A method and system for detecting the continuity of enameled wire paint film online

Abstract

The present invention relates to the technical field of online detection of paint film, and is a method and system for online detection of the continuity of enameled wire paint film, comprising: obtaining an initial enameled wire set, detecting the initial enameled wire set, obtaining a detection image set, extracting features of the detection image, identifying a surface feature vector set, obtaining a surface feature value set, extracting a surface feature value from the surface feature value set in turn, and performing the following operations on the extracted surface feature values: if it is confirmed that the surface feature value is not within a preset feature value interval, obtaining a defect site set, constructing a defect database, classifying the surface feature vectors of the defect site set, obtaining a repair coating, repairing the initial enameled wire set, treating the initial enameled wire corresponding to the standard feature value set as a qualified enameled wire set, and completing online detection of the continuity of the enameled wire paint film based on the defect database and the qualified enameled wire set. The present invention can improve the quality and efficiency of enameled wire production.

Description

Technical Field

[0001] The present invention relates to the technical field of paint film online detection, and in particular to an online detection method and system based on the continuity of enameled wire paint film. Background Art

[0002] Enameled wire film refers to a layer of insulating paint coated on the surface of metal wires, which is used to maintain good insulation performance and provide electrical insulation under high temperature conditions. Continuous online detection refers to a method of real-time and continuous detection of the quality of the paint film during the enameled wire production process to ensure that the paint film meets the quality standards.

[0003] Traditional enameled wire inspection methods are usually conducted offline, that is, sampling inspection of products after the production process is completed. This method has low inspection efficiency and insufficient coverage. Secondly, traditional methods usually rely on manual visual inspection or simple instrument inspection, which makes it difficult to find tiny defects. Therefore, how to improve the quality and efficiency of enameled wire production is a technical problem that needs to be solved urgently. Summary of the invention

[0004] The present invention provides an online detection method based on the continuity of enameled wire paint film and a computer-readable storage medium, the main purpose of which is to improve the quality and efficiency of enameled wire production and ensure the reliability and safety of the product.

[0005] To achieve the above object, the present invention provides an online detection method for the continuity of an enameled wire paint film, comprising:

[0006] Receive an enameled wire detection instruction, obtain an initial enameled wire set, and obtain the material and paint film thickness of the initial enameled wire set;

[0007] Calculating the required power of the pre-built laser scanning sensor according to the material and the thickness of the paint film, and detecting the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set;

[0008] Extract a detection image from the detection image set one by one, and perform the following operations on the extracted detection images:

[0009] Extract features from the detection image to obtain a surface feature vector set, and use a pre-built anomaly recognition model to identify the surface feature vector set to obtain a surface feature value set;

[0010] Extract one surface eigenvalue from the surface eigenvalue set in turn, and perform the following operations on the extracted surface eigenvalues:

[0011] Determine whether the surface eigenvalue is within a preset eigenvalue interval;

[0012] If it is confirmed that the surface eigenvalue is not within the preset eigenvalue interval, the surface eigenvector corresponding to the surface eigenvalue is located to obtain the defect site;

[0013] Summarize the defect sites to obtain the defect site set corresponding to the surface feature value set, build a defect database based on the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, where the defect types include: peeling and cracks;

[0014] Acquire a repair coating, repair the initial enameled wire set in the detection image set according to the defect type set and the repair coating to obtain a repaired initial image set, use the repaired initial image set as the detection image set, return to the step of extracting one detection image from the detection image set in turn, until the surface feature value corresponding to each detection image in the detection image set is within a preset feature value interval, summarize the surface feature values ​​within the preset feature value interval to obtain a standard feature value set, and confirm the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set;

[0015] Based on the defect database and qualified enameled wire set, online detection of the continuity of the enameled wire paint film is completed.

[0016] Optionally, the required power of the pre-built laser scanning sensor is calculated according to the material and the paint film thickness. ， include:

[0017] Obtain the ambient temperature and laser wavelength, obtain the material absorption coefficient based on the ambient temperature, and obtain the attenuation coefficient based on the laser wavelength and paint film thickness;

[0018] Calculate the required power of the laser scanning sensor based on the ambient temperature, laser wavelength, material absorption coefficient, attenuation coefficient and pre-built laser power formula;

[0019] The laser power formula is as follows:

[0020]

[0021] in, represents the required power, Indicates the basic power, Indicates the film thickness, represents the material absorption coefficient, represents the temperature correction factor, Indicates the ambient temperature, represents the laser wavelength, Represents the attenuation coefficient.

[0022] Optionally, the initial enameled wire set is fixed by using a plurality of pre-built fixtures to obtain the enameled wire set to be tested;

[0023] Extract one enameled wire to be detected from the set of enameled wires to be detected in turn, and perform the following operations on the extracted enameled wires to be detected:

[0024] The image of the enameled wire to be inspected is acquired by using a plurality of pre-built rotating platforms, the required power and a pre-built image acquisition device, and a point cloud data set is obtained, wherein the fixture is located on the rotating platform, and one rotating platform corresponds to one fixture, wherein each point cloud data in the point cloud data set represents the coordinate of the paint film of the enameled wire, and the inspection time is preset in the image acquisition device;

[0025] Stitch the point cloud data set to obtain a 3D model image;

[0026] The 3D model images are aggregated to obtain a 3D model image set, and the 3D model image set is confirmed as a detection image set.

[0027] Optionally, the step of stitching the point cloud data set to obtain a 3D model image includes:

[0028] Each point cloud data in the point cloud data set is normalized to obtain a normalized point cloud data set, a reference coordinate system is constructed, and a point cloud coordinate set is obtained based on the reference coordinate system and the normalized point cloud data set, wherein the point cloud data set includes multiple point cloud data, and one point cloud data corresponds to one normalized point cloud data, and one normalized point cloud data corresponds to one point cloud coordinate;

[0029] Extract a target point cloud coordinate from the point cloud coordinate set in turn, and perform the following operations on the extracted target point cloud coordinates:

[0030] Using the preset detection time and the target point cloud coordinates, the next point cloud coordinates adjacent to the target point cloud coordinates are obtained to obtain the adjacent point cloud coordinates, wherein the adjacent point cloud coordinates are the point cloud coordinates whose detection time is later than the target point cloud coordinates;

[0031] Construct an initial point cloud path based on the target point cloud coordinates and the adjacent point cloud coordinates, take the adjacent point cloud coordinates as the target point cloud coordinates, return to the step of extracting one target point cloud coordinate from the point cloud coordinate set in turn, until it is confirmed that the target point cloud coordinates in the point cloud coordinate set have been extracted, summarize and splice the initial point cloud path to obtain the point cloud path;

[0032] Import the point cloud path into the pre-built 3D model to obtain a 3D model image.

[0033] Optionally, the extracting features from the detection image to obtain a surface feature vector set includes:

[0034] Extract the point cloud coordinate set from the detection image, and perform the following operations on each point cloud coordinate in the point cloud coordinate set:

[0035] Use the pre-built K nearest neighbors to obtain the normal vector of the point cloud coordinates, and obtain the light source intensity, reflected light intensity, and the angle between the incident light and the surface normal of the point cloud coordinates;

[0036] The incident intensity is obtained according to the light source intensity and the angle between the incident light and the surface normal, and the reflectivity is obtained according to the incident intensity and the reflected light intensity;

[0037] Get the height difference of point cloud coordinates;

[0038] The normal vector, reflectivity and height difference are fused to obtain the surface feature vector;

[0039] The surface feature vectors are summarized to obtain a surface feature vector set.

[0040] Optionally, obtaining the height difference of the point cloud coordinates includes:

[0041] The height difference is calculated using the pre-built height difference formula and point cloud coordinates, where the height difference formula is as follows:

[0042]

[0043] in, Indicates The height difference of the point cloud coordinates, Represents the component of the point cloud coordinates on the X-axis, Represents the component of the point cloud coordinate on the Y axis, Represents the component of the point cloud coordinates on the Z axis, The normal vector representing the point cloud coordinates, Represents the intercept term of the reference plane.

[0044] Optionally, the surface feature vector set is expressed as:

[0045]

[0046] in, represents the set of surface feature vectors, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, represents the first surface eigenvector.

[0047] Optionally, the surface feature vector set is identified by using a pre-built anomaly recognition model to obtain a surface feature value set, including:

[0048] Obtain a historical surface feature vector set;

[0049] The pre-built SVM model is trained according to the historical surface feature vector set to obtain the SVM mature model;

[0050] Obtaining model parameters based on the SVM mature model, wherein the model parameters include: support vector, label of support vector, Lagrange multiplier and bias term;

[0051] A surface feature vector is extracted from the surface feature vector set in turn, and the following operations are performed on the extracted surface feature vectors:

[0052] The surface feature value is calculated according to the model parameters and the anomaly recognition model, wherein the anomaly recognition model is as follows:

[0053]

[0054] in, represents the surface characteristic value, represents the symbolic function, represents the exponential function with base e, Indicates The labels of the support vectors, represents the parameters of the Gaussian kernel function, represents the bias term, Indicates support vectors, represents the input surface feature vector, Indicates Lagrange multipliers of the support vectors;

[0055] The surface eigenvalues ​​are summarized to obtain a surface eigenvalue set.

[0056] Optionally, obtaining the repair coating comprises:

[0057] A coating monomer mixture is obtained by using a preset weighing ratio, and the coating monomer mixture is introduced into a pre-constructed mixing unit to obtain a coating to be heated, wherein the coating monomer mixture includes: a water-based functional monomer, a soft monomer, a hard monomer and a branched functional monomer;

[0058] The coating to be heated is subjected to step-wise temperature increase to obtain a prepolymerized coating, and the acidity value of the prepolymerized coating is obtained using a preset detection period and a preset acidity meter;

[0059] Determine whether the acidity value is within a preset acidity range;

[0060] If it is confirmed that the acid value is not within the preset acidity range, the prepolymerized coating is continuously heated until the acid value is within the preset acidity range to obtain a polymerized coating, and a pre-constructed solvent is added to the polymerized coating to obtain a modified polymer coating;

[0061] The modified polymer coating is subjected to cooling and curing operations to obtain a repair coating.

[0062] To achieve the above object, the present invention also provides an online detection system based on the continuity of enameled wire paint film, comprising:

[0063] A feature extraction module is used to receive an enameled wire detection instruction, obtain an initial enameled wire set, obtain the material and paint film thickness of the initial enameled wire set, calculate the required power of a pre-built laser scanning sensor according to the material and the paint film thickness, detect the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set, extract a detection image from the detection image set in turn, and perform the following operations on the extracted detection images: extract features from the detection image to obtain a surface feature vector set, and identify the surface feature vector set using a pre-built abnormality recognition model to obtain a surface feature value set;

[0064] The abnormality identification module is used to extract a surface feature value from the surface feature value set in turn, and perform the following operations on the extracted surface feature values: determine whether the surface feature value is within a preset feature value interval, if it is confirmed that the surface feature value is not within the preset feature value interval, then locate the surface feature vector corresponding to the surface feature value to obtain the defect site, summarize the defect sites, obtain the defect site set corresponding to the surface feature value set, build a defect database according to the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, wherein the defect types include: peeling and cracks;

[0065] a defect repair module, for obtaining a repair coating, repairing the initial enameled wire set in the detection image set according to the defect type set and the repair coating, obtaining a repaired initial image set, taking the repaired initial image set as the detection image set, returning to the step of extracting one detection image from the detection image set in turn, until the surface feature values ​​corresponding to each detection image in the detection image set are within a preset feature value interval, summarizing the surface feature values ​​within the preset feature value interval, obtaining a standard feature value set, and confirming the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set;

[0066] The data aggregation module is used to complete the online detection of the continuity of the enameled wire paint film based on the defect database and the qualified enameled wire set.

[0067] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0068] A memory storing at least one instruction;

[0069] The processor executes the instructions stored in the memory to implement the above-mentioned online detection method for continuity of the enameled wire paint film.

[0070] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned online detection method for continuity of enameled wire paint film.

[0071] The present invention is to solve the problem described in the background technology. The present invention receives an enameled wire detection instruction, obtains an initial enameled wire set, and obtains the material and paint film thickness of the initial enameled wire set. The present invention ensures that each batch of enameled wires has a consistent initial state before detection, avoiding detection deviations caused by different materials and paint film thicknesses. The required power of a pre-built laser scanning sensor is calculated according to the material and paint film thickness, and the initial enameled wire set is detected using the enameled wire detection instruction and the required power to obtain a detection image set. The present invention dynamically adjusts the power of the laser scanning sensor according to the material and paint film thickness of the enameled wire to ensure the clarity and accuracy of the detection image, and automatically collects and splices point cloud data through a fixture and a rotating platform using the required power and a preset detection time. , generate a high-precision 3D model image set to ensure comprehensive detection and accurate defect positioning of the enameled wire paint film, extract a detection image from the detection image set in turn, and perform the following operations on the extracted detection images: extract features from the detection image to obtain a surface feature vector set. The present invention automatically extracts a high-precision surface feature vector set by fusing the normal vector, reflectivity and height difference of the point cloud coordinates, comprehensively captures the subtle features of the enameled wire paint film, and provides accurate data support for subsequent abnormality identification. The surface feature vector set is identified using a pre-built abnormality identification model to obtain a surface feature value set. The present invention automatically identifies abnormal features in the surface feature vector set through a mature SVM model trained based on historical data, providing accurate surface The surface feature value set supports efficient quality inspection and defect location. A surface feature value is extracted from the surface feature value set in turn, and the following operations are performed on the extracted surface feature values. The present invention analyzes each surface feature value one by one to ensure that each abnormal point can be accurately identified and located, and judges whether the surface feature value is in a preset feature value interval. The present invention quickly judges whether the surface feature value is abnormal, discovers paint film defects in time, and prevents unqualified products from entering the market. If it is confirmed that the surface feature value is not in the preset feature value interval, the surface feature vector corresponding to the surface feature value is located to obtain the defect site. The present invention quickly locates defects, shortens problem processing time, improves production efficiency, summarizes defect sites, and obtains the surface feature value set corresponding to the surface feature value set. Defect site set, construct a defect database according to the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, wherein the defect types include: peeling and cracking. The present invention realizes long-term accumulation and analysis of defect information by constructing a defect database, provides data support for quality improvement, obtains repair coating, repairs the initial enameled wire set in the detection image set according to the defect type set and the repair coating, obtains a repair initial image set, takes the repair initial image set as the detection image set, and returns to the step of extracting a detection image from the detection image set in turn, until the surface feature values ​​corresponding to each detection image in the detection image set are within a preset feature value interval,The surface characteristic values ​​within the preset characteristic value interval are summarized to obtain the standard characteristic value set, and the initial enameled wire set corresponding to the standard characteristic value set is confirmed as a qualified enameled wire set. The present invention selects a suitable repair coating according to the defect type, automatically repairs the paint film, and improves the repair efficiency and quality. By summarizing the standard characteristic values, it is confirmed that the quality of the enameled wire meets the standard requirements to ensure that qualified products flow into the market. Based on the defect database and the qualified enameled wire set, the continuous online detection based on the enameled wire paint film is completed. The present invention ensures that each batch of products meets the quality standards through continuous online detection, thereby improving production efficiency and product quality. Therefore, the present invention can improve the quality and efficiency of enameled wire production and ensure the reliability and safety of the product. , BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 A schematic flow chart of an online detection method for continuity of an enameled wire paint film provided by an embodiment of the present invention;

[0073] Figure 2 A functional module diagram of an online detection system for continuity of enameled wire paint film provided by an embodiment of the present invention;

[0074] Figure 3 A schematic structural diagram of an electronic device for implementing the online detection method for continuity of enameled wire paint film provided by one embodiment of the present invention.

[0075] Description of reference numerals:

[0076] 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0077] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0078] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0079] The embodiment of the present application provides an online detection method for the continuity of the enameled wire paint film. The execution subject of the online detection method for the continuity of the enameled wire paint film includes but is not limited to at least one of the electronic devices such as the server and the terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the online detection method for the continuity of the enameled wire paint film can be executed by software or hardware installed in the terminal device or the server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0080] Reference Figure 1FIG. 1 is a flow chart of an online detection method for continuity of an enameled wire paint film provided by an embodiment of the present invention. In this embodiment, the online detection method for continuity of an enameled wire paint film includes:

[0081] S1. Receive an enameled wire detection instruction, obtain an initial enameled wire set, and obtain the material and paint film thickness of the initial enameled wire set.

[0082] It should be explained that the enameled wire inspection instruction is usually initiated by the quality inspector to perform quality inspection on the produced enameled wire. For example, Xiao Zhang is a quality inspector of an enameled wire production plant. The plant has just produced a batch of enameled wires of different specifications and materials. In order to ensure that the quality of this batch of enameled wires meets the standards, especially the continuity and integrity of the paint film, Xiao Zhang therefore initiates the enameled wire inspection instruction.

[0083] It should be explained that the initial enameled wire set refers to the set of untested enameled wires obtained by the system before quality inspection or analysis. The step of obtaining the initial enameled wire set is: using an automated conveyor system to convey the enameled wire batches just produced on the production line to the inspection area. The step of obtaining the material and paint film thickness of the initial enameled wire set is: the production management system records the production parameters of each batch of enameled wires, including the material and paint film thickness, and the online inspection system can directly read the material and paint film thickness from the production management system.

[0084] S2. Calculate the required power of the pre-built laser scanning sensor according to the material and the thickness of the paint film, and use the enameled wire detection instruction and the required power to detect the initial enameled wire set to obtain a detection image set.

[0085] In detail, the required power of the pre-built laser scanning sensor is calculated according to the material and the thickness of the paint film. ， include:

[0086] Obtain the ambient temperature and laser wavelength, obtain the material absorption coefficient based on the ambient temperature, and obtain the attenuation coefficient based on the laser wavelength and paint film thickness;

[0087] Calculate the required power of the laser scanning sensor based on the ambient temperature, laser wavelength, material absorption coefficient, attenuation coefficient and pre-built laser power formula;

[0088] The laser power formula is as follows:

[0089]

[0090] in, represents the required power, Indicates the basic power, Indicates the film thickness, represents the material absorption coefficient, represents the temperature correction factor, Indicates the ambient temperature, represents the laser wavelength, Represents the attenuation coefficient.

[0091] It is understood that the laser wavelength refers to the wavelength of the laser emitted by the laser scanning sensor. The material absorption coefficient refers to the material's ability to absorb lasers of a specific wavelength, which is related to the material's properties and temperature. The attenuation coefficient refers to the degree of attenuation of the laser when it passes through the paint film, which is related to the laser wavelength and the thickness of the paint film. The required power refers to the minimum power of the laser scanning sensor under specific conditions. The basic power refers to the reference power of the laser scanning sensor under standard conditions. The temperature correction factor refers to a coefficient related to the ambient temperature, which is used to adjust the changes in the material absorption coefficient caused by temperature changes.

[0092] It should be explained that the steps of obtaining the ambient temperature and laser wavelength are: using a temperature sensor to measure the temperature of the detection environment in real time to obtain the ambient temperature, and obtaining the laser wavelength from the specification of the laser scanning sensor to obtain the laser wavelength. The steps of obtaining the material absorption coefficient according to the ambient temperature are: measuring the absorption rate of the material to a laser of a specific wavelength at different temperatures, and obtaining the material absorption coefficient according to the current temperature. The steps of obtaining the attenuation coefficient according to the laser wavelength and the paint film thickness are: measuring the intensity of the laser before and after passing through the paint film when the paint film thickness is known, and calculating the attenuation coefficient using the following formula:

[0093]

[0094] in, Indicates the intensity of the laser when it does not pass through the paint film. Indicates the intensity of the laser after passing through the paint film.

[0095] In detail, the method of detecting an initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set includes:

[0096] Fixing the initial enameled wire set by using a plurality of pre-built fixtures to obtain the enameled wire set to be tested;

[0097] Extract one enameled wire to be detected from the set of enameled wires to be detected in turn, and perform the following operations on the extracted enameled wires to be detected:

[0098] The image of the enameled wire to be inspected is acquired by using a plurality of pre-built rotating platforms, the required power and a pre-built image acquisition device, and a point cloud data set is obtained, wherein the fixture is located on the rotating platform, and one rotating platform corresponds to one fixture, wherein each point cloud data in the point cloud data set represents the coordinate of the paint film of the enameled wire, and the inspection time is preset in the image acquisition device;

[0099] Stitch the point cloud data set to obtain a 3D model image;

[0100] The 3D model images are aggregated to obtain a 3D model image set, and the 3D model image set is confirmed as a detection image set.

[0101] It should be explained that the fixture is a device used to fix the enameled wire to ensure that the position of the enameled wire is stable and will not move during the inspection process. The enameled wire set to be inspected refers to the set obtained by fixing the initial enameled wire set. Each enameled wire has been fixed on the rotating platform by the fixture, ready for image acquisition and inspection. The rotating platform refers to a device used to carry the fixture and enameled wire, which can rotate around a certain axis so that the enameled wire can be imaged from different angles.

[0102] Importantly, the image acquisition device refers to a device used to capture the image of the enameled wire. The point cloud dataset refers to a three-dimensional coordinate dataset generated by the image acquisition device, and each point represents a coordinate position on the paint film of the enameled wire. The 3D model image refers to a three-dimensional model obtained by splicing and processing multiple point cloud data, representing the complete geometric shape of the enameled wire. The detection image set refers to a set of 3D model images of all enameled wires to be detected.

[0103] Furthermore, the step of stitching the point cloud data set to obtain a 3D model image includes:

[0104] Each point cloud data in the point cloud data set is normalized to obtain a normalized point cloud data set, a reference coordinate system is constructed, and a point cloud coordinate set is obtained based on the reference coordinate system and the normalized point cloud data set, wherein the point cloud data set includes multiple point cloud data, and one point cloud data corresponds to one normalized point cloud data, and one normalized point cloud data corresponds to one point cloud coordinate;

[0105] Extract a target point cloud coordinate from the point cloud coordinate set in turn, and perform the following operations on the extracted target point cloud coordinates:

[0106] Using the preset detection time and the target point cloud coordinates, the next point cloud coordinates adjacent to the target point cloud coordinates are obtained to obtain the adjacent point cloud coordinates, wherein the adjacent point cloud coordinates are the point cloud coordinates whose detection time is later than the target point cloud coordinates;

[0107] Construct an initial point cloud path based on the target point cloud coordinates and the adjacent point cloud coordinates, take the adjacent point cloud coordinates as the target point cloud coordinates, return to the step of extracting one target point cloud coordinate from the point cloud coordinate set in turn, until it is confirmed that the target point cloud coordinates in the point cloud coordinate set have been extracted, summarize and splice the initial point cloud path to obtain the point cloud path;

[0108] Import the point cloud path into the pre-built 3D model to obtain a 3D model image.

[0109] It is understandable that the stitching described in the embodiment of the present invention refers to the operation of sequentially connecting the initial point cloud paths end to end in the order in which the initial point cloud paths are obtained.

[0110] It should be explained that normalization is the operation of converting the coordinates of each point in the point cloud data to a uniform scale and range. The reference coordinate system refers to a three-dimensional coordinate system. The point cloud coordinate set refers to the set of all normalized point cloud data, and each point has a three-dimensional coordinate. The target point cloud coordinates refer to the points extracted one by one from the point cloud coordinate set as the target point cloud coordinates. The adjacent point cloud coordinates refer to the next point cloud coordinates that are adjacent to the current target point cloud coordinates in the time series. The initial point cloud path refers to the trajectory connecting the target point cloud coordinates and its adjacent point cloud coordinates. The point cloud path refers to connecting all the initial point cloud path segments to form a complete path. A three-dimensional model refers to a model that can construct a three-dimensional surface model based on the point cloud path.

[0111] S3. Extract one detection image from the detection image set in turn, and perform the following operations on the extracted detection images: extract features from the detection images to obtain a surface feature vector set, and use a pre-built anomaly recognition model to recognize the surface feature vector set to obtain a surface feature value set.

[0112] In detail, the feature extraction of the detection image to obtain a surface feature vector set includes:

[0113] Extract the point cloud coordinate set from the detection image, and perform the following operations on each point cloud coordinate in the point cloud coordinate set:

[0114] Use the pre-built K nearest neighbors to obtain the normal vector of the point cloud coordinates, and obtain the light source intensity, reflected light intensity, and the angle between the incident light and the surface normal of the point cloud coordinates;

[0115] The incident intensity is obtained according to the light source intensity and the angle between the incident light and the surface normal, and the reflectivity is obtained according to the incident intensity and the reflected light intensity;

[0116] Get the height difference of point cloud coordinates;

[0117] The normal vector, reflectivity and height difference are fused to obtain the surface feature vector;

[0118] The surface feature vectors are summarized to obtain a surface feature vector set.

[0119] Importantly, fusion refers to the operation of combining multiple different features into a comprehensive feature vector. The steps of fusing the normal vector, reflectivity and height difference to obtain the surface feature vector described in the embodiment of the present invention are: combining the normal vector, reflectivity and height difference into a single vector to better describe the surface characteristics of each point in the point cloud.

[0120] Importantly, the step of obtaining the light source intensity of the point cloud coordinates is: based on the light source intensity of the corresponding point of the point cloud coordinates in the detection image. The step of obtaining the reflected light intensity of the point cloud coordinates and the angle between the incident light and the surface normal is: similar to the method of obtaining the light source intensity of the point cloud coordinates, and the method of obtaining the light source intensity of the actual point, the reflected light intensity and the angle between the incident light and the surface normal is the prior art. Point cloud coordinates refer to a coordinate.

[0121] It should be explained that K nearest neighbors refers to finding the K nearest neighboring points to a certain point in the point cloud data. The K nearest neighbors described in the embodiment of the present invention are prior art and will not be described here. The normal vector refers to the vector perpendicular to the surface. The light source intensity refers to the intensity of the laser beam emitted by the laser scanner when scanning the enameled wire. The reflected light intensity refers to the intensity of the light reflected back after the laser beam is irradiated to the surface of the object. The angle between the incident light and the surface normal refers to the angle between the light emitted from the laser scanner and irradiated to the surface of the object and the vector perpendicular to the surface at a certain point on the surface of the object.

[0122] In detail, obtaining the height difference of the point cloud coordinates includes:

[0123] The height difference is calculated using the pre-built height difference formula and point cloud coordinates, where the height difference formula is as follows:

[0124]

[0125] in, Indicates The height difference of the point cloud coordinates, Represents the component of the point cloud coordinates on the X-axis, Represents the component of the point cloud coordinate on the Y axis, Represents the component of the point cloud coordinates on the Z axis, The normal vector representing the point cloud coordinates, Represents the intercept term of the reference plane.

[0126] It should be explained that the component refers to the value of the point cloud coordinates on each coordinate axis in the three-dimensional space. The intercept term refers to a constant, which represents the distance between the reference plane and the origin. The X-axis, Y-axis and Z-axis refer to the Cartesian coordinate system in which the 3D model corresponding to the initial enameled wire is a cylinder, in which the center of mass is the origin, the radial direction of the cylinder is the X-axis, the axial direction is the Y-axis, and the plane perpendicular to the X-axis and the Y-axis is the Z-axis.

[0127] Furthermore, the surface feature vector set is expressed as:

[0128]

[0129] in, represents the set of surface feature vectors, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, represents the first surface eigenvector.

[0130] Importantly, the surface feature vector set is identified using the pre-built anomaly recognition model to obtain a surface feature value set, including:

[0131] Obtain a historical surface feature vector set;

[0132] The pre-built SVM model is trained according to the historical surface feature vector set to obtain the SVM mature model;

[0133] Obtaining model parameters based on the SVM mature model, wherein the model parameters include: support vector, label of support vector, Lagrange multiplier and bias term;

[0134] A surface feature vector is extracted from the surface feature vector set in turn, and the following operations are performed on the extracted surface feature vectors:

[0135] The surface feature value is calculated according to the model parameters and the anomaly recognition model, wherein the anomaly recognition model is as follows:

[0136]

[0137] in, represents the surface characteristic value, represents the symbolic function, represents the exponential function with base e, Indicates The labels of the support vectors, represents the parameters of the Gaussian kernel function, represents the bias term, Indicates support vectors, represents the input surface feature vector, Indicates Lagrange multipliers of the support vectors;

[0138] The surface eigenvalues ​​are summarized to obtain a surface eigenvalue set.

[0139] It should be explained that the historical surface feature vector set refers to the set of object surface feature vectors recorded in past data sets. The SVM model refers to a binary classification model used to find an optimal hyperplane in a high-dimensional space to maximize the interval between two classes. The training SVM model described in the embodiment of the present invention is a prior art and will not be repeated here. The SVM mature model refers to the SVM model obtained by training the historical surface feature vectors. The step of obtaining the model parameters based on the SVM mature model is: extracting the support vector, the label of the support vector, the Lagrange multiplier and the bias term from the SVM mature model.

[0140] It can be understood that the support vector refers to the set of points closest to the hyperplane. The label of the support vector refers to the class label corresponding to the support vector. The Lagrange multiplier refers to the parameters used to optimize the SVM model. The bias term refers to the bias term of the hyperplane, which determines the position of the hyperplane in space.

[0141] S4. Extract one surface feature value from the surface feature value set in turn, and perform the following operation on each of the extracted surface feature values: determine whether the surface feature value is within a preset feature value interval.

[0142] It should be explained that the surface eigenvalue refers to the value obtained by calculating the surface eigenvector based on the anomaly recognition model and model parameters. The surface eigenvalue set refers to a set that contains multiple surface eigenvalues. The eigenvalue interval refers to a range of possible values ​​of the surface eigenvalue, which is set based on historical data.

[0143] S5. If it is confirmed that the surface feature value is not within the preset feature value interval, the surface feature vector corresponding to the surface feature value is located to obtain the defect site.

[0144] It should be explained that the step of locating the surface feature vector corresponding to the surface feature value is: obtaining the surface feature vector according to the surface feature value, each feature vector corresponds to a point cloud coordinate, and confirming the point cloud coordinate as the defect site. The defect site refers to the specific location of the defect on the surface of the enameled wire.

[0145] S6. Summarize the defect sites to obtain a defect site set corresponding to the surface feature value set, build a defect database based on the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set.

[0146] Specifically, the defect types include: peeling and cracks.

[0147] It should be explained that the defect site set refers to a set of all defect sites detected on the surface of the enameled wire. The steps of constructing a defect database based on the defect site set are:

[0148] Obtain a blank database and design a blank database schema, wherein the blank database schema includes: tables, fields, and relationships;

[0149] The defect site set is imported into a blank database to obtain a defect database, wherein the defect database includes: defect sites and defect types.

[0150] It should be explained that a blank database refers to a database without any data. The blank database mode described in the embodiment of the present invention refers to a database structure that has not been filled with data, which defines all tables, fields and the relationships between them contained in the database.

[0151] S7. Obtain repair coating.

[0152] In detail, the step of obtaining the repair coating comprises:

[0153] A coating monomer mixture is obtained by using a preset weighing ratio, and the coating monomer mixture is introduced into a pre-constructed mixing unit to obtain a coating to be heated, wherein the coating monomer mixture includes: a water-based functional monomer, a soft monomer, a hard monomer and a branched functional monomer;

[0154] The coating to be heated is subjected to step-wise temperature increase to obtain a prepolymerized coating, and the acidity value of the prepolymerized coating is obtained using a preset detection period and a preset acidity meter;

[0155] Determine whether the acidity value is within a preset acidity range;

[0156] If it is confirmed that the acid value is not within the preset acidity range, the prepolymerized coating is continuously heated until the acid value is within the preset acidity range to obtain a polymerized coating, and a pre-constructed solvent is added to the polymerized coating to obtain a modified polymer coating;

[0157] The modified polymer coating is subjected to cooling and curing operations to obtain a repair coating.

[0158] It should be understood that the weighing ratio refers to the ratio in which various raw materials are weighed and mixed in a certain ratio during the coating preparation process. The coating monomer mixture refers to a mixture of various monomers mixed in a certain ratio. The mixing unit described in the embodiment of the present invention refers to a unit for mixing and heating the coating monomer mixture. The steps of cooling and curing the modified polymer coating to obtain the repair coating are: using a cooling device to cool the modified polymer coating, and monitoring the temperature of the modified polymer coating to ensure that it gradually drops to a level close to room temperature, adding a curing agent and stirring to obtain a repair coating.

[0159] It is understood that a water-based functional monomer refers to a functional monomer that can be dissolved in water. A soft monomer refers to a soft polymer formed in a polymerization reaction. A hard monomer refers to a hard polymer formed in a polymerization reaction. A branched functional monomer refers to a monomer that can form a branched structure in a polymerization reaction. A coating to be heated refers to a mixture in which a coating monomer mixture has been added to a mixing unit during the coating preparation process but has not yet been subjected to a heating reaction. A pre-polymerized coating refers to a coating that has undergone a preliminary heating reaction but has not yet completed the polymerization process. A detection period refers to the time interval for detecting acidity during the coating preparation process. Acidometer refers to a device for measuring the pH of a solution. Acidity interval refers to a specific range of pre-set acidity values ​​that should be maintained during the coating preparation process. A polymerized coating refers to a coating formed by a polymerization reaction. A modified polymer coating refers to a coating that is further treated by adding a solvent on the basis of a polymerized coating to improve the performance and characteristics of the coating.

[0160] S8. Repair the initial enameled wire set in the detection image set according to the defect type set and the repair coating to obtain a repaired initial image set, take the repaired initial image set as the detection image set, return to the step of extracting one detection image from the detection image set in turn, until the surface eigenvalues ​​corresponding to each detection image in the detection image set are within a preset eigenvalue interval, summarize the surface eigenvalues ​​within the preset eigenvalue interval to obtain a standard eigenvalue set, and confirm the initial enameled wire set corresponding to the standard eigenvalue set as a qualified enameled wire set.

[0161] It should be explained that the step of repairing the initial enameled wire set in the detection image set according to the defect type set and the repair coating is: perform a cleaning operation on the initial enameled wire set according to the defect type set, use a spray gun to evenly apply the repair coating to the defective part, and wait for sufficient time to solidify it according to the curing requirements of the repair coating. The initial repair image set refers to the image set obtained after preliminary repair of the initial enameled wire set in the detection image set. The standard feature value refers to the final feature value confirmed when the surface feature value of the enameled wire reaches the preset feature value range after multiple repairs and inspections during the repair process. The standard feature value set refers to the set obtained by summarizing the standard feature values ​​of all qualified enameled wires.

[0162] S9. Complete the online detection of the continuity of the enameled wire paint film based on the defect database and the qualified enameled wire set.

[0163] It should be explained that a qualified enameled wire set refers to a set of enameled wires whose surface characteristic values ​​meet the requirements of the standard characteristic value set after repair and testing.

[0164] The present invention is to solve the problem described in the background technology. The present invention receives an enameled wire detection instruction, obtains an initial enameled wire set, and obtains the material and paint film thickness of the initial enameled wire set. The present invention ensures that each batch of enameled wires has a consistent initial state before detection, avoiding detection deviations caused by different materials and paint film thicknesses. The required power of a pre-built laser scanning sensor is calculated according to the material and paint film thickness, and the initial enameled wire set is detected using the enameled wire detection instruction and the required power to obtain a detection image set. The present invention dynamically adjusts the power of the laser scanning sensor according to the material and paint film thickness of the enameled wire to ensure the clarity and accuracy of the detection image, and automatically collects and splices point cloud data through a fixture and a rotating platform using the required power and a preset detection time. , generate a high-precision 3D model image set to ensure comprehensive detection and accurate defect positioning of the enameled wire paint film, extract a detection image from the detection image set in turn, and perform the following operations on the extracted detection images: extract features from the detection image to obtain a surface feature vector set. The present invention automatically extracts a high-precision surface feature vector set by fusing the normal vector, reflectivity and height difference of the point cloud coordinates, comprehensively captures the subtle features of the enameled wire paint film, and provides accurate data support for subsequent abnormality identification. The surface feature vector set is identified using a pre-built abnormality identification model to obtain a surface feature value set. The present invention automatically identifies abnormal features in the surface feature vector set through a mature SVM model trained based on historical data, providing accurate surface The surface feature value set supports efficient quality inspection and defect location. A surface feature value is extracted from the surface feature value set in turn, and the following operations are performed on the extracted surface feature values. The present invention analyzes each surface feature value one by one to ensure that each abnormal point can be accurately identified and located, and judges whether the surface feature value is in a preset feature value interval. The present invention quickly judges whether the surface feature value is abnormal, discovers paint film defects in time, and prevents unqualified products from entering the market. If it is confirmed that the surface feature value is not in the preset feature value interval, the surface feature vector corresponding to the surface feature value is located to obtain the defect site. The present invention quickly locates defects, shortens problem processing time, improves production efficiency, summarizes defect sites, and obtains the surface feature value set corresponding to the surface feature value set. Defect site set, construct a defect database according to the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, wherein the defect types include: peeling and cracking. The present invention realizes long-term accumulation and analysis of defect information by constructing a defect database, provides data support for quality improvement, obtains repair coating, repairs the initial enameled wire set in the detection image set according to the defect type set and the repair coating, obtains a repair initial image set, takes the repair initial image set as the detection image set, and returns to the step of extracting a detection image from the detection image set in turn, until the surface feature values ​​corresponding to each detection image in the detection image set are within a preset feature value interval,The surface characteristic values ​​within the preset characteristic value interval are summarized to obtain the standard characteristic value set, and the initial enameled wire set corresponding to the standard characteristic value set is confirmed as a qualified enameled wire set. The present invention selects a suitable repair coating according to the defect type, automatically repairs the paint film, and improves the repair efficiency and quality. By summarizing the standard characteristic values, it is confirmed that the quality of the enameled wire meets the standard requirements to ensure that qualified products flow into the market. Based on the defect database and the qualified enameled wire set, the continuous online detection based on the enameled wire paint film is completed. The present invention ensures that each batch of products meets the quality standards through continuous online detection, thereby improving production efficiency and product quality. Therefore, the present invention can improve the quality and efficiency of enameled wire production and ensure the reliability and safety of the product. ,

[0165] like Figure 2 , which is a functional module diagram of an online detection system for continuity of enameled wire paint film provided by an embodiment of the present invention.

[0166] The online continuity detection system 100 based on the enameled wire paint film of the present invention can be installed in an electronic device. According to the functions to be implemented, the online continuity detection system 100 based on the enameled wire paint film can include a feature extraction module 101, an abnormality identification module 102, a defect repair module 103 and a data aggregation module 104. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device;

[0167] The feature extraction module 101 is used to receive an enameled wire detection instruction, obtain an initial enameled wire set, obtain the material and paint film thickness of the initial enameled wire set, calculate the required power of a pre-built laser scanning sensor according to the material and paint film thickness, detect the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set, extract a detection image from the detection image set in turn, and perform the following operations on the extracted detection images: extract features from the detection image to obtain a surface feature vector set, and identify the surface feature vector set using a pre-built abnormality recognition model to obtain a surface feature value set;

[0168] The abnormality identification module 102 is used to extract a surface feature value from the surface feature value set in turn, and perform the following operations on the extracted surface feature values: determine whether the surface feature value is within a preset feature value interval, if it is confirmed that the surface feature value is not within the preset feature value interval, then locate the surface feature vector corresponding to the surface feature value to obtain the defect site, summarize the defect sites, obtain the defect site set corresponding to the surface feature value set, build a defect database according to the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all the classified surface feature vectors to obtain a defect type set, wherein the defect types include: peeling and cracks;

[0169] The defect repair module 103 is used to obtain a repair coating, to obtain a repair coating, to repair the initial enameled wire set in the detection image set according to the defect type set and the repair coating, to obtain a repaired initial image set, to use the repaired initial image set as the detection image set, to return to the step of extracting one detection image from the detection image set in turn, until the surface feature values ​​corresponding to each detection image in the detection image set are within a preset feature value interval, to summarize the surface feature values ​​within the preset feature value interval, to obtain a standard feature value set, and to confirm the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set;

[0170] The data aggregation module 104 is used to complete the online detection of the continuity of the enameled wire paint film based on the defect database and the qualified enameled wire set.

[0171] In detail, each module in the online detection system 100 based on the continuity of the enameled wire paint film in the embodiment of the present invention is used in the same manner as described above. Figure 1 The technical means are the same as the online detection method for the continuity of the enameled wire paint film described in, and can produce the same technical effects, so they will not be repeated here.

[0172] like Figure 3 , which is a schematic diagram of the structure of an electronic device for implementing an online detection method for continuity of an enameled wire paint film according to an embodiment of the present invention.

[0173] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as an online detection method program for continuity of enameled wire paint film.

[0174] The memory 11 includes at least one type of readable storage medium, including flash memory, mobile hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the program of the online detection method for the continuity of the enameled wire paint film, but also can be used to temporarily store data that has been output or is to be output.

[0175] In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and a combination of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules stored in the memory 11 (for example, a program for an online detection method for the continuity of the paint film of an enameled wire, etc.), and calls data stored in the memory 11, so as to execute various functions of the electronic device 1 and process data.

[0176] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0177] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0178] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0179] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0180] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0181] The program of the method for detecting the continuity of the enameled wire paint film stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, the following can be achieved:

[0182] Receive an enameled wire detection instruction, obtain an initial enameled wire set, and obtain the material and paint film thickness of the initial enameled wire set;

[0183] Calculating the required power of the pre-built laser scanning sensor according to the material and the thickness of the paint film, and detecting the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set;

[0184] Extract a detection image from the detection image set one by one, and perform the following operations on the extracted detection images:

[0185] Extract features from the detection image to obtain a surface feature vector set, and use a pre-built anomaly recognition model to identify the surface feature vector set to obtain a surface feature value set;

[0186] Extract one surface eigenvalue from the surface eigenvalue set in turn, and perform the following operations on the extracted surface eigenvalues:

[0187] Determine whether the surface eigenvalue is within a preset eigenvalue interval;

[0188] If it is confirmed that the surface eigenvalue is not within the preset eigenvalue interval, the surface eigenvector corresponding to the surface eigenvalue is located to obtain the defect site;

[0189] Summarize the defect sites to obtain the defect site set corresponding to the surface feature value set, build a defect database based on the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, where the defect types include: peeling and cracks;

[0190] Acquire a repair coating, repair the initial enameled wire set in the detection image set according to the defect type set and the repair coating to obtain a repaired initial image set, use the repaired initial image set as the detection image set, return to the step of extracting one detection image from the detection image set in turn, until the surface feature value corresponding to each detection image in the detection image set is within a preset feature value interval, summarize the surface feature values ​​within the preset feature value interval to obtain a standard feature value set, and confirm the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set;

[0191] Based on the defect database and qualified enameled wire set, online detection of the continuity of the enameled wire paint film is completed.

[0192] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0193] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0194] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0195] Receive an enameled wire detection instruction, obtain an initial enameled wire set, and obtain the material and paint film thickness of the initial enameled wire set;

[0196] Calculating the required power of the pre-built laser scanning sensor according to the material and the thickness of the paint film, and detecting the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set;

[0197] Extract a detection image from the detection image set one by one, and perform the following operations on the extracted detection images:

[0198] Extract features from the detection image to obtain a surface feature vector set, and use a pre-built anomaly recognition model to identify the surface feature vector set to obtain a surface feature value set;

[0199] Extract one surface eigenvalue from the surface eigenvalue set in turn, and perform the following operations on the extracted surface eigenvalues:

[0200] Determine whether the surface eigenvalue is within a preset eigenvalue interval;

[0201] If it is confirmed that the surface eigenvalue is not within the preset eigenvalue interval, the surface eigenvector corresponding to the surface eigenvalue is located to obtain the defect site;

[0202] Summarize the defect sites to obtain the defect site set corresponding to the surface feature value set, build a defect database based on the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, where the defect types include: peeling and cracks;

[0203] Acquire a repair coating, repair the initial enameled wire set in the detection image set according to the defect type set and the repair coating to obtain a repaired initial image set, use the repaired initial image set as the detection image set, return to the step of extracting one detection image from the detection image set in turn, until the surface feature value corresponding to each detection image in the detection image set is within a preset feature value interval, summarize the surface feature values ​​within the preset feature value interval to obtain a standard feature value set, and confirm the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set;

[0204] Based on the defect database and qualified enameled wire set, online detection of the continuity of the enameled wire paint film is completed.

[0205] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0206] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0207] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0208] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0209] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. An online detection method for the continuity of enameled wire paint film, characterized in that: The method comprises: Receive an enameled wire detection instruction, obtain an initial enameled wire set, and obtain the material and paint film thickness of the initial enameled wire set; Calculating the required power of the pre-built laser scanning sensor according to the material and the thickness of the paint film, and detecting the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set; The method of using the enameled wire detection instruction and the required power to detect the initial enameled wire set to obtain a detection image set includes: Fixing the initial enameled wire set by using a plurality of pre-built fixtures to obtain the enameled wire set to be tested; Extract one enameled wire to be detected from the set of enameled wires to be detected in turn, and perform the following operations on the extracted enameled wires to be detected: The image of the enameled wire to be inspected is acquired by using a plurality of pre-built rotating platforms, the required power and a pre-built image acquisition device, and a point cloud data set is obtained, wherein the fixture is located on the rotating platform, and one rotating platform corresponds to one fixture, wherein each point cloud data in the point cloud data set represents the coordinate of the paint film of the enameled wire, and the inspection time is preset in the image acquisition device; Stitch the point cloud data set to obtain a 3D model image; Summarize the 3D model images to obtain a 3D model image set, and confirm the 3D model image set as a detection image set; The step of stitching the point cloud data set to obtain a 3D model image includes: Each point cloud data in the point cloud data set is normalized to obtain a normalized point cloud data set, a reference coordinate system is constructed, and a point cloud coordinate set is obtained based on the reference coordinate system and the normalized point cloud data set, wherein the point cloud data set includes multiple point cloud data, and one point cloud data corresponds to one normalized point cloud data, and one normalized point cloud data corresponds to one point cloud coordinate; Extract a target point cloud coordinate from the point cloud coordinate set in turn, and perform the following operations on the extracted target point cloud coordinates: Using the preset detection time and the target point cloud coordinates, the next point cloud coordinates adjacent to the target point cloud coordinates are obtained to obtain the adjacent point cloud coordinates, wherein the adjacent point cloud coordinates are the point cloud coordinates whose detection time is later than the target point cloud coordinates; Construct an initial point cloud path based on the target point cloud coordinates and the adjacent point cloud coordinates, take the adjacent point cloud coordinates as the target point cloud coordinates, return to the step of extracting one target point cloud coordinate from the point cloud coordinate set in turn, until it is confirmed that the target point cloud coordinates in the point cloud coordinate set have been extracted, summarize and splice the initial point cloud path to obtain the point cloud path; Import the point cloud path into the pre-built 3D model to obtain a 3D model image; Extract a detection image from the detection image set one by one, and perform the following operations on the extracted detection images: Extract features from the detection image to obtain a surface feature vector set, and use a pre-built anomaly recognition model to identify the surface feature vector set to obtain a surface feature value set; Extract one surface eigenvalue from the surface eigenvalue set in turn, and perform the following operations on the extracted surface eigenvalues: Determine whether the surface eigenvalue is within a preset eigenvalue interval; If it is confirmed that the surface eigenvalue is not within the preset eigenvalue interval, the surface eigenvector corresponding to the surface eigenvalue is located to obtain the defect site; Summarize the defect sites to obtain the defect site set corresponding to the surface feature value set, build a defect database based on the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, where the defect types include: peeling and cracks; Acquire a repair coating, repair the initial enameled wire set in the detection image set according to the defect type set and the repair coating to obtain a repaired initial image set, use the repaired initial image set as the detection image set, return to the step of extracting one detection image from the detection image set in turn, until the surface feature value corresponding to each detection image in the detection image set is within a preset feature value interval, summarize the surface feature values ​​within the preset feature value interval to obtain a standard feature value set, and confirm the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set; Based on the defect database and qualified enameled wire set, online detection of the continuity of the enameled wire paint film is completed.

2. The method for detecting the continuity of the enameled wire paint film online according to claim 1, characterized in that: The required power of the pre-built laser scanning sensor is calculated based on the material and paint film thickness. ， include: Obtain the ambient temperature and laser wavelength, obtain the material absorption coefficient based on the ambient temperature, and obtain the attenuation coefficient based on the laser wavelength and paint film thickness; Calculate the required power of the laser scanning sensor based on the ambient temperature, laser wavelength, material absorption coefficient, attenuation coefficient and pre-built laser power formula; The laser power formula is as follows: in, represents the required power, Indicates the basic power, Indicates the film thickness, represents the material absorption coefficient, represents the temperature correction factor, Indicates the ambient temperature, represents the laser wavelength, Represents the attenuation coefficient.

3. The method for detecting the continuity of the enameled wire paint film online according to claim 2, characterized in that: The feature extraction of the detection image to obtain a surface feature vector set includes: Extract the point cloud coordinate set from the detection image, and perform the following operations on each point cloud coordinate in the point cloud coordinate set: Use the pre-built K nearest neighbors to obtain the normal vector of the point cloud coordinates, and obtain the light source intensity, reflected light intensity, and the angle between the incident light and the surface normal of the point cloud coordinates; The incident intensity is obtained according to the light source intensity and the angle between the incident light and the surface normal, and the reflectivity is obtained according to the incident intensity and the reflected light intensity; Get the height difference of point cloud coordinates; The normal vector, reflectivity and height difference are fused to obtain the surface feature vector; The surface feature vectors are summarized to obtain a surface feature vector set.

4. The method for detecting the continuity of the enameled wire paint film online according to claim 3, characterized in that: The step of obtaining the height difference of the point cloud coordinates includes: The height difference is calculated using the pre-built height difference formula and point cloud coordinates, where the height difference formula is as follows: in, Indicates The height difference of the point cloud coordinates, Represents the component of the point cloud coordinates on the X-axis, Represents the component of the point cloud coordinate on the Y axis, Represents the component of the point cloud coordinates on the Z axis, The normal vector representing the point cloud coordinates, Represents the intercept term of the reference plane.

5. The method for detecting the continuity of the enameled wire paint film online according to claim 4, characterized in that: The surface feature vector set is expressed as: in, represents the set of surface feature vectors, represents the first normal vector, represents the first reflectivity, Indicates the first height difference, represents the second normal vector, represents the second reflectivity, Indicates the second height difference, Indicates normal vectors, Indicates The reflectivity, Indicates height difference, Represents the first surface eigenvector.

6. The method for detecting the continuity of the enameled wire paint film online according to claim 5, characterized in that: The surface feature vector set is identified by using the pre-built anomaly recognition model to obtain a surface feature value set, including: Obtain a historical surface feature vector set; The pre-built SVM model is trained according to the historical surface feature vector set to obtain the SVM mature model; Obtaining model parameters based on the SVM mature model, wherein the model parameters include: support vector, label of support vector, Lagrange multiplier and bias term; A surface feature vector is extracted from the surface feature vector set in turn, and the following operations are performed on the extracted surface feature vectors: The surface feature value is calculated according to the model parameters and the anomaly recognition model, wherein the anomaly recognition model is as follows: in, represents the surface characteristic value, represents the symbolic function, represents the exponential function with base e, Indicates The labels of the support vectors, represents the parameters of the Gaussian kernel function, represents the bias term, Indicates support vectors, represents the input surface feature vector, Indicates Lagrange multipliers of the support vectors; The surface eigenvalues ​​are summarized to obtain a surface eigenvalue set.

7. The method for detecting the continuity of the enameled wire paint film online according to claim 6, characterized in that: The method of obtaining the repair coating comprises: A coating monomer mixture is obtained by using a preset weighing ratio, and the coating monomer mixture is introduced into a pre-constructed mixing unit to obtain a coating to be heated, wherein the coating monomer mixture includes: a water-based functional monomer, a soft monomer, a hard monomer and a branched functional monomer; The coating to be heated is subjected to step-wise temperature increase to obtain a prepolymerized coating, and the acidity value of the prepolymerized coating is obtained using a preset detection period and a preset acidity meter; Determine whether the acidity value is within a preset acidity range; If it is confirmed that the acid value is not within the preset acidity range, the prepolymerized coating is continuously heated until the acid value is within the preset acidity range to obtain a polymerized coating, and a pre-constructed solvent is added to the polymerized coating to obtain a modified polymer coating; The modified polymer coating is subjected to cooling and curing operations to obtain a repair coating.

8. A system using the online detection method for continuity of enameled wire paint film according to claim 1, characterized in that: The system comprises: A feature extraction module is used to receive an enameled wire detection instruction, obtain an initial enameled wire set, obtain the material and paint film thickness of the initial enameled wire set, calculate the required power of a pre-built laser scanning sensor according to the material and the paint film thickness, detect the initial enameled wire set using the enameled wire detection instruction and the required power to obtain a detection image set, extract a detection image from the detection image set in turn, and perform the following operations on the extracted detection images: extract features from the detection image to obtain a surface feature vector set, and identify the surface feature vector set using a pre-built abnormality recognition model to obtain a surface feature value set; The abnormality identification module is used to extract a surface feature value from the surface feature value set in turn, and perform the following operations on the extracted surface feature values: determine whether the surface feature value is within a preset feature value interval, if it is confirmed that the surface feature value is not within the preset feature value interval, then locate the surface feature vector corresponding to the surface feature value to obtain the defect site, summarize the defect sites, obtain the defect site set corresponding to the surface feature value set, build a defect database according to the defect site set, classify all surface feature vectors corresponding to the defect site set, and identify all classified surface feature vectors to obtain a defect type set, wherein the defect types include: peeling and cracks; a defect repair module, for obtaining a repair coating, repairing the initial enameled wire set in the detection image set according to the defect type set and the repair coating, obtaining a repaired initial image set, taking the repaired initial image set as the detection image set, returning to the step of extracting one detection image from the detection image set in turn, until the surface feature values ​​corresponding to each detection image in the detection image set are within a preset feature value interval, summarizing the surface feature values ​​within the preset feature value interval, obtaining a standard feature value set, and confirming the initial enameled wire set corresponding to the standard feature value set as a qualified enameled wire set; The data aggregation module is used to complete the online detection of the continuity of the enameled wire paint film based on the defect database and the qualified enameled wire set.

Images