Bridge damage detection method and system

Abstract

The invention discloses a bridge damage detection method and system, and the method comprises the steps: collecting a bridge vibration signal through an acceleration sensor, carrying out the band-pass filtering preprocessing, converting a time domain signal into a frequency domain vibration signal through fast Fourier transform, and extracting a feature frequency through gradient change detection and threshold screening; a pairing difference value sequence of a monitoring characteristic frequency and a healthy baseline frequency is constructed, and a pairing sample T test is applied to perform statistical significance analysis on a difference value, so that the abnormal frequency change caused by structural rigidity degradation is effectively identified. According to the method, complex modal analysis is replaced by simplified frequency domain features, environmental interference influences are eliminated in combination with a pairing inspection mechanism, the micro-damage detection sensitivity and the false alarm resistance are remarkably improved, and the method is suitable for engineering scenes of long-term bridge monitoring and rapid damage screening.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of bridge damage detection, and relates to a bridge damage detection method and system. BACKGROUND

[0002] With the continuous growth of the service life of bridge structures and the increasing complexity of the environment, the risk of structural damage is significantly increased. Traditional damage detection methods mainly rely on periodic manual inspection and maintenance, which has problems such as large workload, low efficiency, limited coverage, and is difficult to achieve continuous real-time monitoring of the structure state. Under this background, the structural health monitoring (SHM) technology rapidly develops in the field of civil engineering due to its automation and real-time advantages. This technology continuously collects acceleration signals of the dynamic response of the structure by deploying a distributed vibration sensor network at key positions of the bridge, forming time series data, which provides a basis for damage identification. In order to extract effective structural features, the original time-domain vibration signal needs to be converted to the frequency domain by Fast Fourier Transform (FFT), and then the main frequency components in the frequency spectrum are analyzed - these frequency characteristics are highly related to the modal parameters (such as natural frequency) of the structure, and have become an important basis for damage identification. Studies have shown that stiffness degradation damage such as bridge pier cracking and foundation settlement often leads to a decrease in the natural frequency of the structure, so monitoring of frequency domain features can achieve early warning of early damage.

[0003] The current mainstream damage identification method is mainly based on modal parameter analysis and statistical learning technology. For example, combining Principal Component Analysis (PCA) and artificial neural network can improve the damage positioning accuracy; using nonlinear Support Vector Machine (SVM) to classify weak frequency changes can enhance the sensitivity to small damage. However, these methods still face significant challenges: on the one hand, vibration signals are easily disturbed by environmental factors (such as wind speed, temperature fluctuations), leading to non-damage drift of modal parameters, affecting the reliability of identification; on the other hand, the complete modal parameter identification process is computationally complex and time-consuming, limiting the operability and real-time performance in actual engineering. Therefore, developing a damage identification method that is efficient, robust, and can effectively suppress environmental interference has become a key problem that needs to be broken through in the field of bridge structural health monitoring. SUMMARY

[0004] The purpose of the present application is to solve the technical problems in the prior art that vibration signals are easily affected by environmental factors, causing instability of modal parameters, and the complete modal identification process is computationally complex, limiting the operability in actual engineering, and to provide a bridge damage detection method and system.

[0005] To achieve the above object, the present application adopts the following technical solutions to achieve: The first aspect of the present application provides a bridge damage detection method, comprising the following steps: Collecting bridge vibration signals, preprocessing the collected vibration signals; The preprocessed signal is subjected to time-frequency conversion to obtain a frequency domain vibration signal; the characteristic frequency is extracted by peak value detection; Calculate the paired difference sequence of the extracted characteristic frequency and the baseline frequency; Paired T test is performed on the difference sequence to determine whether the bridge is damaged.

[0006] Further, the preprocessing of the collected vibration signals adopts a filtering processing method.

[0007] Further, the characteristic frequency is extracted by peak value detection, comprising: The local peak value of the frequency domain vibration signal is detected by gradient change; Find the peak value point where the gradient changes from positive to negative before and after the local peak value point to obtain the maximum peak value point; Based on the maximum peak value point, the effective peak value is screened out; The natural frequency of the bridge structure corresponding to the effective peak value is taken as the characteristic frequency.

[0008] Further, the paired difference sequence of the extracted characteristic frequency and the baseline frequency is calculated, specifically: Select vibration data at the same position and in the same environment as baseline data, and obtain baseline frequency based on the baseline data; Pair the characteristic frequency and the baseline frequency to obtain the paired difference sequence.

[0009] Further, the paired T test is performed on the difference sequence to determine whether the bridge is damaged, specifically: Based on the paired difference sequence, calculate the T test statistic and the effect size; the paired difference sequence satisfies the normal distribution; Based on the T test statistic, calculate the frequency change probability p; According to the effect size and the frequency change probability p, determine whether the bridge is damaged.

[0010] Further, the T test statistic calculation formula is as follows:

[0011] Wherein, is the mean of the paired difference sequence, is the standard deviation of the paired difference sequence, is the number of characteristic frequencies.

[0012] Further, the expression of p is:

[0013] wherein, is the value of the calculated T test statistic, is the standardized sample mean difference, denotes the gamma function.

[0014] Further, the calculation formula of the effect size is:

[0015] wherein, is the mean of the paired difference sequence, is the standard deviation of the paired difference sequence.

[0016] The second aspect of the present application provides a computer readable storage medium, the computer readable storage medium stores a computer program, the computer program is executed by a processor to realize the bridge damage detection method.

[0017] The third aspect of the present application provides a bridge damage detection system, comprising: The signal processing module pre-processes the collected bridge vibration signal; The characteristic frequency extraction module performs time-frequency conversion on the pre-processed signal to obtain a frequency domain vibration signal; and extracts the characteristic frequency through peak value detection; The paired detection module calculates the paired difference sequence of the extracted characteristic frequency and the baseline frequency; The T detection module performs paired T test on the difference sequence to determine whether the bridge is damaged.

[0018] Compared with the prior art, the present application has the following beneficial effects: The application discloses a bridge damage detection method, which extracts key frequency characteristics (first, third and fifth order natural frequencies) to replace complete modal analysis, avoids high-dimensional modeling and complex calculation, and significantly reduces the computing power requirement of a real-time monitoring system. Compared with traditional methods that rely on machine learning training or modal shape libraries, the method does not require a large number of sample accumulation and can be directly applied to online damage detection. By constructing a frequency difference sequence (i.e., a paired difference sequence) of the monitoring data and the health baseline, the paired sample T test mechanism effectively eliminates non-damage frequency drift caused by environmental noise (such as temperature and wind speed fluctuations). This design significantly reduces the false positive rate (for example, the "structure does not change but the frequency fluctuates" phenomenon), is particularly suitable for strong noise scenes, and can accurately capture the frequency changes caused by minor damage. Based on the standardized statistical test process (T test + Cohen's d effect size), the path is clear, the calculation is lightweight, and can be seamlessly embedded in edge computing devices or Internet of Things platforms. It supports the rapid deployment of distributed sensor networks and meets the long-term, low-cost monitoring needs of bridges. In actual engineering, the method is particularly suitable for long-term health monitoring, periodic state evaluation and rapid screening of sudden damage events of bridge structures. For example, a mobile sensor array can be deployed every quarter or every half year to collect vibration data, and paired analysis can be performed with baseline state data to statistically distinguish periodic structural changes. This way not only can continuously track the health status of the bridge, but also can timely find early damage to the structure, making up for the shortcomings of traditional manual inspection. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0020] Figure 1 The bridge damage detection method flowchart of the application; Figure 2 The first, third and fifth order characteristic frequency changes of Z-24 bridge at different stages; Figure 3 The first, third and fifth order frequency difference histogram and its normal distribution fitting curve. DETAILED DESCRIPTION

[0021] So that the objects, technical solutions and advantages of the embodiments of the present application are more apparent, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0022] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative labor are within the scope of protection of the present application.

[0023] It should be noted that: similar reference numerals and letters represent similar items in the following drawings, therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0024] The present application will be described in further detail below with reference to the drawings: Referring to Figure 1 The bridge damage detection method of the present application proposes a bridge damage identification method based on frequency-domain paired T-test (referred to as FDPTT). The method includes signal preprocessing, frequency domain transformation and feature extraction, sample pairing and statistical test. The overall concept of the present application is to take frequency as the core feature parameter, construct paired samples before and after damage, and use paired T-test to evaluate whether the difference is significant, so as to judge whether the structure is damaged. Unlike the traditional T-test method, the paired T-test of the present application does not require the variances of the two groups of data to be equal, but analyzes the difference value sequence, which is suitable for long-term monitoring or repeated measurement of the structure. In addition, the effect size (Cohen's d) is reported at the same time on the basis of paired T-test, which helps to supplement the statistical significance information reflected by the p value, further quantifies the practical significance of the frequency difference, and enhances the discriminant interpretation of the structural state change. Specifically, the bridge structure damage identification method combining frequency domain feature extraction and paired sample T-test proposed by the present application has a whole process mainly including four stages. The first stage is original signal processing, the second stage is frequency domain feature extraction, the third stage is data pairing and modeling, and the fourth stage is statistical test and damage determination.

[0025] 1. First stage (original signal processing) The acceleration sensor is arranged at the key position of the bridge to collect the vibration signal. The sampling frequency is usually set to 100 Hz or above to ensure sufficient time resolution to obtain the structural response characteristics. After the original signal is collected, band-pass filtering is required to remove low-frequency drift and high-frequency noise and retain the effective frequency band related to the structural modal response. In practical applications, the frequency response of the Butterworth band-pass filter is smooth at the cutoff frequency, so it is very suitable for the relatively stable frequency characteristics in the structural vibration signal. Its frequency response function can be expressed as: (1) wherein, represents the frequency of the input signal, and are the lower and upper limit frequencies of the band-pass filter, is the center frequency of the filter, is the order of the filter. After the original signal is collected, band-pass filtering is required to remove low-frequency drift and high-frequency noise and retain the effective frequency band related to the structural modal response. In practical applications, the frequency response of the Butterworth band-pass filter is smooth at the cutoff frequency, so it is very suitable for the relatively stable frequency characteristics in the structural vibration signal. Its frequency response function can be expressed as: and between

[0026] 2, second stage (frequency domain feature extraction) Step 2-1: Apply fast Fourier transform to the filtered signal to convert it from time domain to frequency domain and obtain the amplitude-frequency characteristics of the acceleration response. The formula of fast Fourier transform is: (2) wherein, is the index of the frequency component, usually corresponding to the discrete frequency , is the sampling time interval. is the complex amplitude of the th frequency component in the frequency domain, representing the frequency response of the signal at frequency . is the sampling value of the signal in the time domain, is the sampling point index in the time domain. is the total number of sampling points, representing the length of the time domain signal. In the frequency domain, the amplitude-frequency characteristics of different frequencies in the signal can be easily analyzed to help identify the vibration characteristics of the structure.

[0027] Step 2-2: Through spectral analysis and combined with peak detection method, the local maximum value in the vibration response can be identified, which corresponds to the natural frequency of the bridge structure under different modes. The spectral peak search algorithm detects the local maximum value by identifying the gradient change in the signal, and further sets the filtering condition to screen out the effective peak value. Its main process includes the following steps: 1) Gradient change detection: First, find potential peak locations by calculating the gradient (or difference) of the signal. In the signal, the gradient changes from positive to negative usually corresponds to a local peak, which can be located by detecting the zero-crossing points.

[0028] 2) Local maximum identification: By analyzing the gradient change direction before and after the zero-crossing points, determine the specific location of the local maximum, that is, find the possible peak near the place where the gradient changes from positive to negative.

[0029] 3) Peak filtering: To exclude meaningless small fluctuations, set filtering conditions to filter effective peaks. Common filtering criteria include: a. Minimum height: Only keep peaks with amplitude higher than the set threshold to avoid misidentifying weak noise fluctuations as peaks; b. Minimum distance: Require enough distance between adjacent peaks to prevent multiple false peaks from being detected on the same main peak.

[0030] 4) Peak determination: Finally, output two main results, the amplitude of the peak and the corresponding position index, for subsequent feature extraction and analysis.

[0031] Step 2-3: In the frequency spectrum, local peaks are usually numbered by frequency from low to high and used as modal frequencies. Generally, the first, third and fifth order frequencies are selected as analysis objects. This order division method is a common convention, and the specific order and its physical meaning may vary depending on the structure characteristics and sensor distribution.

[0032] To ensure the consistency and comparability of feature extraction results, it is necessary to ensure that the environmental conditions (such as temperature, wind speed, load, etc.) during the collection of monitoring data and reference baseline data are basically the same. It is recommended to establish a vibration baseline data of no less than one year under the bridge health state as a reference for subsequent paired analysis. For in-service bridges that need to be evaluated, sensors can be temporarily deployed during the short-term monitoring phase to collect real-time structural vibration signals, which can be paired with baseline data. The characteristic frequencies extracted from both will be used as paired samples for statistical testing.

[0033] 3, third stage (data pairing and modeling) The damage discrimination method used in this invention is based on the modeling of the change in frequency before and after the structure state. Under the condition that other conditions remain unchanged, the decrease in frequency can reflect the weakening of the structural stiffness, thereby inferring the possible existence of damage.

[0034] Let the characteristic frequency of the baseline state be , and the characteristic frequency of the monitoring state be . Define the paired difference as: (3) Select vibration data from the same sensor location, consistent environment, and construct sample pairs to eliminate the influence of individual differences and external interference. This difference sequence is used for subsequent significance analysis.

[0035] 4, the fourth stage (statistical test and damage determination) Step 4-1: After the paired sample is established, use the paired T test to assess whether the mean difference is statistically significant. The T test statistic calculation formula is as follows: (4) Where, is the sample mean difference, is the standard deviation of the sample mean difference, is the number of paired samples.

[0036] Null hypothesis H0: There is no significant difference between the two groups of frequencies, i.e. the bridge structure has not been damaged; Alternative hypothesis H1: There is a significant change in frequency, which may indicate damage.

[0037] The following three prerequisites must be met to use the T test: 1) Paired samples: each group of data comes from the same sensor measurement point; 2) Difference independence: each sample pair is independent of each other; 3) Normal distribution: the difference approximately follows a normal distribution.

[0038] The first two can be guaranteed by experimental design, and the third can be verified by drawing a difference histogram or performing a Shapiro-Wilk test.

[0039] Step 4-2: If the difference distribution satisfies the normality, then the paired T test can be used to determine whether the difference is significant. The p value in the paired T test is the probability of observing the current sample result or a more extreme result under the premise that the null hypothesis is true, and is commonly used to measure the significance level of the sample difference. Its calculation formula is as follows: (5) Where, is the value of the calculated T test statistic, is the standardized sample mean difference. denotes the gamma function, which is mathematically defined as: (6) If the calculated p value is less than the preset significance level (usually set to 0.05), the null hypothesis is rejected, and the frequency difference before and after the damage is considered statistically significant, indicating that the bridge structure may have experienced stiffness degradation or other forms of damage.

[0040] Step 4-3: To further evaluate the degree of change, Cohen's d effect size is introduced, which is calculated as follows: (7) Cohen's d reflects the standardization of the difference, and the interpretation standard is as follows: When d < 0.2: the difference is small and has no practical impact and can be ignored; when 0.2 < d < 0.5: the difference is small and may have some impact; when 0.5 < d < 0.8: the difference is moderate and usually has practical impact; when d > 0.8: the difference is significant and usually has obvious practical impact.

[0041] The combination of p value and effect size makes the damage determination have both statistical significance basis and practical engineering interpretation, and provides quantitative criteria for bridge structure health state identification.

[0042] One embodiment of the present application provides a bridge damage detection method, which is implemented as follows: The present application extracts the characteristic frequencies of the pre-damage and post-damage states by performing frequency domain transformation on the structural vibration signal, and analyzes whether the frequency change has statistical significance by using paired sample T test method, so as to determine whether the bridge structure is damaged. The method combines the physical sensitivity of frequency characteristics with the objectivity of statistical test, and can realize efficient identification without relying on complex modal analysis. The overall process of the method is shown in Figure 1 , which includes four main links of signal processing, feature extraction, hypothesis verification and statistical test.

[0043] In order to verify the superiority of the present application, the series of experimental data of Z-24 bridge are analyzed in this embodiment. The bridge is a prestressed concrete bridge, which is used to carry out a series of controlled damage tests before being demolished. Multi-channel acceleration sensors are deployed on the bridge body to collect vibration response data. The experimental data include multiple groups of vibration response information of the bridge in the undamaged state, pier subsidence (20 mm, 40 mm, 80 mm, 95 mm) and pier top lifting state.

[0044] ​​​​The sampling frequency of the vibration signal is 100 Hz, and each set of data lasts 655 seconds with 65,500 sampling points, covering 9 acceleration channels. Considering that the vibration energy is mainly concentrated in the vertical direction, the first, third and fifth modal frequencies in this direction are selected as the observation objects for damage identification. Table 1 lists the key test stages and corresponding time nodes during the Progressive Damage Test (PDT) of the Z-24 bridge, and briefly describes the damage conditions of each stage. The table covers data from the undamaged state to multiple controlled damage stages, which are used for subsequent paired sample construction and frequency comparison analysis.

[0045] Table 1

[0046] The specific method implemented in this embodiment is described as follows: 1. Data pairing analysis To increase the accuracy of paired sample construction, three sets of undamaged state data are selected as the baseline, corresponding to the stage where the structure has not settled or lifted. These are paired with data from multiple gradual damage stages, and the corresponding first, third and fifth order frequency characteristics of each set are extracted for subsequent statistical testing.

[0047] Figure 2 The first, third and fifth order characteristic frequency changes of the Z-24 bridge at different stages are shown. As can be seen from the figure, the first and third order frequencies show a clear downward trend at the gradual damage stage, indicating that the structural stiffness is gradually weakening. In contrast, the fifth order frequency fluctuates greatly and shows an unstable trend, which may be affected by high-order modal coupling effects and environmental factors such as temperature, wind speed, etc. The frequency changes reflect the different degrees of structural damage, providing an effective reference for subsequent statistical testing.

[0048] 2. Normality test To evaluate the effectiveness of the proposed method, the normality of the three sets of characteristic frequency differences is observed, and the difference histogram is used for visual inspection. Figure 3 The histogram of the first, third and fifth order frequency differences and their normal distribution fitting curves are shown. According to the results in the figure, the first and third order frequency differences approximately follow a normal distribution, meeting the prerequisites for T-test and having good statistical properties, making them suitable for subsequent statistical analysis. In contrast, the fifth order frequency difference shows a discrete long-tailed distribution, indicating that this order of frequency is greatly affected by noise interference and may not fully meet the normality assumption, which needs to be paid attention to in subsequent analysis.

[0049] 3. Statistical test and damage determination After the normality test is completed, paired sample T tests are carried out on the characteristic frequencies of the three groups respectively, and p values and effect size Cohen's d are calculated, and the results are shown in Table 2. As can be seen from the table, the average difference of the first-order frequency between the baseline and the monitoring state is 0.193 Hz, the p value is much smaller than 0.05, and the effect size is 2.441, which belongs to a very strong effect, indicating that the frequency is highly sensitive to structural damage. The average difference of the third-order frequency is 0.104 Hz, and the p value is also much smaller than 0.05, and the effect size is 0.894, which is a medium to strong effect, and also has significant identification ability. The difference of the fifth-order frequency is close to 0, the p value is greater than 0.05, the null hypothesis cannot be rejected, indicating that it cannot be used as a reliable damage identification indicator under the current data.

[0050] Table 2

[0051] The comprehensive analysis results show that the method of the present application has good identification ability and statistical reliability in the Z-24 bridge damage identification experiment. Among them, the first-order modal frequency change is the most significant, which can be used as the main identification index; the third-order frequency also has an auxiliary role; and the fifth-order frequency is easily disturbed by errors, and is only recommended to be used in auxiliary verification. In addition, the design of paired sample test effectively controls the interference of environmental noise and sensor error, so that the statistical result is more stable. The method of the present application combines frequency trend analysis and statistical significance verification, and can realize reliable identification of early and small damage of bridge structure.

[0052] An embodiment of the present application provides a bridge damage detection system, comprising: A signal processing module pre-processes the collected bridge vibration signals; A characteristic frequency extraction module performs time-frequency conversion on the pre-processed signals to obtain frequency domain vibration signals; and extracts characteristic frequencies through peak value detection; A paired detection module calculates a paired difference sequence of the extracted characteristic frequencies and baseline frequencies; A T detection module performs paired T test on the difference sequence to determine whether the bridge is damaged.

[0053] In still another embodiment of the present application, a storage medium, specifically a computer readable storage medium (Memory) is also provided. The computer readable storage medium is a memory device in the terminal equipment, for storing programs and data. It can be understood that the computer readable storage medium herein can include the built-in storage medium in the terminal equipment, and of course can also include the expansion storage medium supported by the terminal equipment, and can be any tangible medium containing or storing programs, which can be used by or in combination with the instruction execution system, device or apparatus. The computer readable storage medium provides a storage space, which stores the operating system of the terminal. Moreover, one or more instructions suitable for being loaded and executed by the processor are also stored in the storage space, and the instructions can be one or more computer programs (including program codes). It should be noted that more specific examples (non-exhaustive list) of the computer readable storage medium herein include: an electrical connection with one or more conductive wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0054] The computer readable storage medium also includes a data signal carrying the readable program code in a baseband or as a part of a carrier, where the readable program code is carried by the data signal. Such a propagated data signal can take any of a variety of forms, including but not limited to electro-magnetic, optical, or any suitable combination thereof. The readable storage medium can also be any readable medium that can be used to carry the readable program code in any suitable manner, and the readable program code can be executed by or in combination with an instruction execution system, device or apparatus. The readable program code contained in the readable storage medium can be transmitted in any suitable manner, including but not limited to wireless, wired, optical, RF, or any suitable combination thereof.

[0055] The program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, C++, or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computing device, partly on the user's computing device, as a stand-alone software package, partly on the user's computing device and partly on a remote computing device or entirely on the remote computing device or server. In the latter scenario, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computing device, such as through the Internet using an Internet Service Provider.

[0056] One or more instructions stored in the computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the bridge damage detection method described in the above embodiments.

[0057] The above only is the preferred embodiment of the present application, and is not used to limit the present application, for the person skilled in the art, the present application can have various changes and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A bridge damage detection method, characterized in that: The following steps are involved: Collect bridge vibration signals and pre-process the collected vibration signals; The pre-processed signal is converted into a time-frequency signal to obtain a frequency domain vibration signal; the characteristic frequency is extracted by peak detection; Calculate the paired difference sequence between the extracted characteristic frequencies and the baseline frequencies; Perform a paired T test on the difference sequence to determine whether the bridge is damaged.

2. The bridge damage detection method according to claim 1, characterized in that: The preprocessing of the collected vibration signal adopts a filtering processing method.

3. The bridge damage detection method according to claim 1, characterized in that: The extracting of characteristic frequencies by peak detection comprises: Detect local peaks of frequency domain vibration signals through gradient changes; Find the peak point where the gradient changes from positive to negative before and after the local peak point, and get the maximum peak point; Based on the maximum peak point, filter out the effective peak; The natural frequency of the bridge structure corresponding to the effective peak is taken as the characteristic frequency.

4. The bridge damage detection method according to claim 1, characterized in that: The paired difference sequence of the extracted characteristic frequency and the baseline frequency is specifically: Select vibration data at the same location and in the same environment as baseline data, and obtain the baseline frequency based on the baseline data; The characteristic frequency and the baseline frequency are paired and the difference is calculated to obtain a paired difference sequence.

5. The bridge damage detection method according to claim 1, characterized in that: The paired T test is performed on the difference sequence to determine whether the bridge is damaged, specifically: Calculating a T-test statistic and an effect size based on a paired difference sequence, wherein the paired difference sequence satisfies a normal distribution; Based on the T-test statistic, calculate the probability of change in frequency p; Determine whether the bridge is damaged based on the effect size and the probability p of frequency change.

6. The bridge damage detection method according to claim 5, characterized in that: The T-test statistic calculation formula is as follows: in, is the mean of the paired difference sequence, is the standard deviation of the paired difference sequence, is the number of characteristic frequencies.

7. The bridge damage detection method according to claim 6, characterized in that: The expression of p is: in, is the value of the calculated T-test statistic, is the standardized sample mean difference, represents the gamma function.

8. The bridge damage detection method according to claim 5, characterized in that: The effect size calculation formula is: in, is the mean of the paired difference sequence, is the standard deviation of the paired difference sequence.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the bridge damage detection method according to any one of claims 1 to 8 is implemented.

10. A bridge damage detection system, characterized in that: include: Signal processing module, which pre-processes the collected bridge vibration signals; The characteristic frequency extraction module performs time-frequency conversion on the preprocessed signal to obtain the frequency domain vibration signal; the characteristic frequency is extracted through peak detection; Paired detection module, which calculates the paired difference sequence between the extracted characteristic frequency and the baseline frequency; The T detection module performs a paired T test on the difference sequence to determine whether the bridge is damaged.

Images