the preset optimization parameters to obtain the optimal numerical model applicable to different test wind speeds. The objective function of the parameter optimization is constructed based on the simulated vibration response and the measured vibration response. The simulated vibration response is calculated through the initial numerical model.
[0040] The steps for constructing the multivariate multinomial regression model include:
[0041] Step A100: Based on the measured vertical amplitude and measured vibration frequency, establish the vertical damping coefficient equation and the vertical stiffness coefficient equation respectively.
[0042] Step A200: Based on the measured amplitude and measured vibration frequency in the torsional direction, establish the damping coefficient equation and the stiffness coefficient equation in the torsional direction, respectively.
[0043] Step A300: Construct a multivariate polynomial regression model for the optimization parameters using the vertical damping coefficient equation, the vertical stiffness coefficient equation, the torsional damping coefficient equation, and the torsional stiffness coefficient equation.
[0044] In this embodiment, since it is difficult to accurately measure the damping coefficient and stiffness coefficient of the bridge deck of a long-span bridge in wind tunnel tests, but they have a significant impact on the numerical simulation results of flutter, they are selected as optimization parameters. Considering the nonlinear relationship between the damping coefficient and stiffness coefficient and the amplitude and frequency of vibration in the torsional direction, the number of optimization parameters, the optimization accuracy of the polynomial regression iterative model, and the relatively low computational resources required, the polynomial regression iterative model is selected as the optimization algorithm. Expressions for the damping coefficient and stiffness coefficient in the vertical and torsional directions are established respectively, resulting in the mathematical expression of the multivariate polynomial regression model, which is:
[0045]
[0046] In the formula, c h and kh c represents the vertical damping coefficient and stiffness coefficient, respectively. α and k α A represents the damping coefficient and stiffness coefficient in the torsional direction, respectively. h and f h A represents the measured vertical amplitude and measured vibration frequency, respectively. α and f α a represents the measured amplitude and measured vibration frequency in the torsional direction, respectively. hb′ a αb′ b hb′ and b αb′ All represent the coefficient parameters of the multivariate polynomial regression model, b′=1,2,…,5.
[0047] The calculation steps for the simulated vibration response include:
[0048] Step B100: Obtain wind tunnel test conditions and input the wind tunnel test conditions into the initial numerical model. The wind tunnel test conditions include model size, flow field characteristics and boundary conditions. The wind tunnel test conditions are consistent with the conditions for soft flutter wind tunnel tests on the physical model.
[0049] Step B200: Select the turbulence model of the initial numerical model;
[0050] Step B300: Perform sensitivity analysis on the initial numerical model for grid size and time step to obtain the optimal grid size and optimal time step of the initial numerical model;
[0051] Step B400: Extract aerodynamic lift and aerodynamic lift torque based on the initial numerical model;
[0052] Step B500: Obtain the structural dynamics equations in the initial numerical model, solve the structural dynamics equations based on the aerodynamic lift and the aerodynamic lift moment, and obtain the simulated vibration response, which includes the simulated amplitude and simulated vibration frequency in the vertical direction and the simulated amplitude and simulated vibration frequency in the torsional direction.
[0053] In this embodiment, the expression for the structural dynamics equation is:
[0054]
[0055]
[0056] In the formula, m represents the mass of the bridge plate, I represents the moment of inertia, and c h and k h c represents the vertical damping coefficient and stiffness coefficient, respectively. α and k αF represents the damping coefficient and stiffness coefficient in the torsional direction, respectively. L (t) represents the aerodynamic lift with respect to time t, M T (t) represents the aerodynamic lift torque at time t, h(t), and Let α(t) represent the vertical displacement, vertical velocity, and vertical acceleration of the bridge deck of a long-span bridge at time t, respectively. and These represent the torsional displacement, torsional velocity, and torsional acceleration of the bridge deck at time t, respectively.
[0057] Because the air-added damping and air-added mass are unclear, it is impossible to accurately obtain the true values of the damping coefficient and stiffness coefficient required for numerical simulation. Therefore, the c in the structural dynamics equations... h k h c α and k α Set as optimization parameters.
[0058] The steps for constructing the objective function include:
[0059] Step C100: Calculate the first relative error based on the measured and simulated vertical amplitudes;
[0060] Step C200: Calculate the second relative error based on the measured amplitude and simulated amplitude in the torsional direction;
[0061] Step C300: Calculate the third relative error based on the measured and simulated vertical vibration frequencies;
[0062] Step C400: Calculate the fourth relative error based on the measured and simulated vibration frequencies in the torsional direction;
[0063] Step C500: Using the first relative error, the second relative error, the third relative error, and the fourth relative error, form a set of relative error equations, and use the set of relative error equations as the objective function.
[0064] In this embodiment, the vibration response of soft flutter based on wind tunnel testing of physical models and numerical simulation based on initial numerical models is compared. Significant differences are found between the two. In order to reduce the differences, it is necessary to optimize the uncertainty parameters affecting bridge deck soft flutter in the initial numerical model, that is, to optimize the optimization parameters set above.
[0065] Therefore, the optimization objective is determined to be that, under the same wind speed and soft flutter conditions, the relative error between the amplitude and vibration frequency in the numerical simulation and wind tunnel test is less than 1%, so that the numerical simulation results match the wind tunnel test results as closely as possible. In other words, the objective function converges when its value is less than 1%. The expression for the objective function is:
[0066]
[0067] In the formula, R h R fh R a and R fα Let A represent the first relative error, the second relative error, the third relative error, and the fourth relative error, respectively. h and f h A represents the measured vertical amplitude and measured vibration frequency, respectively. α and f α These represent the measured amplitude and measured vibration frequency in the torsional direction, respectively. and These represent the vertical simulated amplitude and simulated vibration frequency, respectively. and represents the simulated amplitude and simulated vibration frequency in the torsional direction, respectively, and |·| represents taking the absolute value.
[0068] Step S400 specifically includes:
[0069] Step S401: Based on the vibration data under each test wind speed, fit the multivariate polynomial regression model of the preset optimization parameters to obtain the coefficient parameters of the multivariate polynomial regression model under each test wind speed. The vibration data used in the first iteration includes the measured vibration response, preset damping coefficient and preset stiffness coefficient.
[0070] Step S402: Substitute the coefficient parameters and the measured vibration response into the multivariate polynomial regression model for calculation to obtain the predicted damping coefficient and predicted stiffness coefficient under each test wind speed;
[0071] Step S403: Substitute the predicted damping coefficient and predicted stiffness coefficient at each test wind speed into the initial numerical model to calculate the simulated vibration response at each test wind speed.
[0072] Step S404: Based on the simulated vibration response and measured vibration response at each test wind speed, calculate the function value of the objective function at all test wind speeds;
[0073] Step S405: Determine whether the function value at each test wind speed is less than the preset threshold. If so, stop the iteration, obtain the optimal numerical model at the test wind speed, and use the predicted damping coefficient and predicted stiffness coefficient at the test wind speed as the damping coefficient and stiffness coefficient in the corresponding optimal numerical model. Otherwise, use the simulated vibration response, predicted damping coefficient, and predicted stiffness coefficient at the test wind speed as the vibration data for the next iteration and proceed with the next iteration.
[0074] In this embodiment, the multivariate polynomial regression model is first fitted using the measured vibration response, preset damping coefficient, and preset stiffness coefficient to obtain the coefficient parameters of the multivariate polynomial regression model. The measured amplitude and frequency of the wind tunnel test at the specified wind speed are then input into the multivariate polynomial regression model to calculate the predicted stiffness coefficient and predicted damping coefficient. These predicted stiffness coefficients and predicted damping coefficients are then substituted into the initial numerical model for numerical simulation to obtain the simulated vibration response, and the function value of the objective function is calculated. It is then determined whether the function value is less than a preset threshold (1% in this embodiment). If the threshold is met, the iteration stops, and the optimal numerical model is obtained; otherwise, the simulated vibration response, predicted damping coefficient, and predicted stiffness coefficient are used as the vibration data for the next iteration, and the above steps are repeated.
[0075] The above operation was performed on the wind tunnel test at each test wind speed to ensure that the numerical simulation results were consistent with the measured vibration response of the wind tunnel test, and to obtain the damping coefficient and stiffness coefficient corresponding to each wind speed.
[0076] Step S500: Interpolate the optimal numerical model to obtain a digital twin of the long-span bridge;
[0077] In this embodiment, a digital twin system for wind speed is established using the Kriging interpolation method. This method can be used to predict the damping coefficient and stiffness coefficient at wind speeds that have not been tested in a wind tunnel or exceed the measurement capabilities of a wind tunnel test, thereby obtaining the vibration response of simulated soft flutter, that is, obtaining a digital twin at any wind speed, and obtaining the vibration response at that wind speed through the digital twin.
[0078] Step S500 specifically includes:
[0079] Step S501: Calculate the covariance matrix of each test wind speed based on the Gaussian covariance function;
[0080] Step S502: Calculate the covariance vector of each test wind speed based on the preset wind speed;
[0081] Step S503: Calculate the weighting factor for all test wind speeds based on the inverse matrix of the covariance matrix and the covariance vector;
[0082] Step S504: Based on the damping coefficient, stiffness coefficient and weighting factor of all test wind speeds in the optimal numerical model, calculate the damping coefficient and stiffness coefficient at the preset wind speed to obtain a digital twin of the long-span bridge.
[0083] In this embodiment, the expressions for the damping coefficients or structural stiffness coefficients predicted in the vertical and torsional directions under the preset wind speed are as follows:
[0084]
[0085]
[0086] In the formula, λ represents the damping coefficient or structural stiffness coefficient in the vertical and torsional directions at a preset wind speed. i z represents the weighting factor for the i-th test wind speed. i The damping coefficients or structural stiffness coefficients in the vertical and torsional directions represent the i-th test wind speed, n represents the number of test wind speeds, and c represents the number of test wind speeds. ij Let c represent the covariance matrix. io This represents the covariance vector.
[0087] Wherein, the covariance matrix c ij With covariance vector c io It can be expressed using the Gaussian covariance function as follows:
[0088]
[0089]
[0090] In the formula, σ 2 V represents the variance of a spatial variable, v i Let v represent the wind speed of the i-th test. j Let v represent the j-th test wind speed, l represent the length scale used to control the rate of spatially correlated attenuation, and v o This indicates the preset wind speed, and |·| indicates taking the absolute value.
[0091] Step S600: Predict the soft flutter response of long-span bridges based on the digital twin.
[0092] In this embodiment, a digital twin system for wind speed in wind tunnel testing was established. This digital twin system allows for the generation of digital twins at arbitrary wind speeds, thus eliminating uncertainties in the numerical simulation of wind-bridge coupled vibration algorithms and turbulence models.
[0093] By using digital twins at arbitrary wind speeds, we can gain a deeper understanding of the airflow characteristics around the deck of long-span bridges under different wind speeds. Simultaneously, by interpolating untested wind speeds within the established digital twin system, we can obtain the digital twin and its corresponding damping and stiffness coefficients, thereby enabling numerical simulations to obtain the soft flutter response.
[0094] It can also be used to extrapolate the damping coefficient and stiffness coefficient at high wind speeds exceeding the measurement range of wind tunnel tests, thereby predicting the ultimate soft flutter response that cannot be measured by wind tunnel tests. Furthermore, it can eliminate the influence of the blocking ratio under large amplitude, thus obtaining an accurate prediction of the aerodynamic response of long-span bridges and improving the wind resistance design capability of long-span bridges.
[0095] Example 2:
[0096] This embodiment provides a device for predicting the soft flutter response of long-span bridges, the device comprising:
[0097] The information acquisition module is used to acquire parameter information of long-span bridges;
[0098] The construction module is used to construct a physical model and an initial numerical model based on the parameter information of the long-span bridge. The optimization parameters of the initial numerical model include the damping coefficient and the stiffness coefficient.
[0099] The test module is used to conduct soft flutter wind tunnel tests based on the physical model to obtain the measured vibration response under different test wind speeds. The measured vibration response includes the measured amplitude and frequency in the vertical direction and the measured amplitude and frequency in the torsional direction.
[0100] The optimization module is used to perform parameter optimization on a multivariate multinomial regression model with preset optimization parameters to obtain the optimal numerical model suitable for different test wind speeds. The objective function of the parameter optimization is constructed based on the simulated vibration response and the measured vibration response. The simulated vibration response is calculated through the initial numerical model.
[0101] An interpolation module is used to interpolate the optimal numerical model to obtain a digital twin of the long-span bridge;
[0102] The prediction module is used to predict the soft flutter response of long-span bridges based on the digital twin.
[0103] The optimization module includes:
[0104] The first building unit is used to establish the vertical damping coefficient equation and the vertical stiffness coefficient equation based on the measured vertical amplitude and measured vibration frequency, respectively.
[0105] The second building unit is used to establish the damping coefficient equation and the stiffness coefficient equation in the torsional direction based on the measured amplitude and measured vibration frequency in the torsional direction, respectively.
[0106] The third building unit is used to construct a multivariate polynomial regression model of the optimization parameters using the vertical damping coefficient equation, the vertical stiffness coefficient equation, the torsional damping coefficient equation, and the torsional stiffness coefficient equation.
[0107] The optimization module also includes:
[0108] The fitting unit is used to fit the multivariate polynomial regression model with preset optimization parameters based on the vibration data under each test wind speed, and obtain the coefficient parameters of the multivariate polynomial regression model under each test wind speed. The vibration data used in the first iteration includes the measured vibration response, preset damping coefficient and preset stiffness coefficient.
[0109] The first calculation unit is used to substitute the coefficient parameters and the measured vibration response into the multivariate polynomial regression model for calculation, so as to obtain the predicted damping coefficient and predicted stiffness coefficient under each test wind speed.
[0110] The second calculation unit is used to substitute the predicted damping coefficient and predicted stiffness coefficient at each test wind speed into the initial numerical model to calculate the simulated vibration response at each test wind speed.
[0111] The third calculation unit is used to calculate the function value of the objective function under all test wind speeds based on the simulated vibration response and measured vibration response under each test wind speed.
[0112] The judgment unit is used to determine whether the function value under each test wind speed is less than the preset threshold. If it is, the iteration stops, the optimal numerical model under the test wind speed is obtained, and the predicted damping coefficient and predicted stiffness coefficient under the test wind speed are used as the damping coefficient and stiffness coefficient in the corresponding optimal numerical model. Otherwise, the simulated vibration response, predicted damping coefficient and predicted stiffness coefficient under the test wind speed are used as the vibration data for the next iteration, and the next iteration is performed.
[0113] The interpolation module includes:
[0114] The fourth calculation unit is used to calculate the covariance matrix of each test wind speed based on the Gaussian covariance function;
[0115] The fifth calculation unit is used to calculate the covariance vector of each test wind speed based on the preset wind speed;
[0116] The sixth calculation unit is used to calculate the weighting factor for all test wind speeds based on the inverse matrix of the covariance matrix and the covariance vector.
[0117] The seventh calculation unit is used to calculate the damping coefficient and stiffness coefficient at the preset wind speed based on the damping coefficient, stiffness coefficient and weighting factor of the optimal numerical model under all test wind speeds, so as to obtain a digital twin of the long-span bridge.
[0118] It should be noted that the specific manner in which each module performs its operation in the apparatus described in the above embodiments has been described in detail in the embodiments of the method, and will not be elaborated here.
[0119] Example 3:
[0120] Corresponding to the above method embodiments, this embodiment also provides a device for predicting the soft flutter response of long-span bridges. The device for predicting the soft flutter response of long-span bridges described below and the method for predicting the soft flutter response of long-span bridges described above can be referred to in correspondence.
[0121] Figure 2 This is a block diagram illustrating a soft flutter response prediction device 800 for a long-span bridge according to an exemplary embodiment. Figure 2 As shown, the long-span bridge soft flutter response prediction device 800 may include: a processor 801 and a memory 802. The long-span bridge soft flutter response prediction device 800 may also include one or more of the following: a multimedia component 803, an I / O interface 804, and a communication component 805.
[0122] The processor 801 controls the overall operation of the long-span bridge flutter response prediction device 800 to complete all or part of the steps in the aforementioned long-span bridge flutter response prediction method. The memory 802 stores various types of data to support the operation of the long-span bridge flutter response prediction device 800. This data may include, for example, instructions for any application or method operating on the long-span bridge flutter response prediction device 800, as well as application-related data such as contact data, sent and received messages, images, audio, video, etc. The memory 802 can be implemented using any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. I / O interface 804 provides an interface between processor 801 and other interface modules, such as a keyboard, mouse, and buttons. These buttons can be virtual or physical. Communication component 805 is used for wired or wireless communication between the long-span bridge soft flutter response prediction device 800 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 805 may include a Wi-Fi module, a Bluetooth module, or an NFC module.
[0123] In an exemplary embodiment, the long-span bridge soft flutter response prediction device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described long-span bridge soft flutter response prediction method.
[0124] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the above-described method for predicting the soft flutter response of long-span bridges. For example, the computer-readable storage medium may be the memory 802 including the program instructions, which may be executed by the processor 801 of the long-span bridge soft flutter response prediction device 800 to complete the above-described method for predicting the soft flutter response of long-span bridges.
[0125] Example 4:
[0126] Corresponding to the above method embodiments, this embodiment also provides a readable storage medium. The readable storage medium described below can be referred to in conjunction with the method for predicting the soft flutter response of a long-span bridge described above.
[0127] A readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the method for predicting the soft flutter response of long-span bridges as described in the above-described method embodiments.
[0128] Specifically, the readable storage medium can be a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, or any other readable storage medium capable of storing program code.
[0129] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
[0130] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for predicting soft flutter response of a long-span bridge, characterized by, The method comprises the following steps: obtaining parameter information of a long-span bridge; constructing a physical model and an initial numerical model based on the parameter information of the long-span bridge, wherein the optimization parameters of the initial numerical model include damping coefficients and stiffness coefficients; performing a soft buffeting wind tunnel test based on the physical model to obtain measured vibration responses under different test wind speeds, wherein the measured vibration responses include vertical measured amplitudes and measured vibration frequencies as well as torsional direction measured amplitudes and measured vibration frequencies; performing parameter optimization on a preset optimization parameter multiple polynomial regression model to obtain an optimal numerical model applicable to different test wind speeds, wherein a target function of the parameter optimization is constructed based on simulated vibration responses and the measured vibration responses, and the simulated vibration responses are calculated by the initial numerical model; performing interpolation on the optimal numerical model to obtain a digital twin of the long-span bridge; performing soft buffeting response prediction on the long-span bridge based on the digital twin.
2. The method of predicting soft flutter response of a long-span bridge according to claim 1, characterized in that The construction steps of the multiple polynomial regression model are as follows: based on the vertical measured amplitudes and the measured vibration frequencies, a vertical damping coefficient equation and a vertical stiffness coefficient equation are respectively established; based on the torsional direction measured amplitudes and the measured vibration frequencies, a torsional direction damping coefficient equation and a torsional direction stiffness coefficient equation are respectively established; the multiple polynomial regression model of the optimization parameters is constructed by the vertical damping coefficient equation, the vertical stiffness coefficient equation, the torsional direction damping coefficient equation and the torsional direction stiffness coefficient equation.
3. The method of predicting soft flutter response of a long-span bridge according to claim 1, characterized in that The calculation steps of the simulated vibration responses are as follows: obtaining wind tunnel test conditions and inputting the wind tunnel test conditions into the initial numerical model, wherein the wind tunnel test conditions include model size, flow field characteristics and boundary conditions, and the wind tunnel test conditions are consistent with the conditions of the soft buffeting wind tunnel test of the physical model; selecting a turbulence model of the initial numerical model; performing sensitivity analysis on the initial numerical model in terms of grid size and time step to obtain optimal grid size and optimal time step of the initial numerical model; extracting aerodynamic lift and aerodynamic lift moment based on the initial numerical model; obtaining a structural dynamics equation in the initial numerical model, solving the structural dynamics equation based on the aerodynamic lift and the aerodynamic lift moment to obtain simulated vibration responses, wherein the simulated vibration responses include vertical simulated amplitudes and simulated vibration frequencies as well as torsional direction simulated amplitudes and simulated vibration frequencies.
4. The method of predicting soft flutter response of a long-span bridge according to claim 3, characterized in that The construction steps of the target function are as follows: calculating a first relative error based on the vertical measured amplitudes and the simulated amplitudes; calculating a second relative error based on the torsional direction measured amplitudes and the simulated amplitudes; calculating a third relative error based on the vertical measured vibration frequencies and the simulated vibration frequencies; calculating a fourth relative error based on the torsional direction measured vibration frequencies and the simulated vibration frequencies; composing a relative error equation set by the first relative error, the second relative error, the third relative error and the fourth relative error, and taking the relative error equation set as the target function.
5. The method of predicting soft flutter response of a long-span bridge according to claim 1, wherein The parameter optimization on the preset optimization parameter multiple polynomial regression model to obtain the optimal numerical model applicable to different test wind speeds comprises the following steps: The coefficient parameters of the multivariate polynomial regression model under each test wind speed are obtained by fitting the preset optimization parameter multivariate polynomial regression model based on the vibration data under each test wind speed, wherein the vibration data used in the first iteration includes the measured vibration response, the preset damping coefficient and the preset stiffness coefficient; The predicted damping coefficient and the predicted stiffness coefficient under each test wind speed are obtained by substituting the coefficient parameters and the measured vibration response into the multivariate polynomial regression model for calculation; The predicted damping coefficient and the predicted stiffness coefficient under each test wind speed are substituted into the initial numerical model to calculate the simulated vibration response under each test wind speed; The function values of the objective function under all test wind speeds are calculated based on the simulated vibration response and the measured vibration response under each test wind speed; It is judged whether the function value under each test wind speed is less than the preset threshold value, if yes, the iteration is stopped, the optimal numerical model under the test wind speed is obtained, and the predicted damping coefficient and the predicted stiffness coefficient under the test wind speed are taken as the damping coefficient and the stiffness coefficient in the corresponding optimal numerical model, otherwise, the simulated vibration response, the predicted damping coefficient and the predicted stiffness coefficient under the test wind speed are taken as the vibration data for the next iteration, and the next iteration is performed.
6. The method of predicting soft flutter response of a long-span bridge according to claim 1, wherein The optimal numerical model is interpolated to obtain the digital twin of the long-span bridge, including: The covariance matrix of each test wind speed is calculated based on the Gaussian covariance function; The covariance vector of each test wind speed is calculated based on the preset wind speed; The weight factor of all test wind speeds is calculated based on the inverse matrix of the covariance matrix and the covariance vector; The damping coefficient and the stiffness coefficient under the preset wind speed are calculated based on the damping coefficient, the stiffness coefficient in the optimal numerical model under all test wind speeds and the weight factor of all test wind speeds, to obtain the digital twin of the long-span bridge.
7. A device for predicting the soft flutter response of long-span bridges, characterized in that, It includes: An information acquisition module for acquiring parameter information of a long-span bridge; A construction module for constructing a physical model and an initial numerical model based on the parameter information of the long-span bridge, the optimization parameters of the initial numerical model including damping coefficients and stiffness coefficients; A test module for performing a soft buffeting wind tunnel test based on the physical model to obtain measured vibration responses under different test wind speeds, the measured vibration responses including measured amplitudes and measured vibration frequencies in the vertical direction and measured amplitudes and measured vibration frequencies in the torsional direction; An optimization module for parameter optimization of a preset optimization parameter multivariate polynomial regression model to obtain optimal numerical models applicable to different test wind speeds, the objective function of the parameter optimization being constructed based on simulated vibration responses and the measured vibration responses, the simulated vibration responses being calculated by the initial numerical model; An interpolation module for interpolating the optimal numerical model to obtain the digital twin of the long-span bridge; A prediction module for predicting the soft buffeting response of the long-span bridge according to the digital twin.
8. The large-span bridge soft flutter response prediction device according to claim 7, characterized by, The optimization module includes: A first construction unit for establishing a vertical damping coefficient equation and a vertical stiffness coefficient equation based on the measured amplitudes and the measured vibration frequencies in the vertical direction, respectively; The second construction unit is configured to establish a damping coefficient equation in the torsion direction and a stiffness coefficient equation in the torsion direction based on the measured amplitude and the measured vibration frequency in the torsion direction, respectively. The third construction unit is configured to construct a multiple polynomial regression model of the optimization parameters by using the damping coefficient equation in the vertical direction, the stiffness coefficient equation in the vertical direction, the damping coefficient equation in the torsion direction, and the stiffness coefficient equation in the torsion direction.
9. The large-span bridge soft flutter response prediction device according to claim 7, characterized by, The optimization module further comprises: The fitting unit is configured to fit the multiple polynomial regression model of the optimization parameters based on the vibration data at each test wind speed, to obtain coefficient parameters of the multiple polynomial regression model at each test wind speed, wherein the vibration data used in the initial iteration includes the measured vibration response, the preset damping coefficient, and the preset stiffness coefficient. The first calculation unit is configured to calculate the multiple polynomial regression model by using the coefficient parameters and the measured vibration response, to obtain the predicted damping coefficient and the predicted stiffness coefficient at each test wind speed. The second calculation unit is configured to calculate the simulated vibration response at each test wind speed by using the predicted damping coefficient and the predicted stiffness coefficient at each test wind speed in the initial numerical model. The third calculation unit is configured to calculate the function value of the objective function at all test wind speeds based on the simulated vibration response and the measured vibration response at each test wind speed. The judging unit is configured to determine whether the function value at each test wind speed is less than a preset threshold value, if yes, stop the iteration, obtain the optimal numerical model at the test wind speed, and use the predicted damping coefficient and the predicted stiffness coefficient at the test wind speed as the damping coefficient and the stiffness coefficient in the corresponding optimal numerical model, otherwise, use the simulated vibration response, the predicted damping coefficient, and the predicted stiffness coefficient at the test wind speed as the vibration data for the next iteration, and perform the next iteration.
10. The large-span bridge soft flutter response prediction device according to claim 7, characterized by, The interpolation module comprises: The fourth calculation unit is configured to calculate the covariance matrix of each test wind speed based on a Gaussian covariance function. The fifth calculation unit is configured to calculate the covariance vector of each test wind speed based on a preset wind speed. The sixth calculation unit is configured to calculate the weight factor of all test wind speeds based on the inverse matrix of the covariance matrix and the covariance vector. The seventh calculation unit is configured to calculate the damping coefficient and the stiffness coefficient at the preset wind speed based on the damping coefficient, the stiffness coefficient in the optimal numerical model at all test wind speeds, and the weight factor of all test wind speeds, to obtain the digital twin of the long-span bridge.
