SYSTEM AND METHOD FOR TESTING STRUCTURE MODE OF VIBRATION BASED ON DIGITAL IMAGE RECOGNITION

Abstract

Disclosed is a system and method for testing a structure mode of vibration based on digital image recognition, which comprises a camera, targets, a bridge, a vertical acceleration sensor and a lateral acceleration sensor; the camera is arranged near the bridge head of the bridge; the bridge is equipped with a plurality of targets equidistantly inside guardrails on both sides; and the vertical acceleration sensor and the lateral acceleration sensor are fixedly arranged on the camera. The present application avoids the arrangement of a large number of sensors and complicated wiring in the bridge vibration detection, saves time and reduces economic cost, is convenient to operate, has relatively high precision, and has broad application prospects.

Claims

1. A system for testing a structure mode of vibration based on digital image recognition, comprising a camera (1), targets (2), a bridge (3), a vertical acceleration sensor (4) and a lateral acceleration sensor (5); the camera (1) is arranged near a bridge head of the bridge (3); the bridge (3) is equipped with a plurality of targets (2) equidistantly inside guardrails on both sides; and the vertical acceleration sensor (4) and the lateral acceleration sensor (5) are fixedly arranged on the camera (1).

2. The system for testing the structure mode of vibration based on digital image recognition according to claim 1, the camera (1) is arranged on a camera tripod (8), and the camera tripod (8) is arranged near the bridge head of the bridge (3).

3. The system for testing the structure mode of vibration based on digital image recognition according to claim 1, the targets (2) on both sides of the bridge (3) are aligned correspondingly in the transverse direction of the bridge (3), and the target planes is perpendicular to the line of sight of the camera.

4. The system for testing the structure mode of vibration based on digital image recognition according to claim 1, also comprising a data acquisition system (6) and a computer (7), the camera (1), the vertical acceleration sensor (4) and the lateral acceleration sensor (5) being all connected to the data acquisition system (6), and the computer (7) being connected to the data acquisition system (6).

5. A method for testing a structure mode of vibration based on digital image recognition, being performed by adopting the system for testing the structure mode of vibration based on digital image recognition according to claim 1, comprising the following process that: the camera (1) captures images containing all targets (2) in real time, and the vertical acceleration sensor (4) and the lateral acceleration sensor (5) monitor the vibration of the camera (1) itself in real time; and when the bridge (3) is subjected to random excitation to produce vibration, the targets (2) and the camera (1) vibrate together with the bridge (3), and the lateral vibration curve, longitudinal vibration curve and torsional vibration curve of the bridge (3) are calculated by the images containing the targets (2) and the camera (1) vibration information to realize the testing of bridge (3) vibration.

6. The method for testing the structure mode of vibration based on digital image recognition according to claim 5, the process of calculating the lateral vibration curve, longitudinal vibration curve and torsional vibration curve of the bridge (3) by the images containing the targets (2) and the camera (1) vibration information comprises: capturing the center of each target (2) from the images containing the targets (2), obtaining the pixel displacements of the target (2) center at different moments, and converting into a unit of length, and then obtaining the lateral relative vibration displacements and longitudinal relative vibration displacements of the measuring points where the targets are located; unifying the units of the lateral relative vibration displacement and longitudinal relative vibration displacement of each target (2) and the units of the lateral relative vibration displacement and longitudinal vibration displacement unit of the camera (1), and subtracting the reverse displacement caused by camera vibration from the relative vibration displacement of the measuring point to obtain the absolute vibration displacement of the measuring point; drawing a lateral time history curve and a vertical time history curve of the measuring point corresponding to each target; using the lateral time history curve and the longitudinal time history curve of the measuring point corresponding to each target to obtain the natural frequency of each order, and obtain the lateral vibration amplitude-frequency characteristics and longitudinal vibration amplitude-frequency characteristics of the bridge (3); calculating the real-time torsion angle of the measuring point and drawing a time history curve through the longitudinal absolute vibration displacements of measuring points of the two lateral targets (2) in the longitudinal direction of the bridge (3), and obtaining the torsion amplitude-frequency characteristic curve of each measuring point by using the real-time torsion angles and time history curves of all measuring points; and using the obtained torsion amplitude-frequency characteristic curve of each measuring point to obtain the lateral vibration curve, longitudinal vibration curve and torsional vibration curve of the bridge (3).

7. The method for testing the structure mode of vibration based on digital image recognition according to claim 6, performing denoising preprocessing on the absolute vibration displacement data of the measuring point corresponding to each target by filtering, and drawing the lateral time history curve and longitudinal time history curve of the measuring point corresponding to each target.

8. The method for testing the structure mode of vibration based on digital image recognition according to claim 6, performing the fast Fourier transform and spectrum analysis on the lateral time history curve and longitudinal time history curve of the measuring point corresponding to each target to identify the natural frequency of each order, and obtain the lateral vibration amplitude-frequency characteristic and longitudinal vibration amplitude-frequency characteristic of the bridge (3).

9. The method for testing the structure mode of vibration based on digital image recognition according to claim 6, after calculating the real-time torsion angle of the measuring point and drawing the time history curves, performing the fast Fourier transform and frequency spectrum analysis on the time history curves, and performing the same processing on the data of the measuring points corresponding to all targets to obtain the torsional amplitude-frequency characteristic curve of each measuring point.

10. The method for testing the structure mode of vibration based on digital image recognition according to claim 6, using the obtained amplitude-frequency characteristic curve of each measuring point, calculating the damping coefficient based on a half-power spectral density method, calculating the impact coefficient of the bridge by a weighting method, and drawing the lateral vibration curve, longitudinal vibration curve and torsional vibration curve of the bridge (3).

Description

BRIEF DESCRIPTION OF DRAWINGS

[0020] FIG. 1 is a structural schematic diagram of a system for testing a structure mode of vibration based on digital image recognition according to the present application;

[0021] FIG. 2 is a structural schematic diagram of a camera, a vertical acceleration sensor and a lateral acceleration sensor according to the present application;

[0022] FIG. 3 is a three-dimensional diagram of the system for testing the structure mode of vibration based on digital image recognition according to the present application during image acquisition;

[0023] FIG. 4 is a partial top view of the system for testing the structure mode of vibration based on digital image recognition according to the present application during image acquisition;

[0024] FIG. 5 is a schematic diagram of the layout of targets during the longitudinal mode of vibration test of a simply supported steel beam in the embodiment of the present application;

[0025] FIG. 6(a) is a longitudinal displacement time history curve of a measuring point 2 in the embodiment of the present application;

[0026] FIG. 6(b) is a longitudinal displacement time history curve of a measuring point 3 in the embodiment of the present application;

[0027] FIG. 6(c) is a longitudinal displacement time history curve of a measuring point 4 in the embodiment of the present application;

[0028] FIG. 7 is a frequency amplitude curve of the measuring point 2 in the embodiment of the present application;

[0029] FIG. 8 is a frequency amplitude curve of the measuring point 3 in the embodiment of the present application;

[0030] FIG. 9 is a frequency amplitude curve of the measuring point 4 in the embodiment of the present application;

[0031] FIG. 10(a) is a first-order mode of vibration diagram of a bridge in the embodiment of the present application;

[0032] FIG. 10(b) is a second-order mode of vibration diagram of the bridge in the embodiment of the present application;

[0033] FIG. 11 (a) is a damping ratio diagram of the measuring point 2 in the embodiment of the present application;

[0034] FIG. 11(b) is a damping ratio diagram of the measuring point 3 in the embodiment of the present application;

[0035] FIG. 11(c) is a damping ratio diagram of the measuring point 4 in the embodiment of the present application;

[0036] FIG. 12 is a flowchart of a method for testing a structure mode of vibration based on digital image recognition according to the present application.

REFERENCE SIGNS

[0037] 1—camera, 2—target, 3—bridge, 4—vertical acceleration sensor, 5—lateral acceleration sensor, 6—data acquisition system, 7—computer, 8—camera tripod.

DESCRIPTION OF EMBODIMENTS

[0038] Hereinafter, the structure principle and analysis principle of the present application will be described in detail with reference to the accompanying drawings:

[0039] As shown in FIG. 1 and FIG. 3, a system for testing a structure mode of vibration based on digital image recognition according to the present application includes a camera 1, targets 2, a bridge 3, a vertical acceleration sensor 4, a lateral acceleration sensor 5, a data acquisition system 6, a computer 7, and a camera tripod 8; the camera 1 is arranged near the bridge head of the bridge 3, and the camera tripod 8 is located at the bottom of the camera 1, and is used for supporting the camera 1 and adjusting the position of camera 1; the bridge 3 is equipped with a plurality of targets 2 equidistantly inside the guardrails on both sides; as shown in FIG. 2, the vertical acceleration sensor 4 and the lateral acceleration sensor 5 are fixedly arranged on the camera 1 without relative displacement; and the data acquisition system 6 is connected to the camera 1, and the computer 7 is connected to the data acquisition system 6.

[0040] Wherein the camera 1 is Canon 5D SR, which requires 3 frames per second or more high-speed burst, 12 times or more magnification display, effective pixels of above 4000 W, not less than 40 cross-type AF points, a shutter speed of 1/8000 to 30 seconds, and a CMOS image sensor, and has a mirror vibration control function, and a DIGIC6 digital image processor. The target 2 is a special light circular foam target with a cross bullseye, the target thickness is 20-50 mm, the square target length and width range is 100-200 mm, the cross line width is 5-10 mm, and it is required that the line color and the background color (white) contrast obviously. The target 2 is a special light circular foam target with a cross bullseye.

[0041] The vertical acceleration sensor 4 has voltage sensitivity of 200-500 mV, an operating frequency range of 0.2-8 KHz, a maximum range of not less than 10 g, resolution of not less than 0.0004 g, an operating temperature range of −20˜80° C., a dead weight of not more than 100 g, and a size of less than 50×50 mm.

[0042] The lateral acceleration sensor 5 has voltage sensitivity of 200-500 mV, an operating frequency range of 0.2-8 KHz, a maximum range of not less than 10 g, resolution of not less than 0.0004 g, an operating temperature range of −20˜80° C., a dead weight of not more than 100 g, and a size of less than 50×50 mm.

[0043] The data acquisition system 6 is used for receiving the information acquired by the camera 1, the vertical acceleration sensor 4 and the lateral acceleration sensor 5, and transmitting the acquired information to the computer 7 for data processing.

[0044] The computer 7 is Lenovo ThinkPad T490, which has a CPU speed of 3.1 GHz, memory of 8 G or more, a hard disk capacity of 1 T or more, RJ45, HDMI interfaces, display resolution of 4×4096×2160/4×5120×2880@60 Hz, and video memory of above 2 GB.

[0045] The tripod 8 has an unfolded size of not more than 1000 mm, a load capacity of not less than 10 kg, and a weight of not more than 5 kg. A celestron electric head is matched with the tripod.

[0046] With reference to FIG. 12, when the system for testing the structure mode of vibration based on digital image recognition according to the present application is in use, the bridge 3 is excited by wind load, water flow, vehicle load or even earthquakes to produce vibration, and the target 2 and the camera 1 which are arranged on the bridge 3 are also vibrated. As shown in FIG. 2, the camera 1 can simultaneously monitor the pixel displacements of the plurality of targets 2 at each measuring point by real-time imaging, obtain the vibration situation of each target 2 measuring point relative to the camera 1, and store the image in real time, and transmit through the acquisition card of the data acquisition system 6 and store in the computer 7. The vibration pixel displacement of each target 2 measuring point relative to the camera 1 is converted into the lateral and longitudinal relative vibration displacements expressed in unit of length.

[0047] Before starting the measurement, a certain number of targets 2 are arranged inside guardrails on both sides of the bridge deck. The positions of the targets 2 on both sides of the bridge deck should be aligned correspondingly in the transverse direction of the bridge, and all target planes are perpendicular to the camera's line of sight, and all targets are numbered.

[0048] The camera 1 is arranged at one end of the bridge 3, the camera is facing the center of the target 2, a vertical acceleration sensor 4 and a lateral acceleration sensor 5 are arranged on the camera, and the vertical acceleration sensor 4 and the lateral acceleration sensor 5 are used for detecting the vibration of the camera 1. The data measured by the vertical acceleration sensor 4 and the lateral acceleration sensor 5 can be used to eliminate the effect of the vibration of the camera itself on the detection result in the subsequent calculation process.

[0049] After starting the measurement, the real-time image data containing all the targets 2 monitored by the camera is transmitted through the acquisition card of the data acquisition system 6 and stored in the computer 7.

[0050] Capturing the center of each target 2 in the real-time image transmitted from the camera, the computer 7 obtains the pixel displacement of the target center at different moments, and converts it into a unit of length, and then obtains the lateral and longitudinal relative vibration displacements of the measuring points where all the targets are located.

[0051] The computer 7 unifies the units of the lateral and longitudinal relative vibration displacements of each target 2 and the units of the lateral and longitudinal vibration displacements of the camera 1 at the same time, and subtracts the lateral and longitudinal vibration displacements of the camera 1 from the lateral and longitudinal relative vibration displacements of each target 2 to obtain the lateral and longitudinal absolute vibration displacements of each target 2.

[0052] The computer 7 performs denoising preprocessing on the lateral and longitudinal absolute vibration displacement data of each target 2 by a filtering method, and draws the lateral and longitudinal time history curves of each target 2 measuring point, and performs the fast Fourier transform and spectrum analysis on the lateral and longitudinal time history curves of each target 2 measuring point to identify the natural frequency of each order, can calculate the damping coefficient based on an improved half-power spectral density method, can obtain the impact coefficient of the bridge by adopting a weighting method, and draws lateral and longitudinal mode of vibration curves of the first three orders of the bridge 3.

[0053] The fast Fourier transform and frequency spectrum analysis are performed on the lateral and longitudinal time history curves of each target measuring point to identify the natural frequency of each order, and obtain the lateral and longitudinal vibration amplitude-frequency characteristics of the bridge.

[0054] As shown in FIG. 4, through the longitudinal absolute vibration displacements of measuring points of the two lateral targets 2 in the longitudinal direction of the bridge, the real-time torsion angle of the cross section of the bridge at these two measuring points can be calculated and the time history curve can be drawn. The fast Fourier transform and frequency spectrum analysis are performed on the above-mentioned time history curves. After the above-mentioned processing is performed on the data of all target measuring points, the torsional amplitude-frequency characteristic curve of each measuring point can be obtained, that is, the torsional mode of vibration of the bridge 3.

[0055] Using the amplitude-frequency characteristic curve obtained above, the damping coefficient is calculated based on the half-power spectral density method, the impact coefficient of the bridge is obtained by the weighting method, and the lateral vibration curve, longitudinal vibration curve and torsional vibration curve of the bridge can be obtained to realize the testing of bridge vibration.

[0056] It can be seen from the above that the present application can be a real bridge vibration detection method for testing a full-scale bridge, which has small data acquisition and analysis workload, wide applicability and no damage to the bridge.

Example

[0057] To further illustrate the above method, the longitudinal mode of vibration testing of a simply supported steel beam as shown in FIG. 5 is taken as an example. The total length of the beam is 6 m, and 5 measuring points are arranged equidistantly. Because the measuring point 1 and the measuring point 5 are arranged at the fulcrums, only the data of the measuring point 2, the measuring point 3 and the measuring point 4 are used. The data acquisition frequency of the acquisition system is 59.94 Hz, and the highest recognizable frequency is 29.97 Hz.

[0058] In the present example, the measured longitudinal displacement time history curves of the measuring point 2, the measuring point 3, and the measuring point 4 are shown in FIG. 6(a) to FIG. 6(c), the frequency-amplitude curves of the measuring point 2, the measuring point 3 and the measuring point 4 are shown in FIG. 7˜FIG. 9, and the maximum amplitude of each measuring point along the length of the bridge is shown in FIG. 10(a) and FIG. 10(b). Through the above calculation process of the present application, the damping ratio diagrams of the measuring point 2, the measuring point 3 and the measuring point 4 are shown in FIG. 11(a)˜FIG. 11(c). From the results of FIG. 11(a)˜FIG. 11(c), it can be seen that the natural frequency of the simply supported steel beam is about 3.44 Hz. Therefore, the present method can accurately identify the first two-order mode of vibrations of the simply supported beam, which also shows that the principle and operation of the method are feasible. When the acquisition frequency is increased appropriately, the third-order later mode of vibrations can be identified.

[0059] In summary, it can be seen that compared with the prior art, the present application has the advantages that the number of sensors required for detection is greatly reduced, and the measurement cost is reduced, the bridge does not need to be equipped with acceleration or displacement sensors, a large amount of wiring work is not required, the bridge deck is not damaged, and the high-precision detection of the lateral (bridge width direction), longitudinal (vertical bridge deck direction) vibration and torsional vibration of the bridge can be realized by laying out once, and the synergy is relatively good.