A box girder crack detection method based on multi-point excitation and multi-point measurement

Abstract

A box girder crack detection method with multi-point excitation and multi-point measurement includes two hammer generators fixedly mounted on the box girder and several vibration detection devices. The hammer generators perform hammer excitation on the box girder to be detected, and the vibration detection devices collect audio signals above 250 Hz related to the response of the box girder to the hammer excitation transmission path, and detect information such as stress magnitude and stress distribution in the cement concrete of the box girder. When a crack in the prestressed box girder develops, even if the crack is still in a closed state, structural characteristics such as the continuity of the cement concrete structure, the magnitude of the cement concrete stress, and the stress distribution in the path from the excitation point of the box girder to the signal collection point will change, thereby causing changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection device. By analyzing the changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection device, crack detection of the prestressed box girder of the bridge can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic detection of cracks in prestressed box girder bridges, and in particular to a box girder crack detection method based on multi-point excitation and multi-point measurement. Background Art

[0002] During the use of bridge structures, due to the influence of various factors, there are certain degrees of damage. When these damages accumulate to a certain extent, the performance of the structure will show a significant decline, thereby affecting the normal use of the bridge and even causing major bridge accidents. Therefore, dynamic monitoring of bridge structures to achieve early warning and repair of bridge structure damage can not only prevent the occurrence of major bridge accidents, but also greatly extend the service life of bridge structures. Prestressed box girder bridges account for the majority of bridge structures in my country. Therefore, dynamic monitoring of prestressed box girder structures has great economic and social significance. Structural damage of prestressed box girders initially develops from primary cracks and eventually forms irreversible structural damage. Therefore, dynamic monitoring of prestressed box girder structures is actually monitoring the development of cracks in prestressed box girders.

[0003] Extensive research has been conducted both domestically and internationally on monitoring crack development in prestressed box girders, primarily focusing on the impact of cracks on the natural frequencies, vibration modes, and stiffness of bridge structures. This research has demonstrated the feasibility of crack detection in prestressed box girders. However, to date, automated crack detection in prestressed box girders has not been truly applied in practice. This is due to the following: existing studies on the impact of cracks on prestressed box girders have focused on the low- and medium-order natural frequencies and vibration modes, specifically the impact of cracks on the overall structure of the prestressed box girder. In reality, due to the inherent prestress in prestressed box girders, when cracks have developed but are still closed, their impact on the low- and medium-order natural frequencies and vibration modes, which determine the overall structure, is minimal. Furthermore, the installation location of the detection sensor has a significant impact on the detection results. Therefore, applying the existing research results on the impact of cracks on the natural frequencies, vibration modes, and stiffness of bridge structures to the actual implementation of crack detection in prestressed box girders has encountered a significant technical bottleneck: the inability to accurately detect cracks and their locations. Summary of the Invention

[0004] In order to overcome the shortcomings of the background technology, the present invention discloses a box girder crack detection method with multi-point excitation and multi-point measurement, which includes a hammer generator and a vibration detection device. The hammer generator performs hammer excitation on the box girder to be detected, and the vibration detection device collects audio signals above 250Hz of the box girder to which the detected box girder responds to the hammer excitation. The frequency spectrum characteristics of the audio signal are only related to the characteristics of the hammer generator and the structural characteristics of the audio transmission path from the excitation point of the box girder to the signal collection point, but have nothing to do with the overall structure of the prestressed box girder. The structural characteristics of the path from the excitation point of the box girder to the signal collection point include cement concrete material. Quality characteristics, continuity of cement concrete structure, configuration structure of steel bars in cement concrete, stress magnitude and stress distribution of cement concrete, etc.; when cracks in prestressed box girder develop, even if the cracks are still in a closed state, the continuity of cement concrete structure, stress magnitude and stress distribution and other structural characteristics in the path from the excitation point of the detected box girder to the signal collection point will change, which will cause the frequency spectrum characteristics of the audio signal collected by the vibration detection device to change; conversely, by analyzing the changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection device, the cracks in the prestressed box girder of the bridge can be detected.

[0005] In order to achieve the above-mentioned purpose of the invention, the present invention adopts the following technical scheme: a box girder crack detection method with multi-point excitation and multi-point measurement, including a hammer generator and a vibration detection device, the hammer generator performs hammer excitation on the box girder to be detected, and the vibration detection device collects audio signals above 250Hz of the box girder to which the detected box girder responds to the hammer excitation; when the crack in the prestressed box girder develops, the structural continuity, cement concrete stress magnitude, stress distribution and other structural characteristics in the path from the excitation point of the detected box girder to the signal collection point change, which will cause the frequency spectrum characteristics of the audio signal collected by the vibration detection device to change. Therefore, by analyzing the changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection device, the cracks in the prestressed box girder of the bridge can be detected.

[0006] Furthermore, two hammer generators are provided, which are fixedly arranged at both ends of the box beam in the length direction; a plurality of vibration detection devices are provided, which are fixedly arranged between the two hammer generators along the length direction of the box beam; when the prestressed box beam of the bridge is detected, the two hammer generators successively perform hammer excitation on the box beam at both ends in a set order, and a plurality of vibration detection devices successively detect two groups of audio signals of hammer excitation response; the two groups of audio signals of hammer excitation response are converted into spectrum signals containing the fundamental frequency of the sound and a plurality of frequencies and amplitudes, and then the first N orders of the spectrum signal are converted into a spectrum signal matrix table (that is, a matrix table containing frequencies and corresponding amplitudes, for the specific matrix table structure, see the attached figure of the specification), cluster analysis is performed on the spectrum signal matrix table, and according to the cluster analysis The results can be used to judge whether cracks appear in the prestressed box girder of the bridge and the location of the cracks; in the present invention, two hammer generators and multiple vibration detection devices are set for two purposes: 1. By excitation in two directions, the accuracy of the detection results of cracks on the box girder is guaranteed, because there are countless transmission paths from the hammer generator to the vibration detection device in theory, but the shortest path between the two usually has the greatest impact on the spectral characteristics of the audio signal. As the transmission path becomes longer, its impact on the spectral characteristics of the audio signal gradually decreases. Therefore, when a crack appears between the two vibration detection devices, the spectral characteristics of the audio signal collected by the vibration detection device between the crack and the hammer generator change relatively little, but the vibration detection device outside the crack and the hammer generator changes slightly. The changes in the audio signal spectrum characteristics collected by the dynamic detection device are relatively large. Based on this, theoretically, only one hammer generator is required to detect cracks and their positions. However, in the present invention, the detection of changes in the audio signal spectrum characteristics is achieved through cluster analysis, without any prior conditions or threshold settings. Especially in the early stage of crack detection, when the accumulation of audio signal spectrum characteristic data is small, even if the cracks in the box girder have not developed, the detection analysis results may be misjudged due to the influence of data collection errors and drift. However, after setting the excitation in two directions, the cluster analysis results of the audio signal spectrum characteristics collected by the excitation in the two directions can be compared and verified. Only when the comparison and verification results are consistent, the detection result will be output. The results of the cracks and crack locations are obtained, thereby avoiding misjudgment and ensuring the accuracy of the crack detection results on the box girder; 2. By setting multiple vibration detection devices between the two hammer generators, the sensitivity of crack detection is improved; when the cracks on the box girder develop, the cracks have the greatest impact on the changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection device closest to it. When the number of vibration detection devices is small, the distance between the vibration detection devices and the cracks on the box girder will be relatively large, resulting in relatively small changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection devices; if the number of vibration detection devices set between the two hammer generators is increased, the distance between the detection devices and the cracks on the box girder will be relatively shortened, thereby improving the sensitivity of crack detection;The specific number of vibration detection devices to be installed in actual implementation is determined by the balance between actual detection accuracy requirements and cost. The recommended distance between two vibration detection devices is between 1-4 meters.

[0007] Furthermore, a method for judging whether cracks appear in the prestressed box girder of a bridge based on the cluster analysis results is as follows: the cluster analysis results of the two groups of audio signals simultaneously have temporal stability, temporal synchronization, and stability of the inflection point position of the vibration detection device, that is, the cluster analysis results of the two groups of audio signals are valid only when they simultaneously have temporal stability, temporal synchronization, and stability of the inflection point position of the vibration detection device in the spectrum signal matrix table; hereby explained: temporal stability means that when the cluster analysis results are sorted by date, a certain date is used as the dividing line, and thereafter the classification results no longer have the phenomenon of cross-classification results; temporal synchronization means that when the cluster analysis results are sorted by date, the two groups of audio signal cluster analysis results are consistent in date; the stability of the inflection point position of the vibration detection device means that the inflection point position of the cluster analysis results remains on a certain vibration detection device and does not change.

[0008] Furthermore, a method for determining the location of cracks in the prestressed box girder of a bridge based on the cluster analysis results is as follows: in the spectrum signal matrix table of the cluster analysis results of the two sets of audio signals, the area included between the inflection point positions of the two sets of vibration detection devices is the range where cracks in the prestressed box girder of the bridge appear; when the inflection point positions of the two sets of vibration detection devices overlap, it indicates that the cracks appear near the inflection points of the two sets of vibration detection devices.

[0009] Furthermore, when performing cluster analysis on the spectrum signal, the amplitude of the spectrum signal is normalized. The actual operation is: when converting the first N orders of the spectrum signal into a spectrum signal matrix table, the amplitude of the first N orders of the spectrum signal is normalized.

[0010] Furthermore, the two groups of spectral signals of hammer excitation response are clustered analyzed independently, that is, when the first N orders of the spectral signal are converted into a spectral signal matrix table, the spectral signals of the two groups of hammer excitation response are divided into two spectral signal matrix tables; when performing cluster analysis, cluster analysis is performed on the two spectral signal matrix tables respectively, and finally the two cluster analysis results are compared to analyze whether the cluster analysis results of the two groups of audio signals have temporal stability, temporal synchronization, stability of the inflection point position of the vibration detection device, and the range of occurrence of cracks in the prestressed box girder of the bridge.

[0011] Preferably, the spectral signals of the two groups of hammer excitation responses are merged and then cluster analysis is performed, that is, when the first N orders of the spectral signals are converted into a spectral signal matrix table, the spectral signals of the two groups of hammer excitation responses are combined into a spectral signal matrix table; when performing cluster analysis, cluster analysis is performed on the merged spectral signal matrix tables of the two groups of hammer excitation responses.

[0012] Furthermore, the hammer generator and the vibration detection device are fixedly arranged at the bottom of the box beam bottom plate or the side of the box beam web.

[0013] Preferably, the hammer generators are fixed at both ends of the bottom of the box beam bottom plate; the vibration detection devices are provided in several rows, which are fixed at the bottom of the box beam bottom plate or the side of the box beam web along the length direction of the box beam, and are located between the two hammer generators.

[0014] Due to the adoption of the technical solution as described above, the present invention has the following beneficial effects: the present invention discloses a box girder crack detection method with multi-point excitation and multi-point measurement, including a hammer generator and a vibration detection device. The hammer generator performs hammer excitation on the box girder to be detected, and the vibration detection device collects audio signals above 250Hz of the box girder to which the detected box girder responds to the hammer excitation. The frequency spectrum characteristics of the audio signal are only related to the characteristics of the hammer generator and the structural characteristics of the audio transmission path from the excitation point of the box girder to the signal collection point, but have nothing to do with the overall structure of the prestressed box girder. The structural characteristics of the path from the excitation point of the box girder to the signal collection point include cement concrete material characteristics, the continuity of the cement concrete structure, The configuration structure of steel bars in cement concrete, the stress magnitude and stress distribution of cement concrete, etc.; when cracks in prestressed box girder develop, even if the cracks are still in a closed state, the structural continuity, cement concrete stress magnitude, stress distribution and other structural characteristics in the path from the excitation point of the detected box girder to the signal collection point will change, which will cause the frequency spectrum characteristics of the audio signal collected by the vibration detection device to change; conversely, by monitoring the changes in the frequency spectrum characteristics of the audio signal collected by the vibration detection device, the detection of cracks in the prestressed box girder of the bridge can be achieved; in addition, setting up multi-point excitation and multi-point vibration detection devices can effectively improve the accuracy of box girder crack testing and the sensitivity of crack detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 Schematic diagram of the location of the hammer generator and vibration detection device Figure 1 ;

[0016] Figure 2 Schematic diagram of the location of the hammer generator and vibration detection device Figure 2 ;

[0017] Figure 3 Schematic diagram of the location of the hammer generator and vibration detection device Figure 3

[0018] Figure 4 Schematic diagram of the location of the hammer generator and vibration detection device Figure 4 ;

[0019] Figure 5 Schematic diagram of crack location of prestressed box girder Figure 1 ;

[0020] Figure 6 Schematic diagram of crack location of prestressed box girder Figure 2 ;

[0021] Figure 7 This is the first example of matrix representation of the spectrum signal of the hammer generator A;

[0022] Figure 8 This is the first example of matrix representation of the spectrum signal of the hammer generator B;

[0023] Figure 9 This is the second example of matrix representation of the spectrum signal of the hammer generator A;

[0024] Figure 10 This is the second example of matrix representation of the spectrum signal of the hammer generator B;

[0025] Figure 11 This is the third example of matrix representation of the spectrum signal of the hammer generator A;

[0026] Figure 12 This is the third example of the matrix representation of the spectrum signal of the hammer generator B.

[0027] In the figure: 1. Hammer generator; 2. Vibration detection device; 3. Prestressed box girder; 4. Crack. DETAILED DESCRIPTION

[0028] The present invention can be explained in detail by the following examples, the purpose of which is to disclose the present invention and to protect all technical improvements within the scope of the present invention.

[0029] A method for detecting cracks in a box girder using multi-point excitation and multi-point measurement, based on a hardware device comprising a hammer generator 1 and a vibration detection device 2. The hammer generator 1 performs hammer excitation on the box girder to be detected, and the vibration detection device 2 collects the response signal of the box girder to the hammer excitation. In the present invention, the response signal of the box girder to the hammer excitation collected by the vibration detection device 2 is an audio signal above 250Hz, that is, the detection is of a high-order excitation response of the box girder after receiving the hammer excitation. Two hammer generators 1 are provided, which are fixedly arranged on the lower side of the bottom plate near the end on both sides of the prestressed box girder 3. The number of vibration detection devices 2 is determined according to the actual detection accuracy requirements and the cost balance. It is recommended that the distance between the two vibration detection devices be between 1 and 4 meters. If higher detection accuracy is required, a smaller value is selected for the distance between the two vibration detection devices 2. The vibration detection device 2 is fixedly arranged on the bottom or side of the web of the prestressed box girder 3 along the length direction of the prestressed box girder 3, between the two hammer generators 1. For the specific setting positions of the hammer generator 1 and the vibration detection device 2, please refer to the appendix of the instruction manual. Figure 1 、 2 , 3, 4;

[0030] To simplify the subsequent description, Figure 5 、 6 In the embodiment of the invention, two hammer generators 1 and four vibration detection devices 2 are provided at the bottom of the prestressed box girder. Figure 5 、 6 In the embodiment, the left hammer generator 1 is named hammer generator A, the right hammer generator 1 is named hammer generator B, and the four vibration detection devices 2 are named 1#, 2#, 3#, and 4# from left to right;

[0031] When testing the prestressed box girder 3 of the bridge, the hammer generator A on the left first performs three hammer excitations on the prestressed box girder 3, and the 1#, 2#, 3#, and 4# vibration detection devices 2 detect the audio signals of the three hammer excitation responses, convert the audio signals of the three hammer excitation responses into spectrum signals, take the average values ​​of the frequencies and amplitudes of the first five orders of the spectrum signals and add the test date to make a matrix table of the spectrum signals excited by the hammer generator A; then, the hammer generator B on the right performs three hammer excitations on the prestressed box girder 3, and the same method is used to make a matrix table of the spectrum signals excited by the hammer generator B; when making the spectrum signal matrix table, the amplitude signals of the first five orders of the spectrum signals are all normalized, that is, the amplitude of the fundamental frequency signal in the spectrum signal detected by the vibration detection device 2 is adjusted to 1, and the amplitudes of the remaining four wideband signals are adjusted in the same proportion;

[0032] As the test date accumulates, the final result is as shown in the appendix of the manual. Figure 7 、 8The hammer generator A and hammer generator B are shown in the figure. Cluster analysis is performed on the two groups of excitation spectrum signal matrix tables, and the classification results are 20221012-20221011 and 20221012-20221020. Figure 7 The main factors of the hammer generator A excitation spectrum signal matrix table are the amplitude signals in the spectrum signals detected by the 3# and 4# vibration detection devices 2, where the specification is attached. Figure 8 The main factors of the hammer generator B excitation spectrum signal matrix table are the amplitude signals in the spectrum signals detected by the 1# and 2# vibration detection devices 2; the cluster analysis results of the two groups of excitation spectrum signal matrix tables of this embodiment are analyzed as follows: 1. Figure 7 Hammer generator A excitation spectrum signal matrix table and instruction manual Figure 8 The hammer generator B excitation spectrum signal matrix table shows that the cluster analysis results are sorted by date, with 20221012 as the boundary, showing the stability of the classification in date sorting. Therefore, the cluster analysis results have temporal stability and temporal synchronization; 2. Figure 7 The hammer generator A excitation spectrum signal matrix table, vibration detection device 2 is sorted from left to right, starting from 20221012, the cluster analysis result shows that 3# vibration detection device 2 is the inflection point position (the shaded part, and the closest to hammer generator A), and the inflection point position has not changed in subsequent dates, and is always 3# vibration detection device 2; In addition, the instruction manual is attached Figure 8 The hammer generator B excitation spectrum signal matrix table shows that the vibration detection device 2 is sorted from right to left. Starting from 20221012, the cluster analysis result shows that the 2# vibration detection device 2 is the inflection point position (the shaded part, which is closest to the hammer generator B), and the inflection point position has not changed in subsequent dates and is always 2# vibration detection device 2. Therefore, the cluster analysis result has the stability of the inflection point position of the vibration detection device 2; In summary, the appendix of the manual shows that Figure 7 、 8 The cluster analysis results of the two groups of excitation spectrum signal matrices of hammer generator A and hammer generator B shown are valid, and it can be determined that cracks have appeared in the prestressed box girder of the bridge.

[0033] It is to be noted that the temporal stability and temporal synchronization of the cluster analysis results may be different from the above embodiments. Figure 9 、 10Taking the two groups of excitation spectrum signal matrices of hammer generator A and hammer generator B as examples, it is explained that in terms of the temporal stability and temporal synchronization of the cluster analysis results, there is no obvious dividing line of 20221012, but there may be a continuous period of dates as the dividing line. After a period of time, the classification results show stability in the date sorting; this situation still meets the judgment requirements of the cluster analysis results in terms of temporal stability and temporal synchronization.

[0034] The location of cracks in the prestressed box girder of the bridge is still determined by the instructions in the appendix. Figure 7 、 8 The hammer generator A and hammer generator B are shown as two sets of excitation spectrum signal matrix tables for example. Figure 7 Hammer generator A excitation spectrum signal matrix table, cluster analysis results 3# vibration detection device 2 is the inflection point position, the instruction manual is attached Figure 8 The hammer generator B excitation spectrum signal matrix table, cluster analysis results 2# vibration detection device 2 is the inflection point position, so the location of the bridge prestressed box girder crack is located between 2# vibration detection device 2 and 3# vibration detection device 2, see the appendix of the manual Figure 5 .

[0035] There is another situation to judge the crack position of the prestressed box girder of the bridge. Figure 11 、 12 The hammer generator A and hammer generator B are shown in the following table as an example: Figure 11 Hammer generator A excitation spectrum signal matrix table, cluster analysis results 2# vibration detection device 2 is the inflection point position, the instruction manual is attached Figure 12 The hammer generator B excitation spectrum signal matrix table, the cluster analysis result also shows that the 2# vibration detection device 2 is the inflection point position, so the location of the bridge prestressed box girder crack is near the 2# vibration detection device 2, see the attached manual Figure 6 .

[0036] In the above discussion of the clustering processing and analysis results of the excitation spectrum signal matrix tables of hammer generator A and hammer generator A, the excitation spectrum signal matrix tables of hammer generator A and hammer generator B are two independent matrix tables; however, in the actual processing of collected data, the excitation spectrum signal matrix tables of hammer generator A and hammer generator A can also be merged into one matrix table, and then clustering processing and analysis can be performed.

[0037] The parts not described in detail in this invention are prior art.

Claims

1. A method for detecting cracks in a box girder using multi-point excitation and multi-point measurement, comprising a hammer generator (1) and a vibration detection device (2), wherein the hammer generator (1) performs hammer excitation on the box girder to be detected, and the vibration detection device (2) collects a response signal of the box girder to the hammer excitation; the method is characterized in that: The vibration detection device (2) collects the response signal of the box girder under inspection to the hammer excitation, which is an audio signal of 250 Hz or above; Two hammer generators (1) are provided, which are fixedly arranged at both ends of the box girder in the longitudinal direction; a plurality of vibration detection devices (2) are provided, which are fixedly arranged between the two hammer generators (1) along the longitudinal direction of the box girder; when the prestressed box girder of the bridge is detected, the two hammer generators (1) successively perform hammer excitation on the box girder at both ends in a set order, and the plurality of vibration detection devices (2) successively detect two groups of audio signals of hammer excitation response; the two groups of audio signals of hammer excitation response are converted into frequency spectrum signals, the frequency spectrum signals are clustered and analyzed, and whether cracks occur in the prestressed box girder of the bridge and the location of the cracks are determined according to the cluster analysis results; The method for judging whether cracks occur in the prestressed box girder of a bridge based on the cluster analysis results is as follows: when the cluster analysis results of the two groups of audio signals have temporal stability, temporal synchronization, and stability of the inflection point position of the vibration detection device (2), it can be judged that cracks occur in the prestressed box girder of the bridge.

2. The box girder crack detection method using multi-point excitation and multi-point measurement according to claim 1 is characterized by: The method for determining the location of cracks in the prestressed box girder of a bridge based on the cluster analysis results is as follows: in the cluster analysis results of the two groups of audio signals, the area included between the inflection point positions of the two groups of vibration detection devices (2) is the range where the cracks in the prestressed box girder of the bridge appear.

3. The box girder crack detection method using multi-point excitation and multi-point measurement according to claim 1 is characterized by: When performing cluster analysis on spectrum signals, the amplitude of the spectrum signals is normalized.

4. The box girder crack detection method using multi-point excitation and multi-point measurement according to claim 1 is characterized by: The spectral signals of the two groups of hammer excitation responses were clustered and analyzed independently.

5. The box girder crack detection method using multi-point excitation and multi-point measurement according to claim 1 is characterized by: The spectral signals of the two groups of hammer excitation responses were merged and cluster analysis was performed.

6. The box girder crack detection method using multi-point excitation and multi-point measurement according to claim 1 is characterized by: The hammer generator (1) and the vibration detection device (2) are fixedly arranged at the bottom of the box beam bottom plate or the side of the box beam web plate.

7. The box girder crack detection method using multi-point excitation and multi-point measurement according to claim 6 is characterized by: The hammer generators (1) are fixedly arranged at both ends of the bottom of the box beam bottom plate; the vibration detection devices (2) are provided in a plurality of rows and are fixedly arranged at the bottom of the box beam bottom plate or the side of the box beam web along the length direction of the box beam, and are located between the two hammer generators (1).

Images