An adaptive integral method and system for bridge buffeting response in the frequency domain

By employing the method of reducing the frequency endpoints and using non-uniform point allocation in the frequency domain analysis of bridge buffeting response, combined with adaptive densification of error contribution, the problems of low calculation accuracy and efficiency in the existing technology are solved, and a more efficient bridge buffeting response assessment is achieved.

CN123068751APending Publication Date: 2026-11-10CHINA RAILWAY MAJOR BRIDGE RECONNAISSANCE & DESIGN INSTITUTE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611189407.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-06
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

In existing technologies, frequency domain analysis methods for bridge buffeting response rely on manual experience in point allocation, making it difficult to unify calculation accuracy and efficiency, and global uniform point allocation leads to a waste of computational resources.

Method used

Using the converted frequency as the endpoint, a frequency domain integration interval is constructed, and low-density and high-density grids are generated by the endpoint-clustered non-uniform collocation. Adaptive refinement is performed by combining error contribution, and iterative calculation is performed until the result converges.

Benefits of technology

It improves the accuracy and efficiency of bridge buffet response calculation, reduces reliance on subjective experience, and is applicable to bridge structural analysis under complex aerodynamic models.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123068751A_ABST
    Figure CN123068751A_ABST
Patent Text Reader

Abstract

This application relates to an adaptive integral method for the frequency domain of bridge flutter response, comprising: calculating the reduced frequencies of each mode, using the reduced frequencies as endpoints to construct multiple integration sub-intervals; performing endpoint-clustered non-uniform point allocation within each integration sub-interval to divide it into low-density and high-density reduced frequency grids; performing integration calculations to obtain the corresponding generalized displacement response covariance matrix; constructing an integration convergence error criterion to determine whether the integration result converges; if the integration result converges, using the high-density reduced frequency grid as the final integration grid; if the integration result does not converge, calculating the error contribution of each integration sub-interval and performing adaptive densification point allocation; repeating the iteration until the integration result converges to obtain the final integration grid. This application employs an endpoint-clustered non-uniform point allocation method to automatically concentrate the reduced frequency points near the reduced frequencies of modes with drastic changes in the response spectrum, reducing redundant point allocation in smooth frequency bands and reducing reliance on experience.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of bridge wind-induced vibration assessment technology, specifically to a frequency domain adaptive integral method and system for bridge buffeting response. Background Technology

[0002] Long-span bridges experience significant wind-induced random vibrations under natural wind conditions. Random buffeting is the most common form of wind-induced vibration encountered during the design return period of long-span bridges. Unlike forced vibrations, random buffeting requires consideration of the effects of self-excited aerodynamic damping and aerodynamic stiffness. Furthermore, the aerodynamic force has a wide frequency band distribution, making it difficult to directly control its amplitude using common vibration suppression measures such as aerodynamic and mechanical control. Although random buffeting does not cause catastrophic damage to the structure, its excessive response during the most unfavorable structural state before the structure is closed during construction can endanger the safety of the main structure, construction personnel, and construction equipment. During the operational period, random buffeting under high wind speeds not only raises safety and comfort concerns for passengers but also reduces the service life of bridge components.

[0003] In related technologies, frequency domain analysis methods for random chattering typically involve calculating the generalized displacement matrix at discrete frequency points and obtaining the generalized displacement and the response spectrum at a specified bridge deck location by integrating the discrete frequencies. Generally, the response spectrum is obtained using either manual empirical frequency allocation or globally uniform frequency allocation, and then the statistical characteristics of the response are obtained through numerical integration. If manual empirical frequency allocation is used, the frequency point distribution depends on the engineer's experience. Different bridge structures, wind speeds, bridge deck widths, damping levels, and modal combinations result in variations in the peak position and amplitude of the response spectrum, making it difficult to guarantee a balance between computational accuracy and efficiency using manual allocation. If globally uniform frequency allocation is used, a large number of frequency points need to be distributed across the entire integration interval to capture spectral peaks near the modal frequencies, leading to over-calculation of smooth intervals far from the spectral peaks, resulting in low computational efficiency. Summary of the Invention

[0004] This application provides a frequency domain adaptive integral method and system for bridge flutter response, which solves the technical problems of manual point allocation relying on experience and low calculation efficiency of global uniform point allocation in related technologies.

[0005] This application provides a frequency domain adaptive integration method for bridge flutter response, which includes the following steps: Step S1: Calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals. Step S2: Perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid. Step S3: Perform integration calculations based on the low-density and high-density reduced frequency grids respectively to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix respectively. Step S4: Construct an integration convergence error criterion based on the integration result to determine whether the integration result has converged; If the integration result converges, the high-density reduced frequency grid is used as the final integration grid. If the integration result does not converge, proceed to step S5; Step S5: Calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive densification point allocation for the converted frequency integral sub-intervals with large error contributions; Repeat steps S2-S5 iteratively until the integration result converges or reaches the maximum number of iterations or the maximum number of collocation points to obtain the final integration grid.

[0006] In one embodiment, step S1, calculating the reduced frequencies corresponding to each mode of the bridge structure, and using the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals, includes: Step S11: Calculate the reduced frequencies corresponding to each mode of the bridge structure. ; The calculation formula is: ; in, For each modal frequency; B This refers to the width of the bridge deck. U Average wind speed; Step S12: Using the converted frequency as the endpoint, set the converted frequency integration range and determine multiple converted frequency integration sub-intervals to construct the frequency domain integration interval.

[0007] In one embodiment, step S2, performing endpoint-clustered non-uniform point allocation within each reduced frequency integral sub-interval to generate a reduced frequency grid, and dividing it into a low-density reduced frequency grid and a high-density reduced frequency grid, includes: Step S21: For the first converted frequency integral sub-interval, the collocation points are clustered towards its right endpoint; for the last converted frequency integral sub-interval, the collocation points are clustered towards its left endpoint; for the converted frequency integral sub-interval located in the middle, the collocation points are clustered towards its left and right endpoints respectively; and a corresponding converted frequency grid is generated for each converted frequency integral sub-interval. Step S22: Construct low-density and high-density reduced frequency grids within each reduced frequency integral sub-interval.

[0008] In one implementation, the non-uniform collocation points are generated based on the hyperbolic tangent function, and the degree of collocation point aggregation is controlled by stretching parameters.

[0009] In one embodiment, step S3, performing integration calculations based on the low-density and high-density reduced frequency grids respectively, to obtain integration results, namely the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix, includes: Step S31: Calculate the corresponding generalized displacement response spectrum matrix based on the low-density and high-density reduced frequency grids, respectively; Step S32: Perform numerical integration along the reduced frequency direction based on the corresponding generalized displacement response spectrum matrix to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix, respectively.

[0010] In one embodiment, during step S31, the generalized displacement response spectrum matrix corresponding to the reduced frequency grid is cached during the calculation of the generalized displacement response spectrum matrix, and the cached matrix is ​​called in subsequent calculations of the same reduced frequency grid.

[0011] In one implementation, step S4, constructing an integral convergence error criterion based on the integral result, and determining whether the integral result has converged, includes: Step S41: Calculate the matrix relative error and the modal root mean square response relative error based on the integration results; Step S42: Using the larger value between the matrix relative error and the modal root mean square response relative error as the comprehensive error, determine whether the comprehensive error is less than or equal to the preset error tolerance. If the integrated error is less than or equal to the preset error tolerance, the integration result is considered to have converged. Otherwise, the integral result is considered not to have converged.

[0012] In one embodiment, step S5, calculating the error contribution of each reduced frequency integral sub-interval and adaptively refining the points for the reduced frequency integral sub-intervals with larger error contributions, includes: Step S51: Calculate the error contribution of each reduced frequency integral sub-interval; Step S52: Determine whether the error contribution exceeds a preset threshold; For the sub-intervals of the converted frequency integral where the error contribution exceeds a preset threshold, at least the sub-interval of the converted frequency integral with the largest error contribution is encrypted to increase the number of points in the sub-interval. For the frequency integral sub-interval where the error contribution does not exceed a preset threshold, the number of points within that sub-interval is maintained.

[0013] In one embodiment, the bridge flutter response frequency domain adaptive integration method further includes: Step S6: For the response at a specified location or a specified degree of freedom of the bridge, construct the response projection matrix and integrate it based on the generalized response spectrum matrix to obtain the response covariance matrix and the root mean square value of the response at the specified location.

[0014] This application also provides a frequency-domain adaptive integral system for bridge buffeting response, which applies the frequency-domain adaptive integral method for bridge buffeting response as described in any of the above claims, and includes: The integration interval construction module is configured to: calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals; The non-uniform point allocation module is configured to perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid. The integration calculation module is configured to perform integration calculations based on the low-density reduced frequency grid and the high-density reduced frequency grid respectively, so as to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix respectively. The convergence error judgment module is configured to: construct an integral convergence error criterion based on the integral result, and determine whether the integral result has converged; if the integral result has converged, then use the high-density reduced frequency grid as the final integral grid. The adaptive encryption module is configured to: if the integration result does not converge, calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive encryption point allocation for the converted frequency integral sub-interval with a large error contribution; The iteration module is configured to repeatedly iterate until the integration result converges or reaches the maximum number of iterations or the maximum number of collocations to obtain the final integration grid.

[0015] The beneficial effects of the technical solutions provided in this application include: This application provides an adaptive integration method for the frequency domain of bridge buffeting response. First, the converted frequency is used as the endpoint to construct a frequency domain integration interval formed by multiple converted frequency integration sub-intervals. Then, within each integration sub-interval, an endpoint-clustered non-uniform point allocation method is adopted, automatically concentrating the converted frequency points near the modal converted frequencies where the response spectrum changes drastically. This improves the integration resolution of the response spectrum peak and rapidly changing intervals, while reducing redundant point allocation in smooth frequency bands and reducing reliance on subjective experience. Next, the method combines the subsequent calculation of the low-density and high-density generalized displacement response covariance matrices with adaptive point densification based on error contributions. Through these steps, the final integration grid is obtained collaboratively. This method ensures the accuracy of bridge buffeting response calculation while reducing the calculation of invalid frequency points, improving the accuracy, efficiency, stability, and engineering applicability of bridge frequency domain buffeting response calculation. This application has been successfully applied to the buffeting response assessment of the Xihoumen Highway-Railway Bridge, the world's largest cable-stayed suspension bridge spanning both road and rail. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the steps of a frequency-domain adaptive integral method for bridge buffeting response in one embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of an endpoint-clustered non-uniform point allocation in one embodiment of the present invention.

[0019] Figure 3 This is the displacement response spectrum of a typical position of the main beam under non-uniform point distribution in one embodiment of the present invention.

[0020] Figure 4 This is a comparison chart of calculation accuracy and efficiency in one embodiment of the present invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0022] This application provides a frequency domain adaptive integral method for bridge flutter response, which can solve the technical problems of manual point allocation relying on experience and low calculation efficiency of global uniform point allocation in related technologies.

[0023] like Figure 1 As shown, Figure 1 This is a flowchart illustrating the steps of a frequency-domain adaptive integral method for bridge buffeting response in one embodiment of the present invention.

[0024] This embodiment provides a frequency domain adaptive integral method for bridge buffeting response, which includes the following steps: Step S1: Calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals. Step S2: Perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid. Step S3: Perform integration calculations based on the low-density and high-density reduced frequency grids respectively to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix respectively. Step S4: Construct an integral convergence error criterion based on the integral result to determine whether the integral result has converged; If the integration result converges, the high-density reduced frequency grid is used as the final integration grid. If the integral result does not converge, proceed to step S5; Step S5: Calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive densification point allocation for the converted frequency integral sub-intervals with large error contributions; Repeat steps S2-S5 iteratively until the integration result converges or the maximum number of iterations or the maximum number of collocations is reached, to obtain the final integration grid.

[0025] This embodiment provides an adaptive integration method for the frequency domain of bridge buffeting response. First, the converted frequency is used as the endpoint to construct multiple frequency integration sub-intervals. Then, within each integration sub-interval, an endpoint-clustered non-uniform point allocation method is adopted, automatically concentrating the converted frequency points near the modal converted frequencies where the response spectrum changes drastically. This improves the integration resolution of the response spectrum peak and rapidly changing intervals, while reducing redundant point allocation in smooth frequency bands and reducing reliance on subjective experience. Next, the method combines the subsequent calculation of the low-density and high-density generalized displacement response covariance matrices with adaptive point densification based on error contributions. Through these steps, the final integration grid is obtained collaboratively. This method ensures the accuracy of bridge buffeting response calculation while reducing invalid frequency point calculations, improving the accuracy, efficiency, stability, and engineering applicability of bridge frequency domain buffeting response calculation. This application has been successfully applied to the buffeting response assessment of the Xihoumen Highway-Railway Bridge, the world's largest cable-stayed suspension bridge spanning both road and rail.

[0026] The following provides a detailed explanation of each step.

[0027] In one embodiment, step S1, calculating the reduced frequencies corresponding to each mode of the bridge structure, and using the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals, includes: Step S11: Calculate the reduced frequencies corresponding to each mode of the bridge structure. ; The calculation formula is: ; in, For each modal frequency; B This refers to the width of the bridge deck. U Average wind speed; Step S12: Using the converted frequency as the endpoint, set the conversion frequency integration range and determine multiple conversion frequency integration sub-intervals to construct the frequency domain integration interval.

[0028] Specifically, a lower and upper limit for integration of the converted frequencies are set. The lower limit, each converted frequency, and the upper limit are merged, sorted, and deduplicated to obtain multiple sub-intervals for the integration of converted frequencies. When the preset upper limit for integration does not cover the range of converted frequencies, the upper limit for integration is automatically expanded; when the preset lower limit for integration is higher than the range of converted frequencies, the lower limit for integration is automatically lowered so that the range of integration of converted frequencies covers the contribution interval of the main modal responses.

[0029] The above scheme constructs a frequency domain integration interval based on the modal characteristics of the bridge structure, ensuring the integrity of the coverage area.

[0030] In one embodiment, step S2, performing endpoint-clustered non-uniform point allocation within each reduced frequency integral sub-interval to generate a reduced frequency grid, and dividing it into a low-density reduced frequency grid and a high-density reduced frequency grid, includes: Step S21: For the first reduced frequency integral sub-interval, make the collocation points gather towards its right endpoint; for the last reduced frequency integral sub-interval, make the collocation points gather towards its left endpoint; for the reduced frequency integral sub-interval located in the middle, make the collocation points gather towards its left and right endpoints respectively; generate the corresponding reduced frequency grid for each reduced frequency integral sub-interval.

[0031] Furthermore, the non-uniform collocation points are generated based on the hyperbolic tangent function, and the degree of collocation point aggregation is controlled by the stretching parameter.

[0032] Step S22: Construct low-density and high-density reduced frequency grids within each reduced frequency integral sub-interval.

[0033] Specifically, the number of collocation points in a high-density reduced frequency grid is greater than that in a low-density reduced frequency grid. Furthermore, when the number of collocation points in a low-density reduced frequency grid is n, the number of collocation points in a high-density reduced frequency grid is 2n. 1.

[0034] The above scheme addresses the issue that bridge flutter response spectra typically exhibit peaks or rapidly changing intervals near the modal frequencies, while frequency bands far from the modal characteristic frequencies show relatively gentle response changes. Therefore, the endpoints of each sub-interval of the conversion frequency are set to include the conversion frequencies corresponding to each mode. This endpoint-clustered non-uniform point allocation method automatically concentrates the conversion frequency points near the peak of the modal response spectrum, thereby improving the integration resolution of the peak and rapidly changing intervals of the response spectrum.

[0035] In one embodiment, step S3, performing integration calculations based on both the low-density and high-density reduced frequency grids to obtain integration results, namely the corresponding low-density generalized displacement response covariance matrix and the high-density generalized displacement response covariance matrix, includes: Step S31: Calculate the corresponding generalized displacement response spectrum matrix based on the low-density and high-density reduced frequency grids, respectively; Step S32: Perform numerical integration along the reduced frequency direction based on the corresponding generalized displacement response spectrum matrix to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix, respectively.

[0036] Using the above scheme, integral calculations are performed based on low-density and high-density reduced frequency grids respectively to obtain integral results, which facilitates the subsequent construction of integral convergence error criteria.

[0037] In one embodiment, in step S31, during the calculation of the generalized displacement response spectrum matrix, the generalized displacement response spectrum matrix corresponding to the reduced frequency grid is cached and then called upon in subsequent calculations of the same reduced frequency grid.

[0038] The above scheme caches the generalized displacement response spectrum matrix corresponding to the calculated reduced frequency points during the calculation process. When the same or tolerance-compliant reduced frequency points reappear in subsequent low-density, high-density, or iteratively refined frequency grids, the cached results are directly called. That is, the cache mechanism reuses the generalized response spectrum matrix corresponding to the already calculated reduced frequency points, reducing the repeated calculation of aerodynamic spectrum matrix, aerodynamic stiffness matrix, aerodynamic damping matrix, and generalized response spectrum matrix during iterative refinement, and improving the computational efficiency under multimodal, multi-response point, and complex aerodynamic model conditions.

[0039] In one embodiment, step S4, constructing an integration convergence error criterion based on the integration result, and determining whether the integration result has converged, includes: Step S41: Calculate the relative error of the matrix and the relative error of the modal root mean square response based on the integration results.

[0040] Specifically, the relative error of the matrix is ​​determined by the ratio of the Frobenius norm of the difference between the low-density generalized response covariance matrix and the high-density generalized response covariance matrix to the Frobenius norm of the high-density generalized response covariance matrix. The relative error of the modal root mean square response is obtained by comparing the modal root mean square responses calculated from the diagonal elements of the low-density and high-density generalized response covariance matrices respectively.

[0041] Step S42: Using the larger value between the matrix relative error and the modal root mean square response relative error as the comprehensive error, determine whether the comprehensive error is less than or equal to the preset error tolerance. If the overall error is less than or equal to the preset error tolerance, the integration result is considered to have converged. The high-density reduced frequency grid is used as the final integration grid for the current calculation, and the generalized response spectrum matrix, generalized response covariance matrix, modal root mean square response and convergence information (such as whether it is less than or equal to the preset error tolerance) are output. If the overall error exceeds the preset error tolerance, the integral result is considered not to have converged, and the process proceeds to step S5.

[0042] Related techniques often use local spectral changes or differences in function values ​​at adjacent frequency points as the basis for encryption. However, bridge frequency domain buffeting analysis ultimately focuses not on the response spectrum value at a specific frequency point, but on the integrated generalized response covariance matrix, modal root-mean-square response, and the engineering response at a specified location. Therefore, the above scheme uses the integrated generalized displacement response covariance matrix and modal root-mean-square response as the convergence criteria, rather than simply comparing the response spectrum function values. This ensures that the integration error criterion directly corresponds to the final engineering objective of bridge buffeting response analysis, improving the reliability of the results.

[0043] In one embodiment, step S5, calculating the error contribution of each reduced frequency integral sub-interval, and adaptively refining the points for the reduced frequency integral sub-intervals with larger error contributions, includes: Step S51: Calculate the error contribution of each reduced frequency integral sub-interval; Step S52: Determine whether the error contribution exceeds a preset threshold; For the sub-intervals of the converted frequency integral where the error contribution exceeds a preset threshold, at least the sub-interval of the converted frequency integral with the largest error contribution is densified to increase the number of points in that sub-interval.

[0044] For the frequency integral sub-interval where the error contribution does not exceed the preset threshold, the number of points within that sub-interval is maintained.

[0045] Furthermore, in this approach, only the sub-intervals of the converted frequency integral where the error contribution exceeds a preset proportional threshold are encrypted, resulting in superior computational efficiency. Of course, in other embodiments, non-uniform encryption can be applied to all converted frequency sub-intervals, leading to faster convergence.

[0046] In related technologies, some encryption methods employ a globally uniform encryption strategy, which increases the number of points across all frequency intervals when the overall error does not meet requirements. While this method can improve integration accuracy, it cannot distinguish the contribution of different frequency intervals to the final response error. When the error is mainly concentrated near a few modal frequencies, globally uniform encryption results in a large amount of redundant computation, especially under conditions of multimodal coupling, multi-response point output, and complex aerodynamic derivative models, leading to high computational costs. The above scheme, however, performs local adaptive encryption based on the error contribution of each integration sub-interval, increasing the number of points only for the frequency intervals with significant error contributions, thus avoiding the large amount of invalid computation caused by globally uniform encryption.

[0047] In one embodiment, the frequency domain adaptive integration method for bridge buffeting response further includes: Step S6: For the response at a specified location or a specified degree of freedom of the bridge, construct the response projection matrix and integrate it based on the generalized response spectrum matrix to obtain the response covariance matrix and the root mean square value of the response at the specified location.

[0048] Specifically, for the response at a specified location or a specified degree of freedom of the bridge, a response projection matrix is ​​constructed, the generalized response spectrum matrix is ​​projected into the response spectrum matrix at the specified location, and the response spectrum matrix at the specified location is integrated to obtain the response covariance matrix and the root mean square value of the response at the specified location. The specified degree of freedom includes at least one of vertical displacement, lateral displacement, and torsional angle response.

[0049] The above scheme will output the final integral grid as an indicator of interest in actual engineering.

[0050] This application also provides a frequency-domain adaptive integral system for bridge buffeting response, which applies the above-described frequency-domain adaptive integral method for bridge buffeting response, and includes: The integration interval construction module is configured to: calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals; The non-uniform point allocation module is configured to perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid. The integration calculation module is configured to perform integration calculations based on low-density and high-density reduced frequency grids respectively, so as to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix respectively. The convergence error judgment module is configured to: construct an integral convergence error criterion based on the integral result, and determine whether the integral result has converged; if the integral result has converged, then use the high-density reduced frequency grid as the final integral grid. The adaptive encryption module is configured to: if the integration result does not converge, calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive encryption point allocation for the converted frequency integral sub-interval with a large error contribution; The iteration module is configured to repeatedly iterate until the integration result converges or the maximum number of iterations or the maximum number of collocations is reached, in order to obtain the final integration grid.

[0051] The functions of each module have been described above, and the same content will not be repeated here.

[0052] The following is a specific example for illustration.

[0053] Taking a super-long-span cable-stayed-suspension bridge with a main span of 1800m as an example, its first two lateral bending, vertical bending, and torsional frequencies are 0.051Hz, 0.114Hz, 0.126Hz, 0.135Hz, 0.329Hz, and 0.491Hz, respectively. The bridge structural parameters include the bridge deck width.B Natural frequency Structural damping ratio Generalized mass (related to linear displacement) and generalized mass moment of inertia Parameters (related to torsional displacement), main beam vibration mode ,in , and These represent the dimensionless vibration modes in the vertical, lateral, and torsional directions, respectively. Wind field parameters include average wind speed. U air density Downwind and vertical pulsating wind patterns and Spatial coherence parameter coh. Aerodynamic parameters include static three-component force coefficients. , and and its corresponding derivatives, flutter derivatives in different degrees of freedom directions and ( ), and aerodynamic admittance function .

[0054] This embodiment provides a frequency domain adaptive integral method for bridge buffeting response, which includes the following steps: Step S1: Calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals.

[0055] Specifically, take the average wind speed The dimensionless reduced frequencies of the sixth dominant mode were calculated. The values ​​are 0.360, 0.806, 0.891, 0.954, 2.326, and 3.471, respectively. The preset discrete conversion frequency is... Lower limit of points The maximum score is 0.1. If it is 5.0, If the value exceeds the preset value, the limit will be updated. , By using the converted frequency as the endpoint, multiple sub-intervals of the converted frequency integration are determined to construct the frequency domain integration interval.

[0056] Step S2: Perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid.

[0057] Specifically, such as Figure 2 The non-uniform grid 81 shown has the leftmost reduced frequency integral sub-interval. ,according to Perform right-end clustered non-uniform point allocation; where... This represents the endpoint stretching factor. The larger the value, the higher the degree of endpoint encryption. It is preferably set to 4.0. a and b These represent the two endpoints of the sub-interval of the frequency integral; u for Uniform grid between them. The central frequency integral sub-interval. , , , and ,according to Perform a non-uniform point allocation with two clusters at both ends. The rightmost reduced frequency integral subinterval. ,according to Perform a non-uniform point allocation with clustered points on the left end. This generates a reduced frequency grid. The initial number of points for the low-density reduced frequency grid in each sub-interval is n=5, and the number of points for the high-density grid is 2n-1.

[0058] like Figure 2 As shown, a comparison condition was also set up, with a high-density uniform point reference condition of 8193 points (corresponding to...). Figure 2 The reference uniform grid (8193) has a resolution of This is sufficient to cover the inherent dimensionless reduced frequency of the structure. ② Uniform point distribution working conditions with point numbers of 33, 65, 129, 257, 513, 1025, 2049 and 4097 (corresponding to Figure 2 The uniform grid is 33, 65, 129, 257.

[0059] from Figure 2 As can be seen, the differences between non-uniform and uniform meshes with similar point counts can be compared. Taking a uniform mesh with 129 points as an example, the points are distributed at the discrete reduced frequency. K 1. K 2. K 3. K Around 4, 129 points are uniformly distributed across the grid. K 2, K 3) With only two points within the interval, it cannot reflect the expected change in the response spectrum, which should first decrease and then increase within this interval. However, the endpoint-clustered non-uniform point allocation method provided in this embodiment adaptively addresses this issue. K 2, K 3) Multiple discrete reduced frequency points are inserted, which significantly increases the number of discrete reduced frequency points near the structure's inherent frequency point. Fewer grid points are needed to cover and improve the integral resolution of the response spectrum peak and rapidly changing range, while reducing redundant grid points in the smooth frequency band.

[0060] Step S3: Perform integration calculations based on the low-density and high-density reduced frequency grids respectively to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrices and high-density generalized displacement response covariance matrices.

[0061] Specifically, the formula for calculating the generalized displacement response spectrum matrix is ​​as follows: ; in, ; Discrete frequency points Total damping at the point: ; Discrete frequency points The overall stiffness matrix at: ; ; in, For the Kronecker delta function; The imaginary unit; and For flutter derivatives in different degrees of freedom directions; and represent , or . and middle i and j The index corresponds to the row and column of the matrix.

[0062] ; ; ; ; in, , and and , and These are the static three-component force coefficients and their derivatives, respectively; It is the aerodynamic admittance function; The relevant parameters are taken as coherent functions, based on the aerodynamic parameters of the main bridge beam and the wind and climate type of the engineering site.

[0063] The corresponding generalized displacement response spectrum covariance matrix .

[0064] Step S4: Construct an integral convergence error criterion based on the integral result to determine whether the integral result has converged; If the integral result does not converge, proceed to step S5.

[0065] Step S5: Calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive densification point allocation for the converted frequency integral sub-intervals with large error contributions; Repeat steps S2-S5 iteratively until the integration result converges to obtain the final integration grid.

[0066] Step S6: For the response at a specified location or a specified degree of freedom of the bridge, construct the response projection matrix and integrate it based on the generalized response spectrum matrix to obtain the response covariance matrix and the root mean square value of the response at the specified location.

[0067] Finally, the generalized displacement response spectrum is projected onto the response spectrum at a specified location on the main beam. The displacement response spectra at the mid-span and quarter-span points of the main span under different point allocation strategies are as follows: Figure 3 As shown. Under each collocation strategy, the maximum error of the root mean square displacement relative to the reference uniform grid at typical locations and the relative computational efficiency are as follows. Figure 4 As shown. Figure 4 In the non-uniform grid (897) and non-uniform grid (481), the comprehensive error between the high-density grid and the low-density grid is set to not exceed 1%, which are the results of encryption of all sub-intervals and encryption of only sub-intervals exceeding the error threshold, respectively.

[0068] like Figure 4 As shown, this embodiment employs a response spectrum calculation method based on preset tolerance iterative encryption. In terms of calculation accuracy and efficiency, it reproduces the rapid changes in the displacement response spectrum within the spectral peak and spectral gap by using a limited number of adaptive non-uniform discrete frequency points. Moreover, the response spectrum shape is much smoother than that under the same uniform grid point arrangement. Under the same number of points, the relative error is reduced by 47.7%, and the accuracy is improved by more than 5 times; under the same accuracy, the efficiency is improved by more than 10 times. It significantly improves the calculation accuracy and efficiency, providing an effective method for conducting probabilistic assessment of wind-induced vibration risk of long-span bridges under extreme weather conditions.

[0069] It should be noted that the sequence numbers of the embodiments in this application are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not represent a sequential order, nor do they limit "first," "second," and "third" to different types.

[0070] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0071] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0072] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0073] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A frequency-domain adaptive integral method for bridge buffeting response, characterized in that, It includes the following steps: Step S1: Calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals. Step S2: Perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid. Step S3: Perform integration calculations based on the low-density and high-density reduced frequency grids respectively to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix respectively. Step S4: Construct an integration convergence error criterion based on the integration result to determine whether the integration result has converged; If the integration result converges, the high-density reduced frequency grid is used as the final integration grid. If the integration result does not converge, proceed to step S5; Step S5: Calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive densification point allocation for the converted frequency integral sub-intervals with large error contributions; Repeat steps S2-S5 iteratively until the integration result converges or reaches the maximum number of iterations or the maximum number of collocation points to obtain the final integration grid.

2. The frequency domain adaptive integral method for bridge buffeting response as described in claim 1, characterized in that, Step S1 involves calculating the reduced frequencies corresponding to each mode of the bridge structure, using these reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals, including: Step S11: Calculate the reduced frequencies corresponding to each mode of the bridge structure. ; The calculation formula is: ; in, For each modal frequency; B This refers to the width of the bridge deck. U Average wind speed; Step S12: Using the converted frequency as the endpoint, set the converted frequency integration range and determine multiple converted frequency integration sub-intervals to construct the frequency domain integration interval.

3. The frequency domain adaptive integral method for bridge buffeting response as described in claim 1, characterized in that, Step S2, performing endpoint-clustered non-uniform point allocation within each reduced frequency integral sub-interval to generate a reduced frequency grid, and dividing it into low-density and high-density reduced frequency grids, includes: Step S21: For the first converted frequency integral sub-interval, the collocation points are clustered towards its right endpoint; for the last converted frequency integral sub-interval, the collocation points are clustered towards its left endpoint; for the converted frequency integral sub-interval located in the middle, the collocation points are clustered towards its left and right endpoints respectively; and a corresponding converted frequency grid is generated for each converted frequency integral sub-interval. Step S22: Construct low-density and high-density reduced frequency grids within each reduced frequency integral sub-interval.

4. The frequency domain adaptive integral method for bridge buffeting response as described in claim 3, characterized in that, The non-uniform point distribution is generated based on the hyperbolic tangent function, and the degree of point aggregation is controlled by the stretching parameter.

5. The frequency domain adaptive integral method for bridge buffeting response as described in claim 1, characterized in that, Step S3 involves performing integration calculations based on the low-density and high-density reduced frequency grids, respectively, to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrices and high-density generalized displacement response covariance matrices, including: Step S31: Calculate the corresponding generalized displacement response spectrum matrix based on the low-density and high-density reduced frequency grids, respectively; Step S32: Perform numerical integration along the reduced frequency direction based on the corresponding generalized displacement response spectrum matrix to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix, respectively.

6. The frequency domain adaptive integral method for bridge buffeting response as described in claim 5, characterized in that, In step S31, during the calculation of the generalized displacement response spectrum matrix, the generalized displacement response spectrum matrix corresponding to the reduced frequency grid is cached and called in subsequent calculations of the same reduced frequency grid.

7. The frequency domain adaptive integral method for bridge buffeting response as described in claim 1, characterized in that, Step S4, constructing an integration convergence error criterion based on the integration result, and determining whether the integration result has converged, includes: Step S41: Calculate the matrix relative error and the modal root mean square response relative error based on the integration results; Step S42: Using the larger value between the matrix relative error and the modal root mean square response relative error as the comprehensive error, determine whether the comprehensive error is less than or equal to the preset error tolerance. If the integrated error is less than or equal to the preset error tolerance, the integration result is considered to have converged. Otherwise, the integral result is considered not to have converged.

8. The frequency domain adaptive integral method for bridge buffeting response as described in claim 1, characterized in that, Step S5, calculating the error contribution of each converted frequency integral sub-interval, and adaptively refining the points for the converted frequency integral sub-intervals with larger error contributions, includes: Step S51: Calculate the error contribution of each reduced frequency integral sub-interval; Step S52: Determine whether the error contribution exceeds a preset threshold; For the sub-intervals of the converted frequency integral where the error contribution exceeds a preset threshold, at least the sub-interval of the converted frequency integral with the largest error contribution is encrypted to increase the number of points in the sub-interval. For the frequency integral sub-interval where the error contribution does not exceed a preset threshold, the number of points within that sub-interval is maintained.

9. The frequency domain adaptive integral method for bridge buffeting response as described in claim 1, characterized in that, The frequency domain adaptive integral method for bridge buffeting response also includes: Step S6: For the response at a specified location or a specified degree of freedom of the bridge, construct the response projection matrix and integrate it based on the generalized response spectrum matrix to obtain the response covariance matrix and the root mean square value of the response at the specified location.

10. A frequency-domain adaptive integral system for bridge buffeting response, employing the frequency-domain adaptive integral method for bridge buffeting response as described in any one of claims 1 to 9, characterized in that, It includes: The integration interval construction module is configured to: calculate the reduced frequencies corresponding to each mode of the bridge structure, and use the reduced frequencies as endpoints to construct a frequency domain integration interval formed by multiple reduced frequency integration sub-intervals; The non-uniform point allocation module is configured to perform endpoint-clustered non-uniform point allocation within each sub-interval of the reduced frequency integral to generate a reduced frequency grid, which is divided into a low-density reduced frequency grid and a high-density reduced frequency grid. The integration calculation module is configured to perform integration calculations based on the low-density reduced frequency grid and the high-density reduced frequency grid respectively, so as to obtain the integration results, which are the corresponding low-density generalized displacement response covariance matrix and high-density generalized displacement response covariance matrix respectively. The convergence error judgment module is configured to: construct an integral convergence error criterion based on the integral result, and determine whether the integral result has converged; If the integration result converges, the high-density reduced frequency grid is used as the final integration grid. The adaptive encryption module is configured to: if the integration result does not converge, calculate the error contribution of each converted frequency integral sub-interval, and perform adaptive encryption point allocation for the converted frequency integral sub-interval with a large error contribution; The iteration module is configured to repeatedly iterate until the integration result converges or reaches the maximum number of iterations or the maximum number of collocations to obtain the final integration grid.