A Smart Inspection Method for Precast Concrete Components Based on 3D Laser Scanning and Mold Stage Linkage

Abstract

This invention discloses an intelligent inspection method for precast concrete components based on three-dimensional laser scanning and mold platform linkage, comprising the following steps: S1, the mold platform carrying the precast component is transferred to the scanning station; S2, the three-dimensional laser scanning unit scans the entire mold platform to obtain point cloud data of the precast component; S3, the point cloud data of the entire mold platform is preprocessed, and the preprocessed precast component is accurately identified and separated into point cloud data files of each independent component; S4, the point cloud data of each independent component is automatically registered with the standard BIM design model of the component, key dimensions and geometric tolerances are calculated, and compared with preset tolerance thresholds to determine the component quality level; S5, the mold platform identification information, inspection results, and quality level are uploaded to the MES system. This invention has the advantages of reducing manual intervention, improving inspection efficiency, and realizing automated, intelligent, and information-based quality inspection of precast components.

Description

Technical Field

[0001] This invention relates to the field of precast concrete component testing technology, specifically to an intelligent testing method for precast concrete components based on three-dimensional laser scanning and mold linkage. Background Technology

[0002] Against the backdrop of rapid development in building industrialization and prefabricated construction, the production scale of precast concrete components is expanding daily, and the quality requirements are becoming increasingly stringent. Traditional quality inspection methods mainly rely on manual sampling using tools such as calipers, straightedges, and feeler gauges. This approach suffers from low efficiency, strong subjectivity, easy omissions, and difficulty in data traceability, becoming a bottleneck restricting the high-quality and efficient production of precast components. In recent years, 3D laser scanning technology has begun to be explored for quality inspection in the construction field. However, existing applications are mostly limited to scanning individual demolded components or fixed scenes, failing to deeply integrate with modern precast component production lines; and suffer from problems such as excessive manual intervention, low inspection efficiency, and inconvenient production management. Summary of the Invention

[0003] The technical problem to be solved by this invention is to overcome the existing defects and provide an intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage. By linking three-dimensional laser scanning with the mold table, the entire process from scanning triggering, data acquisition to result uploading is automated, which greatly reduces manual intervention, improves detection efficiency, and effectively avoids missed detections. It can scan multiple precast components of different specifications on the transfer mold table, improving scanning detection efficiency. It can effectively solve the problems in the background technology.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent inspection method for precast concrete components based on three-dimensional laser scanning and mold table linkage, comprising an intelligent inspection system, the intelligent inspection system including a transfer mold table, a mold table identification unit, a three-dimensional laser scanning unit, and an industrial control computer, the industrial control computer including a point cloud processing module, an intelligent identification and segmentation module, a quality inspection and analysis module, and a data communication module, the intelligent inspection method for precast components comprising the following steps: S1. When the transfer mold table carrying the prefabricated components moves to a preset distance from the center point of the three-dimensional laser scanning unit, the industrial control computer obtains the mold table identity information through the mold table identity recognition unit and triggers the three-dimensional laser scanning unit to work. S2, the three-dimensional laser scanning unit performs segmented scanning of the entire transfer mold platform, obtains panoramic cloud data containing all prefabricated components on the transfer mold platform, and transmits it to the industrial control computer; S3: The industrial control computer uses the point cloud processing module to denoise, filter, and stitch the point cloud data of the entire site, reconstruct the actual three-dimensional model of the prefabricated components, and accurately identify the boundaries of each independent component through the intelligent recognition and splitting module, and generate point cloud data files of each independent component. S4, the quality inspection and analysis module automatically registers the point cloud data of each independent component with its corresponding standard BIM design model, and calculates the deviation value of the component through comparative analysis, generating a visual inspection report. S5. The industrial control computer automatically uploads the mold identification information, component information, component inspection report, and timestamp to the MES system through the data communication module via a standard interface. S6. The MES system controls the flow direction of the transfer mold table based on the received quality results.

[0005] Furthermore, the intelligent recognition and segmentation module mentioned in step S3 is a pre-trained PointRCNN or a pre-trained 3D instance segmentation model.

[0006] Furthermore, in step S, the automatic registration is performed by the quality inspection and analysis module using the iterative nearest point algorithm to call the standard BIM design model corresponding to the component to be inspected in the MES system for automatic registration. After registration, the industrial control computer compares and analyzes to calculate the key dimensional deviations and geometric tolerances of the component, and compares them with the preset tolerance thresholds to automatically determine the component's quality level. At the same time, defects such as honeycomb and pitting on the component's surface can be detected by the change in point cloud density.

[0007] Furthermore, the three-dimensional laser scanning unit includes a scanning frame set at the transfer mold table, a scanning robotic arm on the scanning frame, the scanning robotic arm having six rotational degrees of freedom, a laser scanning head on the scanning robotic arm, and a tracker on one side of the transfer mold table; the segmented scanning process in step S2 is as follows: the transfer mold table moves forward a certain distance along the transfer direction and then stops moving, the scanning robotic arm drives the camera to perform an S-shaped running trajectory to scan this segment; after completing the scanning of this segment, the transfer mold table continues to move forward a certain distance along the transfer direction and then stops moving, the scanning robotic arm drives the camera to scan the second segment, and so on to scan multiple segments until the entire transfer mold table is scanned.

[0008] Furthermore, the laser scanning head includes a camera fixed to the drive end of the scanning robotic arm, and the camera is surrounded by multiple reflective markers. The multiple reflective markers are connected by a connecting rod to form a hollowed-out multi-faceted spherical structure.

[0009] Furthermore, the mold identification unit includes an ultra-high frequency RFID reader / writer disposed on one side of the transfer mold platform.

[0010] Compared with existing technologies, the beneficial effects of this invention are as follows: This intelligent inspection method for precast concrete components based on three-dimensional laser scanning and mold table linkage, by linking the three-dimensional laser scanning unit with the transfer mold table, realizes the real-time uploading of inspection results to the MES system from scanning triggering and data acquisition to inspection results. This achieves automated, intelligent, and information-based quality inspection of precast components and closed-loop management of precast component quality data, significantly reducing manual intervention, improving inspection efficiency, and effectively avoiding missed inspections; it also improves production efficiency and quality management level, breaks down information silos, and provides accurate data support for production management, quality traceability, and capacity analysis; by using the three-dimensional laser scanning unit to perform segmented scanning of the transfer mold table, all precast components on the transfer mold table can be obtained. The system collects panoramic point cloud data, which is then processed by a point cloud processing module to remove noise, filter, and stitch the data. This allows for scanning of multiple prefabricated components of different specifications on the transfer platform, improving scanning and inspection efficiency. The intelligent recognition and segmentation module accurately identifies the boundaries of each independent component on the transfer platform and generates point cloud data files for each component for inspection and analysis, further improving inspection efficiency. A scanning robotic arm with six degrees of rotational freedom drives a laser scanning head to perform omnidirectional scanning of the prefabricated components. Multiple marker reflective points around the camera form a hollowed-out multifaceted spherical structure, enabling precise 3D scanning of the prefabricated components and comprehensively collecting 3D point cloud data, further improving scanning and inspection accuracy. Attached Figure Description

[0011] Figure 1 This is a flowchart of the workflow of the present invention; Figure 2 This is a schematic diagram of the intelligent detection system of the present invention; Figure 3 This is a schematic diagram of the three-dimensional laser scanning unit structure of the present invention; Figure 4 This is a partial enlarged view of the three-dimensional laser scanning unit of the present invention; Figure 5 This is a schematic diagram of the industrial control computer structure of the present invention.

[0012] In the diagram: 1. Transfer mold table; 2. 3D laser scanning unit; 21. Scanning frame; 22. Scanning robotic arm; 23. Laser scanning head; 231. Camera; 232. Marking reflective point; 233. Link; 24. Tracker; 3. Mold table identification unit; 4. Industrial control computer; 41. Point cloud processing module; 42. Intelligent identification and segmentation module; 43. Quality inspection and analysis module; 44. Data communication module. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] Example Please see Figures 1-5 This invention provides a technical solution: an intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage. The system includes an intelligent inspection system, which comprises a transfer mold platform 1, a mold platform identification unit 3, a three-dimensional laser scanning unit 2, and an industrial control computer 4. The industrial control computer 4 includes a point cloud processing module 41, an intelligent identification and segmentation module 42, a quality inspection and analysis module 43, and a data communication module 44. The three-dimensional laser scanning unit 2 includes a scanning frame 21 set on the transfer mold table 1. The scanning frame 21 is a gantry structure. A scanning robotic arm 22 is set at the center of the crossbeam of the scanning frame 21. The scanning robotic arm 22 has six rotational degrees of freedom. A laser scanning head 23 is set on the scanning robotic arm 22. A tracker 24 is set on one side of the transfer mold table 1. The laser scanning head 23 includes a camera 231 fixed on the driving end of the scanning robotic arm 22. Multiple marker reflective points 232 are set around the camera 231. The multiple marker reflective points 232 are connected by connecting rods 233 to form a hollow multi-faceted spherical structure. In this embodiment, the size of the transfer mold table 1 is 5m × 9m (width × length). The maximum active diameter of the scanning robotic arm 22 driving the laser scanning head 23 is 5m. The intelligent detection method for precast components includes the following steps: S1. Mold identification and scanning trigger: When the transfer mold 1 carrying the prefabricated components moves to a distance of 1.2m from the center point of the three-dimensional laser scanning unit 2 (the center point is the center point of the scanning frame 21), the transfer mold 1 stops rotating. The industrial control computer 4 obtains the mold identity information through the mold identification unit 3 and triggers the three-dimensional laser scanning unit 2 to work. S2. Full-view cloud data acquisition of the formwork platform: The three-dimensional laser scanning unit 2 scans the entire transfer formwork platform 1 to obtain full-view cloud data containing all prefabricated components on the transfer formwork platform 1, and transmits it to the industrial control computer 4. The specific scanning process is as follows: the transfer formwork platform 1 moves forward 1m along the transfer direction and then stops. The scanning robot arm 22 drives the camera 231 to perform an S-shaped running trajectory to scan this segment. After completing the scanning of this segment, the transfer formwork platform 1 continues to move forward 1m along the transfer direction and then stops. The scanning robot arm 22 drives the camera 231 to scan the second segment. This process is repeated to scan multiple segments until the entire transfer formwork platform 1 is completely scanned.

[0015] S3. Intelligent recognition and point cloud splitting of multiple components: The industrial control computer 4 performs noise reduction, filtering and splicing processing on the point cloud data of the entire site through the point cloud processing module 41 to reconstruct the actual three-dimensional model of the prefabricated components; the intelligent recognition and splitting module 42 analyzes the pre-processed point cloud data of the prefabricated components, accurately identifies the boundaries of each independent component, and generates point cloud data files of each independent component. S4. High-precision quality inspection of single components: The quality inspection and analysis module 43 uses the iterative nearest point algorithm to call the standard BIM design model corresponding to the component to be inspected in the MES system for automatic registration; after registration, the calculation software of the industrial control computer 4 compares and analyzes to calculate the key dimensional deviations (such as length, width, and diagonal difference) and geometric tolerances (such as flatness and warping) of the component, and compares them with the preset tolerance thresholds. At the same time, it can detect defects such as honeycomb and pitting on the surface of the component through changes in point cloud density, automatically determine the quality level of the component, and generate a visual inspection report. S5 and industrial control computer 4 automatically package and upload the mold identity information, component information, quality level, timestamp, inspection report and corresponding point cloud data snapshot to the MES system through standard interfaces such as OPC UA or Restful API via data communication module 44. S6. The quality of the components can be seen intuitively through the MES system, and the MES system controls the flow direction of the transfer mold 1 based on the received quality results. For example, qualified products flow into the curing kiln, and unqualified products flow into the rework area.

[0016] In this embodiment, the mold platform identification unit 3 is an ultra-high frequency RFID reader / writer set on one side of the transfer mold platform 1; the intelligent identification and segmentation module 42 is a pre-trained PointRCNN or a pre-trained 3D instance segmentation model. This invention discloses an intelligent inspection method for precast concrete components based on 3D laser scanning and mold table linkage. By linking the 3D laser scanning unit 2 with the transfer mold table 1, it realizes real-time uploading of inspection results to the MES system from scanning triggering and data acquisition to inspection results. This achieves automated, intelligent, and information-based quality inspection of precast components and closed-loop management of precast component quality data, significantly reducing manual intervention, improving inspection efficiency, and effectively avoiding missed inspections. It enhances production efficiency and quality management, breaks down information silos, and provides accurate data support for production management, quality traceability, and capacity analysis. By using the 3D laser scanning unit 2 to perform segmented scanning of the transfer mold table 1, it can obtain full-view cloud data of all precast components on the transfer mold table 1, and then... The point cloud processing module 41 performs noise reduction, filtering, and stitching processing on the point cloud data of the entire site, enabling scanning of multiple prefabricated components of different specifications on the transfer mold platform 1, thereby improving scanning and detection efficiency. The intelligent recognition and splitting module 42 can accurately identify the boundaries of each independent component on the transfer mold platform 1 and generate point cloud data files of each independent component for detection and analysis, further improving detection efficiency. The scanning robotic arm 22 with six degrees of rotational freedom drives the laser scanning head 23 to perform omnidirectional scanning of the prefabricated components. Multiple marker reflective points 232 around the camera 231 form a hollowed-out multi-faceted spherical structure, realizing precise three-dimensional scanning of the prefabricated components and comprehensively collecting the three-dimensional point cloud data of the prefabricated components, further improving scanning and detection accuracy.

[0017] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for intelligent inspection of precast concrete components based on three-dimensional laser scanning and mold table linkage, comprising an intelligent inspection system, the intelligent inspection system including a transfer mold table, a mold table identification unit, a three-dimensional laser scanning unit, and an industrial control computer, the industrial control computer including a point cloud processing module, an intelligent identification and segmentation module, a quality inspection and analysis module, and a data communication module, characterized in that, The intelligent detection method for precast components includes the following steps: S1. When the transfer mold table carrying the prefabricated components moves to a preset distance from the center point of the three-dimensional laser scanning unit, the industrial control computer obtains the mold table identity information through the mold table identity recognition unit and triggers the three-dimensional laser scanning unit to work. S2, the three-dimensional laser scanning unit performs segmented scanning of the entire transfer mold platform, obtains panoramic cloud data containing all prefabricated components on the transfer mold platform, and transmits it to the industrial control computer; S3. The industrial control computer uses the point cloud processing module to denoise, filter, and stitch the point cloud data of the entire site, reconstruct the actual three-dimensional model of the prefabricated components, and accurately identify the boundaries of each independent component through the intelligent recognition and splitting module, and generate point cloud data files of each independent component. S4, the quality inspection and analysis module automatically registers the point cloud data of each independent component with its corresponding standard BIM design model, and calculates the deviation value of the component through comparative analysis, generating a visual inspection report. S5. The industrial control computer automatically uploads the mold identification information, component information, component inspection report, and timestamp to the MES system through the data communication module via a standard interface. S6. The MES system controls the flow direction of the transfer mold table based on the received quality results.

2. The intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage according to claim 1, characterized in that: The intelligent recognition and segmentation module mentioned in step S3 is a pre-trained PointRCNN or a pre-trained 3D instance segmentation model.

3. The intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage according to claim 1, characterized in that: The automatic registration in step S4 is achieved by the quality inspection and analysis module using the iterative nearest point algorithm to call the standard BIM design model corresponding to the component to be inspected in the MES system for automatic registration; the industrial control computer compares and analyzes to calculate the key dimensional deviations and geometric tolerances of the component, and compares them with the preset tolerance thresholds to automatically determine the quality level of the component; at the same time, defects such as honeycomb and pitting on the surface of the component can be detected by the change of point cloud density.

4. The intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage according to claim 1, characterized in that: The three-dimensional laser scanning unit includes a scanning frame set at the transfer mold table, a scanning robotic arm on the scanning frame, a laser scanning head on the scanning robotic arm, and a tracker on one side of the transfer mold table. The segmented scanning process in step S2 is as follows: the transfer mold table moves forward a certain distance along the transfer direction and then stops moving. The scanning robotic arm drives the camera to perform an S-shaped running trajectory to scan this segment. After completing the scanning of this segment, the transfer mold table continues to move forward a certain distance along the transfer direction and then stops moving. The scanning robotic arm drives the camera to scan the second segment. This process is repeated to scan multiple segments until the entire transfer mold table is scanned.

5. The intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage according to claim 4, characterized in that: The laser scanning head includes a camera fixed to the drive end of the scanning robotic arm. The camera has multiple reflective markers around its periphery, and the multiple reflective markers are connected by a connecting rod to form a hollowed-out multi-faceted spherical structure.

6. The intelligent detection method for precast concrete components based on three-dimensional laser scanning and mold table linkage according to claim 1, characterized in that: The mold identification unit includes an ultra-high frequency RFID reader / writer located on one side of the transfer mold.

Images