Bridge expansion joint online damage early warning method and system based on dynamic fingerprint tracking

Abstract

This invention provides an online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking, comprising: acquiring vertical vibration signals at the expansion joint measuring points according to sampling frequency and sampling time; determining the excitation effectiveness of vehicle impact on the expansion joint within a time unit based on the impact significance factor; if the determination result is that the excitation is effective, performing fast Fourier transform on the vibration acceleration time-domain signal in the vertical vibration signal according to different frequency resolutions to obtain minute-level acceleration spectra and second-level acceleration spectra respectively; based on the minute-level acceleration spectrum, performing peak picking within a specified frequency band to obtain a type I dynamic fingerprint; based on the second-level acceleration spectrum, selecting multiple segments of equal-scale spectrum for mean smoothing to obtain a type II dynamic fingerprint; comparing the type I dynamic fingerprint with an energy threshold and comparing the type II dynamic fingerprint with a correlation threshold; if the type I dynamic fingerprint exceeds or the type II dynamic fingerprint is below the threshold, a damage early warning is issued.

Description

Technical Field

[0001] This invention relates to the field of bridge monitoring technology, and in particular to an online damage early warning method and system for bridge expansion joints based on dynamic fingerprint tracking. Background Technology

[0002] Modular expansion joints are commonly used in medium- and long-span bridge structures, serving as crucial components for ensuring smooth vehicle traffic, releasing beam end displacement, and coordinating bridge deformation. Operating in complex and variable outdoor environments, they are subjected to repeated impacts from vehicle loads over extended periods, making them one of the most vulnerable and susceptible to damage within bridge structures. Affected by factors such as rubber aging and steel fatigue, the actual service life of existing modular expansion joints on bridges is typically less than 5 years, far below their designed lifespan. Failure of the expansion joint not only causes vehicle bouncing, severely reducing driving comfort, but can also endanger traffic safety, leading to bridge traffic disruptions and causing significant socio-economic losses.

[0003] Currently, the operation and maintenance of modular expansion joints relies on traditional manual inspections. This method has significant drawbacks: First, due to the long inspection cycle, it is difficult to provide early warning and timely response to sudden expansion joint damage incidents between adjacent inspection periods; second, manual inspection results rely too heavily on the subjective experience of engineers, and are insufficient in identifying early damage hidden inside the expansion device. Generally, the damage can only be detected when it has developed to a severe stage visible on the surface of the expansion device (such as broken central beams, loose bolts, or sheared rubber waterstops), leading to missed opportunities for optimal treatment.

[0004] Existing online monitoring methods for expansion joints generally rely on time-domain damage diagnosis indicators. By installing strain gauges and displacement gauges on the beams and crossbeams of modular expansion joints, the stress and deformation state is monitored. However, such methods only reflect the local response at the measuring point and cannot quantify the overall performance degradation of the expansion joint. Furthermore, time-domain indicators such as strain, displacement, and amplitude are severely affected by factors such as temperature changes, environmental noise, axle load, the relative position of the measuring point and the tire, and multi-vehicle parallel events, which easily leads to a high false alarm rate. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method and system for online damage early warning of bridge expansion joints based on dynamic fingerprint tracking, which solves the technical problem of poor damage identification capability when providing online damage early warning of bridge expansion joints in existing technologies.

[0006] The technical solution adopted in this invention is as follows: Firstly, a method for online damage early warning of bridge expansion joints based on dynamic fingerprint tracking is provided, including the following steps: The vertical vibration signal at the expansion joint measuring point is obtained according to the sampling frequency and sampling time; The effectiveness of vehicle impact on expansion joint excitation within a time unit is determined based on the impact significance factor. If the excitation is effective, the vibration acceleration time domain signal in the vertical vibration signal is subjected to fast Fourier transform at different frequency resolutions to obtain minute-level acceleration spectrum and second-level acceleration spectrum, respectively. Based on the minute-level acceleration spectrum, peak picking is performed within the specified frequency band to obtain a Class I dynamic fingerprint; Based on the second-level acceleration spectrum, multiple segments of equal-scale spectrum are selected and mean smoothing is performed to obtain a type II dynamic fingerprint. The Class I dynamic fingerprint is compared with the energy threshold, and the Class II dynamic fingerprint is compared with the correlation threshold. If the Class I dynamic fingerprint exceeds the energy threshold, or the Class II dynamic fingerprint is below the correlation threshold, a damage warning is issued.

[0007] Furthermore, when acquiring the vertical vibration signal at the expansion joint measuring point according to the sampling frequency and sampling time, the sampling frequency is not less than 1 kHz, the sampling time is not less than 1 minute, and the expansion joint measuring point is the end of the middle beam of the modular expansion joint in the emergency lane of the bridge deck.

[0008] Furthermore, when determining the excitation effectiveness of vehicle impact on expansion joints within a time unit based on the impact significance factor, excitation effectiveness is determined when the impact significance factor exceeds the effectiveness threshold; the impact significance factor is calculated as follows: in, This is the rounding operator. Sampling frequency, This is the calibration time interval; For the floor operator, N is the data length of a fixed time unit; Within a fixed time unit, according to the chronological order at intervals The extracted root mean square acceleration sequence.

[0009] Furthermore, the validity threshold is calculated as follows: Where median(·) is the median operator. MED is the threshold for impact significance factor, which is the sample sequence. The median, MAD is the median absolute deviation between the calculated SIF value and MED.

[0010] Furthermore, after normalizing the minute-level acceleration spectrum, peak picking is performed within a specified frequency band to obtain a Type I dynamic fingerprint; the specified frequency band is the (0, 50] Hz band; the energy threshold is... .

[0011] Furthermore, based on the second-level acceleration spectrum, multiple segments of equal-scale spectrum are selected for mean smoothing to obtain the edge features of the current minute segment. Then, the correlation between the edge features of the current minute segment and the edge features corresponding to the previous minute segment is analyzed to obtain the type II dynamic fingerprint; the correlation threshold is 0.7.

[0012] Secondly, a bridge expansion joint online damage early warning system based on dynamic fingerprint tracking is provided. This system is used to implement the aforementioned bridge expansion joint online damage early warning method based on dynamic fingerprint tracking, including: The data acquisition module is used to acquire vertical vibration signals at the expansion joint measuring points according to the sampling frequency and sampling time. The verification module is used to determine the excitation effectiveness of vehicle impact on expansion joints within a time unit based on the impact significance factor. The analysis module is used to perform fast Fourier transform on the vibration acceleration time-domain signal in the vertical vibration signal according to different frequency resolutions to obtain minute-level acceleration spectrum and second-level acceleration spectrum respectively; it is also used to normalize the minute-level acceleration spectrum and perform peak picking in a specified frequency band to obtain a type I dynamic fingerprint; and based on the second-level acceleration spectrum, it selects multiple equal-scale spectrum segments for mean smoothing to obtain a type II dynamic fingerprint. The early warning module compares the Type I dynamic fingerprint with the energy threshold and the Type II dynamic fingerprint with the correlation threshold. If the Type I dynamic fingerprint exceeds the energy threshold or the Type II dynamic fingerprint is lower than the correlation threshold, a damage warning is issued.

[0013] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention introduces the impact significance factor as a quantitative basis for judging the excitation effect of vibration acceleration signal, and adaptively selects effective excitation signals for subsequent frequency domain conversion, thereby ensuring the accuracy and usability of dynamic fingerprint.

[0014] (2) This invention uses Type I dynamic fingerprints that reflect low-frequency energy and Type II dynamic fingerprints that reflect spatiotemporal correlation to consider the degree of decrease in overall stiffness of expansion joints caused by fracture failure and the degree of evolution of spectral profile caused by local damage, respectively. This can ensure zero missed reports and zero false reports for online damage warning of modular expansion joints in complex traffic and field environments. Attached Figure Description

[0015] Figure 1This is a flowchart of the online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking, according to an embodiment of the present invention. Figure 2 This is a time history diagram of vibration test when a vehicle drives over an expansion joint according to Embodiment 1 of the present invention; Figure 3 This is a time-frequency spectrum diagram of the acceleration of each lane in Embodiment 1 of the present invention; Figure 4 This is a diagram of the re-acquired vibration acceleration signal from Embodiment 1 of the present invention; Figure 5 This is a statistical graph of the impact significance factor of acceleration at different time units at different lane measuring points in Embodiment 1 of the present invention; Figure 6 This is a diagram showing the minute-level acceleration spectrum and edge feature extraction results of the third lane measuring point in Embodiment 1 of the present invention; Figure 7 This is a diagram showing the minute-level acceleration spectrum and edge feature extraction results of the emergency lane measuring points in Embodiment 1 of the present invention; Figure 8 As in Embodiment 1 of the present invention Figure 6 Autocorrelation heatmap of edge features of each unit in the middle; Figure 9 As in Embodiment 1 of the present invention Figure 7 Autocorrelation heatmap of edge features of each unit in the middle; Figure 10 As in Embodiment 1 of the present invention Figure 6 , Figure 7 The heatmap shows the correlation between edge features of different lane measuring points. Figure 11 This is a minute-level acceleration spectrum and a type I dynamic fingerprint of Embodiment 2 of the present invention; Figure 12 This is a diagram showing the synchronous vibration test record of the same type of expansion joint at the bridge deck inlet and outlet in Embodiment 3 of the present invention. Figure 13 This is a diagram showing the effectiveness of vibration excitation at the bridge deck inlet and outlet expansion joints of the same type in Embodiment 3 of the present invention. Figure 14 This is a diagram showing the minute-level acceleration spectrum and edge feature extraction results of a normal expansion joint on the bridge deck entrance side in Embodiment 3 of the present invention. Figure 15 This is a diagram showing the minute-level acceleration spectrum and edge feature extraction results of the damaged expansion joint on the bridge deck exit side in Embodiment 3 of the present invention. Figure 16 This is an autocorrelation thermogram of each measuring point in Embodiment 3 of the present invention; Figure 17 This is a thermogram showing the correlation between measuring points of normal and damaged expansion joints in Embodiment 3 of the present invention. Detailed Implementation

[0016] The technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments. Example 1:

[0017] like Figure 1 As shown in the figure, this invention proposes an online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking, including: This embodiment utilizes a vibration test of the 160mm modular expansion joint beam at the normal operating entrance of a continuous beam bridge (right span) on a highway in Hubei Province to analyze its measured characteristics. Acceleration measurement points are located at the intersection of the expansion joint beam and the center lines of the third lane and emergency lane, with MEMS sensors installed at these points. The signal sampling frequency is 1kHz. In some implementations, filtering can be used to eliminate low-drift acceleration measurement errors.

[0018] During the process of using sensors to collect vertical vibration signals of the beams in the modular expansion joint, a camera is also installed at the outer edge of the emergency lane to capture video images of vehicles passing by at the same time. For example... Figure 2 As shown, taking a vertical vibration signal record collected at a measuring point for a period of 68 seconds, spanning the passage of 18 vehicles, as an example, the vibration acceleration signal in the vertical vibration signal is expanded into a short-time Fourier transform, yielding the acceleration time spectrum as shown below. Figure 3 As shown, it is evident that the modal energies in the acceleration frequency spectrum only reach a perceptible and measurable significance when a vehicle passes by, indicating that the dynamic fingerprint of modular expansion joints is difficult to identify solely through random environmental vibrations. (Comparison) Figure 2 , Figure 3 It can be seen that the heavier the axle load and the stronger the amplitude, the better the impact excitation effect on the modular expansion joint, and the clearer the frequency domain information obtained.

[0019] Therefore, this invention directly selects the common tire diameter of modern heavy-duty trucks. =1.04m, and obtain the average driving speed of this type of vehicle from the traffic management department. The calibration time interval is calculated using the following formula: Then, using a fixed time unit of 1 minute, vibration acceleration signals were re-acquired on-site for a total of 6 time units at different lane measuring points, such as... Figure 4 As shown. The driving speed is taken as v=80km / h, so the calibrated time interval Δt ≈0.05s when the tire passes through the modular expansion joint.

[0020] The vertical vibration signals collected at each measuring point were then verified. This invention introduces an impact significance factor for verification. The acceleration time history of the vehicle passing each measuring point within one minute was obtained. The acceleration time history was divided into multiple acceleration segments by a calibrated time interval. The root mean square of acceleration in each acceleration segment was then calculated, and all root mean squares were combined into an acceleration change sequence. The impact significance factor was then calculated based on the acceleration change sequence. The specific calculation formula is as follows: in, This is the rounding operator. For floor operation, Sampling frequency, For calibration time interval, The data length of the acceleration time history, Within a fixed time unit, according to the chronological order at intervals The extracted root mean square acceleration sequence, The impact significance factor at the measuring point.

[0021] Based on this, the impact significance factors of the measuring points in the third lane and the emergency lane were calculated respectively. ), columnar analysis such as Figure 5 As shown.

[0022] United Figure 2 , Figure 3 , Figure 5 It can be seen that because no heavy vehicles passed over the bridge surface in the 5th time unit, the maximum amplitudes at the third and emergency lane measuring points did not exceed 2g and 0.6g respectively. Therefore, this unit... The calculated value is the minimum; that is, if the current time unit is the minimum. If the value is much smaller than the calculated values ​​of other units, it can be qualitatively determined that the excitation effect of vehicle impact during this period is poor. Under the same traffic flow, the emergency lane measuring point... All values ​​are much smaller than the calculated values ​​for the third lane, meaning that the farther the measuring point is from the tire pressure application point, the smaller the values ​​become. The smaller, therefore, The quantitative setting of the effectiveness threshold is significantly influenced by the relative position of the test point and the tire's action. It is necessary to combine on-site manual or remote video recordings of multiple heavy vehicles passing through the test point, and then calculate and statistically analyze the data before setting the same test point. Validity threshold.

[0023] This invention calculates the same measurement point within multiple time units. Sample sequence composed of values Based on statistical control theory, the sample sequence was subjected to small outlier detection to obtain the impact significance factor at this measurement point. Validity threshold as follows: in, It is a sample sequence the median of It is the median of the absolute deviation between the calculated SIF value and the MED value.

[0024] Then the time units calculated from this measuring point The value is compared with the validity threshold. If the value is greater than the validity threshold, the verification is deemed successful. Valid excitation signals are then adaptively selected for subsequent frequency domain conversion, ensuring the accuracy and usability of the dynamic fingerprint.

[0025] For powered fingerprint sensors, there are two types: Type I powered fingerprint sensors and Type II powered fingerprint sensors.

[0026] Type I dynamic fingerprinting is based on minute-level acceleration spectrum, and peak picking is performed within a specified frequency band to obtain Type I dynamic fingerprints. Essentially, it uses the minute-level frequency domain transformation result of the current time unit as the analysis object, and considers the peak picking result of the normalized Fourier spectrum exceeding the energy threshold in the (0,50]Hz frequency band as the measured ultra-low frequency significant mode. The implementation method of peak picking is not limited and can be implemented using any feasible method in the existing technology.

[0027] Type II dynamic fingerprinting is based on the second-level acceleration spectrum. Multiple segments of the spectrum at the same scale are selected and mean-smoothed to obtain the edge features of the current minute segment. Then, the correlation between the edge features of the current minute segment and the edge features corresponding to the previous minute segment is analyzed to obtain the Type II dynamic fingerprint. Type II dynamic fingerprinting represents the temporal and spatial correlation of edge features. Temporal correlation refers to the correlation of edge features of the same expansion joint in different time units. Spatial correlation refers to the correlation of edge features between two expansion joints of the same type at the bridge entrance and exit within the same time unit.

[0028] For Type I dynamic fingerprints, in a specific implementation, the vibration acceleration signals from the third and emergency lane measuring points within each time unit where validity verification has passed are subjected to minute-level frequency domain conversion with a time history length of 1 minute. Since the vibration acceleration signal is only 1 minute long, it can be directly converted into the frequency domain after obtaining the vibration acceleration signal. Figure 6 , Figure 7 The minute-level acceleration spectrum (Fourier spectrum) shows that for a normally operating modular expansion joint, no significant measured vertical modes are observed in any time unit below 50Hz, meaning the normalized energy is less than [value missing]. .

[0029] Based on the above analysis, according to the design principle of modular expansion joints, to avoid reducing service life due to "vehicle-expansion joint" resonance, it should be ensured that the measured vertical acceleration of the central beam has no significant modes in the ultra-low frequency band (0, 50] Hz. Based on the inherent characteristics of the modular expansion joint, following the frequency domain determination principle for significant vibrations in slender structures, the energy in the normalized spectrum below [a certain value] should be considered. All information is considered noise and insignificant components, therefore the energy threshold for Class I dynamic fingerprints is... .

[0030] For Type II dynamic fingerprint sensors, in a specific implementation, the vibration acceleration signals from the third and emergency lane measuring points within each time unit are segmented into frequency domains at the second level, with a time span of 1 second. By segmenting the 1-minute vibration acceleration signal into second-level segments (i.e., 60 seconds per minute), there are 60 segments, resulting in 60 sets of results after frequency domain transformation. The average of these 60 sets of results is then taken. Figure 6 , Figure 7 Edge features extracted in this way ensure both the similarity of the spectral contour and the accuracy of the peak features (modal frequencies). For different measurement points on the third and emergency lanes, the minute-level acceleration spectrum and edge features of the fifth unit differ significantly from the corresponding information of other time units. This is due to the poor excitation effect of vehicle impact in the fifth unit.

[0031] Furthermore, autocorrelation thermodynamic analysis was performed on the edge characteristics of each unit at the same measuring point. The analysis results for the third and emergency lane measuring points are as follows: Figure 8 , Figure 9 As shown; where CZ1~CZ6 represent the edge features of the first to sixth time units of the third lane measuring point, and CB1~CB6 represent the edge features of the first to sixth time units of the emergency lane measuring point. The results show that, since the emergency lane measuring point is farther from the tire's point of action, the autocorrelation of CB is weaker than that of CZ; furthermore, the Pearson correlation coefficients between CB5 and other units CB are all below 0.7, so it is necessary to screen the effective time history based on the impact excitation effect (for example, it is determined that unit 5 should be skipped).

[0032] Based on the above analysis, it can be seen that, according to the inherent characteristics of the modular expansion joint, the frequency domain characteristics of the same or the same type of expansion joint remain stable during normal operation. Combined with the statistical quantification of the strong correlation of the Pearson correlation coefficient, the correlation threshold corresponding to the Class II dynamic fingerprint is set to 0.7.

[0033] Furthermore, a thermal analysis was performed on the correlation between the edge features of different measuring points, such as... Figure 10As shown, the edge features between different measuring points have weak or almost no correlation, indicating that modular expansion joint dynamic fingerprint tracking should be based on vibration monitoring data from fixed measuring points.

[0034] In this embodiment, vibration test analysis of normal modular expansion joints shows that, for the scenario of "heavy vehicles passing by", the effectiveness of impact excitation should be determined at the hardware level for the current time unit, and effective signal segments should be selected for subsequent damage determination to ensure the accuracy and usability of dynamic fingerprint. Example 2:

[0035] like Figure 1 As shown in the figure, this invention proposes an online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking, including: This embodiment uses a 160mm modular expansion joint at the exit of a continuous beam bridge (left span) on a highway in Hubei Province, where the central beam fractures and misaligns. Signal acquisition, validity verification, and frequency domain conversion are all performed using the method described in Embodiment 1, yielding the corresponding minute-level acceleration spectrum. Peak values ​​are then extracted from the minute-level acceleration spectrum to obtain the corresponding Type I dynamic fingerprint, such as... Figure 11 As shown.

[0036] Furthermore, based on the sensors on both sides (near and far ends) of the fracture in the middle beam of the third lane, the acceleration signal of this unit is subjected to minute-level frequency domain conversion to extract Type I dynamic fingerprints, such as... Figure 11 As shown. Therefore, it can be seen that, compared with Example 1... Figure 6 Compared to a normally operating modular expansion joint of the same type, the vertical vibration of the central beam in the third lane exhibits significant ultra-low frequency modal characteristics at 30.5 Hz, meaning the normalized energy is greater than [value missing]. This triggers a Class I dynamic fingerprint threshold warning.

[0037] In summary, the vibration test analysis of the modular expansion joint after beam fracture and misalignment described in this embodiment shows that, compared with the same type of modular expansion joint in normal operation, the measured vibration signal after damage exhibits significant ultra-low frequency modal characteristics, proving the early warning effectiveness of Class I dynamic fingerprint. Example 3:

[0038] like Figure 1 As shown in the figure, this invention proposes an online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking, including: To balance the sensor's range margin with ease of installation and debugging, the accelerometer should be placed at the end of the beam in the expansion joint within the emergency lane (near the outer edge guardrail). This embodiment utilizes 160mm modular expansion joints on the entrance and exit sides of a continuous beam bridge (left lane) on a highway in Hubei Province. The expansion joint on the bridge entrance side is still in normal operation, while the beams in the expansion joints in the second and third lanes on the bridge exit side are broken or damaged. The vibration test signal acquisition, validity verification, and frequency domain conversion for each modular expansion joint are performed using the method described in Embodiment 1, yielding the corresponding minute-level acceleration spectrum and edge characteristics, such as... Figures 12-17 As shown.

[0039] Combination Figure 12 and Figure 13 For the same type of modular expansion joint at the same location, the working conditions of the same traffic flow passing through different expansion joints were analyzed. The results show that the amplitude and SIF calculated value of the expansion joint after damage are significantly different from those under normal operation.

[0040] In this embodiment, L=9, and for the normal expansion joint on the bridge deck entrance side, MED = 451, MAD = 49. Therefore, the valid unit numbers are determined to be 2, 3, 4, 5, 6, 7, and 9. For the damaged expansion joint on the bridge deck exit side, MED = 83, MAD = 16. Therefore, the valid unit numbers are determined to be 2, 3, 4, 5, 7, and 9. Figure 13 The shaded area represents the region where impact excitation is ineffective, i.e. Therefore, units 2, 3, 4, 5, 7, and 9 (a total of 6 units) were selected for comparative analysis.

[0041] The vertical acceleration signals of the same type of expansion joints on the bridge deck at the inlet and outlet sides within each time unit are frequency domain converted (integer minutes) to obtain... Figure 14 , Figure 15 The minute-level acceleration spectrum (see Fourier spectrum); where the shaded region represents the range where the frequency domain energy is not significant, i.e., the normalized energy < The results show that, compared with the normal state, the same type of modular expansion joint with fractured or damaged central beam exhibits significant ultra-low frequency modal characteristics, triggering the warning threshold of Class I dynamic fingerprinting.

[0042] In addition, the vertical acceleration signals of the bridge deck inlet and outlet expansion joints within each time unit are frequency domain transformed (segmented second by second) to obtain Figure 14 , Figure 15 Edge characteristics within the measurement point. Autocorrelation thermodynamic analysis was performed on the edge characteristics of each unit at the same measurement point, yielding the analysis results for the bridge deck inlet and outlet expansion joint measurement points, as shown below. Figure 16As shown in the figure, J1 to J6 represent the edge features of units 2, 3, 4, 5, 7, and 9 of the normal expansion joint measuring point at the bridge deck entrance, respectively, while H1 to H6 represent the edge features of units 2, 3, 4, 5, 7, and 9 of the damaged expansion joint measuring point at the bridge deck exit, respectively. It can be seen that for steady-state expansion joints in normal operation or that are already damaged, the autocorrelation between the edge features of each measuring point unit is extremely strong.

[0043] Furthermore, a thermal analysis was performed on the correlation between the edge features of different measuring points, such as... Figure 17 As shown. Analysis of the same traffic flow passing through different expansion joints shows that the edge characteristics of damaged modular expansion joints and normal expansion joints of the same type are not correlated (Pearson correlation coefficient is much lower than 0.7), triggering a Class II dynamic fingerprint threshold warning. Example 4:

[0044] This invention also provides an online damage early warning system for bridge expansion joints based on dynamic fingerprint tracking. This system is used to implement the aforementioned online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking, and includes: The data acquisition module is used to acquire vertical vibration signals at the expansion joint measuring points according to the sampling frequency and sampling time. The verification module is used to determine the excitation effectiveness of vehicle impact on expansion joints within a time unit based on the impact significance factor. The analysis module is used to perform fast Fourier transform on the vibration acceleration time-domain signal in the vertical vibration signal according to different frequency resolutions to obtain minute-level acceleration spectrum and second-level acceleration spectrum respectively; it is also used to perform peak picking in a specified frequency band based on the minute-level acceleration spectrum to obtain type I dynamic fingerprint; and based on the second-level acceleration spectrum, it selects multiple equal-scale spectrum segments for mean smoothing to obtain type II dynamic fingerprint; The early warning module compares the Type I dynamic fingerprint with the energy threshold and the Type II dynamic fingerprint with the correlation threshold. If the Type I dynamic fingerprint exceeds the energy threshold or the Type II dynamic fingerprint is lower than the correlation threshold, a damage warning is issued.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for online damage early warning of bridge expansion joints based on dynamic fingerprint tracking, characterized in that, Includes the following steps: The vertical vibration signal at the expansion joint measuring point is obtained according to the sampling frequency and sampling time; The effectiveness of vehicle impact on expansion joint excitation within a time unit is determined based on the impact significance factor. If the excitation is effective, the vibration acceleration time domain signal in the vertical vibration signal is subjected to fast Fourier transform at different frequency resolutions to obtain minute-level acceleration spectrum and second-level acceleration spectrum, respectively. Based on the minute-level acceleration spectrum, peak picking is performed within the specified frequency band to obtain a Class I dynamic fingerprint; Based on the second-level acceleration spectrum, multiple segments of equal-scale spectrum are selected and mean smoothing is performed to obtain a type II dynamic fingerprint. The Class I dynamic fingerprint is compared with the energy threshold, and the Class II dynamic fingerprint is compared with the correlation threshold. If the Class I dynamic fingerprint exceeds the energy threshold or the Class II dynamic fingerprint is lower than the correlation threshold, a damage warning is issued.

2. The online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking as described in claim 1, characterized in that, When obtaining the vertical vibration signal at the expansion joint measuring point according to the sampling frequency and sampling time, the sampling frequency shall not be less than 1 kHz, the sampling time shall not be less than 1 minute, and the expansion joint measuring point shall be the end of the middle beam of the modular expansion joint in the emergency lane of the bridge deck.

3. The online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking as described in claim 1, characterized in that, When determining the excitation effectiveness of vehicle impacts on expansion joints within a time unit based on the impact significance factor, excitation effectiveness is determined when the impact significance factor exceeds the effectiveness threshold; the impact significance factor is calculated as follows: in, This is the rounding operator. Sampling frequency, This is the calibration time interval; For floor operation, It is the data length of a fixed time unit; Within a fixed time unit, according to the chronological order at intervals The extracted root mean square acceleration sequence.

4. The online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking as described in claim 3, characterized in that, The validity threshold is calculated as follows: Where median(·) is the median operator, MED is the threshold for impact significance factor, which is the sample sequence. The median, MAD is the median absolute deviation between the calculated SIF value and MED.

5. The online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking as described in claim 1, characterized in that, After normalizing the minute-level acceleration spectrum, peak picking is performed within a specified frequency band to obtain a Type I dynamic fingerprint; the specified frequency band is the (0, 50] Hz band; the energy threshold is... .

6. The online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking as described in claim 1, characterized in that, Based on the second-level acceleration spectrum, multiple segments of equal-scale spectrum are selected and mean smoothed to obtain the edge features of the current minute segment. Then, the correlation between the edge features of the current minute segment and the edge features corresponding to the previous minute segment is analyzed to obtain the type II dynamic fingerprint; the correlation threshold is 0.

7.

7. A bridge expansion joint online damage early warning system based on dynamic fingerprint tracking, characterized in that, This system is used to implement the online damage early warning method for bridge expansion joints based on dynamic fingerprint tracking as described in any one of claims 1-6, comprising: The data acquisition module is used to acquire vertical vibration signals at the expansion joint measuring points according to the sampling frequency and sampling time. The verification module is used to determine the excitation effectiveness of vehicle impact on expansion joints within a time unit based on the impact significance factor. The analysis module is used to perform fast Fourier transform on the vibration acceleration time-domain signal in the vertical vibration signal according to different frequency resolutions to obtain minute-level acceleration spectrum and second-level acceleration spectrum respectively; it is also used to normalize the minute-level acceleration spectrum and perform peak picking in a specified frequency band to obtain a type I dynamic fingerprint; and based on the second-level acceleration spectrum, it selects multiple equal-scale spectrum segments for mean smoothing to obtain a type II dynamic fingerprint. The early warning module compares the Type I dynamic fingerprint with the energy threshold and the Type II dynamic fingerprint with the correlation threshold. If the Type I dynamic fingerprint exceeds the energy threshold or the Type II dynamic fingerprint is lower than the correlation threshold, a damage warning is issued.

Images