Composite homogeneous plate quality evaluation method and system based on nondestructive testing

Abstract

The invention provides a composite homogeneous plate quality evaluation method and system based on nondestructive testing, and relates to the technical field of quality evaluation, and the method comprises the steps: building a sample defect plate library; determining a plate detection area; performing region division according to the defect space distribution characteristics to obtain a plurality of plate detection sub-regions; when the first plate detection sub-region is aligned with the nondestructive detection execution region, activating a plurality of first nondestructive detection devices to carry out data acquisition according to the first defect type distribution characteristics to obtain a first multi-dimensional detection data set; performing internal structure quality analysis and surface structure quality analysis to generate a first plate quality analysis result; and drawing a plate quality cloud picture according to the plurality of plate quality analysis results. The method solves the technical problems that the quality evaluation method of the composite homogeneous plate in the prior art mostly focuses on detection of a single defect type and lacks comprehensive evaluation of multiple defect types of the composite homogeneous plate, so that missing detection or misjudgment is caused, and the product quality reliability is insufficient.

Description

Technical Field

[0001] This invention relates to the field of quality assessment technology, and specifically to a method and system for quality assessment of composite homogeneous plates based on nondestructive testing. Background Technology

[0002] Composite homogeneous panels possess excellent properties such as lightweight, high strength, and high rigidity. However, during the production process, various defects often occur due to the influence of material properties, process control, and environmental factors, such as interlayer delamination, bubbles, uneven density, and surface defects. These defects can seriously affect the overall performance of composite homogeneous panels and even lead to structural failure. Therefore, how to accurately assess the quality of composite homogeneous panels and effectively detect and analyze their defects has become a pressing technical problem in the industry. However, existing composite homogeneous panel quality assessment methods mostly focus on the detection of single defect types, lacking a comprehensive assessment of multiple defect types. This makes it impossible for the detection system to comprehensively assess the quality of composite homogeneous panels, leading to missed detections or misjudgments, and consequently, unreliable product quality assurance. Summary of the Invention

[0003] This application provides a method and system for quality assessment of composite homogeneous plates based on non-destructive testing, aiming to solve the technical problem that existing composite homogeneous plate quality assessment methods mostly focus on the detection of single defect types and lack comprehensive assessment of multiple defect types in composite homogeneous plates, leading to missed detections or misjudgments, and thus insufficient product quality reliability.

[0004] The first aspect disclosed in this application provides a method for quality assessment of composite homogeneous plates based on nondestructive testing (NDT). The method includes: establishing a sample defect plate library based on the production process and design requirements of the target composite homogeneous plate; determining plate inspection areas based on the defect analysis results of the sample defect plate library, wherein the defect analysis results include spatial distribution characteristics and defect type distribution characteristics; performing region division based on the defect spatial distribution characteristics, combined with the geometric shape and material properties of the plate inspection areas as spatial constraints, to obtain multiple plate inspection sub-regions; controlling the target composite homogeneous plate to move at a uniform speed within the NDT execution area using a motion control device; when a first plate inspection sub-region aligns with the NDT execution area, activating several first NDT devices to collect data based on the first defect type distribution characteristics, to obtain a first multidimensional inspection dataset; performing internal structural quality analysis and surface structural quality analysis of the target composite homogeneous plate based on the first multidimensional inspection dataset, generating a first plate quality analysis result; and sequentially traversing the multiple plate inspection sub-regions, drawing a plate quality cloud map of the target composite homogeneous plate based on the multiple plate quality analysis results.

[0005] The second aspect of this application discloses a quality assessment system for composite homogeneous plates based on nondestructive testing (NDT). This system is used in the aforementioned NDT-based quality assessment method for composite homogeneous plates. The system includes: a sample library establishment module for establishing a sample defect plate library based on the production process and design requirements of the target composite homogeneous plate; a plate inspection area determination module for determining the plate inspection area based on the defect analysis results of the sample defect plate library, wherein the defect analysis results include defect spatial distribution characteristics and defect type distribution characteristics; and a region division module for performing region division based on the defect spatial distribution characteristics, combined with the geometric shape and material properties of the plate inspection area as spatial constraints, to obtain multiple plates. The system comprises: a material inspection sub-region; a data acquisition module, used to control the target composite homogeneous plate to move at a constant speed within the non-destructive testing execution area via a motion control device; when the first plate inspection sub-region is aligned with the non-destructive testing execution area, activating several first non-destructive testing devices to acquire data based on the distribution characteristics of the first defect type, thereby obtaining a first multidimensional inspection dataset; a quality analysis module, used to perform internal structural quality analysis and surface structural quality analysis of the target composite homogeneous plate based on the first multidimensional inspection dataset, generating a first plate quality analysis result; and a plate quality cloud map establishment module, used to sequentially traverse the multiple plate inspection sub-regions and draw a plate quality cloud map of the target composite homogeneous plate based on the multiple plate quality analysis results.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects: By establishing a sample defect plate library based on the production process and design requirements of the target composite homogeneous plate, different defect types and their distribution characteristics can be comprehensively and systematically collected and classified, providing a data foundation for subsequent testing and enhancing the accuracy and targeting of the testing. Based on the defect analysis results of the sample defect plate library, a comprehensive understanding of the common defect types of composite homogeneous plates and their spatial distribution characteristics can be achieved. This analysis allows for precise determination of the plate's testing area, optimizing the allocation of testing resources and making the testing work more efficient and targeted. Combining the spatial distribution characteristics of defects, the geometric shape of the plate's testing area, and material properties, region division is performed to obtain multiple plate testing sub-regions. This process uses spatial constraints to meticulously divide the target plate into regions, ensuring that each testing sub-region effectively covers areas where defects may exist, avoiding blind spots and improving testing efficiency. The process is controlled by a motion control device. The target composite homogeneous plate moves at a constant speed, and corresponding non-destructive testing (NDT) equipment is activated based on the defect type distribution characteristics to collect data. This process, through precise alignment of the inspection area and selection of appropriate NDT equipment, efficiently collects high-quality multidimensional datasets. Based on the multidimensional dataset, the internal and surface structures of the target composite homogeneous plate are analyzed to generate plate quality analysis results, ensuring a comprehensive assessment of plate quality. This analysis not only examines surface defects but also delves into the internal structure, enabling accurate evaluation of the overall quality of the composite homogeneous plate. By sequentially traversing multiple plate inspection sub-regions, a quality cloud map of the target composite homogeneous plate is finally generated. This allows quality analysis to move beyond a single inspection area, presenting the quality status of each sub-region through overall visualization. This improves the intuitiveness and comprehensibility of quality assessment, enabling managers and engineers to quickly identify areas with quality problems and take targeted measures.

[0007] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the process for evaluating the quality of composite homogeneous plates based on nondestructive testing, as provided in an embodiment of this application.

[0009] Figure 2 This is a schematic diagram of the composite homogeneous plate quality assessment system based on non-destructive testing provided in an embodiment of this application.

[0010] Explanation of reference numerals in the attached diagram: Sample library establishment module 10, board material detection area determination module 20, area division module 30, data acquisition module 40, quality analysis module 50, board material quality cloud map establishment module 60. Detailed Implementation

[0011] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0012] Example 1, as Figure 1 As shown in the embodiments of this application, a method for quality assessment of composite homogeneous plates based on nondestructive testing is provided, the method comprising: Based on the production process and design requirements of the target composite homogeneous board, a sample defect board library was established.

[0013] The production process and design requirements for the target composite homogeneous board include the board's structural characteristics, thickness, material, and possible usage environment. Based on key process steps such as curing, molding, and processing, possible defect modes are defined, such as interlayer delamination defects and bubble defects. According to different defect modes, relevant defect samples are collected through historical data, experimental data, or on-site test results. These defect samples include defects of different degrees and types to comprehensively cover defects that may occur during the production process. These defect samples are integrated to establish a sample defect board library, which serves as the basic dataset for subsequent defect analysis and detection.

[0014] Based on the defect analysis results of the sample defect board library, the board inspection area is determined. The defect analysis results include defect spatial distribution characteristics and defect type distribution characteristics.

[0015] By analyzing a sample defect library of sheet metal, the spatial distribution characteristics of different defects on the sheet metal are extracted. For example, delamination defects are mainly concentrated in certain areas, while bubble defects are more evenly distributed. By analyzing the spatial distribution patterns of various defects, it is possible to identify which areas are prone to specific types of defects, thus providing a reference for subsequent inspection area division. By analyzing the distribution of defect types, the probability distribution of each type of defect is determined, including the frequency, size, and shape of each defect, as well as the changes of these defects under different production processes and design requirements. Different defect types have different requirements for inspection equipment and methods; therefore, it is necessary to rationally allocate inspection resources based on the distribution characteristics of defect types. Based on the spatial distribution characteristics and defect type distribution characteristics of defects, combined with the design requirements of the target composite homogeneous board, the sheet metal inspection areas that need to be focused on are determined. These inspection areas can be high-risk areas where defects frequently occur, or areas with complex structures and high process requirements.

[0016] Based on the spatial distribution characteristics of the defects, and combined with the geometric shape and material properties of the plate inspection area as spatial constraints, region division is performed to obtain multiple plate inspection sub-regions.

[0017] When dividing the material into regions, the geometry of the sheet material must be considered. For example, the sheet material may be rectangular, circular, or other shapes. Different shapes require different division methods. Simultaneously, material properties, such as stiffness and coefficient of thermal expansion, affect the setup of the testing equipment and the testing results. Therefore, material properties are considered as one of the constraints. Based on the spatial distribution characteristics of defects, combined with the geometry and material properties of the target sheet material, the sheet material is divided into multiple sub-regions for testing. Each sub-region corresponds to a specific defect detection method. For example, some regions require more intensive testing, while others only require rapid scanning. After region division, detailed testing can be performed on each sub-region to obtain more accurate defect analysis results and conduct targeted quality assessments.

[0018] The target composite homogeneous plate is controlled to move at a constant speed within the non-destructive testing execution area by a motion control device. When the first plate inspection sub-area is aligned with the non-destructive testing execution area, several first non-destructive testing devices are activated to collect data according to the distribution characteristics of the first defect type, thereby obtaining the first multidimensional inspection dataset.

[0019] The motion control device consists of actuators, a drive system, and a control unit. It precisely controls the speed, position, and orientation of the target composite homogeneous plate. By controlling the target composite homogeneous plate to move at a uniform speed within the non-destructive testing (NDT) execution area, this uniform movement ensures that the plate in each testing sub-region passes through the testing range of the equipment within a suitable time window, thereby guaranteeing the integrity and accuracy of the test data. The NDT execution area refers to the spatial region where the equipment performs testing; within this area, the plate interacts with the NDT equipment.

[0020] During uniform motion, a portion of the sheet metal will align with the non-destructive testing (NDT) area. When the first sub-region of the sheet metal enters this area, the testing equipment begins operation. Based on the defect type distribution characteristics corresponding to the first sub-region of the sheet metal, a suitable NDT device is activated for data acquisition. For example, if a region is prone to interlayer delamination defects, an ultrasonic flaw detector will be activated; if there are many surface defects, an image acquisition device will be activated. By activating different NDT devices to acquire data from the first sub-region of the sheet metal, the results form the first multi-dimensional inspection dataset. This dataset includes detection data from different devices, such as ultrasonic flaw detection curves, image acquisition data, and position coordinates.

[0021] Based on the first multidimensional detection dataset, the internal structure quality analysis and surface structure quality analysis of the target composite homogeneous plate are performed to generate the first plate quality analysis result.

[0022] The internal structure of the sheet metal is analyzed based on data collected by an ultrasonic flaw detector. This includes: determining the presence of internal defects such as delamination, bubbles, and cracks based on the characteristics of the echo signals; estimating the size of defects by analyzing the intensity and propagation time of the ultrasonic signals; and determining the precise location of the defects within the sheet metal by combining ultrasonic flaw detection coordinates. Based on these analysis results, an internal structure quality coefficient is generated to characterize the internal quality level of the target sheet metal.

[0023] For the surface of the board material, multi-angle images acquired by an image acquisition device are used for defect identification. For example, image processing techniques such as edge detection and pattern recognition are used to identify surface defects such as cracks, pits, and scratches. Combined with image acquisition coordinates, the location of surface defects is accurately marked. Based on these analysis results, a surface structure quality coefficient is generated to quantify the quality of the board material surface. The internal structure quality coefficient and the surface structure quality coefficient are integrated to obtain a comprehensive quality analysis result, generating the first board material quality analysis result.

[0024] The multiple sub-regions for board material detection are traversed sequentially, and a board quality cloud map of the target composite homogeneous board is drawn based on the multiple board material quality analysis results.

[0025] The process sequentially traverses multiple sub-regions of the target composite homogeneous board, performing the aforementioned detection and analysis procedures on each sub-region. Each sub-region generates a set of quality analysis results, including the internal structural quality coefficient, surface structural quality coefficient, and overall quality result. The board quality cloud map is a visualization tool used to display the quality distribution of the entire composite homogeneous board. By comprehensively plotting the quality analysis results of multiple sub-regions on a two-dimensional or three-dimensional coordinate system, the quality status of the board can be visually displayed. In the board quality cloud map, color and size are typically used to represent quality levels; for example, poor quality areas are represented by red or other prominent colors, while better quality areas may be represented by green or blue. This allows operators to quickly identify weak or potentially defective areas of the board for subsequent quality control and optimization.

[0026] Furthermore, this includes: Traverse the sample defect plate library, extract the set of interlayer peeling defect events based on the interlayer peeling defect pattern; extract the set of bubble defect events based on the bubble defect pattern; extract the set of density unevenness defect events based on the density unevenness defect pattern; extract the set of surface defect events based on the surface defect pattern; perform defect location labeling and defect type identification on the set of interlayer peeling defect events, the set of bubble defect events, the set of density unevenness defect events, and the set of surface defect events, and generate the defect spatial distribution features and defect type distribution features.

[0027] The sample defect plate library contains various defect samples generated during the production of composite homogeneous plates. Each record in the library is traversed to analyze its defect type and corresponding defect characteristics. Delamination defects are a common type of defect in composite materials, typically occurring between interfaces of different layers. By analyzing defect samples related to delamination in the sample library, a set of all delamination defect events conforming to this pattern is extracted. Each delamination defect event contains multiple data dimensions, such as defect location, delamination area, delamination depth, and the process conditions under which it occurred.

[0028] Bubble defects refer to the formation of bubbles caused by the failure of air or gas to be completely eliminated during resin impregnation, curing, or molding. These defects often appear on the surface or inner layer of composite materials. By traversing the sample defect plate library, all defect events that match the bubble defect pattern are extracted. Each bubble defect event typically includes the size, distribution characteristics, location, and cause of the bubble.

[0029] Uneven density refers to the uneven density distribution in different regions of a composite material. This may be caused by uneven raw material ratios, improper molding process control, or uneven temperature control during curing. By traversing the sample defect plate library, all defect events that conform to the uneven density defect pattern are extracted. These defects are usually detected by scanning imaging. The set of uneven density defect events includes the spatial distribution characteristics of the defects, such as density anomalies in local areas, detected density changes, and the potential impact of defects on the structural strength of the plate.

[0030] Surface defects typically include scratches, cracks, pits, etc., which may occur during processing (such as grinding, cutting, polishing, etc.) or during transportation and storage. By traversing the sample defect plate library, defect events that match the surface defect pattern are extracted. Each surface defect event includes the defect type, location, shape, size, and possible external factors.

[0031] The defect locations within each of the aforementioned defect event sets are precisely calibrated. This includes locating the defect positions in a two-dimensional plane or three-dimensional model of the board material. During calibration, a coordinate system is used to accurately mark each defect within the inspection area of ​​the board material. Each defect pattern has a specific identifier to distinguish different types of defects; for example, delamination, bubbles, density inhomogeneity, and surface defects are labeled separately and stored in a categorized manner. After calibrating and identifying defects, spatial distribution features of defects are generated based on this data. These features describe the spatial distribution of defects within the board material; for example, a certain type of defect may frequently occur in certain areas of the board material, while other areas may lack this type of defect. Defect type distribution features describe the frequency and proportion of different types of defects in the sample board material; for example, under certain processing conditions, bubble defects are more common, while under other processing conditions, delamination defects are more prominent.

[0032] Furthermore, the influencing factors of the interlayer delamination defect mode include adhesive performance, environmental factors, and mechanical stress; the influencing factors of the bubble defect mode include resin impregnation uniformity, curing process control precision, and environmental factors; the influencing factors of the density unevenness defect mode include raw material ratio uniformity, molding process control, and curing process temperature distribution uniformity; and the influencing factors of the surface defect mode include surface friction and resin curing uniformity.

[0033] Delamination is one of the most common problems in composite materials, especially in laminates or multilayer structures. Adhesives are the bonding agents between the layers in a composite material, and their performance significantly affects delamination. Insufficient adhesive strength or poor adhesion to the substrate can lead to delamination. Adhesive performance is influenced by factors such as its composition, curing speed, viscosity, and surface treatment methods. Environmental factors refer to the effects of temperature, humidity, and air pressure on composite materials. High temperature and high humidity conditions can cause the adhesive to soften, degrade, or decrease its adhesion, thus leading to delamination. Mechanical stress is another key factor causing delamination. During the use of composite materials, the layers are subjected to different forces, including tension, compression, and bending. If these mechanical stresses exceed the strength of the adhesive, or if the interfaces between the composite layers are uneven, delamination will occur.

[0034] Bubble defects typically occur during resin impregnation. They form when residual bubbles in the resin are not completely removed. The uniformity of resin impregnation directly affects bubble formation. Uneven resin penetration, especially in low-density areas, easily leads to bubble formation, which may be due to insufficient resin flowability or improper prepreg treatment. The curing process is a critical step in resin composite material production. Precise temperature and pressure control during curing is essential. Large fluctuations in temperature and pressure during curing can result in incomplete bubble removal, leaving bubble defects. Environmental factors also influence bubble formation. In high-humidity environments, the resin may react with moisture in the air to form bubbles, while in high-temperature environments, the expansion of bubbles may not be released in time, trapping them within the material.

[0035] Uneven density is a defect caused by the uneven density distribution in different regions of a composite material, often resulting from improper control of the molding process. The proportions of raw materials (such as fibers, resins, and fillers) in a composite material must be uniform. Improper proportions may lead to excessive resin in some areas, resulting in localized over-density, or insufficient resin, resulting in low density. During the molding process, improper control of parameters such as mold temperature, pressure, and time can cause uneven resin flow within the mold, leading to uneven density. Uneven temperature distribution during curing can result in inconsistent resin curing rates, thus affecting density distribution. Areas with excessively high curing temperatures may have excessive resin and excessively high density.

[0036] Surface defects are mainly caused by external factors during processing. Common surface defects include scratches, pits, and cracks. Friction during processing directly affects surface quality. Excessive friction can lead to scratches or localized overheating, resulting in surface defects. The resin curing process also has a significant impact on surface quality. Uneven curing can cause defects such as bubbles, cracks, or pits on the surface.

[0037] Furthermore, based on the distribution characteristics of the first defect type, several first non-destructive testing devices are activated to collect data, thereby obtaining a first multidimensional inspection dataset, including: Based on the defect type distribution characteristics, a standard non-destructive testing (NDT) equipment set and a standard testing control parameter set are interactively obtained from the standard defect sample set. The standard testing control parameters include standard equipment control parameters and standard environmental control parameters. The standard defect sample set is traversed using the first defect type distribution characteristics to select several first NDT equipment and several first standard testing control parameters. The several first NDT equipment are deployed, and the equipment is debugged and the environment is adjusted according to the several first standard testing control parameters. After completion, data is collected from the first plate inspection sub-area to obtain a first multidimensional inspection dataset.

[0038] A standard defect sample set refers to sample defects extracted from historical data. These represent different types of defects, such as delamination, bubbles, and density inhomogeneities. These samples are precisely calibrated to help understand the required testing methods and equipment for different defect occurrences. A set of non-destructive testing (NDT) equipment refers to a collection of commonly used NDT equipment. Each piece of equipment is suitable for detecting specific types of defects; for example, an ultrasonic flaw detector is suitable for detecting internal defects. A set of standard testing control parameters includes equipment control parameters, such as ultrasonic frequency, probe angle, power, and scanning speed, as well as environmental control parameters, such as light intensity, light angle, temperature, and humidity. These parameters are set according to the defect type, material properties, and process requirements to ensure the accuracy and effectiveness of the testing.

[0039] Different types of defects require different detection methods. For example, ultrasonic testing is suitable for internal defects, while surface defects require visual inspection. The appropriate detection equipment should be selected based on the size, shape, and distribution of the defect. For example, smaller surface defects require higher resolution equipment, such as high-definition camera equipment, while larger internal defects require ultrasonic equipment with stronger penetration.

[0040] Based on the geometry and defect distribution characteristics of the inspection area, several primary non-destructive testing (NDT) devices are positioned to best capture defects. The device's position, angle, and distance all affect the final inspection results. Depending on the selected devices and defect types, the device control parameters are adjusted. These parameters include scanning speed, power, frequency, probe angle, and resolution. Simultaneously, environmental parameters are adjusted according to environmental changes to ensure the accuracy of the test data. After equipment debugging, data is acquired from the primary inspection sub-area of ​​the board material. During data acquisition, the NDT devices perform detailed scanning or imaging of the primary inspection sub-area, generating a primary multidimensional inspection dataset.

[0041] Furthermore, it also includes: The quality assessment accuracy constraints and quality assessment efficiency constraints of the target composite homogeneous plate are interactively obtained; based on the quality assessment accuracy constraints and quality assessment efficiency constraints, the adaptive constraints of the several first non-destructive testing devices and several first standard testing control parameters are adjusted.

[0042] The accuracy constraint of quality assessment refers to the required accuracy of the test results when assessing the quality of composite homogeneous plates, while the efficiency constraint of quality assessment refers to the expected speed or time required to complete the assessment.

[0043] Based on the constraints of accuracy and efficiency in quality assessment, appropriate testing equipment should be selected. For example, if high accuracy is required, high-resolution equipment, such as a high-frequency ultrasonic flaw detector or a high-definition image acquisition device, can be chosen. If high efficiency is required, equipment with a faster scanning speed but slightly lower accuracy can be selected. Sometimes, to improve efficiency, the number of testing devices can be increased, especially in large-scale production, where more devices can scan multiple inspection areas simultaneously, thereby reducing the overall inspection time. The operating parameters of the equipment can be adjusted, such as the frequency of the ultrasonic waves, the angle of the probe, and the scanning speed. For example, to improve efficiency, the scanning speed can be increased, but it must be ensured that accuracy requirements are still met. Alternatively, a choice can be made regarding the equipment configuration to balance inspection accuracy and equipment efficiency.

[0044] Furthermore, this includes: The standard non-destructive testing equipment set includes an ultrasonic flaw detector for detecting internal defects and an image acquisition device for detecting surface defects. When the plurality of first non-destructive testing equipment simultaneously includes the ultrasonic flaw detector and the image acquisition device, the ultrasonic flaw detector performs ultrasonic flaw detection on the first plate inspection sub-region to obtain an ultrasonic flaw detection curve, and the image acquisition device performs multi-angle image acquisition on the first plate inspection sub-region to obtain a multi-angle image set, and records the ultrasonic flaw detection coordinates and image acquisition coordinates. The ultrasonic flaw detection curve, the multi-angle image set, the ultrasonic flaw detection coordinates, and the image acquisition coordinates are integrated to obtain the first multi-dimensional detection dataset.

[0045] Ultrasonic flaw detectors are used to detect internal defects, and are particularly suitable for detecting deep or difficult-to-detect defects such as delamination, cracks, and bubbles. Ultrasonic waves pass through the material and are reflected back to the probe; changes in the reflected signal reflect the internal defects of the material. Image acquisition devices are used to detect surface defects. These devices acquire images of the composite material surface to identify surface defects such as scratches, surface bubbles, and corrosion, and typically include high-resolution cameras.

[0046] When the ultrasonic flaw detector and the image acquisition device are used simultaneously, they detect different defects in the composite homogeneous plate. During ultrasonic testing, the ultrasonic signals emitted by the probe are reflected when they encounter different media (such as air, uneven material parts, etc.). The probe receives these reflected waves and records the signals. Based on the time difference and intensity of the reflected waves, specific information about internal defects can be obtained. The image acquisition device focuses on detecting surface defects. By capturing images from different angles, more complete information about surface defects can be obtained. For example, some surface cracks are more obvious at specific angles, while they may not be clearly visible at other angles. Therefore, multi-angle image acquisition can improve the accuracy of defect identification. For each detection data, including ultrasonic flaw detection data and image acquisition data, its coordinate information is also recorded. By recording the specific coordinates of the probe or image acquisition device, the location of each defect can be accurately determined.

[0047] After data acquisition, the two test results are integrated to form the first multidimensional test dataset, which can comprehensively reflect the internal and surface quality of the composite homogeneous plate.

[0048] Furthermore, the first board quality analysis results are generated, including: The ultrasonic flaw detection curve is subjected to feature analysis to determine the type and size of internal defects, and the coordinates of internal defects are determined by combining the ultrasonic flaw detection coordinates. Based on the type and size of internal defects, internal structural quality analysis is performed to obtain an internal structural quality coefficient. Defect identification is performed on the multi-angle image set to determine the type and size of surface defects, and the coordinates of surface defects are determined by combining the image acquisition coordinates. Based on the type and size of surface defects, surface structural quality analysis is performed to obtain a surface structural quality coefficient. When the coordinate distance between the internal defect coordinates and the surface defect coordinates is less than a predetermined threshold, defect correlation analysis is performed, and a correlation defect quality assessment coefficient is generated based on the defect correlation analysis results. The internal structural quality coefficient, surface structural quality coefficient, and correlation defect quality assessment coefficient are integrated to generate the first plate quality analysis result.

[0049] The results of ultrasonic flaw detector testing form one or more ultrasonic flaw detection curves. These curves contain characteristics such as the intensity of reflected waves, waveform changes, and propagation time. These characteristics reflect the internal structure of the material, especially defects hidden beneath the surface. By observing waveform changes, such as variations in signal intensity and the reflection and refraction of the waveform, the type of internal defect can be determined, such as bubbles, cracks, delamination, and porosity. By analyzing the waveform's time delay and reflection intensity, the defect size, such as its depth and width, can be calculated. Combined with the ultrasonic flaw detection coordinates recorded in the flaw detection data, the analyzed defect information is correlated with its spatial location to accurately determine the defect's position within the composite material.

[0050] Based on the determined types and sizes of internal defects, the internal structural quality of the material is evaluated. For example, delamination defects affect the overall strength and durability of the material, while bubble defects lead to local structural fragility and reduce load-bearing capacity. For defects of different types and sizes, appropriate quality analysis models are used, such as finite element analysis and material strength analysis, to assess the impact of defects on the overall structural performance and calculate the internal structural quality coefficient. This is a numerical index that comprehensively evaluates the internal quality of composite materials. The lower the value, the more severe the internal defects and the worse the quality; the higher the value, the fewer the internal defects and the better the quality.

[0051] For multi-angle image collections, image processing techniques are used to identify defects. Typically, image processing includes steps such as grayscale conversion, edge detection, and feature extraction. By classifying abnormal regions in the image, the type of defect is determined, such as cracks, scratches, bubbles, and corrosion. The dimensions of defects in the image, including length, width, and depth, can be measured using image processing software. Similar to ultrasonic flaw detection coordinates, the spatial location of defects in the image is determined by combining the image acquisition coordinates.

[0052] Surface defects directly affect the surface properties of composite materials, especially their appearance, corrosion resistance, coefficient of friction, and thermal conductivity. Based on the identified surface defect types and sizes, surface structure quality analysis is performed. For example, surface cracks can make materials prone to fracture or fatigue failure under load, while surface scratches can affect the aesthetics and corrosion resistance of materials. By analyzing the size and distribution of defects, their impact on the surface properties of materials is assessed, and a surface structure quality coefficient is obtained. Similar to the internal structure quality coefficient, the surface structure quality coefficient is also a quantitative indicator that reflects the overall impact of surface defects in composite materials.

[0053] For the coordinates of internal and surface defects, spatial geometry methods, such as Euclidean distance and Manhattan distance, are used to calculate the distance between them. If this distance is less than a set threshold, it indicates that the two defects are physically related. When a related defect is confirmed, a deeper analysis is performed to assess how the combined effect of the two defects affects the overall quality of the composite material. For example, surface cracks may extend to internal interlaminar delamination, or internal bubbles may cause surface rupture under external forces. The related defect quality assessment coefficient is generated based on the analysis results of these related defects, and this coefficient comprehensively considers the impact of related defects on the overall quality.

[0054] The internal structure quality coefficient, surface structure quality coefficient, and associated defect quality assessment coefficient are integrated into a comprehensive quality analysis result, namely the first board quality analysis result. The integration method can be weighted average, etc., and the calculation is based on the importance of each coefficient and its degree of influence on the board quality. For example, associated defects have a greater impact on quality, so its coefficient can be given a higher weight, and finally a comprehensive evaluation result is generated.

[0055] Furthermore, defect identification of the multi-angle image set includes: After grayscale processing of the multi-angle image set, grayscale value comparison is performed based on the reference grayscale value, and abnormal region images are cropped; the surface defect size is obtained based on the abnormal region images; the detection log data of the composite homogeneous plate is retrieved to obtain a sample plate defect image set and a sample plate defect type label set; a plate surface defect recognition channel is trained based on the sample plate defect image set and the sample defect type label set; the abnormal region image is input into the plate surface defect recognition channel to perform surface defect recognition and obtain the surface defect type.

[0056] Grayscale conversion is the process of converting a collection of images from multiple angles into a grayscale image. In image processing, grayscale conversion is the step of converting a color image into a grayscale image. It is usually done by converting the red, green, and blue (RGB) components of each pixel into a single grayscale value. This helps reduce computational complexity, especially when performing defect detection, as it can effectively focus on changes in brightness and contrast.

[0057] A baseline grayscale value is a reference value used to measure the difference in grayscale levels in an image. Typically, this baseline value is set by analyzing the grayscale distribution of a normal board surface. Any area significantly different from the baseline grayscale value may be a potential defect area. Based on the baseline grayscale value, pixels in the grayscale image are compared to identify areas where the grayscale value differs significantly from the standard value. These areas represent abnormal parts of the image, such as cracks, scratches, and other defects. According to the comparison results, areas with abnormal grayscale values ​​are selected and cropped out to form separate abnormal area images. These cropped images contain potential defect information and are key data for subsequent analysis and identification.

[0058] The dimensions of surface defects can be extracted from the cropped abnormal region image. This can be achieved by calculating the boundary, area, aspect ratio, etc. of the abnormal region.

[0059] The system accesses inspection log data, which includes images of sample sheet metal and defect labels from similar inspections conducted historically. The inspection log data typically contains detailed records of each inspection, including the image set, defect type, defect location, and size. The sample sheet metal defect image set is manually annotated and contains images of different types of defects. Each defect image is accompanied by a defect type label, such as crack or scratch. This data will be used to train the defect recognition model.

[0060] Using a historical set of sample board defect images and corresponding sample defect type labels, a board surface defect recognition channel is trained through a convolutional neural network. This model learns how to identify different types of defects based on features in the image, such as texture, color, and shape. During training, the model automatically adjusts its parameters to make the output prediction results as close as possible to the actual labels. This is an iterative process that optimizes the accuracy of the model through continuous error feedback.

[0061] The cropped abnormal area image is input into the trained board surface defect recognition channel. This channel identifies the defect type based on the previously trained model. By performing operations such as convolution and pooling on the input image, potential defect features are extracted and the defect type of the area is predicted. After the model makes inferences, the surface defect type is returned, which is the classification result of the abnormal area.

[0062] Furthermore, it also includes: Based on the first board quality analysis results, the board quality of the first board detection sub-area is judged. When the board quality judgment fails, a first abnormality reminder message is generated. Abnormal feedback management of the first board detection sub-area is performed based on the first abnormality reminder message.

[0063] The quality of the first board is judged based on the obtained quality analysis results. If the quality analysis results of the first board do not meet the predetermined quality standards, then it can be considered that there is a problem in the corresponding first board inspection sub-area and it does not meet the requirements. At this time, the first abnormality reminder information is generated. This information includes the specific defect type, location and quality indicators that may be affected. This is an automated feedback mechanism designed to promptly remind staff to carry out further processing or adjust the production process.

[0064] Anomaly feedback management is implemented for problematic areas based on the initial anomaly alert information. This includes: relaying the initial anomaly alert information to operators, quality control personnel, or production managers to ensure timely handling; if the cause of the defect is related to the production process, such as uneven raw material distribution or improper operation, the anomaly can be fed back to the production system for adjustment of relevant processes or parameters. Through anomaly feedback management, product quality can be effectively controlled, preventing substandard boards from continuing to enter subsequent production or the final market.

[0065] Example 2, based on the same inventive concept as the composite homogeneous plate quality assessment method based on non-destructive testing in the previous examples, such as... Figure 2 As shown in the embodiment of this application, a composite homogeneous plate quality assessment system based on non-destructive testing is provided. The system includes: The sample library establishment module 10 is used to establish a sample defect plate library based on the production process and design requirements of the target composite homogeneous plate; the plate detection area determination module 20 is used to determine the plate detection area based on the defect analysis results of the sample defect plate library, wherein the defect analysis results include defect spatial distribution characteristics and defect type distribution characteristics; the region division module 30 is used to perform region division based on the defect spatial distribution characteristics, combined with the geometric shape and material properties of the plate detection area as spatial constraints, to obtain multiple plate detection sub-regions; the data acquisition module 40 is used to control the target composite plate through a motion control device. The homogeneous plate moves at a constant speed within the non-destructive testing (NDT) execution area. When the first plate inspection sub-region is aligned with the NDT execution area, several first NDT devices are activated based on the distribution characteristics of the first defect type to collect data and obtain a first multidimensional inspection dataset. The quality analysis module 50 is used to perform internal structure quality analysis and surface structure quality analysis of the target composite homogeneous plate based on the first multidimensional inspection dataset, and generate a first plate quality analysis result. The plate quality cloud map establishment module 60 is used to sequentially traverse the multiple plate inspection sub-regions and draw a plate quality cloud map of the target composite homogeneous plate based on the multiple plate quality analysis results.

[0066] Furthermore, the plate detection area determination module 20 is used to perform the following operation steps: Traverse the sample defect plate library, extract the set of interlayer peeling defect events based on the interlayer peeling defect pattern; extract the set of bubble defect events based on the bubble defect pattern; extract the set of density unevenness defect events based on the density unevenness defect pattern; extract the set of surface defect events based on the surface defect pattern; perform defect location labeling and defect type identification on the set of interlayer peeling defect events, the set of bubble defect events, the set of density unevenness defect events, and the set of surface defect events, and generate the defect spatial distribution features and defect type distribution features.

[0067] Furthermore, the influencing factors of the interlayer delamination defect mode include adhesive performance, environmental factors, and mechanical stress; the influencing factors of the bubble defect mode include resin impregnation uniformity, curing process control precision, and environmental factors; the influencing factors of the density unevenness defect mode include raw material ratio uniformity, molding process control, and curing process temperature distribution uniformity; and the influencing factors of the surface defect mode include surface friction and resin curing uniformity.

[0068] Furthermore, the data acquisition module 40 is used to perform the following operation steps: Based on the defect type distribution characteristics, a standard non-destructive testing (NDT) equipment set and a standard testing control parameter set are interactively obtained from the standard defect sample set. The standard testing control parameters include standard equipment control parameters and standard environmental control parameters. The standard defect sample set is traversed using the first defect type distribution characteristics to select several first NDT equipment and several first standard testing control parameters. The several first NDT equipment are deployed, and the equipment is debugged and the environment is adjusted according to the several first standard testing control parameters. After completion, data is collected from the first plate inspection sub-area to obtain a first multidimensional inspection dataset.

[0069] Furthermore, the data acquisition module 40 is used to perform the following operation steps: The quality assessment accuracy constraints and quality assessment efficiency constraints of the target composite homogeneous plate are interactively obtained; based on the quality assessment accuracy constraints and quality assessment efficiency constraints, the adaptive constraints of the several first non-destructive testing devices and several first standard testing control parameters are adjusted.

[0070] Furthermore, the data acquisition module 40 is used to perform the following operation steps: The standard non-destructive testing equipment set includes an ultrasonic flaw detector for detecting internal defects and an image acquisition device for detecting surface defects. When the plurality of first non-destructive testing equipment simultaneously includes the ultrasonic flaw detector and the image acquisition device, the ultrasonic flaw detector performs ultrasonic flaw detection on the first plate inspection sub-region to obtain an ultrasonic flaw detection curve, and the image acquisition device performs multi-angle image acquisition on the first plate inspection sub-region to obtain a multi-angle image set, and records the ultrasonic flaw detection coordinates and image acquisition coordinates. The ultrasonic flaw detection curve, the multi-angle image set, the ultrasonic flaw detection coordinates, and the image acquisition coordinates are integrated to obtain the first multi-dimensional detection dataset.

[0071] Furthermore, the quality analysis module 50 is used to perform the following operation steps: The ultrasonic flaw detection curve is subjected to feature analysis to determine the type and size of internal defects, and the coordinates of internal defects are determined by combining the ultrasonic flaw detection coordinates. Based on the type and size of internal defects, internal structural quality analysis is performed to obtain an internal structural quality coefficient. Defect identification is performed on the multi-angle image set to determine the type and size of surface defects, and the coordinates of surface defects are determined by combining the image acquisition coordinates. Based on the type and size of surface defects, surface structural quality analysis is performed to obtain a surface structural quality coefficient. When the coordinate distance between the internal defect coordinates and the surface defect coordinates is less than a predetermined threshold, defect correlation analysis is performed, and a correlation defect quality assessment coefficient is generated based on the defect correlation analysis results. The internal structural quality coefficient, surface structural quality coefficient, and correlation defect quality assessment coefficient are integrated to generate the first plate quality analysis result.

[0072] Furthermore, the quality analysis module 50 is used to perform the following operation steps: After grayscale processing of the multi-angle image set, grayscale value comparison is performed based on the reference grayscale value, and abnormal region images are cropped; the surface defect size is obtained based on the abnormal region images; the detection log data of the composite homogeneous plate is retrieved to obtain a sample plate defect image set and a sample plate defect type label set; a plate surface defect recognition channel is trained based on the sample plate defect image set and the sample defect type label set; the abnormal region image is input into the plate surface defect recognition channel to perform surface defect recognition and obtain the surface defect type.

[0073] Furthermore, the quality analysis module 50 is used to perform the following operation steps: Based on the first board quality analysis results, the board quality of the first board detection sub-area is judged. When the board quality judgment fails, a first abnormality reminder message is generated. Abnormal feedback management of the first board detection sub-area is performed based on the first abnormality reminder message.

[0074] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for quality assessment of composite homogeneous plates based on nondestructive testing, characterized in that, The method includes: Based on the production process and design requirements of the target composite homogeneous board, a sample defect board library was established; Based on the defect analysis results of the sample defect plate library, the plate detection area is determined. The defect analysis results include defect spatial distribution characteristics and defect type distribution characteristics. Based on the spatial distribution characteristics of the defects, and combined with the geometric shape and material properties of the plate inspection area as spatial constraints, region division is performed to obtain multiple plate inspection sub-regions. The target composite homogeneous plate is controlled to move at a constant speed in the non-destructive testing execution area by a motion control device. When the first plate detection sub-area is aligned with the non-destructive testing execution area, several first non-destructive testing devices are activated to collect data according to the distribution characteristics of the first defect type, and a first multidimensional detection dataset is obtained. Based on the first multidimensional detection dataset, perform internal structural quality analysis and surface structural quality analysis on the target composite homogeneous plate to generate the first plate quality analysis result. The multiple sub-regions for board material detection are traversed sequentially, and a board quality cloud map of the target composite homogeneous board is drawn based on the multiple board material quality analysis results.

2. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 1, characterized in that, include: Traverse the sample defect plate library and extract the set of interlayer peeling defect events based on the interlayer peeling defect pattern; Extract the bubble defect event set based on the bubble defect pattern; Extract the set of density-uniformity defect events based on the density-uniformity defect pattern; Extract the set of surface defect events based on surface defect patterns; The defect location and defect type are identified for the set of interlayer peeling defect events, the set of bubble defect events, the set of density unevenness defect events, and the set of surface defect events, and the spatial distribution characteristics and defect type distribution characteristics of the defects are generated.

3. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 2, characterized in that, The influencing factors of the interlayer delamination defect mode include adhesive performance, environmental factors, and mechanical stress. The influencing factors of the bubble defect mode include resin impregnation uniformity, curing process control precision, and environmental factors. The influencing factors of the density unevenness defect mode include raw material ratio uniformity, molding process control, and curing process temperature distribution uniformity. The influencing factors of the surface defect mode include surface friction and resin curing uniformity.

4. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 1, characterized in that, Based on the distribution characteristics of the first defect type, several first non-destructive testing devices are activated to collect data, thereby obtaining a first multidimensional inspection dataset, including: Based on the defect type distribution characteristics, a standard non-destructive testing equipment set and a standard testing control parameter set are interactively obtained for the standard defect sample set. The standard testing control parameters include standard equipment control parameters and standard environmental control parameters. The standard defect sample set is traversed using the first defect type distribution characteristics, and several first non-destructive testing devices and several first standard testing control parameters are selected. The first non-destructive testing equipment is deployed, and the equipment is debugged and the environment is adjusted according to the first standard testing control parameters. After completion, data is collected from the first plate testing sub-area to obtain the first multidimensional testing dataset.

5. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 4, characterized in that, Also includes: Interactively acquire the quality assessment accuracy constraints and quality assessment efficiency constraints of the target composite homogeneous plate; Based on the aforementioned quality assessment accuracy constraints and quality assessment efficiency constraints, adaptive constraints are adjusted for the aforementioned first nondestructive testing equipment and several first standard testing control parameters.

6. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 4, characterized in that, include: The standard non-destructive testing equipment set includes an ultrasonic flaw detector for detecting internal defects and an image acquisition device for detecting surface defects. When the plurality of first non-destructive testing devices simultaneously include the ultrasonic flaw detector and the image acquisition device, the ultrasonic flaw detector performs ultrasonic flaw detection on the first plate inspection sub-area to obtain the ultrasonic flaw detection curve, and the image acquisition device performs multi-angle image acquisition on the first plate inspection sub-area to obtain a multi-angle image set, and records the ultrasonic flaw detection coordinates and the image acquisition coordinates. By integrating the ultrasonic flaw detection curves, multi-angle image sets, ultrasonic flaw detection coordinates, and image acquisition coordinates, the first multidimensional detection dataset is obtained.

7. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 6, characterized in that, Generate the first board quality analysis results, including: The ultrasonic flaw detection curve is subjected to feature analysis to determine the type and size of internal defects, and the coordinates of internal defects are determined by combining the ultrasonic flaw detection coordinates. Based on the internal defect type and internal defect size, an internal structural quality analysis is performed to obtain the internal structural quality coefficient. Defect identification is performed on the multi-angle image set to determine the surface defect type and size, and the surface defect coordinates are determined by combining the image acquisition coordinates. Based on the surface defect type and surface defect size, surface structure quality analysis is performed to obtain the surface structure quality coefficient. When the coordinate distance between the internal defect coordinates and the surface defect coordinates is less than a predetermined threshold, a defect correlation analysis is performed, and a quality evaluation coefficient for the associated defect is generated based on the defect correlation analysis results. By integrating the internal structure quality coefficient, surface structure quality coefficient, and associated defect quality assessment coefficient, the quality analysis result of the first plate material is generated.

8. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 7, characterized in that, Defect identification of the multi-angle image set includes: After grayscale processing of the multi-angle image set, grayscale value comparison is performed based on the reference grayscale value, and abnormal region images are cropped. The size of the surface defect is obtained from the image of the abnormal region. Retrieve the inspection log data of the composite homogeneous plate to obtain the sample plate defect image set and the sample plate defect type label set; Based on the sample board defect image set and the sample defect type label set, a board surface defect recognition channel is trained and obtained. The abnormal area image is input into the surface defect recognition channel of the board material to perform surface defect recognition and obtain the surface defect type.

9. The method for quality assessment of composite homogeneous plates based on nondestructive testing as described in claim 1, characterized in that, Also includes: Based on the first board quality analysis results, the board quality of the first board detection sub-area is judged. When the board quality judgment fails, a first abnormality reminder message is generated. Based on the first abnormality alert information, perform abnormality feedback management for the first board material detection sub-area.

10. A composite homogeneous plate quality assessment system based on non-destructive testing, characterized in that, The system is used to implement the composite homogeneous plate quality assessment method based on nondestructive testing as described in any one of claims 1-9, the system comprising: The sample library creation module is used to create a sample defect plate library based on the production process and design requirements of the target composite homogeneous plate. The board material inspection area determination module is used to determine the board material inspection area based on the defect analysis results of the sample defect board material library. The defect analysis results include defect spatial distribution characteristics and defect type distribution characteristics. The region division module is used to perform region division based on the spatial distribution characteristics of the defects, combined with the geometric shape and material properties of the plate inspection area as spatial constraints, to obtain multiple plate inspection sub-regions. The data acquisition module is used to control the target composite homogeneous plate to move at a constant speed in the non-destructive testing execution area through the motion control device. When the first plate detection sub-area is aligned with the non-destructive testing execution area, several first non-destructive testing devices are activated to acquire data according to the distribution characteristics of the first defect type, and a first multidimensional detection dataset is obtained. The quality analysis module is used to perform internal structure quality analysis and surface structure quality analysis of the target composite homogeneous plate based on the first multidimensional detection dataset, and generate the first plate quality analysis result. The board quality cloud map creation module is used to sequentially traverse the multiple board detection sub-regions and draw the board quality cloud map of the target composite homogeneous board based on the multiple board quality analysis results.

Images