Search

CN-122016201-A - Automatic batch identification method for vortex-induced vibration of stay cables

CN122016201ACN 122016201 ACN122016201 ACN 122016201ACN-122016201-A

Abstract

The application relates to a stay cable vortex-induced vibration batch automatic identification method, which comprises the steps of firstly, automatically detecting wave crests in batches and automatically extracting wave crests of a plurality of stay cable vibration signals in batches; and finally, judging the vibration states of all stay cables at one time through the threshold values of the slope and the standard deviation, thereby realizing the automatic batch identification of the vortex-induced vibration of the stay cables. The application solves the problems that the traditional vortex-induced vibration identification method can only identify one stay cable vortex-induced vibration at a time, has low identification efficiency and large resource requirement, can be used for on-line identification of the stay cable vortex-induced vibration and automatic analysis of the stay cable long-term monitoring data of the bridge health monitoring system, and improves the efficiency of the stay cable vortex-induced vibration identification.

Inventors

  • HE MIN
  • WANG LIJUN

Assignees

  • 西安科技大学

Dates

Publication Date
20260512
Application Date
20260122

Claims (8)

  1. 1. The automatic batch identification method for vortex-induced vibration of stay cables is characterized by comprising the following steps of: acquiring stay cable original vibration signals of a plurality of stay cables, and establishing an original signal matrix; The method comprises the steps of numbering elements in a crest matrix, normalizing the numbers to obtain a normalized time matrix, calculating the average value of each column of elements of the crest matrix to form crest average value row vectors, expanding the crest average value row vectors to be the same dimension as the crest matrix to obtain the crest average value matrix, calculating a linear regression slope vector and a linear regression intercept matrix according to the normalized time matrix, the crest average value matrix and the crest matrix, and eliminating trend items of the crest according to the linear regression slope vector and the linear regression intercept matrix to obtain a residual matrix; calculating standard deviation of each column of the residual matrix to obtain a standard residual vector; And comparing each column of elements of the standard residual error vector and each column of elements of the linear regression slope vector with corresponding thresholds respectively to obtain the recognition result of the stay cable vortex-induced vibration of each stay cable.
  2. 2. The method of claim 1, wherein determining peak locations from the original signal matrix, establishing a peak matrix, comprises: Obtaining a forward differential matrix and a backward differential matrix according to the original signal matrix; Setting elements smaller than 0 in the forward differential matrix and the backward differential matrix as 0, and calculating the product of the forward differential matrix and the backward differential matrix to obtain a product matrix; And determining the positions of non-0 elements in the product matrix, and extracting the elements in the corresponding positions in the original signal matrix to obtain a wave crest matrix.
  3. 3. The method of claim 1, wherein computing a linear regression slope vector and a linear regression intercept matrix from the normalized time matrix, the peak-to-average matrix, and the peak matrix comprises: calculating the average value of each column of the normalized time matrix to obtain a time average value vector, and expanding the time average value vector to obtain a time average value matrix; The linear regression slope vector is: Wherein, the Is a linear regression slope vector which is used to determine, In order to normalize the time matrix, In the form of a time-averaged matrix, In the form of a matrix of peaks, As the peak-to-average value matrix, Is that Elements of the ith column of the matrix, sum representing the sum of the columns of the matrix; Calculating a linear regression intercept row vector: Wherein, the As a linear regression of the intercept row vector, Is a linear regression slope vector which is used to determine, Is a time mean vector; And expanding the linear regression intercept row vector to obtain a linear regression intercept matrix.
  4. 4. The method of claim 1, wherein removing trend terms of peaks from the linear regression slope vector and linear regression intercept matrix to obtain a residual matrix comprises: Wherein, the As a residual matrix of the image data, In the form of a matrix of peaks, Is a linear regression slope vector which is used to determine, Is a linear regression intercept matrix.
  5. 5. The method of claim 1, wherein comparing each column of elements of the standard residual vector and each column of elements of the linear regression slope vector with corresponding thresholds, respectively, results in an identification of a stay cable vortex-induced vibration for each stay cable, comprising: If it is The vibration is random vibration, wherein, An ith column element which is a standard residual vector; If it is And is also provided with The vibration is a vortex-induced vibration stabilization stage, wherein, Column i element of the linear regression slope vector; If it is And is also provided with The vibration is a vortex-induced vibration forming stage; If it is And is also provided with The vibration is in the vortex-induced vibration dissipation phase.
  6. 6. The utility model provides a suspension cable vortex-induced vibration batched automatic identification device which characterized in that includes: the system comprises a peak matrix establishing module, a peak matrix determining module and a peak matrix determining module, wherein the peak matrix establishing module is used for acquiring stay cable original vibration signals of a plurality of stay cables and establishing an original signal matrix; The peak trend item removing module is used for numbering elements in the peak matrix and normalizing the numbers to obtain a normalized time matrix, calculating the average value of each column of elements of the peak matrix to form a peak average value row vector, expanding the peak average value row vector to the same dimension as the peak matrix to obtain a peak average value matrix, and calculating a linear regression slope vector and a linear regression intercept matrix according to the normalized time matrix, the peak average value matrix and the peak matrix; The standard residual vector determining module is used for calculating standard deviations of each column of the residual matrix to obtain a standard residual vector; And the vibration identification module is used for comparing each column of elements of the standard residual error vector and each column of elements of the linear regression slope vector with corresponding thresholds respectively to obtain the identification result of the vortex-induced vibration of the stay cable of each stay cable.
  7. 7. A computer readable storage medium, characterized in that the computer readable storage medium stores a computer program which, when executed by a processor, implements the method for automatic mass identification of vortex-induced vibration of a suspension cable according to any one of claims 1-5.
  8. 8. A computer program product comprising computer programs/instructions which, when executed by a processor, implement the method for automated mass identification of stay cable vortex-induced vibration of any one of claims 1 to 5.

Description

Automatic batch identification method for vortex-induced vibration of stay cables Technical Field The application relates to the field of bridge health monitoring stay cable vortex-induced vibration identification, in particular to a batch automatic identification method for stay cable vortex-induced vibration. Background Stay cable vortex-induced vibration is typical abnormal vibration of stay cables, and in bridge health monitoring, monitoring and identifying stay cable vortex-induced vibration are important requirements for bridge health monitoring. The existing stay cable vortex-induced vibration identification method mainly comprises a displacement reconstruction method based on acceleration integration, a decision tree method based on historical data, an automatic clustering method based on time-frequency domain characteristic indexes and the like. The method has good effect in recognizing vortex-induced vibration of the main beam and the single stay cable, and can accurately recognize vortex-induced vibration of the stay cable, but in daily monitoring, the stay cables are monitored in batches, and the existing method can only analyze one stay cable at a time. In the on-line monitoring, the existing method needs to perform parallel analysis to perform vibration states of all monitored stay cables, hardware resources are very consumed, and in the off-line data long-term analysis, all stay cables can be analyzed in a cyclic analysis mode, but the timeliness of the method is low, and the calculation time is very consumed. Disclosure of Invention In order to overcome at least one defect in the prior art, the application provides a stay cable vortex-induced vibration batch automatic identification method. In a first aspect, a method for automatically identifying vortex-induced vibration of stay cables in batches is provided, which comprises the following steps: Acquiring stay cable original vibration signals of a plurality of stay cables, and establishing an original signal matrix; The method comprises the steps of numbering elements in a wave crest matrix, normalizing the numbers to obtain a normalized time matrix, calculating the average value of each column element of the wave crest matrix to form a wave crest average value row vector, expanding the wave crest average value row vector to be the same dimension as the wave crest matrix to obtain the wave crest average value matrix, calculating a linear regression slope vector and a linear regression intercept matrix according to the normalized time matrix, the wave crest average value matrix and the wave crest matrix, and removing trend items of wave crests according to the linear regression slope vector and the linear regression intercept matrix to obtain a residual matrix; calculating standard deviation of each column of the residual matrix to obtain a standard residual vector; And comparing each column of elements of the standard residual error vector and each column of elements of the linear regression slope vector with corresponding thresholds respectively to obtain the recognition result of the stay cable vortex-induced vibration of each stay cable. In one embodiment, determining peak locations from an original signal matrix, establishing a peak matrix, includes: obtaining a forward differential matrix and a backward differential matrix according to the original signal matrix; setting elements smaller than 0 in the forward differential matrix and the backward differential matrix as 0, and calculating the product of the forward differential matrix and the backward differential matrix to obtain a product matrix; And determining the positions of non-0 elements in the product matrix, and extracting the elements at the corresponding positions in the original signal matrix to obtain a wave crest matrix. In one embodiment, calculating a linear regression slope vector and a linear regression intercept matrix from the normalized time matrix, the peak mean matrix, and the peak matrix, comprises: Calculating the average value of each column of the normalized time matrix to obtain a time average value vector, and expanding the time average value vector to obtain a time average value matrix; The linear regression slope vector is: Wherein, the Is a linear regression slope vector which is used to determine,In order to normalize the time matrix,In the form of a time-averaged matrix,In the form of a matrix of peaks,As the peak-to-average value matrix,Is thatElements of the ith column of the matrix, sum representing the sum of the columns of the matrix; Calculating a linear regression intercept row vector: Wherein, the As a linear regression of the intercept row vector,Is a linear regression slope vector which is used to determine,Is a time mean vector; And expanding the linear regression intercept row vector to obtain a linear regression intercept matrix. In one embodiment, removing trend terms of peaks according to the linear regression slope vector and the linea