Search

CN-121996955-A - High-rise stay wire mast modal decoupling and multi-strategy collaborative modal parameter identification method

CN121996955ACN 121996955 ACN121996955 ACN 121996955ACN-121996955-A

Abstract

The invention discloses a method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towline mast, which comprises the steps of collecting acceleration response signals and preprocessing, determining the frequency of a binary circle based on the excellent frequency of a power spectrum, separating a tower body and a fiber rope mode by using an analysis modal decomposition method, adaptively selecting an identification method from an algorithm library containing SSI-COV, FBFFT, RDT and FDD according to the number of monitoring points, the sample scale and the identification accuracy requirement, obtaining modal parameters, and finally outputting a final result through multi-dimensional criteria and cross verification. The invention breaks through the dense coupling mode separation bottleneck of the high-rise guy mast, obviously improves the accuracy and efficiency of the modal parameter identification of the high-rise guy mast, is suitable for the modal parameter identification under the condition of environmental excitation or limited measuring points, and can provide key technical support for dynamic characteristic evaluation, safety monitoring, wind-induced response analysis and full life cycle health management of the high-rise guy mast.

Inventors

  • LIU MUGUANG
  • XIE WANSHENG
  • WANG ZHE

Assignees

  • 华南理工大学

Dates

Publication Date
20260508
Application Date
20260112

Claims (10)

  1. 1. The method for identifying the modal decoupling and multi-strategy collaborative modal parameters of the towering stay wire mast is characterized by comprising the following steps of: S1, arranging acceleration sensors at different heights of a high-rise guy mast structure, collecting acceleration response signals of the structure under environmental excitation, and preprocessing the response signals; S2, analyzing and modal decomposing AMD (advanced data processing) is carried out on the response signals after pretreatment, wherein the AMD comprises the steps of extracting excellent frequencies based on power spectrum analysis, determining binary circle frequencies according to trough between adjacent excellent frequencies; s3, based on the decoupled vibration components of the tower body, selecting an adaptive modal parameter identification method from a preset algorithm library to calculate according to the number of monitoring points, the sample scale and the identification precision requirement by adopting a self-adaptive strategy, and obtaining modal frequency, damping ratio and vibration mode; and S4, establishing multidimensional verification indexes of frequency stability, damping stability and vibration mode similarity, carrying out collaborative cross verification on modal parameters identified by different algorithms, eliminating false modes, and outputting a final modal parameter identification result.
  2. 2. The method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast according to claim 1, wherein in step S1, the preprocessing is specifically: And carrying out frequency domain filtering pretreatment on the acquired response signals by adopting a Kaiser window FIR band-pass filter, and reserving low-frequency dominant vibration frequency band signals of 0-2.5 Hz.
  3. 3. The method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast according to claim 1, wherein step S2 is specifically: Performing power spectrum analysis on the acceleration response signal by a Welch power spectrum estimation method, extracting excellent frequencies in a low frequency range, and determining a frequency interval of a dense mode; according to the distribution of the excellent frequencies, calculating the corresponding frequencies of the trough between adjacent excellent frequencies as binary circle frequencies, and dividing frequency boundaries of a tower body mode and a fiber rope mode; constructing a sub-signal separation function by using Hilbert transform: ; Wherein i=1, 2,..n-1; is the corresponding i-th segment sub-signal; for the ith divide-by-two circle frequency, H [. Cndot ] represents the Hilbert transform, expressed as: ; and decomposing the original signal according to the i-1 binary circle frequencies, so as to eliminate the modal interference of the fiber ropes.
  4. 4. The method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast according to claim 1, wherein in step S3, a preset algorithm library comprises: covariance driving random subspace SSI-COV, fast Bayesian FFT FBFFT, random decrement RDT and frequency domain decomposition FDD; In step S3, the adaptive strategy specifically includes: when the monitoring points are multiple points and high requirements are met on identification precision and vibration mode identification, adopting a method of combining SSI-COV with a stable graph; When the monitoring points are multiple points, the sample data size is large and the calculation efficiency is high, a FBFFT method is adopted; When the monitoring point is a single point and the sample sampling time length meets the statistical requirement, adopting an RDT method or an FDD method.
  5. 5. The method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towline mast according to claim 4, wherein when calculating by adopting the method of combining SSI-COV with a stability diagram, the method comprises: The method comprises the steps of calculating an output covariance matrix based on vibration components of a multi-measuring-point tower body, constructing a Toeplitz matrix, determining an effective order of a system through a stable state of singular entropy increment, carrying out SVD singular value decomposition on the weighted Toeplitz matrix, removing noise components corresponding to micro singular values, solving a system matrix of a discrete state space model, carrying out eigenvalue decomposition on the system matrix to obtain a modal parameter initial value set, constructing a multi-dimensional criterion stability diagram comprising frequency, damping ratio and modal confidence MAC, and screening real modal parameters meeting stability conditions, wherein the judging threshold of the stability conditions is that the frequency tolerance is less than or equal to 5%, the damping ratio tolerance is less than or equal to 15% and the vibration mode tolerance is less than or equal to 5%; The method comprises the following specific steps: calculating and outputting covariance sequence based on vibration components of tower body with multiple measuring points I is a time interval, y k represents an output vector of the system at k time, and a Toeplitz matrix is constructed: ; Wherein, the Obtained from the Hankel matrix, The value range of the row block number i of the Toeplitz matrix is 2 beta-4 beta, and beta is the ratio of the sampling frequency to the basic frequency of the mast structure; SVD singular value decomposition is carried out on the weighted Toeplitz matrix, and then: ; Wherein, the ; Calculating the effective order of a singular entropy increment judging system, and calculating the singular entropy increment The calculation formula of (2) is as follows: ; Wherein, the Is the i-th order discrete time complex eigenvalue; When the singular entropy increment tends to be stable, the corresponding order is the effective order of the system, so that no omission and redundancy of modal information are ensured; calculation of a considerable matrix from the Toeplitz matrix for singular value decomposition results Inversion controllable matrix Identifying system matrix A based on least square method, and decomposing eigenvalue to obtain discrete time complex eigenvalue The system matrix A is identified by adopting a linear least square method and is expressed as: ; Wherein, the Representing a pseudo-inverse operation; Performing eigenvalue decomposition on the system matrix A, and calculating structural modal parameters: ; Wherein, the =Diag (λ i ) is the i-th order diagonal matrix consisting of discrete-time complex eigenvalues; Is a complex eigenvector matrix; ; Wherein Re represents the real part; constructing a multi-dimensional criterion stability diagram comprising frequency stability, damping ratio stability and MAC vibration mode similarity, screening real modal parameters meeting stability conditions, wherein the judgment criterion of stability points is that the following conditions are simultaneously met: ; Wherein j is the system order, f, 、 Respectively representing the frequency, damping ratio and vibration mode corresponding to each order, wherein MAC is a mode confidence factor; And when the deviation of the high-order parameter and the low-order parameter is within the tolerance range, judging the mode as a stable mode, and eliminating the divergent false mode to obtain the final mode parameter.
  6. 6. The method for identifying modal decoupling and multi-strategy collaborative modal parameters for a towering mast according to claim 4, wherein when computing by FBFFT comprises: constructing a spectrum density matrix, carrying out eigenvalue decomposition, taking an eigenvector corresponding to the maximum eigenvalue as an optimal estimation value of a vibration mode, constructing a negative log likelihood function, adopting an unconstrained optimization function to iteratively solve the frequency and damping ratio for minimizing the likelihood function; The method comprises the following specific steps: Performing discrete Fourier transform FFT (fast Fourier transform) on the signal containing the single-mode information obtained in the step S2 to obtain a real part of a response signal And imaginary part The calculation formula is as follows: ; Wherein, the Representing a sampling interval; And Representing the real and imaginary parts of the response signal y j fourier transformed, respectively; k=1, 2,) N, but taking into account the end-point effect and the fourier transform, the conjugated bilateral spectrum is obtained only taking k=2, 3,) the data corresponding to int [ N/2] +1 are subjected to subsequent analysis, with corresponding frequency f k = (k-1)/(N ); Calculating the process quantity and constructing a matrix A 0 : ; Decomposing the characteristic value of A 0 , and taking the characteristic vector corresponding to the maximum characteristic value lambda 0 as the best estimated value of the vibration mode vector And calculate Predicting spectral density errors MPV values of (2): ; Where N f denotes the number of data points used for analysis, i.e., N f = int [ N/2]; Constructing a log-likelihood function, solving a frequency f and a damping ratio : ; Wherein, the ; Optimizing the matlab by adopting an unconstrained optimization function fminunc to obtain the modal frequency and the damping ratio; Assembling the Hessian matrix of the log-likelihood function by the second partial derivative of the log-likelihood function: ; Wherein, the Is that Is characterized in that, Is the corresponding characteristic value; After the calculation of the covariance matrix C, The posterior uncertainty of the identified modal parameters is measured by the coefficient of variation, and a confidence interval reference is provided for engineering application.
  7. 7. The method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast according to claim 4, wherein when calculating by using the RDT method, the method comprises: Calculating the root mean square RMS of the acceleration of the modal component of the tower body, taking the root mean square RMS as a self-adaptive trigger threshold, and intercepting a plurality of sections of random subsamples meeting the threshold requirement in a crossing trigger mode, performing time alignment and arithmetic average on all the random subsamples to obtain a free damping vibration response curve for removing random noise interference; The method comprises the following specific steps: and (2) utilizing the signal of the single-mode information obtained in the step (S2) to set a trigger threshold A to intercept the signal to obtain a series of different intersection points, and obtaining random subsamples, wherein the random subsamples are expressed as follows: ; wherein, D (t) represents the free vibration response of the system with initial displacement of 1 and initial speed of 0, V (t) represents the free vibration response of the system with initial displacement of 0 and initial speed of 1; H (t) is a unit impulse response function, f (t) is external excitation; for random subsamples Performing time alignment and arithmetic average to eliminate the influence of random excitation, and obtaining the free vibration response X (t) with initial displacement A and initial speed 0: ; wherein N is the number of random subsamples; Extracting a free vibration attenuation curve, wherein the wave crests and the wave troughs of the free vibration attenuation curve are exponentially attenuated: ; Zeta is the damping ratio of the structure, w n is the undamped circular frequency of the structure, and the frequency and the damping ratio of the structure are obtained by fitting the wave crest and the wave trough of the free vibration damping curve obtained by the RDT method according to the above formula.
  8. 8. The method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast according to claim 4, wherein when calculating by using the FDD method, comprising: calculating a power spectrum density matrix of the acceleration response signal, and carrying out singular value decomposition on the power spectrum density matrix; Extracting the frequency corresponding to the peak value of the singular value curve as the candidate tower body modal frequency, and judging the dominant modal through the modal confidence MAC; selecting frequency band data near a peak value, performing inverse Fourier transform (IFFT) to obtain a normalized autocorrelation function, and identifying a damping ratio of a corresponding mode through linear fitting of logarithmic attenuation rate; The method comprises the following specific steps: and (2) adopting the signal of the single-mode information obtained in the step (S2), wherein the input and output power spectral density matrixes of the system meet the following relations: ; Wherein, the A structural frequency response function matrix; And The power spectrum density matrixes are respectively input and output by the system; Complex conjugate transpose operation of the representation matrix; For a pair of Singular value decomposition is carried out, and when a certain frequency interval only has the dominant mode of the r th order, the decomposition result is approximately as follows: ; Wherein, the I.e. the mode shape vector of the r-th order mode; d r is a real constant; Re represents the real part; determining the mode frequency of the r-th order by comparing the MAC values of the mode shapes near the peak value of the singular value curve; performing inverse Fourier transform (IFFT) on the corresponding frequency band to obtain a normalized autocorrelation function, and performing logarithmic attenuation rate Obtaining the mode damping ratio of the r order by linear fitting : ; Wherein, the And The initial value of the autocorrelation function and the value after attenuation of k peaks are represented, respectively.
  9. 9. The method for identifying modal decoupling and multi-strategy collaborative modal parameters for a towering mast according to claim 6, wherein an initial iteration value of an unconstrained optimization function is set as: The initial value of the frequency is the corresponding excellent frequency extracted by AMD, and the initial value of the damping ratio is 0.5%.
  10. 10. The method for identifying modal decoupling and multi-strategy collaborative modal parameters for a towering mast according to claim 7, wherein the trigger threshold is set to 0.8 to 1.5 times the root mean square RMS of the acceleration response, and when the number of sub-samples taken is less than a predetermined value, the threshold is automatically reduced and re-taken until no less than the predetermined number of valid sub-samples are obtained.

Description

High-rise stay wire mast modal decoupling and multi-strategy collaborative modal parameter identification method Technical Field The invention belongs to the technical field of structural engineering, and particularly relates to a method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast. Background The high-rise mast structure has the advantages of small windward area, low manufacturing cost, convenient installation and the like, and is widely applied to the fields of communication, meteorological observation, power transportation and the like. However, such structures generally have high flexibility, low damping ratio and remarkable nonlinear dynamic characteristics, and under the action of strong wind such as typhoons, downburst and the like, the structures show complex dynamic response characteristics, and the multimode participation is remarkable, so that the integral bending mode of the tower body and the local vibration mode of the fiber rope are easily excited at the same time, and dense and overlapped mode components appear in the frequency spectrum. When the traditional mode decomposition methods such as EMD and VMD process such dense modes, the problems of mode aliasing, end-point effect, signal reconstruction distortion and the like are commonly existed, the tower body and the fiber rope modes are difficult to effectively separate, and the error of the subsequent mode identification result is increased. In addition, strong wind measured data often accompanies high noise and non-stationary characteristics, and the difficulty of modal identification is further aggravated. The existing modal parameter identification method mostly adopts a single algorithm frame, and is difficult to meet the different requirements of different engineering scenes on identification precision, calculation efficiency, measuring point number and data quality. Therefore, a technical scheme capable of accurately and efficiently separating dense modes and flexibly adapting an identification algorithm according to actual engineering scenes and accurately identifying mode parameters under the action of strong wind is needed, so that accuracy, reliability and engineering applicability of identifying the mode parameters of the high-rise mast structure under the strong wind working condition are improved. Disclosure of Invention The invention mainly aims to overcome the defects and shortcomings of the prior art and provides a high-rise stay wire mast modal decoupling and multi-strategy collaborative modal parameter identification method. In order to achieve the above purpose, the present invention adopts the following technical scheme: A method for identifying modal decoupling and multi-strategy collaborative modal parameters of a towering mast comprises the following steps: S1, arranging acceleration sensors at different heights of a high-rise guy mast structure, collecting acceleration response signals of the structure under environmental excitation, and preprocessing the response signals; S2, analyzing and modal decomposing AMD (advanced data processing) is carried out on the response signals after pretreatment, wherein the AMD comprises the steps of extracting excellent frequencies based on power spectrum analysis, determining binary circle frequencies according to trough between adjacent excellent frequencies; s3, based on the decoupled vibration components of the tower body, selecting an adaptive modal parameter identification method from a preset algorithm library to calculate according to the number of monitoring points, the sample scale and the identification precision requirement by adopting a self-adaptive strategy, and obtaining modal frequency, damping ratio and vibration mode; and S4, establishing multidimensional verification indexes of frequency stability, damping stability and vibration mode similarity, carrying out collaborative cross verification on modal parameters identified by different algorithms, eliminating false modes, and outputting a final modal parameter identification result. Compared with the prior art, the invention has the following advantages and beneficial effects: 1. The method realizes separation of dense modes of the tower body and the fiber ropes by utilizing an AMD method, fundamentally solves the problems of modal aliasing and end-point effect existing in the traditional methods such as EMD, VMD and the like, has extremely low reconstruction error of decomposed signals, completely retains the dynamic characteristics of original signals, innovatively integrates multiple algorithms of SSI-COV, FBFFT, RDT and FDD, breaks the limitation of a single algorithm through multi-strategy self-adaptive selection and collaborative cross verification to form a complementary recognition mechanism, remarkably improves the engineering applicability of the method, designs differential recognition paths aiming at different engineering scenes, gives consider