Abnormal point positioning method and system under laser cutting die machining

Abstract

The invention relates to the technical field of cutting die machining, and discloses an abnormal point positioning method and system under laser cutting die machining, and the method comprises the steps: calculating a feature abnormal index corresponding to a physical quantity feature, and recognizing an abnormal feature in the physical quantity feature; analyzing an abnormal evolution venation between the abnormal features, drawing an abnormal conduction track corresponding to the abnormal features, and determining a to-be-detected area of equipment used under laser cutting die processing; calculating the laser-related material absorption rate of the cutting die material; carrying out point location marking on the to-be-detected area to obtain an initial marked point location, carrying out image acquisition on the laser cutting die of the to-be-detected area to obtain an area cutting die image, and analyzing point location abnormal factors corresponding to the initial marked point location; and carrying out positioning processing on the point location abnormal factors to obtain abnormal factor positioning, and generating a positioning result of the abnormal point location under laser cutting die processing. The invention mainly aims to solve the problem of poor accuracy of a laser cutting die processing abnormal point positioning method.

Description

Technical Field

[0001] The invention relates to a method and a system for locating abnormal points in laser die cutting processing, belonging to the technical field of die cutting processing. Background Art

[0002] Laser die-cutting processing plays an important role in modern manufacturing. With the increasing market demand for high-precision and complex-shaped products, higher requirements are placed on the quality and efficiency of laser die-cutting processing. In order to ensure the accuracy and stability of laser die-cutting processing, it is very important to locate abnormal points in time.

[0003] The existing laser die cutting process abnormal point positioning is mainly achieved through manual inspection and experience judgment. The inspectors use naked eye observation and simple measuring tools to check the processed die cutting to find possible abnormal points. However, this method is inefficient, the accuracy is difficult to guarantee, and it is easily affected by the subjective factors of the inspectors. At the same time, for some tiny abnormal points, manual inspection is often difficult to find, which may lead to product quality problems. Therefore, the existing laser die cutting process abnormal point positioning method has the problem of poor accuracy and cannot meet the high quality and high efficiency requirements of modern manufacturing industry for laser die cutting process. Therefore, a method that can accurately realize the abnormal point positioning of laser die cutting process is needed. Summary of the invention

[0004] The present invention provides a method and system for locating abnormal points in laser die cutting processing, the main purpose of which is to solve the problem of poor accuracy of the abnormal point locating method in laser die cutting processing.

[0005] To achieve the above-mentioned purpose, the present invention provides a method for locating abnormal points in laser die cutting processing, comprising: Collecting physical quantity data of the equipment used in laser die cutting processing, performing feature extraction on the physical quantity data to obtain physical quantity features, calculating feature anomaly indexes corresponding to the physical quantity features, and identifying abnormal features in the physical quantity features based on the feature anomaly indexes; Analyze the abnormal evolution context between the abnormal features, draw the abnormal conduction trajectory corresponding to the abnormal features based on the abnormal evolution context, and determine the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction trajectory; Acquire the laser performance parameters corresponding to the laser cutting die in the area to be detected, query the cutting die material corresponding to the laser cutting die and its corresponding material optical parameters, and calculate the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters; In combination with the preset reference absorption rate and the material absorption rate, the area to be detected is marked to obtain initial marking points, an image of the laser cutting die of the area to be detected is collected to obtain a regional cutting die image, and based on the regional cutting die image, the abnormal factors of the points corresponding to the initial marking points are analyzed; The abnormal factors at the points are positioned to obtain abnormal factor positioning, and the abnormal factors at the points and the abnormal factor positioning are combined to generate positioning results of abnormal points under laser die cutting processing.

[0006] Optionally, the extracting features from the physical quantity data to obtain physical quantity features includes: Smoothing the physical quantity data to obtain smoothed physical quantity data; Analyzing data attributes corresponding to the smoothed physical quantity data, and extracting initial physical quantity features corresponding to the smoothed physical quantity data; Parsing the physical quantity label corresponding to the initial physical quantity feature, and analyzing the correlation between the data attribute and the physical quantity label; Based on the correlation, feature selection processing is performed on the initial physical quantity feature to obtain a physical quantity feature.

[0007] Optionally, the calculating a characteristic anomaly index corresponding to the physical quantity characteristic includes: Calculating a characteristic equilibrium value corresponding to the physical quantity characteristic, and calculating a characteristic standard deviation corresponding to the physical quantity characteristic based on the characteristic equilibrium value; Combining the characteristic equilibrium value and the characteristic standard deviation, the characteristic anomaly index corresponding to the physical quantity characteristic is calculated by the following formula: ; Among them, A represents the characteristic anomaly index corresponding to the physical quantity characteristic, Indicates the value corresponding to the ath feature in the physical quantity feature, represents the characteristic equilibrium value, It represents the characteristic standard deviation, a represents the serial number corresponding to the physical quantity characteristic, and q represents the quantity corresponding to the material quantity characteristic.

[0008] Optionally, analyzing the abnormal evolution context between the abnormal features includes: Performing clustering processing on the abnormal features to obtain clustered abnormal features; Calculating the characteristic distance value between the clustered abnormal features, and analyzing the abnormal evolution trend corresponding to the clustered abnormal features based on the characteristic distance value; Performing feature fusion processing on the clustered abnormal features to obtain fused abnormal features; Calculating the feature interaction strength between the fused abnormal features, and determining the associated abnormal features in the fused abnormal features based on the feature interaction strength; The associated abnormal features and the abnormal evolution trend are combined to generate the abnormal evolution context between the abnormal features.

[0009] Optionally, the calculating the feature interaction strength between the fused abnormal features includes: Standardizing the fused abnormal features to obtain standard abnormal features; Based on the standard abnormal feature, calculating the feature variance corresponding to the fused abnormal feature; Measuring the feature probability corresponding to the standard abnormal feature, and calculating the feature information entropy corresponding to the fused abnormal feature based on the feature probability; The covariance matrix between the fused abnormal features is calculated, and the feature interaction strength between the fused abnormal features is calculated by combining the covariance matrix, the feature variance and the feature information entropy through the following formula: ; Among them, D represents the feature interaction strength between the fused abnormal features, represents the covariance matrix, represents the feature variance corresponding to the cth feature in the fused abnormal feature, represents the feature variance corresponding to the dth feature in the fused abnormal feature, c and d represent the sequence numbers corresponding to the fused abnormal feature, represents the balance coefficient, and They respectively represent the feature information entropy corresponding to the cth feature and the dth feature in the fused abnormal feature, and r represents the number of features of the fused abnormal feature.

[0010] Optionally, determining the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction track includes: Query the equipment performance corresponding to the equipment used in laser die cutting processing, and analyze the performance driving factors and performance output factors in the equipment performance; Combining the efficiency driving factors and the efficiency output factors, analyzing the equipment cause-effect relationship between the equipment used in laser die cutting processing; By combining the causal relationship of the equipment and the abnormal conduction trajectory, the area to be inspected of the equipment used in laser die cutting processing is determined.

[0011] Optionally, the calculating the material absorption rate of the knife mold material with respect to the laser based on the laser performance energy and the material optical parameters includes: Extracting a photon characteristic scale of the laser from the laser performance energy; Based on the optical parameters of the material, determining the material attenuation coefficient corresponding to the knife die material; Combining the photon characteristic scale and the material attenuation coefficient, the material absorption rate of the knife mold material with respect to the laser is calculated by the following formula: ; Among them, G represents the material absorption rate, represents the material attenuation coefficient, Represents the characteristic scale of photons.

[0012] Optionally, determining the material attenuation coefficient corresponding to the cutting die material based on the material optical parameter includes: Dispatching material size information corresponding to the die cutting material, and determining the material thickness corresponding to the die cutting material based on the material size information; Using a preset spectrophotometer to measure the material transmittance corresponding to the knife die material; In combination with the material thickness and the material transmittance, the material attenuation coefficient corresponding to the knife die material is determined by the following formula: ; Among them, K represents the material attenuation coefficient corresponding to the die material, I represents the material transmittance, and L represents the material thickness.

[0013] Optionally, combining the regional die-cutting image to analyze the abnormal factors of the points corresponding to the initial marking points includes: According to the initial marking points, the regional die-cutting image is segmented to obtain a marking point image; Performing image noise reduction processing on the marked point image to obtain a noise-reduced point image, and performing image enhancement processing on the noise-reduced point image to obtain an enhanced point image; Extracting point color features corresponding to the enhanced point image, performing grayscale processing on the enhanced point image, and obtaining a point grayscale image; The point texture features corresponding to the point grayscale image are extracted, and the point abnormality factors corresponding to the initial marked points are analyzed by combining the point texture features, the point color features and a preset standard point library.

[0014] A system for locating abnormal points in laser die cutting processing, the system comprising: An abnormal feature recognition module is used to collect physical quantity data of the equipment used in laser die cutting processing, perform feature extraction on the physical quantity data to obtain physical quantity features, calculate feature abnormality indexes corresponding to the physical quantity features, and identify abnormal features in the physical quantity features based on the feature abnormality indexes; A region division module is used to analyze the abnormal evolution context between the abnormal features, draw the abnormal conduction trajectory corresponding to the abnormal features based on the abnormal evolution context, and determine the area to be detected of the equipment used in the laser die cutting process based on the abnormal conduction trajectory; A material absorption rate calculation module is used to obtain the laser performance parameters corresponding to the laser cutting die of the area to be detected, and query the cutting die material corresponding to the laser cutting die and its corresponding material optical parameters, and calculate the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters; The abnormal factor analysis module is used to mark the points of the area to be detected in combination with the preset benchmark absorption rate and the material absorption rate to obtain initial marking points, collect images of the laser cutting mold of the area to be detected to obtain a regional cutting mold image, and analyze the abnormal factors of the points corresponding to the initial marking points based on the regional cutting mold image; The abnormality positioning module is used to position the abnormal factors of the points to obtain the positioning of the abnormal factors, and to generate the positioning results of the abnormal points under laser die cutting processing by combining the abnormal factors of the points and the positioning of the abnormal factors.

[0015] Compared with the problems described in the background technology, the present invention can obtain key feature information in the physical quantity data by performing feature extraction on the physical quantity data, and convert the physical quantity data into more representative features, laying a foundation for the subsequent calculation of feature anomaly index. The present invention can obtain the evolution law between the abnormal features by analyzing the abnormal evolution context between the abnormal features, and based on the abnormal evolution context, draw the abnormal conduction trajectory corresponding to the abnormal feature, and obtain the abnormal conduction route corresponding to the abnormal feature, thereby providing a basis for the subsequent determination of the area to be detected. The present invention can obtain the performance description parameters of the laser used in the laser die cutting process by obtaining the laser performance parameters corresponding to the laser die cutting area to be detected, and By querying the knife die material corresponding to the laser knife die and its corresponding material optical parameters, the index information of the knife die material's response characteristics to light can be understood. The present invention combines the preset benchmark absorption rate and the material absorption rate to mark the points in the area to be detected, and can mark the points with abnormalities in the area to be detected. By performing image acquisition on the laser knife die in the area to be detected, the image corresponding to the laser knife die in the area to be detected can be obtained, so as to improve the recognition accuracy of subsequent abnormal marked points. The present invention locates the abnormal factors at the points, and can obtain the specific positions corresponding to the abnormal factors at the points. By combining the abnormal factors at the points and the positioning of the abnormal factors, the accurate positioning results of the abnormal points under laser knife die processing can be obtained. Therefore, the present invention proposes a method and system for realizing the positioning of abnormal points under laser knife die processing, so as to improve the accuracy of positioning of abnormal points under laser knife die processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flow chart of a method for locating abnormal points in laser die cutting processing provided by an embodiment of the present invention; Figure 2 A functional module diagram of an abnormal point positioning system for laser die cutting processing provided in one embodiment of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] 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.

[0019] The embodiment of the present application provides a method for locating abnormal points under laser die cutting processing. The execution subject of the method for locating abnormal points under laser die cutting processing includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for locating abnormal points under laser die cutting processing can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0020] Embodiment 1: Reference Figure 1 FIG. 1 is a flow chart of a method for locating an abnormal point in laser die cutting processing according to an embodiment of the present invention. In this embodiment, the method for locating an abnormal point in laser die cutting processing includes: S1. Collect physical quantity data of the equipment used in laser die cutting processing, perform feature extraction on the physical quantity data to obtain physical quantity features, calculate feature anomaly indexes corresponding to the physical quantity features, and identify abnormal features in the physical quantity features based on the feature anomaly indexes.

[0021] The present invention can obtain key feature information in the physical quantity data by performing feature extraction on the physical quantity data, and convert the physical quantity data into more representative features, thus laying a foundation for subsequent calculation of feature anomaly index.

[0022] It should be explained that the laser die cutting is a mold used to process various materials, the physical quantity data is a record of the operating status parameters of the equipment used in the laser die cutting processing, and the physical quantity characteristics are the key attribute performances corresponding to the physical quantity data. Exemplarily, the collection of physical quantity data of the equipment used in the laser die cutting processing can be achieved through sensors, such as temperature sensors, pressure sensors, and vibration sensors.

[0023] In detail, the extracting features of the physical quantity data to obtain physical quantity features includes: Smoothing the physical quantity data to obtain smoothed physical quantity data; Analyzing data attributes corresponding to the smoothed physical quantity data, and extracting initial physical quantity features corresponding to the smoothed physical quantity data; Parsing the physical quantity label corresponding to the initial physical quantity feature, and analyzing the correlation between the data attribute and the physical quantity label; Based on the correlation, feature selection processing is performed on the initial physical quantity feature to obtain a physical quantity feature.

[0024] It should be explained that the smoothed physical quantity data is the data obtained after the noise in the physical quantity data is removed, the data attribute is the data property corresponding to the smoothed physical quantity data, such as statistical properties or data distribution form, etc., the initial physical quantity feature is all the essential characteristics corresponding to the smoothed physical quantity data, the physical quantity label is the characteristic property corresponding to the initial physical quantity feature, and the correlation represents the degree of mutual correlation between the data attribute and the physical quantity label.

[0025] Exemplarily, the smoothing of the physical quantity data can be achieved by a moving average method; the analysis of the data attributes corresponding to the smoothed physical quantity data can be achieved by a principal component analysis method; the extraction of the initial physical quantity features corresponding to the smoothed physical quantity data can be achieved by a feature engineering algorithm; the parsing of the physical quantity labels corresponding to the initial physical quantity features can be achieved by a manual labeling method; the analysis of the synergy between the data attributes and the physical quantity labels can be achieved by a correlation analysis algorithm; based on the level of the synergy, the initial physical quantity features are subjected to feature selection processing to obtain physical quantity features, and the features with high synergy are selected and retained to obtain physical quantity features.

[0026] The present invention calculates the characteristic anomaly index corresponding to the physical quantity feature, and can understand the degree of abnormality corresponding to the physical quantity feature through the characteristic anomaly index, and identify the abnormal features in the physical quantity feature based on the characteristic anomaly index, and then obtain the abnormal feature subset in the material quantity feature, so as to facilitate subsequent abnormal analysis and processing. It should be explained that the characteristic anomaly index represents the degree of abnormality corresponding to the physical quantity feature, and the abnormal feature is the abnormal performance part in the physical quantity feature; exemplarily, based on the comparison result of the characteristic anomaly index and a preset index threshold, the abnormal feature in the physical quantity feature is identified. If the characteristic anomaly index exceeds the preset index threshold, it indicates that the feature has an abnormal situation, and the physical quantity feature is used as the abnormal feature.

[0027] In detail, the calculation of the characteristic anomaly index corresponding to the physical quantity characteristic includes: Calculating a characteristic equilibrium value corresponding to the physical quantity characteristic, and calculating a characteristic standard deviation corresponding to the physical quantity characteristic based on the characteristic equilibrium value; Combining the characteristic equilibrium value and the characteristic standard deviation, the characteristic anomaly index corresponding to the physical quantity characteristic is calculated by the following formula: ; Among them, A represents the characteristic anomaly index corresponding to the physical quantity characteristic, Indicates the value corresponding to the ath feature in the physical quantity feature, represents the characteristic equilibrium value, It represents the characteristic standard deviation, a represents the serial number corresponding to the physical quantity characteristic, and q represents the quantity corresponding to the material quantity characteristic.

[0028] It should be explained that the characteristic equilibrium value is the average value corresponding to the physical quantity characteristic, and the characteristic standard deviation is a description of the degree of discreteness corresponding to the physical quantity characteristic. Exemplarily, the calculation of the characteristic equilibrium value corresponding to the physical quantity characteristic can be achieved through an average function; the calculation of the characteristic standard deviation corresponding to the physical quantity characteristic can be achieved through a standard deviation formula.

[0029] S2. Analyze the abnormal evolution context between the abnormal features, draw the abnormal conduction trajectory corresponding to the abnormal features based on the abnormal evolution context, and determine the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction trajectory.

[0030] The present invention can obtain the evolution law between the abnormal features by analyzing the abnormal evolution context between the abnormal features, and based on the abnormal evolution context, draw the abnormal conduction trajectory corresponding to the abnormal feature, and obtain the abnormal conduction route corresponding to the abnormal feature, thereby providing a basis for the subsequent determination of the area to be detected.

[0031] It should be explained that the abnormal evolution context is the abnormal evolution trend or evolution law between the abnormal features, and the abnormal conduction trajectory is the abnormal conduction path corresponding to the abnormal feature. Exemplarily, based on the abnormal evolution context, the abnormal conduction trajectory corresponding to the abnormal feature is drawn, and combined with drawing software, with time as the axis, the abnormal features are connected in chronological order to form an abnormal conduction trajectory.

[0032] In detail, the analysis of the abnormal evolution context between the abnormal features includes: Performing clustering processing on the abnormal features to obtain clustered abnormal features; Calculating the characteristic distance value between the clustered abnormal features, and analyzing the abnormal evolution trend corresponding to the clustered abnormal features based on the characteristic distance value; Performing feature fusion processing on the clustered abnormal features to obtain fused abnormal features; Calculating the feature interaction strength between the fused abnormal features, and determining the associated abnormal features in the fused abnormal features based on the feature interaction strength; The associated abnormal features and the abnormal evolution trend are combined to generate the abnormal evolution context between the abnormal features.

[0033] It should be explained that the clustered abnormal feature is a feature set obtained by clustering the features of the same nature in the abnormal features together, the feature distance value indicates the degree of separation between the clustered abnormal features, the abnormal evolution trend is the change trend corresponding to the clustered abnormal feature, the fused abnormal feature is a feature obtained by merging the features in the clustered abnormal feature together, the feature interaction intensity indicates the degree of mutual influence between the fused abnormal features, and the associated abnormal feature is a feature with an associated relationship in the fused abnormal feature.

[0034] Exemplarily, the clustering processing of the abnormal features can be implemented by a clustering algorithm, such as the K-Means algorithm; the feature distance value between the clustered abnormal features can be implemented by the Euclidean distance algorithm; based on the feature distance value, the abnormal evolution trend corresponding to the clustered abnormal features is analyzed, such as the feature distance value of two clustered abnormal features gradually increases, which may indicate that the abnormalities are developing in different directions and the differences are expanding. If the feature distance value gradually decreases, it may mean that the abnormalities have a trend of convergent evolution; the feature fusion processing of the clustered abnormal features can be implemented by an automatic encoder; when the feature interaction strength is greater than the preset strength, the associated abnormal features in the fused abnormal features are determined, and the preset strength can be set to 0.8, or it can be set according to the specific application scenario.

[0035] Further, as an optional embodiment of the present invention, the calculating the feature interaction strength between the fused abnormal features includes: Standardizing the fused abnormal features to obtain standard abnormal features; Based on the standard abnormal feature, calculating the feature variance corresponding to the fused abnormal feature; Measuring the feature probability corresponding to the standard abnormal feature, and calculating the feature information entropy corresponding to the fused abnormal feature based on the feature probability; The covariance matrix between the fused abnormal features is calculated, and the feature interaction strength between the fused abnormal features is calculated by combining the covariance matrix, the feature variance and the feature information entropy through the following formula: ; Among them, D represents the feature interaction strength between the fused abnormal features, represents the covariance matrix, represents the feature variance corresponding to the cth feature in the fused abnormal feature, represents the feature variance corresponding to the dth feature in the fused abnormal feature, c and d represent the sequence numbers corresponding to the fused abnormal feature, represents the balance coefficient, and They respectively represent the feature information entropy corresponding to the cth feature and the dth feature in the fused abnormal feature, and r represents the number of features of the fused abnormal feature.

[0036] It should be explained that the standard abnormal feature is a feature obtained after eliminating the differences between the fused abnormal features, the feature information entropy indicates the amount of information corresponding to the fused abnormal feature, the covariance matrix is ​​the covariance value between the fused abnormal features, and the balance coefficient is used to balance and adjust the information entropy. It can be set by empirical method. According to the experience of handling similar problems in the past, or the advice of experts in the field, an initial value is set, and then through continuous experiments and adjustments, the result changes of the feature interaction strength are observed to find a more suitable value. For example, you can try to set it to 0.5 first to observe whether the calculated feature interaction strength meets expectations. If not, you can gradually increase or decrease the value until a satisfactory result is obtained. Exemplarily, the standardization of the fused abnormal feature can be achieved by a normalization method; the calculation of the feature variance corresponding to the fused abnormal feature can be achieved by a variance calculation formula; the feature probability corresponding to the standard abnormal feature can be achieved by a statistical method; the feature information entropy corresponding to the fused abnormal feature can be achieved by a Shannon entropy algorithm; the calculation of the covariance matrix between the fused abnormal features can be achieved by matrix operation software or library (such as the NumPy library in Python).

[0037] The present invention determines the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction trajectory, so as to clarify the area where abnormalities will occur during laser die cutting processing, and then narrow the positioning range of the abnormal points. It should be explained that the area to be detected is a high-incidence area of ​​problems within the processing area corresponding to the equipment used in laser die cutting processing.

[0038] In detail, the method of determining the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction trajectory includes: Query the equipment performance corresponding to the equipment used in laser die cutting processing, and analyze the performance driving factors and performance output factors in the equipment performance; Combining the efficiency driving factors and the efficiency output factors, analyzing the equipment cause-effect relationship between the equipment used in laser die cutting processing; By combining the causal relationship of the equipment and the abnormal conduction trajectory, the area to be inspected of the equipment used in laser die cutting processing is determined.

[0039] It should be explained that the equipment performance is the equipment function description information corresponding to the equipment used in laser die cutting processing, the performance driving factor and the performance output factor are the function input and function output in the equipment performance, respectively, and the equipment causal relationship is the processing logic relationship between the equipment used in laser die cutting processing. For example, the equipment performance corresponding to the equipment used in laser die cutting processing can be obtained by querying the official website information of the equipment manufacturer; the analysis of the performance driving factor and the performance output factor in the equipment performance can be achieved by semantic analysis, identifying the description characters in the equipment performance, and the performance driving factor and the performance output factor according to the semantics of the description characters; the analysis of the equipment causal relationship between the equipment used in laser die cutting processing can be achieved by fishbone diagram method, constructing a fishbone diagram between the performance driving factor and the performance output factor, so as to analyze the equipment causal relationship; according to the equipment causal relationship, determine the associated equipment in the equipment, identify the abnormal equipment according to the abnormal conduction trajectory, combine the abnormal equipment and the associated equipment, determine the final abnormal equipment, obtain the equipment processing area corresponding to the final abnormal equipment, and use the equipment processing area as the area to be detected of the equipment used in laser die cutting processing.

[0040] S3. Obtain the laser performance parameters corresponding to the laser cutting die in the area to be detected, and query the cutting die material corresponding to the laser cutting die and its corresponding material optical parameters, and calculate the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters.

[0041] The present invention can obtain the laser performance parameters corresponding to the laser cutting mold in the area to be detected, thereby obtaining the performance description parameters of the laser used in the laser cutting mold processing process, and querying the cutting mold material corresponding to the laser cutting mold and its corresponding material optical parameters, so as to understand the indicator information of the response characteristics of the cutting mold material to light.

[0042] It should be explained that the laser performance parameters are the performance description parameters corresponding to the laser knife mold in the area to be inspected, the knife mold material is the material used corresponding to the laser knife mold, and the material optical parameters are indicator information of the response characteristics of the knife mold material to light. Exemplarily, the laser performance parameters corresponding to the laser knife mold in the area to be inspected can be obtained through the laser generator control panel of the area; the knife mold material corresponding to the laser knife mold and its corresponding material optical parameters can be queried from the Internet through human-computer interaction.

[0043] The present invention calculates the material absorption rate of the knife mold material with respect to the laser based on the laser performance energy and the optical parameters of the material, so as to obtain a quantitative index of the degree of absorption of the laser by the knife mold material, thereby laying a foundation for the subsequent point marking of the area to be detected. It should be explained that the material absorption rate represents the degree of absorption of the laser by the knife mold material.

[0044] In detail, the calculation of the material absorption rate of the knife mold material with respect to the laser based on the laser performance energy and the material optical parameters includes: Extracting a photon characteristic scale of the laser from the laser performance energy; Based on the optical parameters of the material, determining the material attenuation coefficient corresponding to the knife die material; Combining the photon characteristic scale and the material attenuation coefficient, the material absorption rate of the knife mold material with respect to the laser is calculated by the following formula: ; Among them, G represents the material absorption rate, represents the material attenuation coefficient, Represents the characteristic scale of photons.

[0045] It should be explained that the photon characteristic scale is the wavelength of the laser in the laser performance energy, and the material attenuation coefficient represents the absorption capacity of the knife mold material to the laser. Exemplarily, the extraction of the photon characteristic scale of the laser from the laser performance energy can be achieved through an extraction function, and the extraction function is compiled by a scripting language, such as JS scripting language.

[0046] Further, as an optional embodiment of the present invention, determining the material attenuation coefficient corresponding to the cutting die material based on the material optical parameters includes: Dispatching material size information corresponding to the die cutting material, and determining the material thickness corresponding to the die cutting material based on the material size information; Using a preset spectrophotometer to measure the material transmittance corresponding to the knife die material; In combination with the material thickness and the material transmittance, the material attenuation coefficient corresponding to the knife die material is determined by the following formula: ; Among them, K represents the material attenuation coefficient corresponding to the die material, I represents the material transmittance, and L represents the material thickness.

[0047] It should be explained that the material size information is a description of the physical properties of the material corresponding to the knife die material, the spectrophotometer is a scientific instrument for measuring the transmittance of a substance to light of different wavelengths, and the material transmittance represents the ability of the knife die material to transmit lasers. Exemplarily, the scheduling of the material size information corresponding to the knife die material can be achieved through a material database, which is a database for storing basic information corresponding to the knife die material.

[0048] S4. In combination with the preset benchmark absorption rate and the material absorption rate, the area to be detected is marked to obtain initial marking points, and the laser cutting mold of the area to be detected is imaged to obtain a regional cutting mold image, and in combination with the regional cutting mold image, the point abnormality factors corresponding to the initial marking points are analyzed.

[0049] The present invention marks the points of the area to be detected by combining a preset benchmark absorption rate and the material absorption rate, and can mark the abnormal points in the area to be detected. By performing image acquisition on the laser knife mold of the area to be detected, an image corresponding to the laser knife mold of the area to be detected can be obtained, so as to improve the recognition accuracy of subsequent abnormal marked points. It should be explained that the preset benchmark absorption rate is a reference standard corresponding to the material absorption rate. The actual measured material absorption rate is significantly different from the preset benchmark absorption rate, which indicates that the knife mold material has an abnormal absorption of the laser in this area, which may be caused by problems with the material itself, improper laser parameter settings or other factors in the processing process. The initial marking point is the point corresponding to the laser in the area to be detected where the material absorption rate is not within the preset benchmark absorption rate range. The regional knife mold image is a visualized image corresponding to the laser knife mold of the area to be detected. Further, the point marking of the area to be detected can be achieved by manual annotation; and the image acquisition of the laser knife mold in the area to be detected can be achieved by an industrial camera.

[0050] The present invention analyzes the point abnormality factors corresponding to the initial marking points by combining the regional knife die image, and then understands the abnormal causes corresponding to the initial marking points, providing a basis for subsequently improving the accuracy of identifying abnormal marking points from the initial marking points. It should be explained that the point abnormality factors are the abnormal causes corresponding to the initial marking points.

[0051] In detail, the combining of the regional die-cutting image to analyze the abnormal factors of the points corresponding to the initial marking points includes: According to the initial marking points, the regional die-cutting image is segmented to obtain a marking point image; Performing image noise reduction processing on the marked point image to obtain a noise-reduced point image, and performing image enhancement processing on the noise-reduced point image to obtain an enhanced point image; Extracting point color features corresponding to the enhanced point image, performing grayscale processing on the enhanced point image, and obtaining a point grayscale image; The point texture features corresponding to the point grayscale image are extracted, and the point abnormality factors corresponding to the initial marked points are analyzed by combining the point texture features, the point color features and a preset standard point library.

[0052] It should be explained that the marked point image is the image corresponding to the initial marked point in the regional die cutting image, the denoised point image is the image obtained after noise suppression in the marked point image, the enhanced point image is the image obtained after enhancement processing of the denoised point image, the point color feature is the color attribute corresponding to the enhanced point image, the point grayscale image is the image expressed by only a single color in the enhanced point image, the point texture feature is the microstructure representation corresponding to the point grayscale image, and the preset standard point library is a collection of various feature information of specific points on multiple normal regional die cutting images.

[0053] Exemplarily, the segmentation processing of the regional knife mold image can be achieved by a threshold segmentation method; the image noise reduction processing of the marked point image can be achieved by a low-pass filter; the image enhancement processing of the noise reduction point image can be achieved by a histogram equalization method; the extraction of the point color features corresponding to the enhanced point image can be achieved by a color histogram method; the grayscale processing of the enhanced point image can be achieved by an average value method, and the values ​​of the three channels of the enhanced point image are averaged as the grayscale value; the extraction of the point texture features corresponding to the point grayscale image can be achieved by a grayscale co-occurrence matrix method; the similarities between the point texture features, the point color features and the corresponding features in the preset standard point library are calculated respectively, and the corresponding point abnormal factors are determined according to the feature similarities. If the feature similarities are within a reasonable range, it means that the laser knife mold is normal, and the point abnormal factor is laser abnormality. If the feature similarity is not within a reasonable range, the point abnormal factor is that there are abnormal factors in the laser knife mold.

[0054] S5. Positioning the abnormal factors of the points to obtain abnormal factor positioning, and combining the abnormal factors of the points with the abnormal factor positioning to generate positioning results of abnormal points under laser die cutting processing.

[0055] The present invention can obtain the specific position corresponding to the abnormal point factor by positioning the abnormal point factor, and combine the abnormal point factor and the abnormal factor positioning to obtain the accurate positioning result of the abnormal point under laser die cutting processing. It should be explained that the abnormal factor positioning is the specific position information corresponding to the abnormal point factor. Exemplarily, the positioning processing of the abnormal point factor can be achieved by the time positioning method. According to the timestamp corresponding to the abnormal point factor, the time when the abnormal point appears is determined, and then the positioning processing of the abnormal point factor is achieved to obtain the abnormal factor positioning, such as recording the processing time of different materials on the production line. When an abnormal point appears in a certain area, it is found according to the timestamp that the abnormality occurs when a specific material enters the processing link, and then it is located that the abnormal point may be caused by a problem with the material.

[0056] Compared with the problems described in the background technology, the present invention can obtain key feature information in the physical quantity data by performing feature extraction on the physical quantity data, and convert the physical quantity data into more representative features, laying a foundation for the subsequent calculation of feature anomaly index. The present invention can obtain the evolution law between the abnormal features by analyzing the abnormal evolution context between the abnormal features, and based on the abnormal evolution context, draw the abnormal conduction trajectory corresponding to the abnormal feature, and obtain the abnormal conduction route corresponding to the abnormal feature, thereby providing a basis for the subsequent determination of the area to be detected. The present invention can obtain the performance description parameters of the laser used in the laser die cutting process by obtaining the laser performance parameters corresponding to the laser die cutting area to be detected, and By querying the knife die material corresponding to the laser knife die and its corresponding material optical parameters, the index information of the knife die material's response characteristics to light can be understood. The present invention combines the preset benchmark absorption rate and the material absorption rate to mark the points in the area to be detected, and can mark the points with abnormalities in the area to be detected. By performing image acquisition on the laser knife die in the area to be detected, the image corresponding to the laser knife die in the area to be detected can be obtained, so as to improve the recognition accuracy of subsequent abnormal marked points. The present invention locates the abnormal factors of the points, and can obtain the specific positions corresponding to the abnormal factors of the points. Combine the abnormal factors of the points and the positioning of the abnormal factors, and then obtain the accurate positioning results of the abnormal points under laser knife die processing. Therefore, the present invention proposes a method for locating abnormal points under laser knife die processing, so as to improve the accuracy of abnormal point positioning under laser knife die processing.

[0057] Embodiment 2: like Figure 2 , which is a functional module diagram of an abnormal point positioning system for laser die cutting processing provided by an embodiment of the present invention.

[0058] The abnormal point positioning system 100 for realizing laser die cutting processing described in the present invention can be installed in an electronic device. According to the functions realized, the abnormal point positioning system 100 for realizing laser die cutting processing can include an abnormal feature recognition module 101, a region division module 102, a material absorption rate calculation module 103, an abnormal factor analysis module 104 and an abnormal positioning module 105. The module described in 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.

[0059] In this embodiment, the functions of each module / unit are as follows: The abnormal feature recognition module 101 is used to collect physical quantity data of the equipment used in laser die cutting processing, perform feature extraction on the physical quantity data to obtain physical quantity features, calculate feature abnormality indexes corresponding to the physical quantity features, and identify abnormal features in the physical quantity features based on the feature abnormality indexes; The region division module 102 is used to analyze the abnormal evolution context between the abnormal features, draw the abnormal conduction trajectory corresponding to the abnormal features based on the abnormal evolution context, and determine the area to be detected of the equipment used in the laser die cutting process based on the abnormal conduction trajectory; The material absorption rate calculation module 103 is used to obtain the laser performance parameters corresponding to the laser cutting die of the area to be detected, and query the cutting die material corresponding to the laser cutting die and its corresponding material optical parameters, and calculate the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters; The abnormal factor analysis module 104 is used to mark the points of the area to be detected in combination with the preset benchmark absorption rate and the material absorption rate to obtain initial marking points, collect images of the laser cutting mold of the area to be detected to obtain a regional cutting mold image, and analyze the abnormal factors of the points corresponding to the initial marking points based on the regional cutting mold image; The abnormality positioning module 105 is used to perform positioning processing on the abnormal factors of the points to obtain abnormal factor positioning, and to generate positioning results of abnormal points under laser die cutting processing by combining the abnormal factors of the points and the abnormal factor positioning.

[0060] In detail, each module described in the system 100 for realizing abnormal point positioning in laser die cutting processing described in the embodiment of the present application is used in the same manner as described above. Figure 1 The technical means described in the text for realizing the abnormal point positioning method under laser die cutting processing is the same as that described in the text, and can produce the same technical effect, which will not be repeated here.

[0061] 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.

[0062] 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. A method for locating abnormal points in laser die cutting processing, characterized in that: The method comprises: Collecting physical quantity data of the equipment used in laser die cutting processing, performing feature extraction on the physical quantity data to obtain physical quantity features, calculating feature anomaly indexes corresponding to the physical quantity features, and identifying abnormal features in the physical quantity features based on the feature anomaly indexes; Analyze the abnormal evolution context between the abnormal features, draw the abnormal conduction trajectory corresponding to the abnormal features based on the abnormal evolution context, and determine the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction trajectory; Acquire the laser performance parameters corresponding to the laser cutting die of the area to be detected, query the cutting die material corresponding to the laser cutting die and its corresponding material optical parameters, and calculate the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters; In combination with the preset reference absorption rate and the material absorption rate, the area to be detected is marked to obtain initial marking points, an image of the laser cutting die of the area to be detected is collected to obtain a regional cutting die image, and based on the regional cutting die image, the abnormal factors of the points corresponding to the initial marking points are analyzed; The abnormal factors at the points are positioned to obtain abnormal factor positioning, and the abnormal factors at the points and the abnormal factor positioning are combined to generate positioning results of abnormal points under laser die cutting processing.

2. The method for locating abnormal points in laser die cutting processing according to claim 1, characterized in that: The step of extracting features from the physical quantity data to obtain physical quantity features includes: Smoothing the physical quantity data to obtain smoothed physical quantity data; Analyzing data attributes corresponding to the smoothed physical quantity data, and extracting initial physical quantity features corresponding to the smoothed physical quantity data; Parsing the physical quantity label corresponding to the initial physical quantity feature, and analyzing the correlation between the data attribute and the physical quantity label; Based on the correlation, feature selection processing is performed on the initial physical quantity feature to obtain a physical quantity feature.

3. The method for locating abnormal points in laser die cutting processing according to claim 1, characterized in that: The calculating the characteristic anomaly index corresponding to the physical quantity characteristic comprises: Calculating a characteristic equilibrium value corresponding to the physical quantity characteristic, and calculating a characteristic standard deviation corresponding to the physical quantity characteristic based on the characteristic equilibrium value; Combining the characteristic equilibrium value and the characteristic standard deviation, the characteristic anomaly index corresponding to the physical quantity characteristic is calculated by the following formula: ； Among them, A represents the characteristic anomaly index corresponding to the physical quantity characteristic, Indicates the value corresponding to the ath feature in the physical quantity feature, represents the characteristic equilibrium value, It represents the characteristic standard deviation, a represents the serial number corresponding to the physical quantity characteristic, and q represents the quantity corresponding to the material quantity characteristic.

4. The method for locating abnormal points in laser die cutting processing according to claim 1, characterized in that: The analyzing the abnormal evolution context between the abnormal features includes: Performing clustering processing on the abnormal features to obtain clustered abnormal features; Calculating the characteristic distance value between the clustered abnormal features, and analyzing the abnormal evolution trend corresponding to the clustered abnormal features based on the characteristic distance value; Performing feature fusion processing on the clustered abnormal features to obtain fused abnormal features; Calculating the feature interaction strength between the fused abnormal features, and determining the associated abnormal features in the fused abnormal features based on the feature interaction strength; The associated abnormal features and the abnormal evolution trend are combined to generate the abnormal evolution context between the abnormal features.

5. The method for locating abnormal points in laser die cutting processing according to claim 4, characterized in that: The calculating the feature interaction strength between the fused abnormal features includes: Standardizing the fused abnormal features to obtain standard abnormal features; Based on the standard abnormal feature, calculating the feature variance corresponding to the fused abnormal feature; Measuring the feature probability corresponding to the standard abnormal feature, and calculating the feature information entropy corresponding to the fused abnormal feature based on the feature probability; The covariance matrix between the fused abnormal features is calculated, and the feature interaction strength between the fused abnormal features is calculated by combining the covariance matrix, the feature variance and the feature information entropy through the following formula: ； Among them, D represents the feature interaction strength between the fused abnormal features, represents the covariance matrix, represents the feature variance corresponding to the cth feature in the fused abnormal feature, represents the feature variance corresponding to the dth feature in the fused abnormal feature, c and d represent the sequence numbers corresponding to the fused abnormal feature, represents the balance coefficient, and They respectively represent the feature information entropy corresponding to the cth feature and the dth feature in the fused abnormal feature, and r represents the number of features of the fused abnormal feature.

6. The method for locating abnormal points in laser die cutting processing according to claim 1, characterized in that: The method of determining the area to be detected of the equipment used in laser die cutting processing based on the abnormal conduction track includes: Query the equipment performance corresponding to the equipment used in laser die cutting processing, and analyze the performance driving factors and performance output factors in the equipment performance; Combining the efficiency driving factors and the efficiency output factors, analyzing the equipment cause-effect relationship between the equipment used in laser die cutting processing; By combining the causal relationship of the equipment and the abnormal conduction trajectory, the area to be inspected of the equipment used in laser die cutting processing is determined.

7. The method for locating abnormal points in laser die cutting processing according to claim 1, characterized in that: The step of calculating the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters comprises: Extracting a photon characteristic scale of the laser from the laser performance energy; Based on the optical parameters of the material, determining the material attenuation coefficient corresponding to the knife die material; Combining the photon characteristic scale and the material attenuation coefficient, the material absorption rate of the knife mold material with respect to the laser is calculated by the following formula: ； Among them, G represents the material absorption rate, represents the material attenuation coefficient, Represents the characteristic scale of photons.

8. The method for locating abnormal points in laser die cutting processing according to claim 7, characterized in that: Determining the material attenuation coefficient corresponding to the cutting die material based on the material optical parameters includes: Dispatching material size information corresponding to the die cutting material, and determining the material thickness corresponding to the die cutting material based on the material size information; Using a preset spectrophotometer to measure the material transmittance corresponding to the knife die material; In combination with the material thickness and the material transmittance, the material attenuation coefficient corresponding to the knife die material is determined by the following formula: ； Among them, K represents the material attenuation coefficient corresponding to the die material, I represents the material transmittance, and L represents the material thickness.

9. The method for locating abnormal points in laser die cutting processing according to claim 1, characterized in that: The combining of the regional die-cutting image to analyze the abnormal factors of the points corresponding to the initial marking points includes: According to the initial marking points, the regional die-cutting image is segmented to obtain a marking point image; Performing image noise reduction processing on the marked point image to obtain a noise-reduced point image, and performing image enhancement processing on the noise-reduced point image to obtain an enhanced point image; Extracting point color features corresponding to the enhanced point image, performing grayscale processing on the enhanced point image, and obtaining a point grayscale image; The point texture features corresponding to the point grayscale image are extracted, and the point abnormality factors corresponding to the initial marked points are analyzed by combining the point texture features, the point color features and a preset standard point library.

10. The abnormal point positioning system used to realize laser die cutting processing is characterized in that: The system comprises: An abnormal feature recognition module is used to collect physical quantity data of the equipment used in laser die cutting processing, perform feature extraction on the physical quantity data to obtain physical quantity features, calculate feature abnormality indexes corresponding to the physical quantity features, and identify abnormal features in the physical quantity features based on the feature abnormality indexes; A region division module is used to analyze the abnormal evolution context between the abnormal features, draw the abnormal conduction trajectory corresponding to the abnormal features based on the abnormal evolution context, and determine the area to be detected of the equipment used in the laser die cutting process based on the abnormal conduction trajectory; A material absorption rate calculation module is used to obtain the laser performance parameters corresponding to the laser cutting die of the area to be detected, and query the cutting die material corresponding to the laser cutting die and its corresponding material optical parameters, and calculate the material absorption rate of the cutting die material with respect to the laser based on the laser performance energy and the material optical parameters; The abnormal factor analysis module is used to mark the points of the area to be detected in combination with the preset benchmark absorption rate and the material absorption rate to obtain initial marking points, collect images of the laser cutting mold of the area to be detected to obtain a regional cutting mold image, and analyze the abnormal factors of the points corresponding to the initial marking points based on the regional cutting mold image; The abnormality positioning module is used to position the abnormal factors of the points to obtain the positioning of the abnormal factors, and to generate the positioning results of the abnormal points under laser die cutting processing by combining the abnormal factors of the points and the positioning of the abnormal factors.

Images