CN117473263A - A method and system for automatic identification of frequency and damping ratio for bridge vibration monitoring - Google Patents

Abstract

A frequency and damping ratio automatic identification method and system for bridge vibration monitoring comprises the following steps: s1, carrying out Fourier transformation on actual monitoring data of a bridge to obtain approximate first-order self-vibration frequency of the bridge, and constructing a free damping vibration signal of a damped single-degree-of-freedom spring vibrator by setting an initial damping ratio; s2, positioning and screening out a plurality of signals with the highest degree of correlation with the constructed free damping vibration signals in the actual monitoring data by using a cross-correlation function; s3, dividing the screened first plurality of segments of signals into single-frequency free damping vibration signals and multi-frequency free damping vibration signals according to frequency spectrum analysis, and respectively calculating the single-frequency free damping vibration signals and the multi-frequency free damping vibration signals to obtain frequency and damping ratios. The invention has the advantages of wide applicability, high identification precision, high working efficiency and no need of priori information.

Description

A method and system for automatic identification of frequency and damping ratio for bridge vibration monitoring

Technical field

The invention relates to the field of bridge health monitoring, and in particular to a method and system for automatic identification of frequency and damping ratio for bridge vibration monitoring.

Background technique

Frequency and damping ratio are important modal parameters of the bridge structure, which reflect the overall safety level of the bridge and play an important role in bridge health monitoring. The bridge frequency is usually obtained from the Fourier transform. However, due to the influence of passing vehicles, certain errors often occur in frequency measurement. On the other hand, how to accurately estimate the damping ratio based on monitoring data remains a challenge.

The calculation methods of damping ratio generally include half-power bandwidth method, exponential decay method (EA), stochastic subspace identification method (SSI), eigensystem implementation algorithm (ERA), autoregressive model with external source input, frequency domain decomposition, subspace Numerical algorithm for spatial state space identification, continuous wavelet transform, etc. Among these methods, the EA method is generally considered to be a relatively reliable and accurate method. However, this method requires a free vibration response of the structure and therefore can only be used for vibration tests where controlled excitation can be applied, such as falling weights and jumping trucks.

In fact, during the daily operation of the bridge, the stimulation received by the bridge is usually brought by the traffic flow on the bridge. This is a typical non-stationary random excitation. Bridges often do not vibrate freely under the action of random traffic flow. The excitation transmitted by the traffic flow around the bridge through the coupling effect of the soil structure can be approximately considered as pulse excitation, that is, the time acting on the bridge can be ignored compared to the time of the bridge vibration response. Under the influence of this excitation, the bridge will produce obvious free attenuation vibration. If similar free attenuation vibration signals cannot be found from the monitoring data of the bridge, it will be difficult to accurately extract the damping ratio and frequency of the bridge. However, it is impractical to manually extract such responses from large amounts of data.

Therefore, there are still technical difficulties in identifying the frequency and damping ratio of the bridge from the bridge monitoring data.

Contents of the invention

The main purpose of the present invention is to overcome the above-mentioned defects in the prior art and propose a method and system for automatic identification of frequency and damping ratio for bridge vibration monitoring, which has wide applicability, high identification accuracy, high work efficiency and does not require prior information. advantage.

The present invention adopts the following technical solutions:

An automatic identification method of frequency and damping ratio for bridge vibration monitoring, which is characterized by including:

S1, perform Fourier transform on the actual monitoring data of the bridge to obtain the approximate first-order natural frequency of the bridge, formulate the initial damping ratio and construct the free attenuation vibration signal of the damped single-degree-of-freedom spring oscillator;

S2, use the cross-correlation function to locate and filter out the first several segments of signals in the actual monitoring data that have the greatest degree of correlation with the constructed free attenuation vibration signal;

S3. According to spectrum analysis, the first several segments of the screened signals are divided into single-frequency free attenuation vibration signals and multi-frequency free attenuation vibration signals. For the single-frequency free attenuation vibration signal and the multi-frequency free attenuation vibration signal, respectively Calculate the frequency and damping ratio;

S4, draw the frequency and damping of the first several segments of signals into a frequency-damping scatter plot, and average the calculated frequencies and damping ratios to obtain the final identification result.

Step S1 specifically includes:

Preprocess actual monitoring data;

_The acceleration signal of the preprocessed actual monitoring data is defined as X(t). The acceleration signal X(t) is a time series with _a sampling frequency of f _s and _a sampling time of T f _s T _X ;

Perform Fourier transform on X(t):

Among them, e is the natural logarithm, i is the imaginary unit, and ω is the circular frequency;

Determine the position of the highest peak in the spectrum through peak picking, and obtain the approximate first-order natural frequency of the bridge.

Proposed initial damping ratio And construct the signal Y(t) of the free attenuated vibration of the damped single-degree-of-freedom spring oscillator:

Signal Y(t) is a time series with the same sampling frequency f _s and sampling time _TY . Its signal length is _LY , and _LY =f _{s TY} .

The preprocessing includes:

First, perform detrend processing on the actual monitoring data to remove the offset in the original signal;

Secondly, low-pass filtering is performed on the actual monitoring data to reduce the interference of measurement noise.

Step S2 specifically includes:

Calculate the cross-correlation function of the acceleration signal X(t) and the free attenuated vibration signal Y(t):

From this, the correlation vector between signals X(t) and Y(t) is obtained And at the same time get the lagged index vector of the correlation/>

Take the top k values with the largest absolute values in the correlation vector R, find their corresponding position index in the lag index vector L, and filter out the correlation between the actual monitoring data and the constructed free attenuation vibration signal. The first k-segment signals with the largest degree, X ₁ (t), X ₂ (t), ..., X _k (t).

Step S3 specifically includes:

Perform Fourier transform on the first several segments of signals X ₁ (t), X ₂ (t),..., X _k (t):

Among them, e is the natural logarithm, i is the imaginary unit, ω is the circular frequency, and the value range of k is 10-30;

Peak peaks in each spectrum are obtained through peak picking, where the peak value greater than the initial set threshold is recorded as a valid peak value; a signal with a valid peak number of 1 is the single-frequency free attenuation vibration signal, and a signal with a valid peak number greater than 1 is The multi-frequency freely attenuates vibration signals.

In step S3, for the single-frequency free attenuation vibration signal, the frequency and damping ratio are directly calculated using the exponential decay method, and the wave peak value A _m in the signal and its corresponding time t _m and sequence number m are obtained through peak picking, and the wave peak value is obtained The logarithm of and its corresponding serial number are linearly fitted by the least squares method to obtain:

lnA＝a·n+b

Among them, lnA is the logarithmic value of the wave crest, n is the wave number,

Where, M is the total wave number of the single-frequency free attenuated vibration signal;

Calculate damping ratio:

Calculate frequency:

In step S3, for the multi-frequency free attenuating vibration signal, the signal is decomposed into multiple single-frequency free attenuating vibration signals through a variational nonlinear component decomposition method, and exponential attenuation is used for each of the single-frequency free attenuating vibration signals. method to calculate their respective frequencies and damping ratios.

An automatic identification system of frequency and damping ratio for bridge vibration monitoring, which is characterized by including:

The acceleration sensor is installed under the main beam of the bridge and is used to convert the vibration acceleration of the bridge into a voltage signal to measure the vibration acceleration of the bridge;

The signal acquisition module receives and collects the voltage signal from the accelerometer, converts it from an analog signal to a digital signal, and then records the signal;

The correlation analysis module receives the time course signal recorded by the signal acquisition module, and uses the cross-correlation function to filter out the free attenuation vibration signal in the signal;

The time-frequency analysis module receives the free attenuation vibration signal output from the correlation analysis module, uses Fourier transform to divide the free attenuation vibration signal into a single-frequency free attenuation vibration signal and a multi-frequency free attenuation vibration signal, and then uses the variational nonlinear component Decompose the multi-frequency free attenuation vibration signal into multiple single-frequency free attenuation vibration signals;

The linear fitting module receives the single-frequency free attenuation vibration signal output from the time-frequency analysis module, and uses linear fitting to calculate the frequency and damping of the bridge.

From the above description of the present invention, it can be seen that compared with the prior art, the present invention has the following beneficial effects:

1. The present invention can realize automatic identification of the entire process from free attenuation signal extraction to damping ratio identification in measured signals. It can realize automatic identification without manually screening signal segments to establish a database, and can be well applied to real-time and continuous signal processing. , which has broad application prospects in bridge health detection.

2. Compared with the traditional damping identification method, the present invention constructs a damped single free attenuation signal based on the natural vibration frequency of the structure obtained by Fourier transformation of the actual signal. It is insensitive to the vibration form of the free attenuation signal in the actual measured signal and has powerful The robustness makes it more suitable for working in operational environments.

Description of the drawings

Figure 1 is a flow chart of the method of the present invention;

Figure 2 shows the acceleration time history response curve of a certain distributed measuring point on a certain bridge;

Figure 3 shows the free attenuated vibration signal of the constructed damped single-degree-of-freedom spring oscillator;

Figure 4 shows the correlation coefficient between the actual monitoring data of the 03B0105 distributed measuring points and the constructed attenuation signal;

Figure 5 is a schematic diagram of the actual monitoring data of the 03B0105 distributed measuring points and the identified attenuation section;

Figure 6 shows the damping ratio calculation process;

Figure 7 is a schematic diagram of the frequency-damping ratio scatter distribution and its average value of the 03B0105 distribution measuring points;

Figure 8 shows the correlation coefficient between the measured signal at the 03B0114 distributed measuring point and the constructed attenuated signal;

Figure 9 is a schematic diagram of the measured signal and the identified attenuation section of 03B0114 distributed measuring points;

Figure 10 is a schematic diagram of signal decomposition;

Figure 11 shows the damping ratio calculation process;

Figure 12 is a schematic diagram of the first-order frequency-damping ratio, second-order frequency-damping ratio scatter distribution and their average value.

The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

Detailed ways

The present invention will be further described below through specific embodiments.

The present invention will be described in more depth and detail below with reference to the accompanying diagrams and specific implementation examples. These drawings and implementation examples will provide readers with a more intuitive understanding, so that the structure, working principle and application scenarios of the present invention can be more clearly presented and understood.

Referring to Figure 1, an automatic identification method of frequency and damping ratio for bridge vibration monitoring includes:

S1, perform Fourier transform on the actual monitoring data of the bridge to obtain the approximate first-order natural frequency of the bridge, formulate the initial damping ratio and construct the free attenuation vibration signal of the damped single-degree-of-freedom spring oscillator.

This step specifically includes:

First, preprocess the actual monitoring data. Preprocessing includes: first, detrending the actual monitoring data to remove the offset in the original signal; second, performing low-pass filtering on the actual monitoring data to reduce the interference of measurement noise.

_The acceleration signal of the preprocessed actual monitoring data is defined as X(t). The signal X(t) is a time series with a sampling frequency of f _s and a sampling time of _TX . Its signal length is _{L s} T _X ;

Perform Fourier transform on X(t):

Among them, e is the natural logarithm, i is the imaginary unit, and ω is the circular frequency.

Determine the position of the highest peak in the spectrum through peak picking, and obtain the approximate first-order natural frequency of the bridge.

Proposed initial damping ratio And construct the signal Y(t) of the free attenuated vibration of the damped single-degree-of-freedom spring oscillator:

Signal Y(t) is a time series with the same sampling frequency f _s and sampling time T _Y (the recommended duration is 10 to 20 times the approximate first-order period of the bridge). Its signal length is _LY , _LY = f _{s TY} .

S2, use the cross-correlation function to locate and screen out the first few signals in the actual monitoring data that have the greatest correlation with the constructed free attenuation vibration signal. Specifically include:

Calculate the cross-correlation function between the acceleration signal X(t) and the free attenuated vibration signal Y(t):

From this, the correlation vector between signals X(t) and Y(t) is obtained And at the same time get the lagged index vector of the correlation/>

Take the top k values with the largest absolute value in the correlation vector R, find their corresponding position index in the lag index vector L, and filter out the top k values with the largest degree of correlation between the actual monitoring data and the constructed free attenuation vibration signal. K-segment signal, X ₁ (t), X ₂ (t), ..., X _k (t).

S3, according to the spectrum analysis, the first few selected signals are divided into single-frequency free attenuation vibration signals and multi-frequency free attenuation vibration signals. For the single-frequency free attenuation vibration signals and multi-frequency free attenuation vibration signals, the frequencies and damping ratios are calculated respectively. ;

Specifically include:

Perform Fourier transform on the first several segments of signals X ₁ (t), X ₂ (t),..., X _k (t):

Among them, e is the natural logarithm, i is the imaginary unit, and ω is the circular frequency. The value of k ranges from 10 to 30.

Peak peaks in each spectrum are obtained through peak picking. The peak value greater than the initial set threshold is recorded as a valid peak value; a signal with a valid peak number of 1 is a single-frequency free attenuation vibration signal, and a signal with a valid peak number greater than 1 is a multi-frequency signal. Freely attenuates vibration signals.

For single-frequency free attenuation vibration signals, the frequency and damping ratio are directly calculated using the exponential decay method. The peak value A _m in the signal and its corresponding time t _m and sequence number m are obtained through peak picking. The logarithm of the wave peak value and its corresponding The serial number is linearly fitted by the least squares method to obtain:

lnA＝a·n+b

Among them, lnA is the logarithmic value of the wave crest, n is the wave number,

Among them, M is the total wave number of single-frequency free attenuation vibration signal;

Calculate damping ratio:

Calculate frequency:

For multi-frequency free attenuated vibration signals, the signal is decomposed into multiple single-frequency free attenuated vibration signals through the variational nonlinear component decomposition method, and then the exponential decay method is used to calculate the respective frequency and damping of each single-frequency free attenuated vibration signal. Compare.

S4, draw the frequency and damping of the first several segments of the signal into a frequency-damping scatter plot, and average the calculated frequency and damping ratio to obtain the final identification result.

Based on this, the present invention also proposes an automatic identification system of frequency and damping ratio for bridge vibration monitoring, including:

The acceleration sensor is installed under the main beam of the bridge and is used to convert the vibration acceleration of the bridge into a voltage signal to measure the vibration acceleration of the bridge;

The signal acquisition module receives and collects the voltage signal from the accelerometer, converts it from an analog signal to a digital signal, and then records the signal;

The correlation analysis module receives the time course signal recorded by the signal acquisition module, and uses the cross-correlation function to filter out the free attenuation vibration signal in the signal;

The time-frequency analysis module receives the free attenuation vibration signal output from the correlation analysis module, uses Fourier transform to divide the free attenuation vibration signal into a single-frequency free attenuation vibration signal and a multi-frequency free attenuation vibration signal, and then uses the variational nonlinear component Decompose the multi-frequency free attenuation vibration signal into multiple single-frequency free attenuation vibration signals;

The linear fitting module receives the single-frequency free attenuation vibration signal output from the time-frequency analysis module, and uses linear fitting to calculate the frequency and damping of the bridge.

The system of the present invention adopts the above-mentioned automatic identification method of frequency and damping ratio for bridge vibration monitoring, draws the frequency and damping of the first several segments of signals into frequency-damping scatter diagrams, and compares the calculated frequencies and damping ratios respectively. The average is taken to obtain the final recognition result.

The invention can automatically extract the low-order frequency and damping ratio of the bridge from the structural vibration monitoring signal. It has the advantages of wide applicability, high recognition accuracy, high work efficiency and no need for prior information, and has broad applications in bridge health detection. prospect.

Application examples

Take a certain Z24 bridge as an example. It is a classic post-tensioned prestressed concrete double-unit box girder bridge with a main span of 30 meters and a span of 14 meters on both sides. The bridge was monitored for nearly a year, and artificially controllable damage was gradually applied during the later stages of the monitoring period. Taking the Z24 bridge as the engineering background of civil engineering systems, the researchers determined the research topic SIMCES, established the benchmark problem of the Z24 bridge, and studied a series of issues in the field of health monitoring, including natural frequencies, vibration shapes, and damping ratios under bridge damage. As well as the transformation of structural modal parameters caused by environmental factors, the Z24 bridge has become the foundation bridge for many scholars to study bridge health monitoring. Therefore, choosing this bridge as the research object of the experiment has strong practical significance.

The monitoring of a Z24 bridge is divided into 9 distributions, each distribution has 33 channels, of which 28 channels are located on the bridge deck, 5 fixed channels are reference points, 3 reference point channels are located on the bridge deck, and 2 test channels are located on the bridge deck. pier. A Z24 bridge sensor collected 65536 samples at a sampling rate of 100Hz. After converting the bin format data provided by the SVS website into mat format in MATLAB, the acceleration time history response curve of a distributed measuring point of the Z24 bridge can be obtained, as shown in the figure 2 shown. Under environmental excitation, there are many typical free attenuation signals in vibration signals. It is worth mentioning that the measured data used for identification does not require multiple measurement points. The monitoring data collected by a single acceleration sensor can be used for identification by this method. Therefore, the automatic identification technology using this method can avoid the disadvantages of installing multiple sensors. .

First, Fourier transform is performed on the actual monitoring data to obtain the approximate first-order natural frequency of the bridge, the initial damping ratio is formulated, and the free attenuated vibration signal of the damped single-degree-of-freedom spring oscillator is constructed. Figure 3 shows the constructed damped free decay time history curve.

Next, the cross-correlation function is used to determine the correlation coefficient between the measured signal and the constructed free attenuation vibration signal, as shown in Figure 4, and the correlation between the two signals is used to locate the largest first few segments at the screening point. signal, and the result is shown in Figure 5.

Finally, further screening is performed, and the first few selected signals are divided into single-frequency free attenuation vibration signals and multi-frequency free attenuation vibration signals based on spectrum analysis. The peak value in each spectrum is obtained through peak picking, where the peak value greater than the initial set threshold is recorded as a valid peak value. A signal with an effective peak number of 1 is a single-frequency free attenuation vibration signal, and a signal with an effective peak number greater than 1 is a multi-frequency free attenuation vibration signal.

For single-frequency signals, the damping ratio and frequency are obtained directly using Fourier transform and exponential decay method, and the logarithm of the wave peak value and its corresponding serial number are linearly fitted by the least squares method, as shown in Figure 6, where (a) is a certain extracted attenuation section, (b) is the linear fitting of peak logarithm value and wave number. Further, the damping ratio and frequency are calculated, and the results are shown in Figure 7. The distribution of measuring point frequency-damping ratio scatter point distribution and its average value (★). Since in this example, the original data only contains the first-order free attenuation vibration signal, only the first-order frequency can be obtained after Fourier transform of the attenuation segment extracted from the measured data. The automatic identification results proposed by the present invention are compared with the natural frequencies and damping ratios of the reference-based combined deterministic-random subspace and DBSCAN-based identification, as shown in Table 1.

Table 1

For further testing, this method can also be used to identify the damping ratio of multi-frequency free attenuated vibration signals. Select the acceleration data of channel 5 of the Z24 bridge at a certain moment (03B0114), or use the damped single degree of freedom constructed as shown in Figure 3. The attenuated vibration signal is used to locate and screen the free attenuated signal in the measured signal.

Next, the cross-correlation function is used to determine the correlation coefficient between the measured signal and the constructed free attenuation vibration signal, as shown in Figure 8, and the correlation between the two signals is used to locate the largest first few segments at the screening point. signal, and the result is shown in Figure 9.

For multi-frequency signals, first use the variational nonlinear component decomposition method to decompose them into multiple single-frequency signals, as shown in Figure 10, (a) is the original signal; (b) is the decomposed first-order single-frequency signal; (c) is the decomposed second-order single-frequency signal. Figure 11 shows the process of calculating the damping ratio using the exponential decay method for a decomposed single-frequency signal. (a) is a decomposed single-frequency signal; (b) is the linear fitting of the peak logarithm value and the wave number. Then use Fourier transform to obtain the frequency of each single-frequency signal, and take the arithmetic average. The frequency-damping scatter plot of the free attenuated vibration signal is as shown in Figure 12: first-order frequency-damping ratio and second-order frequency-damping ratio. Schematic diagram of scatter distribution and its mean (★). The automatic identification results proposed by the present invention are compared with the natural frequencies and damping ratios of the reference-based combined deterministic-random subspace and DBSCAN-based identification, as shown in Table 2. The proposed automatic identification method of damping ratio based on structural vibration monitoring data realizes automatic identification of frequency and damping ratio, and can be well applied to real-time and continuous signal processing.

Table 2

It is further explained that the vibration conditions of the free attenuation signal filtered out from the measured signals are not very similar to the vibration conditions of the constructed damped single-degree-of-freedom free attenuation vibration signal, but they can all be identified and extracted in the end. As long as the attenuation response curve existing in the measured data approximately satisfies the shape of the constructed damped free attenuation curve, the attenuation segment in the data can be identified. In this Z24 bridge actual measurement data test, it was proved that the automatic mode recognition method proposed by the present invention has strong robustness and is expected to be applied to real-time health monitoring.

It should be emphasized that the foregoing examples are only for explaining the technical content of the present invention and do not represent its sole definition. Although we have introduced the present invention in detail through preferred examples, technical experts in relevant fields should know that they can make appropriate adjustments or equivalent replacements to these technical contents. These changes should still follow the core concept of the present invention and Without exceeding the scope of its definition, these should be included within the scope of the invention.

Claims (8)

1. An automatic identification method of frequency and damping ratio for bridge vibration monitoring, which is characterized by including: S1, perform Fourier transform on the actual monitoring data of the bridge to obtain the approximate first-order natural frequency of the bridge, formulate the initial damping ratio and construct the free attenuation vibration signal of the damped single-degree-of-freedom spring oscillator; S2, use the cross-correlation function to locate and filter out the first several segments of signals in the actual monitoring data that have the greatest degree of correlation with the constructed free attenuation vibration signal; S3. According to spectrum analysis, the first several segments of the screened signals are divided into single-frequency free attenuation vibration signals and multi-frequency free attenuation vibration signals. For the single-frequency free attenuation vibration signal and the multi-frequency free attenuation vibration signal, respectively Calculate the frequency and damping ratio; S4, draw the frequency and damping of the first several segments of signals into a frequency-damping scatter plot, and average the calculated frequencies and damping ratios to obtain the final identification result. 2. An automatic identification method of frequency and damping ratio for bridge vibration monitoring as claimed in claim 1, characterized in that step S1 specifically includes: Preprocess actual monitoring data; _The acceleration signal of the preprocessed actual monitoring data is defined as X(t). The acceleration signal X(t) is a time series with _a sampling frequency of f _s and _a sampling time of T f _s T _X ; Perform Fourier transform on X(t): Among them, e is the natural logarithm, i is the imaginary unit, and ω is the circular frequency; Determine the position of the highest peak in the spectrum through peak picking, and obtain the approximate first-order natural frequency of the bridge. Proposed initial damping ratio And construct the signal Y(t) of the free attenuated vibration of the damped single-degree-of-freedom spring oscillator: Signal Y(t) is a time series with the same sampling frequency f _s and sampling time _TY . Its signal length is _LY , and _LY =f _{s TY} . 3. An automatic identification method of frequency and damping ratio for bridge vibration monitoring according to claim 2, characterized in that the preprocessing includes: First, perform detrend processing on the actual monitoring data to remove the offset in the original signal; Secondly, low-pass filtering is performed on the actual monitoring data to reduce the interference of measurement noise. 4. An automatic identification method of frequency and damping ratio for bridge vibration monitoring as claimed in claim 2, characterized in that step S2 specifically includes: Calculate the cross-correlation function of the acceleration signal X(t) and the free attenuated vibration signal Y(t): From this, the correlation vector between signals X(t) and Y(t) is obtained And at the same time get the lagged index vector of the correlation/> Take the top k values with the largest absolute values in the correlation vector R, find their corresponding position index in the lag index vector L, and filter out the correlation between the actual monitoring data and the constructed free attenuation vibration signal. The first k-segment signals with the largest degree, X ₁ (t), X ₂ (t), ..., X _k (t). 5. An automatic identification method of frequency and damping ratio for bridge vibration monitoring as claimed in claim 1, characterized in that step S3 specifically includes: Perform Fourier transform on the first several segments of signals X ₁ (t), X ₂ (t),..., X _k (t): where, e is the natural logarithm, i is the imaginary unit, and ω is the circular frequency. The value range of k is 10-30; Peak peaks in each spectrum are obtained through peak picking, where the peak value greater than the initial set threshold is recorded as a valid peak value; a signal with a valid peak number of 1 is the single-frequency free attenuation vibration signal, and a signal with a valid peak number greater than 1 is The multi-frequency freely attenuates vibration signals. 6. The automatic identification method of frequency and damping ratio based on bridge vibration monitoring data according to claim 1, characterized in that: in step S3, the frequency and damping ratio are directly calculated using the exponential decay method for the single-frequency free attenuation vibration signal. , the wave peak value A _m in the signal and its corresponding time t _m and sequence number m are obtained through peak picking. The logarithm of the wave peak value and its corresponding sequence number are linearly fitted by the least squares method to obtain: ln A＝a·n+b Among them, ln A is the logarithmic value of the wave crest, n is the wave number, Where, M is the total wave number of the single-frequency free attenuated vibration signal; Calculate damping ratio: Calculate frequency: 7. The automatic identification method of frequency and damping ratio based on bridge vibration monitoring data according to claim 1, characterized in that: in step S3, for the multi-frequency free attenuation vibration signal, the variational nonlinear component decomposition method is used to decompose the frequency and damping ratio. The signal is decomposed into multiple single-frequency free-attenuated vibration signals, and the exponential decay method is used to calculate the respective frequency and damping ratio of each single-frequency free-attenuated vibration signal. 8. An automatic identification system of frequency and damping ratio for bridge vibration monitoring, which is characterized by including: The acceleration sensor is installed under the main beam of the bridge and is used to convert the vibration acceleration of the bridge into a voltage signal to measure the vibration acceleration of the bridge; The signal acquisition module receives and collects the voltage signal from the accelerometer, converts it from an analog signal to a digital signal, and then records the signal; The correlation analysis module receives the time course signal recorded by the signal acquisition module, and uses the cross-correlation function to filter out the free attenuation vibration signal in the signal; The time-frequency analysis module receives the free attenuation vibration signal output from the correlation analysis module, uses Fourier transform to divide the free attenuation vibration signal into a single-frequency free attenuation vibration signal and a multi-frequency free attenuation vibration signal, and then uses the variational nonlinear component Decompose the multi-frequency free attenuation vibration signal into multiple single-frequency free attenuation vibration signals; The linear fitting module receives the single-frequency free attenuation vibration signal output from the time-frequency analysis module, and uses linear fitting to calculate the frequency and damping of the bridge.