Intelligent quality inspection system and method for precast beam field, electronic equipment and storage medium

Abstract

The invention provides an intelligent quality inspection system and method for a precast beam field, electronic equipment and a storage medium. The intelligent quality inspection system comprises a conveying and positioning subsystem, a detection subsystem and a control subsystem, wherein the conveying and positioning subsystem is used for conveying a to-be-detected prefabricated part to a detection station; the central processing unit is used for sending a detection task instruction, performing multi-dimensional comprehensive analysis processing based on an appearance quality identification result, a geometric dimension identification result and an internal defect identification result, and determining a comprehensive quality evaluation result; the appearance quality detection subsystem is used for carrying out appearance defect detection processing on the surface image information based on an artificial intelligence visual identification algorithm and determining an appearance quality identification result; the geometric dimension detection subsystem is used for performing dimension analysis on the actual geometric parameters to determine a geometric dimension identification result; and the internal defect detection subsystem is used for processing the internal structure information based on an internal defect analysis algorithm and determining an internal defect identification result. The quality inspection efficiency is greatly improved, and the efficiency bottleneck problem of traditional manual detection is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of precast beam yard detection, in particular to an intelligent quality inspection system and method for a precast beam yard, an electronic device and a storage medium. BACKGROUND

[0002] At present, the quality inspection in the precast beam yard generally adopts a traditional quality inspection mode of dispersion, semi-automation and manual combination. This mode usually distributes different projects such as geometric size measurement, appearance quality inspection and internal defect detection at different stations of the production line, and the operating personnel uses handheld devices or visual methods for sampling detection. Specifically, in the geometric size detection link, the detection personnel adopts an electronic ruler or a total station instrument to manually measure the key section of the component, and records in a paper form or a handheld terminal, which is difficult to realize 100% full-size rechecking; in the appearance detection aspect, the workers rely on naked eyes to observe whether there are defects such as honeycomb, pitted surface, exposed reinforcement, color difference and the like on the surface of the component, and supplement with simple tools for judgment, which has strong subjectivity, non-uniform standards and high omission rate; in the internal structure detection aspect, it depends on professional non-destructive testing personnel to use rebound hammer, reinforcement scanner, ultrasonic detector or ground penetrating radar and other independent devices for local sampling inspection, and the detection data is stored in isolation and cannot be automatically associated with the component identity information. All the above detection results finally need to be manually summarized, compared and comprehensively judged by quality engineers to form the final quality inspection report. However, the detection by manual means will lead to the technical problem of inaccurate quality detection of the precast beam yard. Therefore, how to build a precast component quality evaluation system integrated with multiple detection means, realizing full-process automation operation and having intelligent analysis capability has become an urgent technical demand in the current intelligent construction field. SUMMARY

[0003] Therefore, the purpose of the present application is to provide an intelligent quality inspection system and method for a precast beam yard, an electronic device and a storage medium, through integrated equipment and automatic process, the component can complete all quality inspection projects without manual intervention after entering the intelligent quality inspection system, the detection speed matches the production line rhythm, greatly improving the quality inspection efficiency and solving the efficiency bottleneck problem of traditional manual detection.

[0004] The present application provides an intelligent quality inspection system for a precast beam yard, which comprises a conveying and positioning subsystem, a central processor, an appearance quality detection subsystem, a geometric size detection subsystem and an internal defect detection subsystem, the appearance quality detection subsystem, the geometric size detection subsystem and the internal defect detection subsystem are in communication connection with the central processor; wherein, The conveying and positioning subsystem is used for conveying the precast component to be detected to the detection station. The central processing unit is used to send inspection task instructions to each subsystem after detecting that the prefabricated component to be inspected is located on the inspection station. Based on the appearance quality identification results, geometric dimension identification results and internal defect identification results sent by multiple subsystems, it performs multi-dimensional comprehensive analysis and processing to determine the comprehensive quality evaluation result of the prefabricated component to be inspected. The appearance quality inspection subsystem is used to collect surface image information of the prefabricated component to be inspected after receiving the inspection task instruction, and to perform appearance defect detection processing on the surface image information based on artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the prefabricated component to be inspected. The geometric dimension detection subsystem is used to collect the actual geometric parameters of the prefabricated component to be detected after receiving the detection task instruction, and to perform dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the prefabricated component to be detected; The internal defect detection subsystem is used to obtain the internal structural information of the prefabricated component to be inspected after receiving the inspection task instruction, process the internal structural information based on the internal defect analysis algorithm, and determine the internal defect identification result of the prefabricated component to be inspected.

[0005] In one possible implementation, the appearance quality inspection subsystem, when collecting surface image information of the prefabricated component to be inspected and performing appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the prefabricated component to be inspected, is specifically used for: Under controlled lighting conditions, one or more industrial cameras are used to acquire high-definition images of the surface of the prefabricated component to be inspected from multiple angles to obtain surface image information. The surface image information is subjected to denoising, contrast enhancement, and geometric distortion correction to determine the preprocessed image; The preprocessed image is input into the artificial intelligence visual recognition algorithm to automatically identify and locate at least one type of appearance defect on the surface of the component, and generate appearance quality recognition results including the location, type, size and quantity of the defect.

[0006] In one possible implementation, when the geometric dimension detection subsystem is used to collect the actual geometric parameters of the prefabricated component to be inspected and to perform dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the prefabricated component to be inspected, the geometric dimension detection subsystem is specifically used for: Data is collected from the surface of the precast component to be tested from multiple angles using one or more non-contact measuring devices to obtain three-dimensional point cloud data of the precast component to be tested; Based on the three-dimensional point cloud data, the key geometric features of the precast component to be inspected are extracted, and the actual geometric parameters of the precast component to be inspected are determined based on the key geometric features; wherein, the actual geometric parameters include actual length, width, height, dimensions of each section and pre-camber parameters; The actual geometric parameters are automatically compared with the theoretical dimensions to identify deviation values ​​and determine whether they exceed the allowable tolerance range, thereby generating geometric dimension identification results.

[0007] In one possible implementation, when the internal defect detection subsystem is used to acquire the internal structural information of the precast component to be inspected, process the internal structural information based on an internal defect analysis algorithm, and determine the internal defect identification result of the precast component, the internal defect detection subsystem is specifically used for: The scanning device performs a full-area coverage scan of the key areas of the precast component to be inspected, and collects the physical response signals inside the precast component in real time; wherein, the scanning device includes at least one of an ultrasonic imaging array, ground penetrating radar, shock echo meter or infrared thermal imager. The physical response signal is input into a pre-trained internal defect analysis model. The physical response signal is processed by signal feature parsing. The parsed signal features are then processed to identify internal voids, cracks, interlayers, or rebar position offsets to determine the internal defect identification result. The internal defect identification result includes the defect location, size, and severity level.

[0008] In one possible implementation, the central processing unit determines the overall quality evaluation result of the prefabricated component to be tested through the following steps: Based on a pre-set comprehensive evaluation rule base, the appearance quality identification results, the geometric dimension identification results, and the internal defect identification results are weighted and fused to determine the comprehensive quality score. A comprehensive quality evaluation result is generated based on the comprehensive quality score and the scoring threshold. The received identification results of each dimension and the comprehensive quality evaluation result are then associated with the identity identifier of the precast component being inspected, thereby constructing a digital quality file for the precast component being inspected.

[0009] In one possible implementation, the intelligent quality inspection system further includes a rebound strength detection subsystem and a large-screen control subsystem, both of which are communicatively connected to the central processing unit; wherein... The rebound strength detection subsystem is used to collect a set of rebound value sequences under multiple impact tests of the precast component to be tested, perform validity screening on the rebound value sequence, remove abnormal data, calculate the average rebound value, determine the strength estimate based on the average rebound value, compare the strength estimate with the preset strength standard to determine whether it meets the specification requirements, and generate the rebound strength detection result. The large screen control subsystem is used to input information about the prefabricated component to be inspected before starting the inspection, view the inspection process, and view the recognition results.

[0010] In one possible implementation, the central processing unit is further configured to: Receive a request from the operation terminal in the beam storage area to inspect the precast component to be inspected at the gate, and retrieve the digital quality file according to the identification of the precast component to be inspected; The digital quality archive is verified for completeness and compliance based on a pre-defined set of exit verification rules. If the verification passes, an electronic signature prompt is sent to the operation terminal. Upon receiving the electronic signature instruction, a unique electronic certificate for leaving the factory is generated, and the factory status of the prefabricated component to be inspected is updated synchronously.

[0011] This application also provides an intelligent quality inspection method for precast beam yards, the intelligent quality inspection method comprising: After detecting that the prefabricated component to be inspected is located at the inspection station, the surface image information of the prefabricated component to be inspected is collected, and the appearance defect detection processing of the surface image information is performed based on the artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the prefabricated component to be inspected. The actual geometric parameters of the prefabricated component to be inspected are collected, and the geometric dimension identification result of the prefabricated component to be inspected is determined by dimensional analysis of the actual geometric parameters. The internal structural information of the prefabricated component to be inspected is obtained, and the internal structural information is processed based on an internal defect analysis algorithm to determine the internal defect identification result of the prefabricated component to be inspected. The appearance quality identification results, geometric dimension identification results, and internal defect identification results are analyzed and processed in a multi-dimensional manner to determine the comprehensive quality evaluation result of the prefabricated component to be inspected.

[0012] This application also provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, the steps of the intelligent quality inspection method for precast beam yards described above are performed.

[0013] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, performs the steps of the intelligent quality inspection method for precast beam yards as described above.

[0014] The intelligent quality inspection system, method, electronic device, and storage medium for precast beam yards provided in this application embodiment include a conveying and positioning subsystem, a central processing unit, an appearance quality inspection subsystem, a geometric dimension inspection subsystem, and an internal defect detection subsystem. The appearance quality inspection subsystem, the geometric dimension inspection subsystem, and the internal defect detection subsystem are all communicatively connected to the central processing unit. The conveying and positioning subsystem is used to convey the precast component to be inspected to the inspection station. The central processing unit, after detecting that the precast component to be inspected is located at the inspection station, sends inspection task instructions to each subsystem. Based on the appearance quality identification results, geometric dimension identification results, and internal defect identification results sent by multiple subsystems, it performs multi-dimensional comprehensive analysis and processing to determine the precast component to be inspected. The system provides a comprehensive quality evaluation result. The appearance quality inspection subsystem, upon receiving the inspection task instruction, collects surface image information of the precast component to be inspected and performs appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the precast component to be inspected. The geometric dimension inspection subsystem, upon receiving the inspection task instruction, collects the actual geometric parameters of the precast component to be inspected and performs dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the precast component to be inspected. The internal defect detection subsystem, upon receiving the inspection task instruction, acquires the internal structural information of the precast component to be inspected, processes the internal structural information based on an internal defect analysis algorithm, and determines the internal defect identification result of the precast component to be inspected. Through integrated equipment and automated processes, all quality inspection items can be completed without manual intervention after the component enters the intelligent quality inspection system. The inspection speed matches the production line rhythm, greatly improving quality inspection efficiency and solving the efficiency bottleneck problem of traditional manual inspection.

[0015] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 One of the structural schematic diagrams of an intelligent quality inspection system for a precast beam yard provided in an embodiment of this application; Figure 2 A second structural schematic diagram of an intelligent quality inspection system for a precast beam yard provided in an embodiment of this application; Figure 3 A flowchart illustrating an intelligent quality inspection method for a precast beam yard provided in an embodiment of this application; Figure 4 This is a schematic diagram of the intelligent quality inspection method for the precast beam yard after tensioning, provided in an embodiment of this application. Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0018] Icons: 100 - Intelligent quality inspection system for precast beam yard; 110 - Conveying and positioning subsystem; 120 - Central processing unit; 130 - Appearance quality inspection subsystem; 140 - Geometric dimension inspection subsystem; 150 - Internal defect inspection subsystem; 160 - Rebound strength inspection subsystem; 170 - Large screen control subsystem; 500 - Electronic equipment; 510 - Processor; 520 - Memory; 530 - Bus. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. Based on the embodiments of this application, every other embodiment obtained by those skilled in the art without inventive effort falls within the scope of protection of this application.

[0020] First, the applicable application scenarios of this application will be introduced. This application can be applied to the field of precast beam yard testing technology.

[0021] Research has revealed that quality inspection in precast beam yards generally employs a traditional, decentralized, semi-automated, and manual inspection model. This model typically distributes different items, such as geometric dimension measurement, visual quality inspection, and internal defect detection, across different workstations on the production line, with operators using handheld devices or visual methods for sampling inspection. Specifically, in the geometric dimension inspection stage, inspectors often use electronic rulers or total stations to manually measure key sections of components and record the data in paper forms or handheld terminals, making 100% full-dimensional verification difficult. For visual inspection, workers rely on visual observation of the component surface for defects such as honeycomb, pitting, exposed reinforcement, and color differences, supplemented by simple tools. This approach is highly subjective, lacks standardized criteria, and has a high rate of missed inspections. For internal structure inspection, professional non-destructive testing personnel use independent equipment such as rebound hammers, rebar scanners, ultrasonic detectors, or ground-penetrating radar for localized sampling. The inspection data is stored in isolation and cannot be automatically linked to component identification information. All the above inspection results ultimately require manual summarization, comparison, and comprehensive evaluation by quality engineers to form the final quality inspection report. However, relying on manual inspection methods can lead to inaccurate quality checks at precast beam yards. Therefore, constructing a precast component quality evaluation system that integrates multiple inspection methods, achieves fully automated operation, and possesses intelligent analysis capabilities has become an urgent technical need in the field of intelligent construction.

[0022] Based on this, this application provides an intelligent quality inspection system for precast beam yards. Through integrated equipment and automated processes, all quality inspection items can be completed without manual intervention after the components enter the intelligent quality inspection system. The inspection speed matches the production line rhythm, which greatly improves the quality inspection efficiency and solves the efficiency bottleneck problem of traditional manual inspection.

[0023] Please see Figure 1 , Figure 1 This is one of the structural schematic diagrams of an intelligent quality inspection system 100 for a precast beam yard provided in an embodiment of this application. Figure 1 As shown in the figure, the intelligent quality inspection system 100 for precast beam yards provided in this application embodiment includes a conveying and positioning subsystem 110, a central processing unit 120, an appearance quality inspection subsystem 130, a geometric dimension inspection subsystem 140, and an internal defect inspection subsystem 150. The appearance quality inspection subsystem 130, the geometric dimension inspection subsystem 140, and the internal defect inspection subsystem 150 are all communicatively connected to the central processing unit 120.

[0024] Specifically, the transmission and positioning subsystem 110 is used to transmit the prefabricated component to be inspected to the inspection station; the central processing unit 120 is used to send inspection task instructions to each subsystem after detecting that the prefabricated component to be inspected is located at the inspection station, and to perform multi-dimensional comprehensive analysis and processing based on the appearance quality recognition results, geometric dimension recognition results, and internal defect recognition results sent by multiple subsystems to determine the comprehensive quality evaluation result of the prefabricated component to be inspected; the appearance quality inspection subsystem 130 is used to collect surface image information of the prefabricated component to be inspected after receiving the inspection task instructions, and to perform analysis and processing based on artificial intelligence visual recognition algorithms. The surface image information is processed for appearance defect detection to determine the appearance quality identification result of the precast component to be inspected; the geometric dimension detection subsystem 140 is used to collect the actual geometric parameters of the precast component to be inspected after receiving the inspection task instruction, and perform dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the precast component to be inspected; the internal defect detection subsystem 150 is used to acquire the internal structure information of the precast component to be inspected after receiving the inspection task instruction, process the internal structure information based on the internal defect analysis algorithm, and determine the internal defect identification result of the precast component to be inspected.

[0025] It should be noted that the inspection content of the intelligent quality inspection system 100 for precast beam yards in this application is not limited to internal defect inspection, appearance quality inspection, and geometric dimension inspection. The inspection content can be added and selected according to the user's implementation needs.

[0026] Here, the conveying and positioning subsystem 110 includes a track, an electric trolley, and a hydraulic lifting and locking mechanism (to achieve precise positioning and parking). The prefabricated component to be inspected is placed on the trolley and transported to the inspection station by the track.

[0027] It should be noted that the central processing unit 120 receives and synchronizes raw data (point cloud, images, radar wave signals, etc.) from various subsystems. It stores all raw data, processing results, and component identification information. Based on the comprehensive inspection results from multiple dimensions, it intelligently evaluates the component quality according to preset standards and automatically generates a quality inspection report. The system control is responsible for coordinating the work of each subsystem and communicating with the beam yard system to upload inspection data and results in real time.

[0028] The appearance quality inspection subsystem 130 can be configured, depending on budget and equipment capabilities, to either fix the inspection equipment in the workshop or mount it on a mobile platform such as a robotic arm to achieve full coverage inspection of the appearance quality of precast components. Its core components include: a camera to acquire high-definition images of the component surface from different angles and lighting conditions; and a controllable lighting unit to provide a uniform and stable light source, eliminating ambient light interference and ensuring consistent image quality. This subsystem inputs the acquired images into a pre-trained AI visual recognition algorithm model, automatically identifying and framing defects such as color difference, bubbles, honeycomb, pitting, and exposed reinforcement, and then classifying and statistically determining the appearance quality identification results of the precast component to be inspected.

[0029] Here, the geometric dimension inspection subsystem 140 can be installed in a fixed location in the workshop or mounted on a mobile platform such as a robotic arm, depending on the budget and equipment capabilities, to achieve full coverage inspection of the geometric dimensions of precast components. Equipment selection includes, but is not limited to, 3D laser scanners, high-precision distance sensors, and industrial cameras with image processing algorithms. This subsystem compares measured data with the component's design dimensions, automatically calculates the compliance and deviation of geometric parameters such as cross-sectional dimensions, length, width, height, thickness, and pre-camber, and determines the geometric dimension identification result of the precast component to be inspected.

[0030] The internal defect detection subsystem 150 can integrate one or more of the following devices as needed: an ultrasonic imaging array for detecting internal voids, cracks, and delaminations in components; and ground-penetrating radar for non-destructive testing of rebar location, protective layer thickness, and voids beneath the rebar cage. The equipment can be mounted on a robotic arm to achieve automatic path scanning. The collected acoustic or electromagnetic signals are processed and evaluated by a dedicated system to generate internal structural images, thereby identifying and evaluating internal defects such as voids, cracks, and abnormal rebar placement, and determining the internal defect identification results of the precast components.

[0031] In one possible implementation, the appearance quality inspection subsystem 130, when collecting surface image information of the prefabricated component to be inspected and performing appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the prefabricated component to be inspected, is specifically used for: A: Under controlled lighting conditions, one or more industrial cameras are used to acquire high-definition images of the surface of the prefabricated component to be inspected from multiple angles to obtain surface image information.

[0032] Here, under controlled lighting conditions, one or more industrial cameras are used to acquire high-definition images of the surface of the prefabricated component to be inspected from multiple angles, thereby obtaining surface image information.

[0033] B: The surface image information is subjected to denoising, contrast enhancement, and geometric distortion correction to determine the preprocessed image.

[0034] Here, the surface image information is subjected to noise reduction, contrast enhancement, and geometric distortion correction to determine the preprocessed image.

[0035] C: Input the preprocessed image into the artificial intelligence visual recognition algorithm to automatically identify and locate at least one type of appearance defect on the surface of the component, and generate appearance quality recognition results including the location, type, size and quantity of the defect.

[0036] Here, during the inference process, the AI ​​visual recognition algorithm performs pixel-by-pixel analysis and feature extraction on the input image. Through a convolutional neural network, it automatically learns the texture, shape, edge, grayscale distribution, and spatial context features of various defects. Combined with defect pattern information from a prior knowledge base, it achieves accurate classification and spatial localization of appearance defects. Specifically, the algorithm outputs appearance quality recognition results containing the following four key information categories: 1. Defect location: Precisely calibrating the center point or contour range of each defect on the component surface using image coordinates (e.g., pixel coordinates (x, y)) or actual physical coordinates (after spatial mapping transformation, in millimeters); 2. Defect type: Outputting the identified defect category label, such as "honeycomb" or "exposed reinforcement," supporting multi-label recognition; 3. Defect size: Calculating the actual area, length, width, and other geometric dimensions of the defect based on image resolution and spatial calibration parameters; 4. Defect quantity: Counting the total number of defects of the same type or all types in the image.

[0037] The AI ​​visual recognition algorithm model is a composite neural network model combining deep learning-based object detection and semantic segmentation. It employs mainstream architectures such as YOLOv5, Faster R-CNN, or U-Net, and is thoroughly trained and validated on a database of surface defects in real precast components containing a large number of labeled samples. This model can identify at least one type of appearance defect, including but not limited to: color difference, bubbles, honeycomb, pitting, exposed reinforcement, cracks, formwork misalignment, and stains—common concrete appearance defects.

[0038] In one possible implementation, when the geometric dimension detection subsystem 140 is used to collect the actual geometric parameters of the prefabricated component to be inspected and to perform dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the prefabricated component to be inspected, the geometric dimension detection subsystem 140 is specifically used for: (1): Data is collected from the surface of the prefabricated component to be tested from multiple angles using one or more non-contact measuring devices to obtain three-dimensional point cloud data of the prefabricated component to be tested.

[0039] Here, 3D point cloud data is represented as a set of discrete points in a spatial coordinate system. Each point contains 3D coordinate information (x, y, z) and may optionally include auxiliary attributes such as normal vector, color, or reflection intensity.

[0040] (2): Extract the key geometric features of the precast component to be tested based on the three-dimensional point cloud data, and determine the actual geometric parameters of the precast component to be tested based on the key geometric features; wherein, the actual geometric parameters include actual length, width, height, cross-sectional dimensions and pre-camber parameters.

[0041] Here, key geometric features include, but are not limited to: the main axis direction and central symmetry plane of the component; the contour lines of each section, such as typical cross-sections like webs, flanges, and end sections; characteristic edges and corners, such as edges, chamfers, and boundaries of reserved holes; and the design reference surface and theoretical contour surface. The feature extraction process employs a combination of point cloud clustering, plane fitting, RANSAC (Random Sample Consensus) algorithm, PCA (Principal Component Analysis), and curvature analysis. For example, the RANSAC algorithm is used to fit the main planar regions of the component, such as the top surface, bottom surface, and side webs; multiple cross-sectional point sets are generated by slicing along the length direction, and contour reconstruction and least squares fitting are performed on each cross-section to extract its actual width, height, thickness, and other local dimensions.

[0042] (3): The actual geometric parameters are automatically compared with the theoretical dimensions, the deviation values ​​are identified and it is determined whether they exceed the allowable tolerance range, and the geometric dimension identification results are generated.

[0043] Here, the extracted actual geometric parameters are matched with the corresponding theoretical dimensions one by one. For each comparable parameter, its deviation value is calculated. If the deviation value exceeds the allowable tolerance range, the geometric parameter is unqualified.

[0044] Specifically, for geometric parameters deemed to be out of tolerance, their locations can be highlighted in the 3D visualization interface, with deviation values ​​overlaid to assist technicians in quickly locating the problem area. Furthermore, when a type of component exhibits systematic deviations in the same direction multiple times consecutively, an early warning mechanism can be triggered, providing process adjustment suggestions to the production control system to achieve closed-loop quality management.

[0045] In this application, by automatically comparing actual geometric parameters with theoretical dimensions, an objective, efficient, and standardized evaluation of the geometric accuracy of precast components is achieved. This avoids problems such as subjective misjudgment, recording errors, and response delays that exist in the traditional manual comparison process, and significantly improves the intelligence level and traceability of quality control.

[0046] In one possible implementation, when the internal defect detection subsystem 150 is used to acquire the internal structural information of the prefabricated component to be inspected, process the internal structural information based on an internal defect analysis algorithm, and determine the internal defect identification result of the prefabricated component to be inspected, the internal defect detection subsystem 150 is specifically used for: a: The scanning device performs a full-area coverage scan of the key areas of the precast component to be tested, and collects the physical response signals inside the precast component in real time; wherein, the scanning device includes at least one of an ultrasonic imaging array, ground penetrating radar, shock echo meter or infrared thermal imager.

[0047] Here, the scanning equipment performs a full-area coverage scan of the key areas of the precast component to be inspected, and collects the physical response signals inside the precast component in real time.

[0048] b: Input the physical response signal into the pre-trained internal defect analysis model, perform signal feature parsing processing on the physical response signal, and perform internal void, crack, interlayer or rebar position displacement identification processing on the parsed signal features to determine the internal defect identification result; wherein, the internal defect identification result includes defect location, size and severity level.

[0049] Here, the specific signal feature analysis processing includes: Time-domain feature extraction: analyzing echo arrival time, pulse width, peak amplitude, waveform distortion, etc., to preliminarily determine the depth and size of defects; Frequency-domain feature extraction: obtaining the changes in signal frequency components through Fast Fourier Transform (FFT) or Short-Time Fourier Transform (STFT) to identify energy attenuation modes caused by material impedance differences; Spatial distribution feature extraction: combining scan path coordinate information to reconstruct the spatial topological relationship of the signal response and generate thermal maps or tomographic images of abnormal areas within the component cross-section; Multi-channel fusion feature extraction: for data acquired by array sensors, beamforming or synthetic aperture focusing (SAFT) techniques are used to improve resolution and extract cross-channel consistency features. After completing the signal feature analysis, the internal defect analysis model further classifies and locates the obtained high-dimensional feature vectors to achieve automatic identification of at least one of the following types of internal defects: internal voids: manifested as signal loss areas or multiple reflection phenomena behind strong reflection interfaces; cracks: manifested as linear low-energy bands extending in a specific direction or sharp-edged echo anomalies; interlayers: manifested as interlayer debonding or foreign object embedding areas parallel to the surface, with periodic reflection characteristics; rebar position offset: by comparing the measured rebar echo intensity and spatial distribution with the design layout, the model identifies rebar spacing deviations, abnormal protective layer thickness, or overall rebar cage offset.

[0050] The internal defect identification results are encapsulated into electronic data packets, bound to the component identity ID, stored in the data storage module, and participate in the subsequent multi-dimensional quality comprehensive evaluation process executed by the central processor 120.

[0051] In this application, by inputting the physical response signal into a pre-trained internal defect analysis model, intelligent, high-precision, and non-contact identification of internal defects in precast components is achieved. This overcomes the technical bottlenecks of traditional manual interpretation of ultrasonic or radar images, which relies on experience, is highly subjective, and is prone to missed detections. It significantly improves the reliability and automation level of internal quality inspection.

[0052] In one possible implementation, the central processing unit 120 determines the comprehensive quality evaluation result of the prefabricated component to be tested through the following steps: Based on a pre-defined comprehensive evaluation rule base, the appearance quality identification results, geometric dimension identification results, and internal defect identification results are weighted and fused to determine a comprehensive quality score. A comprehensive quality evaluation result is generated based on the comprehensive quality score and the scoring threshold. The received identification results of each dimension and the comprehensive quality evaluation result are associated with the identity identifier of the prefabricated component being inspected to construct a digital quality file for the prefabricated component being inspected.

[0053] The weighted fusion analysis process includes the following steps: 1. Defect classification and weight allocation: The system classifies defects into different levels and assigns corresponding weight coefficients based on the defect type and its impact on structural safety, durability, and appearance. For example: Critical defects (such as exposed main reinforcement, structural cracks, large voids in the web area): high weight (e.g., 0.8~1.0); General defects (such as slight pitting, small air bubbles in non-stressed areas): medium weight (e.g., 0.3~0.5); Minor defects (such as minor edge defects, uneven color): low weight (e.g., 0.1~0.2). 2. Quantitative scoring mechanism: Each type of defect is converted into a quantitative deduction value based on its quantity, area, depth, location sensitivity, etc. For example, 1 point is deducted for every 10cm increase in crack length, 2 points are deducted for a deviation in protective layer thickness exceeding the limit, and points are deducted proportionally for a total honeycomb area exceeding the threshold. Multi-source information fusion calculation: 3. Using weighted summation, fuzzy comprehensive evaluation or machine learning classifiers (such as random forests and support vector machines), the quantitative scores of each dimension are combined into a normalized comprehensive quality score. The value range can be set from 0 to 100 points, where the higher the score, the better the overall quality of the component.

[0054] After obtaining the comprehensive quality score, the system further compares it with the preset scoring threshold to generate the final comprehensive quality evaluation result.

[0055] This application achieves real-time and traceability of quality data: all inspection data is uploaded in real time and stored in a structured format, forming a complete "digital quality archive" for the components. Through integration with BIM models and beam yard systems, it not only facilitates real-time monitoring of quality status but also provides strong data support for subsequent quality traceability and production process optimization. Once a problem is detected, the specific component and defect can be quickly located.

[0056] For further details, please refer to Figure 2 , Figure 2 This is a second structural schematic diagram of an intelligent quality inspection system 100 for a precast beam yard provided in an embodiment of this application. Figure 2 As shown, the intelligent quality inspection system 100 of the precast beam yard also includes a rebound strength detection subsystem 160 and a large screen control subsystem 170, both of which are communicatively connected to the central processing unit 120.

[0057] Specifically, the rebound strength detection subsystem 160 is used to collect a set of rebound value sequences under multiple impact tests of the precast component to be tested, perform validity screening on the rebound value sequence, remove abnormal data, calculate the average rebound value, determine the strength estimate based on the average rebound value, compare the strength estimate with the preset strength standard to determine whether it meets the specification requirements, and generate the rebound strength detection result; the large screen control subsystem 170 is used to input the information of the precast component to be tested before starting the test, view the test progress, and view the recognition results.

[0058] Here, the rebound strength testing subsystem 160 uses a robotic arm equipped with a rebound hammer to automate the rebound strength testing of precast components. The collected data is processed by a dedicated rebound strength analysis algorithm and compared with the design strength for compliance evaluation.

[0059] The large screen control subsystem 170 is used to input the beam information to be inspected before starting the inspection, start the inspection, view the inspection progress, and view the inspection results after the inspection is completed.

[0060] In one possible implementation, the central processing unit 120 is further configured to: The system receives a request from the operation terminal in the beam storage area to inspect the precast component to be inspected upon leaving the factory. It retrieves the digital quality file based on the identification of the precast component to be inspected. The system performs integrity and compliance checks on the digital quality file based on a preset factory verification rule base. If the check passes, the system sends an electronic signature prompt to the operation terminal. Upon receiving the electronic signature instruction, the system generates a unique electronic certificate for leaving the factory and updates the factory status of the precast component to be inspected simultaneously.

[0061] Here, the exit verification rule base contains two core verification rules: 1. Completeness verification rules are used to determine whether the digital quality archive contains data records for all necessary inspection items, preventing omissions of key inspection steps. The system checks whether the following contents have valid data: geometric dimension identification results (such as key parameters such as length, width, height, and pre-camber); appearance quality identification results (including defect identification images, location, type, quantity, etc.); internal defect identification results (such as ultrasonic or radar scan maps, void / crack / rebar offset identification records); rebound strength test results (if there are sampling or full inspection requirements); comprehensive quality evaluation results (including scores and conclusions); maintenance records (such as the completion status of the 28-day maintenance cycle); electronic signature records (confirmation information of the quality responsible person). If any mandatory inspection item is missing or marked as "not inspected", the system determines that the completeness is not met and automatically blocks the subsequent exit process, while prompting "XX inspection data is missing, please supplement the inspection". 2. Compliance verification rules are used to determine whether the various inspection results in the digital quality archive comply with legal or agreed technical standards. The system compares the test data item by item with preset thresholds or allowable ranges. For example: geometric dimension deviation must not exceed ±10 mm; there must be no exposed reinforcement or structural cracks in the main stress areas; internal voids with a diameter greater than 50 mm are considered unqualified; the total area of ​​surface honeycomb must not exceed 0.5% of the component's surface area; and the average measured rebound strength must not be lower than 95% of the design strength grade. For each compliance indicator, the system performs automated comparison and classifies the defect level according to its severity. When all defects pass and the overall quality evaluation result is "qualified," the system determines that compliance is met; otherwise, it outputs a warning message: "Defects exceeding the standard exist, prohibited from leaving the site." Only when both completeness and compliance pass the verification will authorized personnel perform electronic signatures in the management system, and the central processor 120 will generate a unique electronic certificate for leaving the site (such as a QR code or digital signature file), which is used for on-site gantry crane scheduling and transport vehicle verification operations.

[0062] This application provides an intelligent quality inspection system for a precast beam yard. The intelligent quality inspection system includes a conveying and positioning subsystem, a central processing unit (CPU), an appearance quality inspection subsystem, a geometric dimension inspection subsystem, and an internal defect detection subsystem. The appearance quality inspection subsystem, the geometric dimension inspection subsystem, and the internal defect detection subsystem are all communicatively connected to the CPU. The conveying and positioning subsystem is used to convey the precast component to be inspected to the inspection station. The CPU, after detecting that the precast component is located at the inspection station, sends inspection task instructions to each subsystem. Based on the appearance quality identification results, geometric dimension identification results, and internal defect identification results sent by multiple subsystems, it performs multi-dimensional comprehensive analysis and processing to determine the comprehensive quality assessment of the precast component to be inspected. The evaluation results are as follows: The appearance quality inspection subsystem is used to collect surface image information of the precast component to be inspected after receiving the inspection task instruction, and to perform appearance defect detection processing on the surface image information based on artificial intelligence visual recognition algorithms to determine the appearance quality identification result of the precast component to be inspected; The geometric dimension inspection subsystem is used to collect the actual geometric parameters of the precast component to be inspected after receiving the inspection task instruction, and to perform dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the precast component to be inspected; The internal defect detection subsystem is used to acquire the internal structure information of the precast component to be inspected after receiving the inspection task instruction, and to process the internal structure information based on internal defect analysis algorithms to determine the internal defect identification result of the precast component to be inspected. Through integrated equipment and automated processes, all quality inspection items can be completed without manual intervention after the component enters the intelligent quality inspection system. The inspection speed matches the production line rhythm, greatly improving the quality inspection efficiency and solving the efficiency bottleneck problem of traditional manual inspection.

[0063] Please see Figure 3 , Figure 3 A flowchart illustrating an intelligent quality inspection method for a precast beam yard provided in an embodiment of this application. Figure 3 As shown in the embodiments of this application, the intelligent quality inspection method includes: S301: After detecting that the prefabricated component to be inspected is located on the inspection station, the surface image information of the prefabricated component to be inspected is collected, and the appearance defect detection processing of the surface image information is performed based on the artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the prefabricated component to be inspected.

[0064] S302: Collect the actual geometric parameters of the prefabricated component to be tested, and perform dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the prefabricated component to be tested.

[0065] S303: Obtain the internal structural information of the precast component to be inspected, process the internal structural information based on the internal defect analysis algorithm, and determine the internal defect identification result of the precast component to be inspected.

[0066] S304: Perform multi-dimensional comprehensive analysis and processing on the appearance quality identification results, the geometric dimension identification results, and the internal defect identification results to determine the comprehensive quality evaluation result of the prefabricated component to be inspected.

[0067] The descriptions of S301 to S305 can be referenced from the description of the intelligent quality inspection system of the precast beam yard, and can achieve the same technical effect, so they will not be elaborated further.

[0068] For further details, please refer to Figure 4 , Figure 4 This is a schematic diagram of the intelligent quality inspection method for the precast beam yard after tensioning, provided in an embodiment of this application. Figure 4 As shown, Step 1: The component is moved into the intelligent quality inspection system of the precast beam yard. After the tensioning and grouting process, the precast component enters the inspection position through the conveying and positioning subsystem. On-site personnel operate the control screen, input the ID information of the precast component to be inspected, and start the inspection. Subsequent inspection data is then bound to the ID of the precast component. Step 2: Each inspection subsystem starts and executes the inspection task in sequence. Step 3: Multi-dimensional data fusion and intelligent evaluation. The inspection results from Step 2 are collected. Based on a comprehensive evaluation rule base (e.g., assigning different weights to different defects), a multi-dimensional comprehensive evaluation of the component quality is performed to generate a comprehensive quality evaluation result. Step 4: Report generation and data upload. A detailed quality inspection report containing all inspection data, defect images / maps, and evaluation conclusions is automatically generated and bound to the component's identity ID. This report is then uploaded to the management system in real time, completing the full-process quality inspection of a single component. Step 5: Whether the inspection report is qualified. Step Six: If the precast component passes the quality inspection, it will be hoisted to the water curing area for maintenance. Step Seven: If the precast component fails the quality inspection, it will be moved to the re-inspection area for manual re-inspection. Step Eight: Move to the beam storage area. After 28 days of curing, once the component strength meets the requirements, the precast component will be moved to the beam storage area for storage. Step Nine: Pre-departure beam inspection. After the precast component meets the departure requirements in the beam storage area, quality inspectors will retrieve the complete digital quality inspection file of the component through the system terminal for final verification. After confirming that its historical test data, evaluation conclusions, and maintenance records all meet the departure standards, electronic signature will be made in the management system to complete the factory authorization. Step Ten: Beam delivery. After receiving the factory authorization instruction, the management system will generate a unique electronic certificate for delivery. On-site personnel will use equipment such as gantry cranes to hoist the precast component from the beam storage area to the transport vehicle based on the certificate information. The central processor will synchronously update the component status to "departed," marking the end of the full life cycle quality management process for the component in the intelligent quality inspection workshop.

[0069] In one possible implementation, the step of acquiring surface image information of the precast component to be inspected, and performing appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the precast component to be inspected includes: Under controlled lighting conditions, one or more industrial cameras are used to acquire high-definition images of the surface of the prefabricated component to be inspected from multiple angles to obtain surface image information. The surface image information is subjected to denoising, contrast enhancement, and geometric distortion correction to determine the preprocessed image; The preprocessed image is input into the artificial intelligence visual recognition algorithm to automatically identify and locate at least one type of appearance defect on the surface of the component, and generate appearance quality recognition results including the location, type, size and quantity of the defect.

[0070] In one possible implementation, the step of acquiring the actual geometric parameters of the prefabricated component to be inspected and performing dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the prefabricated component to be inspected includes: Data is collected from the surface of the precast component to be tested from multiple angles using one or more non-contact measuring devices to obtain three-dimensional point cloud data of the precast component to be tested; Based on the three-dimensional point cloud data, the key geometric features of the precast component to be inspected are extracted, and the actual geometric parameters of the precast component to be inspected are determined based on the key geometric features; wherein, the actual geometric parameters include actual length, width, height, dimensions of each section and pre-camber parameters; The actual geometric parameters are automatically compared with the theoretical dimensions to identify deviation values ​​and determine whether they exceed the allowable tolerance range, thereby generating geometric dimension identification results.

[0071] In one possible implementation, the step of acquiring the internal structural information of the precast component to be inspected, processing the internal structural information based on an internal defect analysis algorithm, and determining the internal defect identification result of the precast component to be inspected includes: The scanning device performs a full-area coverage scan of the key areas of the precast component to be inspected, and collects the physical response signals inside the precast component in real time; wherein, the scanning device includes at least one of an ultrasonic imaging array, ground penetrating radar, shock echo meter or infrared thermal imager. The physical response signal is input into a pre-trained internal defect analysis model. The physical response signal is processed by signal feature parsing. The parsed signal features are then processed to identify internal voids, cracks, interlayers, or rebar position offsets to determine the internal defect identification result. The internal defect identification result includes the defect location, size, and severity level.

[0072] In one possible implementation, the comprehensive quality evaluation result of the prefabricated component to be tested is determined through the following steps: Based on a pre-set comprehensive evaluation rule base, the appearance quality identification results, the geometric dimension identification results, and the internal defect identification results are weighted and fused to determine the comprehensive quality score. A comprehensive quality evaluation result is generated based on the comprehensive quality score and the scoring threshold. The received identification results of each dimension and the comprehensive quality evaluation result are then associated with the identity identifier of the precast component being inspected, thereby constructing a digital quality file for the precast component being inspected.

[0073] In one possible implementation, the intelligent quality inspection method further includes: A sequence of rebound values ​​is collected from multiple impact tests of the precast component to be tested. The validity of the rebound value sequence is screened, abnormal data is removed, and the average rebound value is calculated. Based on the average rebound value, the strength estimate is determined. The strength estimate is compared with the preset strength standard to determine whether it meets the specification requirements, and the rebound strength test result is generated.

[0074] In one possible implementation, the intelligent quality inspection method further includes: Receive a request from the operation terminal in the beam storage area to inspect the precast component to be inspected at the gate, and retrieve the digital quality file according to the identification of the precast component to be inspected; The digital quality archive is verified for completeness and compliance based on a pre-defined set of exit verification rules. If the verification passes, an electronic signature prompt is sent to the operation terminal. Upon receiving the electronic signature instruction, a unique electronic certificate for leaving the factory is generated, and the factory status of the prefabricated component to be inspected is updated synchronously.

[0075] This application provides an intelligent quality inspection method for precast beam yards, the intelligent quality inspection method comprising: After detecting that the prefabricated component to be inspected is located at the inspection station, the surface image information of the prefabricated component to be inspected is collected, and the appearance defect detection processing of the surface image information is performed based on the artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the prefabricated component to be inspected. The actual geometric parameters of the prefabricated component to be inspected are collected, and the geometric dimension identification result of the prefabricated component to be inspected is determined by dimensional analysis of the actual geometric parameters. The internal structural information of the prefabricated component to be inspected is obtained, and the internal structural information is processed based on an internal defect analysis algorithm to determine the internal defect identification result of the prefabricated component to be inspected. The appearance quality identification results, geometric dimension identification results, and internal defect identification results are analyzed and processed in a multi-dimensional manner to determine the comprehensive quality evaluation result of the prefabricated component to be inspected.

[0076] This application provides an intelligent quality inspection method for a precast beam yard. The method includes: acquiring surface image information of the precast component after detecting it at the inspection station; performing appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm to determine the appearance quality identification result of the precast component; acquiring the actual geometric parameters of the precast component and performing dimensional analysis on the actual geometric parameters to determine the geometric dimension identification result of the precast component; acquiring the internal structural information of the precast component and processing the internal structural information based on an internal defect analysis algorithm to determine the internal defect identification result of the precast component; and performing multi-dimensional comprehensive analysis and processing on the appearance quality identification result, the geometric dimension identification result, and the internal defect identification result to determine the comprehensive quality evaluation result of the precast component. Through integrated equipment and automated processes, all quality inspection items can be completed without manual intervention after the component enters the intelligent quality inspection system. The inspection speed matches the production line cycle, greatly improving quality inspection efficiency and solving the efficiency bottleneck problem of traditional manual inspection.

[0077] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.

[0078] The memory 520 stores machine-readable instructions executable by the processor 510. When the electronic device 500 is running, the processor 510 and the memory 520 communicate via the bus 530. When the machine-readable instructions are executed by the processor 510, they can perform the operations described above. Figure 3 The steps of intelligent quality inspection of the precast beam yard in the method embodiment are shown below. For specific implementation methods, please refer to the method embodiment, which will not be repeated here.

[0079] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described actions. Figure 3 The steps of the intelligent quality inspection method for the precast beam yard in the method embodiment are described in detail in the method embodiment, and will not be repeated here.

[0080] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0081] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the shown or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0082] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0083] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0084] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An intelligent quality inspection system for a precast beam yard, characterized in that, The intelligent quality inspection system comprises a conveying and positioning subsystem, a central processor, an appearance quality detection subsystem, a geometric dimension detection subsystem and an internal defect detection subsystem, the appearance quality detection subsystem, the geometric dimension detection subsystem and the internal defect detection subsystem are in communication connection with the central processor; wherein, The conveying and positioning subsystem is configured to convey a precast component to be detected to a detection station; The central processor is configured to send a detection task instruction to each subsystem after detecting that the precast component to be detected is located on the detection station, perform multi-dimensional comprehensive analysis and processing based on appearance quality recognition results, geometric dimension recognition results and internal defect recognition results sent by multiple subsystems, and determine a comprehensive quality evaluation result of the precast component to be detected; The appearance quality detection subsystem is configured to collect surface image information of the precast component to be detected after receiving the detection task instruction, perform appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm, and determine an appearance quality recognition result of the precast component to be detected; The geometric dimension detection subsystem is configured to collect actual geometric parameters of the precast component to be detected after receiving the detection task instruction, perform dimension analysis on the actual geometric parameters, and determine a geometric dimension recognition result of the precast component to be detected; The internal defect detection subsystem is configured to obtain internal structure information of the precast component to be detected after receiving the detection task instruction, perform processing on the internal structure information based on an internal defect analysis algorithm, and determine an internal defect recognition result of the precast component to be detected.

2. The intelligent quality inspection system of claim 1, wherein, When the appearance quality detection subsystem is used to collect surface image information of the precast component to be detected, and perform appearance defect detection processing on the surface image information based on an artificial intelligence visual recognition algorithm to determine an appearance quality recognition result of the precast component to be detected, the appearance quality detection subsystem is specifically configured to: High-definition image collection of the surface of the precast component to be detected from multiple angles by using one or more industrial cameras in a controlled lighting environment to obtain the surface image information; Denoising processing, contrast enhancement and geometric distortion correction processing are performed on the surface image information to determine preprocessed images; The preprocessed images are input into the artificial intelligence visual recognition algorithm for automatic recognition and positioning of at least one type of appearance defect on the surface of the component to generate an appearance quality recognition result containing defect position, type, size and quantity.

3. The intelligent quality inspection system of claim 1, wherein, When the geometric dimension detection subsystem is used to collect actual geometric parameters of the precast component to be detected, and perform dimension analysis on the actual geometric parameters to determine a geometric dimension recognition result of the precast component to be detected, the geometric dimension detection subsystem is specifically configured to: Data collection of the surface of the precast component to be detected from multiple angles by using one or more non-contact measurement devices to obtain three-dimensional point cloud data of the precast component to be detected; The key geometric features of the prefabricated component to be detected are extracted based on the three-dimensional point cloud data, and actual geometric parameters of the prefabricated component to be detected are determined based on the key geometric features; the actual geometric parameters include actual length, width, height, cross-sectional size and pre-camber parameters; The actual geometric parameters are automatically compared with theoretical sizes, deviation values are identified, and whether the deviation values exceed an allowable tolerance range is determined, to generate a geometric size identification result.

4. The intelligent quality inspection system of claim 1, wherein, When the internal defect detection subsystem is used to acquire internal structure information of the prefabricated component to be detected, processes the internal structure information based on an internal defect analysis algorithm, and determines an internal defect identification result of the prefabricated component to be detected, the internal defect detection subsystem is specifically configured to: The key regions of the prefabricated component to be detected are scanned by a scanning device in a global coverage manner, and physical response signals inside the prefabricated component to be detected are collected in real time; the scanning device includes at least one of an ultrasonic imager array, a ground penetrating radar, an impact echo instrument or an infrared thermal imager; The physical response signals are input into a pre-trained internal defect analysis model, signal feature analysis processing is performed on the physical response signals, internal cavity, crack, interlayer or steel bar position offset identification processing is performed on the analyzed signal features, and the internal defect identification result is determined; the internal defect identification result includes defect position, size and severity level.

5. The intelligent quality inspection system of claim 1, wherein, The central processor determines a comprehensive quality evaluation result of the prefabricated component to be detected by the following steps: Based on a preset comprehensive evaluation rule library, the appearance quality identification result, the geometric size identification result and the internal defect identification result are weighted and fused for analysis, and a comprehensive quality score is determined; Based on the comprehensive quality score and a score threshold, a comprehensive quality evaluation result is generated, and the received identification results in each dimension and the comprehensive quality evaluation result are associated with an identity of the prefabricated component to be detected, to build a digital quality file of the prefabricated component to be detected.

6. The intelligent quality inspection system of claim 1, wherein, The intelligent quality inspection system further includes a rebound strength detection subsystem and a large-screen control subsystem, and the rebound strength detection subsystem and the large-screen control subsystem are in communication connection with the central processor; wherein The rebound strength detection subsystem is configured to collect a group of rebound value sequences of the prefabricated component to be detected under multiple impact tests, perform effectiveness screening on the rebound value sequences, calculate an average rebound value after removing abnormal data, determine a strength estimation value based on the average rebound value, compare the strength estimation value with a preset strength standard to determine whether the strength estimation value meets a specification requirement, and generate a rebound strength detection result; The large-screen control subsystem is configured to input information of the prefabricated component to be detected before starting detection, view a detection process and view an identification result.

7. The intelligent quality inspection system of claim 1, wherein, The central processor is further configured to: Receive an outgoing beam inspection request of the prefabricated component to be detected initiated by an operation terminal in a beam storage area, and retrieve a digital quality file according to an identity of the prefabricated component to be detected; The digital quality archive is checked for integrity and compliance based on a preset appearance verification rule library: If the check passes, an electronic signature prompt is sent to the operation terminal, and after receiving the electronic signature instruction, a unique appearance electronic voucher is generated, and the appearance state of the prefabricated component to be detected is updated synchronously.

8. An intelligent quality inspection method for a precast beam yard, the method comprising: The intelligent quality inspection method is applied to the prefabricated beam yard intelligent quality inspection system of any one of claims 1-7, and the intelligent quality inspection method comprises: After monitoring that the prefabricated component to be detected is located on the detection station, surface image information of the prefabricated component to be detected is collected, and appearance defect detection processing is performed on the surface image information based on an artificial intelligence visual recognition algorithm to determine an appearance quality recognition result of the prefabricated component to be detected; Actual geometric parameters of the prefabricated component to be detected are collected, and a geometric size recognition result of the prefabricated component to be detected is determined by performing size analysis on the actual geometric parameters; Internal structure information of the prefabricated component to be detected is acquired, and an internal defect recognition result of the prefabricated component to be detected is determined by processing the internal structure information based on an internal defect analysis algorithm; A comprehensive quality evaluation result of the prefabricated component to be detected is determined by performing multi-dimensional comprehensive analysis and processing on the appearance quality recognition result, the geometric size recognition result, and the internal defect recognition result.

9. An electronic device, comprising: It comprises: A processor, a memory, and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the prefabricated beam yard intelligent quality inspection method of claim 8.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to perform the steps of the prefabricated beam yard intelligent quality inspection method of claim 8.

Images