CN114777985A - Iron tower bolt complete loosening rapid detection method based on vibration characteristics - Google Patents
Iron tower bolt complete loosening rapid detection method based on vibration characteristics Download PDFInfo
- Publication number
- CN114777985A CN114777985A CN202210237693.0A CN202210237693A CN114777985A CN 114777985 A CN114777985 A CN 114777985A CN 202210237693 A CN202210237693 A CN 202210237693A CN 114777985 A CN114777985 A CN 114777985A
- Authority
- CN
- China
- Prior art keywords
- vibration
- signal
- bolt
- tower
- frequency
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L5/00—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes
- G01L5/24—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes for determining value of torque or twisting moment for tightening a nut or other member which is similarly stressed
- G01L5/246—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes for determining value of torque or twisting moment for tightening a nut or other member which is similarly stressed using acoustic waves
Landscapes
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a method for quickly detecting complete loosening of a bolt of an iron tower based on vibration characteristics, which is used for quickly detecting the fault defect of complete loosening of the bolt of the power transmission iron tower, collecting and analyzing vibration responses of two tower materials connected under pulse excitation before and after the bolt of the power transmission iron tower is completely loosened, processing a vibration signal by using an autocorrelation-VMD (linear velocity difference decomposition) method, and finally detecting whether the bolt connection structure is completely loosened by comparing spectrograms generated by the vibration responses before and after the bolt is completely loosened. Modal aliasing and end-point effects are well avoided by using Variational Modal Decomposition (VMD).
Description
Technical Field
The invention relates to the technical field of power equipment, in particular to a method for quickly detecting complete looseness of an iron tower bolt based on vibration characteristics.
Background
At present, the bolt is used as an important connecting piece on a tower material of a power transmission tower, and plays a very crucial role in ensuring the safe and stable operation of the power transmission tower. The power transmission iron tower is influenced by strong wind all the year round, so that bolts are gradually loosened, the rigidity of the power transmission iron tower is reduced, and the normal operation of the power transmission iron tower is seriously influenced. Therefore, the running condition of the bolt is evaluated, and corresponding maintenance is imperative in time.
Experts and scholars at home and abroad make a large amount of theoretical researches and experiments to detect the loosening fault of a bolt connection structure, wherein detection is mainly carried out by climbing iron towers through maintainers and utilizing torque wrenches, and ultrasonic detection, pretightening force detection, vibration signal detection technology and the like. In the research of the ultrasonic detection method, the ultrasonic detection technology is applied to the monitoring of the bolt state of the GIS basin-type insulator, and the detection of bolt looseness is successfully realized; the chaotic ultrasonic excitation is applied to the detection of the looseness of the single-limb bolt connection structure, and the characteristic parameters of the bolt looseness are obtained. Although the ultrasonic detection technology has the characteristics of nondestructive detection, high resolution and the like, the accuracy of the ultrasonic detection technology needs to be improved. In the research of the pretightening force detection method, the pretightening force detection method comprises the steps of simulating the variation process of the pretightening force of the bolt under the transverse load to judge the bolt loosening process; the anti-loosening performance of the bolt under the action of the transverse vibration load is quantitatively evaluated by using an experimental method. Although the pretightening force detection method can realize accurate evaluation of bolt loosening conditions by measuring the axial pretightening force of a bolt screw, the method is not suitable for quick detection of bolt loosening because the pretightening force of the bolt needs to be monitored for a long time. The method is most widely applied at present, the bolt looseness is detected and predicted according to the difference of vibration signals before and after the bolt looseness, and the prior art comprises the steps of applying external excitation to a power transmission tower, further measuring the vibration signals of a bolt connection structure of the power transmission tower, and evaluating the bolt looseness in terms of the difference of vibration amplitude values; applying a method based on vibration sensitive characteristics and manifold learning to diagnosis of the loosening degree of the bolts of the base of the fan; judging whether the bolt is loosened or not by using the change characteristics of the vibration modal parameters before and after the bolt is loosened; the sub-harmonic resonance identification method is applied to bolt damage detection.
In the engineering practice, external interference noise seriously affects the acquisition of vibration signals, so that further noise reduction processing needs to be carried out on the vibration signals, and the prior art comprises the step of applying a wavelet analysis method to the processing of the vibration signals of the system before and after the bolt is loosened; empirical Mode Decomposition (EMD) is used for processing bolt loosening vibration signals, but EMD algorithm has serious problems, such as defects of low efficiency, mode aliasing and end point effect, and therefore fault identification is often influenced.
Therefore, how to realize the rapid and accurate detection of the defect of the complete loosening fault of the bolt of the power transmission tower is a problem to be solved urgently by the technical personnel in the field.
Disclosure of Invention
In view of the above, the invention provides a method for rapidly detecting complete loosening of a steel tower bolt based on vibration characteristics, which is used for rapidly detecting the defect of complete loosening fault of the steel tower bolt, collecting and analyzing vibration responses of two tower materials connected under pulse excitation before and after the steel tower bolt is completely loosened, processing a vibration signal by using an autocorrelation-VMD method, and finally detecting whether a bolt connection structure is completely loosened by comparing a spectrogram generated by the vibration responses before and after the bolt is completely loosened. Modal aliasing and end-point effects are well avoided by using Variational Modal Decomposition (VMD).
In order to achieve the purpose, the invention adopts the following technical scheme:
a method for rapidly detecting complete looseness of an iron tower bolt based on vibration characteristics comprises the following steps:
step 1: measuring vibration signals of tower materials where vibration excitation points are located and vibration signals of the tower materials connected with the bolts through the bolts in a fastening state after the installation of the bolts of the power transmission iron tower and a current detection state by using a pulse vibration excitation method;
and 2, step: then, carrying out noise reduction, decomposition and reconstruction processing on the vibration signals by adopting an autocorrelation-VMD method to obtain a spectrogram corresponding to each vibration signal;
and step 3: respectively comparing a frequency spectrum diagram corresponding to a vibration signal of a tower material with an excitation point, a tower material connected with a bolt in a fastening state with a frequency spectrum diagram corresponding to a vibration signal in a current detection state, and judging whether the current bolt is completely loosened according to whether the frequency spectrum diagram has a difference exceeding a set threshold value.
Preferably, frequency spectrograms corresponding to vibration signals of the tower material where the vibration excitation point is located and the tower material connected with the bolt in the fastening state and the current detection state are respectively compared to obtain vibration characteristics of the tower materials before and after the bolt is loosened;
when the bolts are fastened, all tower materials generate self-excited vibration under the pulse excitation, the corresponding frequencies of main peaks on frequency spectrograms corresponding to vibration signals of the two tower materials connected by the bolts are different, the difference exceeds a set frequency threshold, and the set frequency threshold can be 10 Hz; when the bolt is completely loosened, the tower material where the excitation point is located is in self-excited vibration, the tower materials connected through the bolt are in forced vibration, the vibration frequency spectrums of the two tower materials connected through the bolt are relatively close, and the vibration frequency spectrum change characteristics of the two tower materials connected through the bolt before and after the bolt is loosened can be used as a judgment basis for judging whether the bolt is completely loosened.
Preferably, in the step 2, an autocorrelation-VMD method is adopted to perform noise reduction and decomposition processing on the vibration signal, and generate a spectrogram, and the specific process is as follows:
step 21: carrying out noise reduction processing on the vibration signal by utilizing an autocorrelation function to obtain a noise reduction signal;
step 22: taking the noise reduction signal as an initial signal, setting the initial decomposition mode number to be K-2, setting the penalty factor alpha to be 2500, and setting the bandwidth to be 0;
step 23: VMD decomposition is carried out on the initial signal to obtain K modal components, and the central frequencies of the K modal components are recorded;
and step 24: if the modal component center frequency of the nth layer of the K modal components is smaller than the modal component center frequency of the (n-1) th layer, or the modal component center frequency of the nth layer is smaller than the modal component center frequency difference of the (n-1) th layer and is smaller than the set difference, the decomposition modal number K is equal to K-1; otherwise, making the decomposition mode number K equal to K +1, and returning to the step 23;
step 25: decomposing the noise reduction signal into K modal components through the steps 22-24;
step 26: and solving a correlation coefficient between each modal component and the noise reduction signal, and reconstructing the modal component of which the correlation number is greater than a set coefficient threshold value to finally obtain a reconstructed signal.
Preferably, the coefficient threshold is set to 0.2; and adding the modal components with the correlation coefficient larger than 0.2 in a time domain to obtain the reconstruction signal, and performing FFT (fast Fourier transform) on the reconstruction signal to obtain a spectrogram of a corresponding vibration signal.
Preferably, the autocorrelation function is expressed as:
wherein x (t) is the acquired initial vibration signal; t denotes the acquisition period.
Preferably, in step S23, the specific process of performing VMD decomposition includes:
s231: performing Hilbert transformation on the noise reduction signal subjected to autocorrelation noise reduction to obtain an analytic signal;
s232: translating the analytic signal to a baseband, estimating the bandwidth of the signal by utilizing the square of the 2 norm of the gradient of the analytic signal, and carrying out variation constraint, wherein the expression is as follows:
the variation constraint process of S231 to S233 described above can be described by the formulas shown in formulas (2) and (3):
s.t.∑Kuk(t)＝f (3)
in the formula uk(t) is the modal component (IMF), ωkIs the center frequency, δ, of each IMF(t)Is a pulse function, and K is the number of modes obtained by decomposition;
represents the time partial derivation;
step 233: in order to solve the variational problem constructed in the above formulas (2) and (3), solve the variational constraint, introduce a Lagrange multiplier lambda and a secondary penalty factor alpha, and introduce an augmented Lagrange function, wherein the augmented Lagrange is as shown in formula (4):
each IMFu can be obtained by searchingkWith its central frequency omegakIs shown in equations (5) to (6):
the iterative manner of the operator λ of the lagrange algorithm is shown in formula (7):
continuously iterating according to the equations (5), (6) and (7), and ending the iteration when the Lagrangian function has minimum value points, so that the initial signal is finally decomposed into K IMFuk。
Preferably, the adopted device comprises a pulse force hammer, a vibration sensor, a signal acquisition instrument and a PC (personal computer) end; the pulse force hammer applies pulse excitation to the tower material excitation point; the vibration sensor is fixed on the power transmission iron tower by using the magnetic suction seat, and collects vibration signals and transmits the vibration signals to the signal collecting instrument; the signal acquisition instrument sampler acquires vibration signals at a set sampling frequency and transmits the vibration signals to the PC terminal; and the PC terminal detects and analyzes the vibration signal.
According to the technical scheme, compared with the prior art, the method for rapidly detecting the complete looseness of the iron tower bolt based on the vibration characteristic is disclosed by the invention, the vibration signal measured on site is subjected to autocorrelation-VMD decomposition noise reduction treatment, the vibration signal of the tower material of the power transmission iron tower is subjected to effective noise reduction, the accuracy of signal analysis is improved, a spectrogram is obtained through decomposition, the change of the vibration spectrogram of a measuring point is used as a judgment basis for the complete looseness of the bolt, when the bolt is fastened, the two tower materials connected through the bolt are equivalent to a unified whole, and the two tower materials are subjected to self-excited vibration under the action of excitation; after the bolt is completely loosened, under the action of excitation, the tower material where the excitation point is located generates self-excited vibration, and the other tower material connected with the bolt generates forced vibration and self-excited vibration, so that the changes of the spectrogram of the two tower materials connected with the bolt before and after the bolt is loosened are different, and the changes are used as judgment bases to realize the rapid detection and the accurate detection of the complete loosening of the bolt of the iron tower.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart of an autocorrelation-VMD decomposition noise reduction method provided by the present invention;
FIG. 2 is a schematic diagram of a bolt to be measured and a vibration measuring point of a power transmission tower provided by the invention;
fig. 3 is a schematic structural diagram of a power transmission tower bolt detection system provided by the invention;
FIG. 4 is a flow chart of an experiment provided by the present invention;
FIG. 5 is a time domain signal diagram after the self-correlation preprocessing of the measuring point 1 during the bolt fastening provided by the invention;
FIG. 6 is a time domain waveform diagram of each modal component after decomposition by the measurement point VMD provided by the present invention;
FIG. 7 is a graph showing a comparison of frequency spectra before and after noise reduction at a point 1 during bolt fastening;
FIG. 8 is a graph of spectrum of point 1 during bolt fastening according to the present invention;
FIG. 9 is a graph of spectrum of the measuring point 1 when the bolt provided by the invention is completely loosened;
FIG. 10 is a graph of spectrum at the 2 measuring points during bolt fastening according to the present invention;
FIG. 11 is a graph of spectrum of 2 measuring points when the bolt provided by the present invention is completely loosened;
FIG. 12 is a graph of spectrum of 3 measuring points during bolt fastening according to the present invention;
FIG. 13 is a graph of spectrum of the 3 measuring points when the bolt provided by the present invention is completely loosened;
FIG. 14 is a graph of spectrum at 4 measuring points during bolt fastening according to the present invention;
FIG. 15 is a graph of spectrum at 4 measuring points when the bolt provided by the invention is completely loosened.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a method for quickly detecting complete looseness of an iron tower bolt based on vibration characteristics, which comprises the following steps:
s1: measuring a vibration signal of a tower material where an excitation point of the power transmission tower bolt in a fastening state after installation and in a current detection state and a vibration signal of the tower material connected with the tower material through the bolt by using a pulse excitation method, and respectively obtaining a vibration signal before loosening and a vibration signal in a current detection state of the tower material where the excitation point is located, and a vibration signal before loosening and a vibration signal in a current detection state of the tower material connected with the bolt; when the tower material at the excitation point is measured, the measuring point should be close to the bolt to be measured as much as possible;
s2: then, noise reduction, decomposition and reconstruction processing are carried out on the vibration signals by adopting an autocorrelation-VMD method to obtain a spectrogram corresponding to each vibration signal;
s21: carrying out noise reduction processing on the vibration signal by utilizing an autocorrelation function to obtain a noise reduction signal;
s22: taking the noise reduction signal as an initial signal, setting the initial decomposition mode number as K-2, setting the penalty factor alpha as 2500 and setting the bandwidth as 0;
s23: performing VMD decomposition on the initial signal, and recording the central frequencies of K modal components;
s231: solving analysis signals for the autocorrelation noise-reduced signals by Hilbert transformation;
s232: translating the analytic signal to a baseband, and estimating the bandwidth of the signal by utilizing the square of the 2 norm of the gradient of the analytic signal to carry out variation constraint;
the variation constraint process can be described by formulas shown in formulas (2) and (3):
s.t.∑Kuk(t)＝f (3)
in the formula uk(t) is the modal component (IMF), ωkIs the center frequency, δ, of each IMF(t)Is a function of the pulseK is the number of modes obtained by decomposition;
s233: in order to solve the diversity problem in the formulas (2) and (3) constructed above, a Lagrange multiplier lambda and a secondary penalty factor alpha are introduced, and an augmented Lagrange function is introduced, wherein the augmented Lagrange is shown as a formula (4):
each IMFu can be obtained by searchingkWith its central frequency omegakIs shown in equations (5) to (6):
the iterative manner of the operator λ of the lagrange algorithm is shown in formula (7):
continuously iterating according to the equations (5), (6) and (7), and ending the iteration when the Lagrangian function has minimum value points, so that the initial signal is finally decomposed into K IMFuk。
S24: if the modal component center frequency of the nth layer in the K IMFs is smaller than the modal component center frequency of the (n-1) th layer, or the modal component center frequency of the nth layer is smaller than the modal component center frequency difference of the (n-1) th layer and is smaller than the set difference, enabling the decomposition modal parameter K to be K-1; otherwise, the decomposition modal parameter K is made to be K +1, and S23 is returned; the set difference value can be 10Hz
S25: finally decomposing the noise reduction signal into K modal components through the steps S22 to S24;
s26: and solving a correlation coefficient between each modal component and the noise reduction signal, reconstructing the modal components with the correlation number larger than 0.2, adding the modal components with the correlation coefficients larger than 0.2 in a time domain to obtain a reconstructed signal, and performing FFT (fast Fourier transform) on the reconstructed signal to obtain a spectrogram of a corresponding vibration signal.
S3: and comparing the frequency spectrogram corresponding to the vibration signal of the tower material with the excitation point and the tower material connected with the bolt, and judging whether the bolt is completely loosened according to the difference that whether the frequency spectrogram exceeds a set threshold value.
When the bolts are fastened, all the tower materials generate self-excited vibration under the impulse excitation, and the frequency spectrogram corresponding to the vibration signals of the two tower materials connected by the bolts has a difference exceeding a set frequency threshold (whether the change value of the frequency corresponding to the main peak value exceeds 10HZ, if the main peak value has obvious difference, the difference is called as large difference; when the bolt is completely loosened, the tower material where the excitation point is located is in self-excited vibration, the tower materials connected through the bolt are in forced vibration, and the vibration frequency spectrums of the two tower materials connected through the bolt are relatively close to each other, so that the change characteristics of the vibration frequency spectrums of the two tower materials connected through the bolt can be used as a basis for judging whether the bolt is completely loosened.
Examples
And (3) denoising by adopting an autocorrelation function, wherein autocorrelation is a process of performing correlation analysis on data corresponding to different moments of the signal. Repeated patterns on time domain signals, such as periodic signals covered by noise, can be well found out by utilizing the autocorrelation function, and fundamental frequency hidden in signal harmonic frequency can be identified. Therefore, the autocorrelation function is widely applied to signal processing such as signal identification, useful signal extraction, noise rejection and the like. The correlation of the autocorrelation function is defined as follows:
and (3) performing signal decomposition by adopting a Variational Modal Decomposition (VMD), wherein the variational modal decomposition is an adaptive signal decomposition algorithm based on classical wiener filtering and Hilbert transformation. The VMD assumes that each modal component (IMF) after its decomposition is a narrow band signal centered around the respective center frequency. VMD decomposition requires a three-step processing:
(1) solving an analysis signal by Hilbert transformation of an original signal;
(2) the signal is shifted to baseband and the bandwidth of the signal is estimated for the square of the 2-norm of the gradient.
The process of the variation constraint described above can be described by the formulas shown in equations (2) to (3):
s.t.∑Kuk(t)＝f (3)
in the formula uk(t) is each IMF,. omegakIs the center frequency, δ, of each IMF(t)Is a pulse function, and K is the number of modes obtained by decomposition;
(3) in order to solve the variation problem constructed above, a lagrangian multiplier λ and a secondary penalty factor α can be introduced, and an augmented lagrangian function is introduced. Wherein, the augmented Lagrange is shown in formula (4):
each IMFu can be obtained by searchingkWith its central frequency omegakIs shown in equations (5) to (6):
the iterative manner of the operator λ of the lagrange algorithm is shown in formula (7):
continuously iterating according to the equations (5), (6) and (7), and ending the iteration when the Lagrange function has a minimum value point, so that the original signal is finally decomposed into K IMFUsk。
In order to obtain signal data generated by vibration when the bolts of the power transmission iron tower are fastened and completely loosened, a field experiment is carried out on the power transmission iron tower of a certain power company, and the power transmission iron tower is a 110kV power transmission iron tower, is completely built and is not put into use. The bolt to be measured and the vibration measuring point selected on the on-site power transmission iron tower are shown in figure 2, and two measuring points which are close to the bolt and far away from the bolt are respectively selected on two sections of tower materials fastened by the bolt.
And analyzing the influence of the position of the fault measuring point when the bolt is completely loosened on the detection result under the condition of ensuring that the excitation position is not changed. The experimental selection of the bolt to be tested is shown in fig. 2, and two measuring points are respectively selected on two tower materials fixed by the bolt for research. The detailed positions of the measuring points and the bolts to be measured are shown in table 1.
TABLE 1 detailed position table of each measuring point
| Measurement point number | Distance from the bolt to be tested |
| Measuring point No. 1 | |
| Number | |
| 2 measuring point | 50mm |
| Measuring point No. 3 | 70mm |
| Measuring point No. 4 | 430mm |
In consideration of the actual detection working condition of the power transmission tower, the excitation source adopts a pulse excitation mode, namely, a pulse force hammer is utilized to apply pulse excitation to the power transmission tower. The sensitivity of the force sensor of the selected pulse force hammer is 4.44pC/N, the gain of the force hammer is 0.500mV/Pv, and the excitation of a local area of an iron tower and the acquisition of signals can be met; the vibration sensor adopts an IEPE piezoelectric acceleration sensor, the axial sensitivity of the vibration sensor is 100.051mv/g respectively, and the vibration sensor is fixed on angle steel of a power transmission tower by using a magnetic attraction seat; and selecting a proper signal acquisition instrument and a signal processing instrument at the PC end. And finally, completing the construction work of the final test system according to the field condition and the selected experimental instrument, as shown in figure 3.
And setting the sampling frequency of the signal acquisition instrument at 5kHz on site, and respectively carrying out two groups of different experiments. Experiment one: adjusting a bolt to be tested to be in a completely fastened state by using a torque wrench, holding an excitation hammer to apply pulse excitation to a power transmission tower material at an excitation point, collecting vibration signals at 4 measurement points by using a vibration acceleration sensor and a dynamic signal collector respectively, and finally performing signal processing on the collected signals by using frequency spectrum analysis software on a PC (personal computer) terminal; experiment two: and adjusting the bolt to be tested to be in a completely loosened state by using the torque wrench, and repeating the operation process of the first experiment. The flow chart of the field experiment is shown in figure 4.
(1) Carrying out noise reduction processing on the transmission tower vibration signal based on an autocorrelation-VMD decomposition noise reduction method;
when the bolt connection looseness fault of the power transmission tower is detected on site, due to the existence of various external interferences, noise reduction processing needs to be carried out on the acquired tower material vibration signals. Fig. 1 is a flow chart of the autocorrelation-VMD decomposition noise reduction processing.
As shown in fig. 1, first, the signal collected by the vibration sensor is subjected to noise reduction processing by using an autocorrelation function. And determining a decomposition mode number K value according to a center frequency principle, and performing VMD decomposition, wherein the method specifically comprises the following steps:
1) and taking the signal subjected to the autocorrelation noise reduction processing as an initial signal, setting the initial modal parameter to be K-2, setting the penalty factor alpha to be 2500, and setting the bandwidth to be a default value of 0.
2) And running a VMD decomposition program by utilizing PC-end signal analysis software, and keeping track of the central frequency of the modal component.
3) And observing the center frequency of the signal decomposition, and determining that the mode number K is K-1 when the center frequency of the nth layer is smaller than the center frequency of the (n-1) th layer or the center frequencies are close to each other. If the above condition is not satisfied, K is K +1, and the operation of (2) is continued.
In order to verify the effectiveness of the autocorrelation-VMD decomposition noise reduction method in reducing the noise of the vibration signals of the tower materials of the power transmission tower. According to the autocorrelation-VMD decomposition noise reduction method flow, VMD decomposition is carried out on the 1 measuring point vibration data subjected to autocorrelation pretreatment, the corresponding central frequencies under different modal component numbers are obtained and are shown in a table 2, the 1 measuring point vibration data subjected to autocorrelation pretreatment is shown in a graph 5, the abscissa represents time, and the ordinate represents acceleration.
TABLE 2 centre frequency of VMD decomposition of 1 measuring point during bolt fastening
As can be seen from table 2, when the VMD decomposition layer number K is 7, the center frequency ω of the IMF4 is equal to41243.92Hz and IMF5 has a center frequency of ω51120.30Hz, the 5 th component center frequency is lower than the 4 th component center frequency, which meets the requirement of VMD decomposition end, so K6 is selected as the VMD decomposition layer number. The time domain waveform of each component obtained by the decomposition is shown in fig. 6.
The correlation coefficient between each IMF of the 1-point measurement points and the vibration signal subjected to autocorrelation preprocessing at the time of bolt fastening is shown in table 3.
TABLE 3 correlation coefficient table between vibration signals
| Modal component | IMF1 | IMF2 | IMF3 | IMF4 | IMF5 | IMF6 |
| Coefficient of correlation | 0.275 | 0.263 | 0.364 | 0.695 | 0.410 | 0.225 |
And solving a correlation coefficient between each IMF and the vibration signal subjected to autocorrelation preprocessing, and adding modal components with the correlation coefficient larger than 0.2 in a time domain for reconstruction as shown in table 3. And finally, performing FFT (fast Fourier transform) on the reconstructed signal and the originally acquired original signal (vibration signal) respectively, and comparing frequency spectrograms of the two signals, as shown in fig. 7, so as to verify the feasibility of the noise reduction method.
As shown in fig. 7, the original signal is compared with the noise-reduced signal after noise reduction in the frequency domain. It can be found that the low-frequency part of the original signal is removed from the signal after noise reduction, the peak value corresponding to the resonance frequency is well preserved, and clutter signals caused by non-resonance noise are suppressed. Therefore, the autocorrelation-VMD noise reduction can effectively remove noise components in the signals while keeping original information of the original signals, can better find out the characteristic frequency of the vibration signals before and after the bolt is completely loosened, and can obtain whether the bolt is completely loosened.
(2) Vibration spectrum analysis of each measuring point before and after complete loosening of bolt of power transmission tower
And sequentially processing the vibration data of the measuring points before and after the bolt is completely loosened according to the autocorrelation-VMD noise reduction method to respectively obtain the frequency spectrogram after noise reduction of the measuring points before and after the bolt is completely loosened.
As shown in the figures 8 and 9, before and after the bolt is completely loosened, the frequency value corresponding to the main peak of the vibration spectrogram of the 1 measuring point is not changed, but after the bolt is completely loosened, the peak value originally at 1660Hz disappears, and small peak values appear at 723.3Hz, 1543Hz and 1883 Hz. The spectrogram changes but slightly before and after complete loosening.
As shown in FIG. 10 and FIG. 11, before and after the bolt is completely loosened, the frequency value corresponding to the main peak of the 2-point vibration spectrogram is changed from 1840Hz to 1257Hz, and after the bolt is completely loosened, the peak value at 1840Hz disappears, and a small peak value at 323.3Hz appears. The spectrogram changes greatly before and after complete loosening.
As shown in FIGS. 12 and 13, before and after the bolt is completely loosened, the frequency value corresponding to the main peak of the 3-point vibration spectrogram is changed from 736.7Hz to 1243Hz, and small peaks appear at 336.7Hz and 2183Hz after the bolt is completely loosened. The spectrogram changes greatly before and after complete loosening.
As shown in fig. 14 and fig. 15, before and after the bolt is completely loosened, the frequency value corresponding to the main peak of the vibration spectrogram with 4 measuring points is changed from 733.3Hz to 1243Hz, and after the bolt is completely loosened, the peak value originally at 1760Hz disappears, and small peak values appear at 336.7Hz, 1633Hz and 2133 Hz. The spectrogram changes greatly before and after complete loosening.
(3) Analysis and discussion
By combining the comparison of the frequency spectrogram, the frequency spectrogram of each measuring point is changed after the bolt of the power transmission tower is completely loosened, so that the change of the vibration frequency spectrogram of each measuring point can be used as a judgment basis for the complete loosening of the bolt. However, by analysis, the following problems were found:
firstly, because the distance between the point 1 and the bolt to be tested is far, the quality of the located angle steel is large, the angle steel is fixed together by a plurality of bolts, and when one bolt is completely loosened, the influence on the rigidity of the bolt to be tested is small, so that the change of the vibration spectrogram of the point 1 is small after the bolt to be tested is completely loosened, and therefore in the subsequent detection process, the selected vibration testing points are selected to be as similar as the testing positions of the points 2, 3 and 4 as possible.
Secondly, when the bolts of the transmission tower are completely loosened, the rigidity k of the tower material of the transmission tower is reduced to a certain degree according to a natural frequency formula
It can be concluded that the natural frequency is also reduced to some extent. However, it can be found through the spectrum comparison that after the bolt is completely loosened, only the frequency corresponding to the spectrum peak of the 2 measuring points shows a descending trend, and the frequencies corresponding to the spectrum peaks of the 3 measuring points and the 4 measuring points have an ascending trend, so that the spectra before and after the bolt is completely loosened are compared with each other to draw the following conclusion:
when the bolt is fastened, the frequency spectrum peak value corresponding frequency among three measuring points of 2, 3 and 4 is compared, and the larger difference is found. Therefore, when the bolts are fastened, the two tower materials connected by the bolts are equivalent to a unified whole, and therefore, each measuring point generates self-excited vibration under the action of the vibration exciting hammer;
after the bolt is completely loosened, the frequency corresponding to the peak values of the frequency spectrum graphs among the three measuring points 2, 3 and 4 are mutually compared, and the main peak values of the three measuring points are all at 1250Hz, and the frequency corresponding to the peak values is similar. Therefore, after the bolt is completely loosened, the vibration of the tower material 1 is still self-excited due to the fact that the vibration exciting point and the measuring point 2 are the same, and therefore the frequency corresponding to the frequency spectrum peak value of the tower material is in a descending trend due to the fact that the rigidity is reduced; and the tower materials 2 with the measuring points 3 and 4 are not unified with the tower material 1, when an impact signal is applied to the tower material 1 by an excitation hammer, the tower material 1 vibrates and collides with the tower material 2, so that the tower material 2 vibrates together, and the vibration generated by the tower material 2 is forced vibration caused by the tower material 1, so that the main peak values of the measuring points 3 and 4 are consistent with the measuring point 2, and in addition, the vibration components before loosening are still remained at the measuring points 3 and 4 after the bolt is completely loosened, so that the vibration of the tower material 2 is influenced by both the forced vibration and the self-excited vibration after the bolt is completely loosened.
(4) Conclusion
The method is characterized in that an excitation hammer is used for artificially exciting the power transmission tower, a piezoelectric acceleration vibration sensor is used for measuring vibration signals before and after the bolt of the power transmission tower is completely loosened, and a spectrogram is compared to obtain a judgment basis for the complete loosening fault of the bolt of the power transmission tower. The main conclusions are as follows:
(1) the method disclosed by the invention is found to effectively realize the noise reduction of the vibration signals of the power transmission tower material by carrying out autocorrelation-VMD decomposition noise reduction treatment on the vibration signals measured on site;
(2) before and after the bolt is completely loosened, vibration frequency spectrograms of the 4 measuring points are compared and analyzed, and changes of the vibration frequency spectrograms of the 4 measuring points can be used as judgment bases for the complete loosening of the bolt, but in comparison, the judgment effect of the 3 measuring points 2, 3 and 4 is better than that of the measuring point 1.
(3) Before and after the bolt is completely loosened, vibration frequency spectrograms of 4 measuring points are mutually compared and analyzed, and the fact that when the bolt is fastened, two tower materials connected by the bolt are equivalent to a unified whole, and the two tower materials vibrate in a self-excitation mode under the action of excitation vibration is obtained; after the bolt is completely loosened, under the action of excitation, the tower material where the excitation point is located generates self-excited vibration, and the other tower material connected with the bolt generates forced vibration and self-excited vibration.
In the present specification, the embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for rapidly detecting complete looseness of an iron tower bolt based on vibration characteristics comprises the following steps:
step 1: measuring vibration signals of tower materials where vibration excitation points are located and vibration signals of the tower materials connected with the bolts through the bolts in a fastening state after the installation of the bolts of the power transmission iron tower and a current detection state by using a pulse vibration excitation method;
and 2, step: denoising, decomposing and reconstructing the vibration signals by adopting an autocorrelation-VMD method to obtain a spectrogram corresponding to each vibration signal;
and 3, step 3: and respectively comparing the spectrogram corresponding to the vibration signal in the fastening state with the spectrogram corresponding to the vibration signal in the current detection state, and judging whether the current bolt is completely loosened according to the difference of whether the spectrogram exceeds a set threshold value.
2. The method for rapidly detecting complete loosening of the iron tower bolt based on the vibration characteristic as claimed in claim 1, wherein in the step 2, an autocorrelation-VMD method is adopted to perform noise reduction and decomposition processing on the vibration signal and generate a spectrogram, and the specific process is as follows:
step 21: carrying out noise reduction processing on the vibration signal by utilizing an autocorrelation function to obtain a noise reduction signal;
step 22: setting an initial decomposition mode number, a penalty factor and a bandwidth by taking the noise reduction signal as an initial signal;
step 23: VMD decomposition is carried out on the initial signal to obtain a modal component, and the central frequency of the modal component is recorded;
step 24: if the modal component center frequency of the nth layer in the modal components is smaller than the modal component center frequency of the (n-1) th layer, or the modal component center frequency of the nth layer is smaller than the modal component center frequency difference of the (n-1) th layer and is smaller than the set difference, enabling the decomposition modal number K to be K-1; otherwise, making the decomposition mode number K equal to K +1, and returning to the step 23;
step 25: finally decomposing the noise reduction signal into K modal components through the steps 22-24;
step 26: and solving a correlation coefficient between each modal component and the noise reduction signal, and reconstructing the modal components of which the correlation number is greater than a set coefficient threshold value to finally obtain a reconstructed signal.
3. The method for rapidly detecting complete loosening of iron tower bolts based on vibration characteristics according to claim 2, wherein in the step 22, the initial decomposition mode number is set to K-2, the penalty factor α is 2500, and the bandwidth is set to 0.
4. The method for rapidly detecting complete loosening of the iron tower bolt based on the vibration characteristic as claimed in claim 2, wherein a coefficient threshold value is set to be 0.2 in the step 26; adding the modal components with the correlation coefficient larger than 0.2 in a time domain to obtain the reconstruction signal, and performing FFT (fast Fourier transform) on the reconstruction signal to obtain a spectrogram of the corresponding vibration signal.
5. The method for rapidly detecting complete loosening of the iron tower bolt based on the vibration characteristics as claimed in claim 2, wherein the autocorrelation function is expressed as:
wherein x (t) is the acquired initial vibration signal; t denotes the acquisition period.
6. The iron tower bolt complete loosening rapid detection method based on vibration characteristics as claimed in claim 2, wherein the specific process of VMD decomposition comprises:
step 231: carrying out Hilbert transformation on the noise reduction signal subjected to the autocorrelation noise reduction to obtain an analytic signal;
step 232: translating the analytic signal to a baseband, estimating the bandwidth of the signal by utilizing the square of the 2 norm of the gradient of the analytic signal, and carrying out variation constraint, wherein the expression is as follows:
s.t.∑Kuk(t)＝f (3)
in the formula uk(t) is the modal component, ωkIs the center frequency, δ, of each modal component(t)Is a pulse function, and K is a mode number obtained by decomposition;
step 233: in order to solve the variation constraint, introducing a Lagrange multiplier lambda and a secondary penalty factor alpha, and introducing an augmented Lagrange function, wherein the augmented Lagrange is shown as a formula (4):
obtaining each modal component u by searchingkWith its central frequency omegakIs shown in equations (5) to (6):
the iterative manner of the operator λ of the lagrange algorithm is shown in formula (7):
continuously iterating according to the formulas (5) to (7), when a minimum value point appears in the Lagrangian function, the iteration is ended, and the initial signal is finally decomposed into K modal components uk。
7. The method for rapidly detecting the complete loosening of the iron tower bolt based on the vibration characteristics as claimed in claim 1, wherein the vibration characteristics of the tower materials before and after the bolt is loosened are obtained by respectively comparing frequency spectrograms corresponding to vibration signals of the tower material where the excitation point is located and the tower material connected with the bolt in the fastening state and the current detection state.
8. The method for rapidly detecting the complete loosening of the iron tower bolt based on the vibration characteristics as claimed in claim 7, wherein when the bolt is fastened, all tower materials generate self-excited vibration under the impulse excitation, and a spectrogram corresponding to a vibration signal of the tower materials connected with the bolt has a difference exceeding a set frequency threshold; when the bolts are completely loosened, the tower material where the excitation points are located is in self-excited vibration, the tower materials connected through the bolts are in forced vibration, and the difference of frequency spectrums corresponding to the vibration signals of the tower materials connected through the bolts is smaller than a set frequency threshold value.
9. The method for rapidly detecting complete loosening of the iron tower bolt based on the vibration characteristics as claimed in claim 8, wherein the frequency threshold value is 10 Hz.
10. The iron tower bolt complete loosening rapid detection method based on the vibration characteristics as claimed in claim 1, wherein the adopted devices comprise a pulse force hammer, a vibration sensor, a signal acquisition instrument and a PC terminal; the pulse force hammer applies pulse excitation to the tower material excitation point; the vibration sensor is fixed on the power transmission iron tower by using the magnetic suction seat, and collects vibration signals and transmits the vibration signals to the signal collecting instrument; the signal acquisition instrument sampler acquires vibration signals at a set sampling frequency and transmits the vibration signals to the PC terminal; and the PC terminal detects and analyzes the vibration signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210237693.0A CN114777985A (en) | 2022-03-11 | 2022-03-11 | Iron tower bolt complete loosening rapid detection method based on vibration characteristics |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210237693.0A CN114777985A (en) | 2022-03-11 | 2022-03-11 | Iron tower bolt complete loosening rapid detection method based on vibration characteristics |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114777985A true CN114777985A (en) | 2022-07-22 |
Family
ID=82423241
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210237693.0A Pending CN114777985A (en) | 2022-03-11 | 2022-03-11 | Iron tower bolt complete loosening rapid detection method based on vibration characteristics |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114777985A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118032937A (en) * | 2024-04-12 | 2024-05-14 | 南京土星信息科技有限公司 | System for detecting looseness voiceprint of bolt of power transmission tower |
-
2022
- 2022-03-11 CN CN202210237693.0A patent/CN114777985A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118032937A (en) * | 2024-04-12 | 2024-05-14 | 南京土星信息科技有限公司 | System for detecting looseness voiceprint of bolt of power transmission tower |
| CN118032937B (en) * | 2024-04-12 | 2024-07-09 | 南京土星信息科技有限公司 | System for detecting looseness voiceprint of bolt of power transmission tower |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hemmati et al. | Roller bearing acoustic signature extraction by wavelet packet transform, applications in fault detection and size estimation | |
| CN108168891B (en) | Method and equipment for extracting weak fault signal characteristics of rolling bearing | |
| Golafshan et al. | SVD and Hankel matrix based de-noising approach for ball bearing fault detection and its assessment using artificial faults | |
| Imaouchen et al. | A frequency-weighted energy operator and complementary ensemble empirical mode decomposition for bearing fault detection | |
| Li et al. | Application of EEMD and improved frequency band entropy in bearing fault feature extraction | |
| Yu et al. | An adaptive sensitive frequency band selection method for empirical wavelet transform and its application in bearing fault diagnosis | |
| Zhang et al. | Acoustic emission detection of rail defect based on wavelet transform and Shannon entropy | |
| Yang et al. | Fault diagnosis of rolling element bearings using basis pursuit | |
| Fan et al. | A hybrid approach for fault diagnosis of planetary bearings using an internal vibration sensor | |
| Yang | Interpretation of mechanical signals using an improved Hilbert–Huang transform | |
| Liu et al. | Application of correlation matching for automatic bearing fault diagnosis | |
| Yan et al. | Harmonic wavelet-based data filtering for enhanced machine defect identification | |
| Yan et al. | Energy-based feature extraction for defect diagnosis in rotary machines | |
| CN113375939B (en) | Mechanical part fault diagnosis method based on SVD and VMD | |
| Zhu et al. | A detection method for bearing faults using null space pursuit and S transform | |
| Jena et al. | Radial ball bearing inner race defect width measurement using analytical wavelet transform of acoustic and vibration signal | |
| Liang et al. | Intelligent bearing fault detection by enhanced energy operator | |
| Chiementin et al. | Early detection of fatigue damage on rolling element bearings using adapted wavelet | |
| Williams et al. | Helicopter transmission fault detection via time-frequency, scale and spectral methods | |
| CN114777985A (en) | Iron tower bolt complete loosening rapid detection method based on vibration characteristics | |
| CN112098093A (en) | Bearing fault feature identification method and system | |
| Jain et al. | A review on vibration signal analysis techniques used for detection of rolling element bearing defects | |
| Rubio et al. | Time-frequency analysis for rotor-rubbing diagnosis | |
| Li et al. | Biphase randomization wavelet bicoherence for mechanical fault diagnosis | |
| Gryllias et al. | A peak energy criterion (pe) for the selection of resonance bands in complex shifted morlet wavelet (csmw) based demodulation of defective rolling element bearings vibration response |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |