CN-122016804-A - Steel-concrete combined bridge crack monitoring system and method based on unmanned aerial vehicle

CN122016804ACN 122016804 ACN122016804 ACN 122016804ACN-122016804-A

Abstract

The invention relates to the technical field of bridge engineering and information, and discloses a steel-concrete combined bridge crack monitoring system and method based on an unmanned aerial vehicle, wherein the system comprises a multi-rotor unmanned aerial vehicle, a ground multi-base station RTK-GPS system, a stereoscopic vision acquisition module, a laser radar scanning module and an infrared thermal imager; the multi-rotor unmanned aerial vehicle is provided with each monitoring module to fly to a region to be detected, a ground multi-base station RTK-GPS system generates a virtual base station through carrier phase difference to realize centimeter-level positioning of the unmanned aerial vehicle, a stereoscopic vision acquisition module acquires a high-definition image, a laser radar scanning module generates point cloud data, the point cloud data and the point cloud data are fused to construct a textured three-dimensional model, and a thermal infrared imager identifies hidden cracks. The invention realizes millimeter-level non-contact monitoring of the crack of the joint surface of the steel web and the concrete, does not need personnel climbing, has high positioning precision and strong adaptability, is suitable for long-term health monitoring of bridges, and provides reliable data support for predicting the structural performance of the bridge.

Inventors

CHEN SHUXIA
XIA JIATING
Bao wanbao
ZHAO TAIPING

Assignees

中国十七冶集团有限公司

Dates

Publication Date: 20260512
Application Date: 20260126

Claims (10)

1. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle is characterized by comprising a multi-rotor unmanned aerial vehicle, a ground multi-base-station RTK-GPS system, a stereoscopic vision acquisition module, a laser radar scanning module and an infrared thermal imager, wherein the multi-rotor unmanned aerial vehicle is used for carrying monitoring equipment and flying to a steel-concrete combined beam to-be-detected area, the ground multi-base-station RTK-GPS system is composed of a plurality of ground base stations, virtual base stations are generated through carrier phase difference to achieve centimeter-level positioning of the unmanned aerial vehicle, the stereoscopic vision acquisition module is used for acquiring a high-definition image of the surface of the steel-concrete combined beam to achieve crack identification, the laser radar scanning module and the stereoscopic vision acquisition module work cooperatively to generate point cloud data of a steel-concrete combined bridge structure to ensure that edge point cloud of cracks are complete, and the infrared thermal imager is used for detecting a heat conduction abnormal area caused by cracks inside the steel-concrete combined beam to identify hidden cracks.
2. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle according to claim 1, wherein in the ground multi-base-station RTK-GPS system, a base station generates a virtual base station coordinate by fusing multi-base-station differential data by adopting a weighted least square method, and a calculation formula is In the following For the base station coordinates, Is the unmanned aerial vehicle to base station distance.
3. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle is characterized by comprising a stereoscopic vision acquisition module, an image preprocessing unit and a gray level equalization processing unit, wherein the stereoscopic vision acquisition module comprises a high-definition optical camera pair, the distortion of the camera is less than 1%, an included angle of 15 degrees is fixed on a tripod head, the baseline distance is 15cm, and the image preprocessing unit is used for correcting distortion based on a Zhang calibration method.
4. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle, which is characterized in that the laser radar scanning module is fused with data acquired by the stereoscopic vision acquisition module to construct a textured three-dimensional model for intuitively displaying the structure and crack distribution of the steel-concrete combined beam.
5. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle, which is characterized by further comprising a data processing module, wherein the data processing module is used for automatically identifying cracks of collected image data by adopting a U-Net++ network model, the U-Net++ network model training set comprises at least 1000 bridge crack images under different types and working conditions, and a crack development prediction model is established in the data processing module: In the middle of For the initial width to be the same, As a function of the temperature-sensitive coefficient, The temperature difference is used for predicting the future development trend of the crack.
6. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle is characterized in that a base station shell in the ground multi-base station RTK-GPS system is made of 316L stainless steel, or a PPK post-processing mode is adopted, and the monitoring difficulty of serious areas shielded by satellite signals such as canyons is overcome by combining at least four ground base station control points at two ends of the bridge.
7. The steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle is characterized in that SSD is built in the multi-rotor unmanned aerial vehicle, images are stored, data storage in a non-network environment is achieved, and steel-concrete combined bridge monitoring of non-network areas such as mountain areas is adapted.
8. A steel-concrete combined bridge crack monitoring method based on an unmanned aerial vehicle is characterized by comprising the following steps of: Step S1, a multi-base station RTK-GPS dynamic differential positioning step, namely initializing a ground multi-base station before the unmanned aerial vehicle flies, generating a virtual base station through carrier phase differential, and receiving differential data in real time during the unmanned aerial vehicle flying process to realize centimeter-level positioning; S2, a three-dimensional vision and laser radar point cloud fusion modeling step, namely, an unmanned aerial vehicle flies to a region to be detected of a steel-concrete composite beam, a three-dimensional vision acquisition module and a laser radar scanning module work synchronously to acquire images and point cloud data, and a textured three-dimensional model is constructed through data fusion; And step S3, in the step of automatic crack identification and trend prediction, a data processing module adopts a U-Net++ network model to automatically identify the acquired image cracks, and the future development trend of the cracks is predicted by combining the crack development prediction model according to the identification result.
9. The method for monitoring the cracks of the steel-concrete composite bridge based on the unmanned aerial vehicle according to claim 8, wherein the step S2 is characterized by further comprising the steps of performing image preprocessing on the acquired data, namely performing distortion correction based on a Zhang' S calibration method, performing gray level equalization processing, performing preprocessing operations such as denoising on laser radar point cloud data, and the like.
10. The steel-concrete combined bridge crack monitoring method based on the unmanned aerial vehicle according to claim 8, wherein in the step S3, the U-Net++ network model periodically updates a training set of a crack identification model so as to improve the accuracy and adaptability of model identification, and meanwhile, the crack development prediction model is optimized by combining data such as ambient temperature, humidity and the like.

Description

Steel-concrete combined bridge crack monitoring system and method based on unmanned aerial vehicle Technical Field The invention relates to the technical field of line construction, in particular to a steel-concrete combined bridge crack monitoring system and method based on an unmanned aerial vehicle. Background The steel web and concrete combined beam is a design method commonly adopted in bridges, and has the advantages of high construction speed, reduced concrete consumption, reduced self weight of the beam slab, different expansion rates of steel and concrete, easiness in generating expansion and contraction cracks at the joint of the steel web and the concrete under the weather condition of large high-temperature and low-temperature change, easiness in entering rainwater into a beam body to damage the beam body structure, and reduced service life of the bridge. The simplest and most common crack observation method at present is visual inspection, and for larger bridges, personnel cannot climb up each part of the beam body to observe. Other observation methods include contact type monitoring such as a crack needle, vibration wire type or electronic crack meters arranged on two sides of the crack to monitor the width change in real time, pre-buried sensors which are pre-buried fiber bragg grating crack sensors on the joint surface in the construction stage and track the crack development for a long time, non-contact type monitoring image recognition technology and three-dimensional laser scanning technology, and ultrasonic method and infrared camera are generally adopted for the crack depth and internal defects. According to the monitoring method, monitoring personnel are required to carry instruments and equipment to be close to the monitored part, and the monitoring personnel cannot be close to the connection position of the steel web and the concrete for the finished steel web-concrete combined bridge to observe cracks. Disclosure of Invention In order to make up for the defects, the invention provides a steel-concrete combined bridge crack monitoring system and method based on an unmanned aerial vehicle, wherein a multi-rotor unmanned aerial vehicle is provided with an optical high-definition camera pair, a laser radar and a thermal infrared imager, and the millimeter-level non-contact monitoring of the cracks of the joint surface of the steel web and the concrete is realized by combining a ground multi-base station RTK-GPS differential positioning technology. The technical scheme is that the steel-concrete combined bridge crack monitoring system based on the unmanned aerial vehicle comprises a multi-rotor unmanned aerial vehicle, a ground multi-base-station RTK-GPS system, a stereoscopic vision acquisition module, a laser radar scanning module and a thermal infrared imager, wherein the multi-rotor unmanned aerial vehicle is used for carrying monitoring equipment and flying to a region to be detected of the steel-concrete combined beam, the ground multi-base-station RTK-GPS system is composed of a plurality of ground base stations, virtual base stations are generated through carrier phase difference to achieve centimeter-level positioning of the unmanned aerial vehicle, the stereoscopic vision acquisition module is used for acquiring high-definition images of the surface of the steel-concrete combined beam to achieve crack identification, the laser radar scanning module and the stereoscopic vision acquisition module are used for cooperatively working to generate point cloud data of the structure of the steel-concrete combined beam to ensure that point cloud of crack edge points is complete, and the thermal infrared imager is used for detecting a heat conduction abnormal region caused by cracks inside the steel-concrete combined beam to identify hidden cracks. As a further description of the technical scheme, in the ground multi-base-station RTK-GPS system, a base station adopts a weighted least square method to fuse multi-base-station differential data to generate a virtual base station coordinate, and a calculation formula is as followsIn the followingFor the base station coordinates,Is the unmanned aerial vehicle to base station distance. The stereoscopic vision acquisition module comprises a high-definition optical camera pair, wherein the camera distortion is less than 1%, the camera distortion is fixed on a holder at an included angle of 15 degrees, the base line distance is 15cm, and the stereoscopic vision acquisition module further comprises an image preprocessing unit, performs distortion correction based on a Zhang calibration method and performs gray level equalization processing. As a further description of the technical scheme, the laser radar scanning module is fused with the data acquired by the stereoscopic vision acquisition module to construct a textured three-dimensional model for intuitively displaying the structure and crack distribution of the steel-concrete composite beam. As