A cutting tool groove detection method and system based on artificial intelligence

The AI-based cutting tool groove detection method and system addresses the limitations of traditional methods by using ultrasonic sensors and point cloud technology for precise and efficient groove parameter capture and classification, enhancing accuracy and reliability in cutting tool inspection.

CN118864465BActive Publication Date: 2024-12-06NANTONG BAIRUILI ELECTRIC TOOLS CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411341422.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-25
Publication Date
2024-12-06
Estimated Expiration
2044-09-25

AI Technical Summary

Technical Problem

Traditional cutting tool groove detection methods rely on manual visual inspection or simple measuring tools, which are time-consuming, prone to human error, and lack accuracy, especially for complex or tiny structures, failing to meet the high-precision and high-efficiency requirements of modern manufacturing.

Method used

A cutting tool groove detection method and system utilizing artificial intelligence, combining ultrasonic sensors for synchronous detection, point cloud technology, and machine learning algorithms to accurately capture and classify groove parameters and shapes, including ultrasonic signal data processing, point cloud data analysis, and three-dimensional model visualization.

Benefits of technology

Enhances detection accuracy, reduces human misjudgment, and improves production efficiency by providing precise capture and classification of groove parameters, ensuring high-quality standards and market competitiveness through reliable quality control.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118864465B_ABST
    Figure CN118864465B_ABST
Patent Text Reader

Abstract

The present invention discloses a cutting tool groove detection method and system based on artificial intelligence, which relates to the field of cutting processing technology. The method includes: obtaining the groove parameters of the cutting tool; obtaining the point cloud data of the groove parameters based on the point cloud technology, and analyzing the groove shape according to the point cloud data of the groove parameters to obtain the classification result of the groove shape; constructing a groove three-dimensional model based on the classification result of the groove shape, and visually displaying it. The present invention uses ultrasonic sensor technology for synchronous detection, which not only significantly improves the accuracy of cutting tool groove detection, but also realizes the precise capture and measurement of groove parameters, improves the reliability of the cutting tool groove detection process, and ensures the high quality standards of cutting tools in industrial production; at the same time, it promotes a significant improvement in production efficiency, and provides strong technical support for the sustainable development and quality control of the manufacturing industry.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of cutting processing technology, and in particular to a cutting tool groove detection method and system based on artificial intelligence. Background Art

[0002] Cutting tools are tools used for cutting in mechanical manufacturing and are widely used in industrial production. Cutting tools are usually made of steel, carbide, ceramics and other materials, and process workpieces by cutting, planing, drilling and other methods. As an important part of modern manufacturing, cutting tools directly affect processing quality and efficiency, and play a key role in improving labor productivity and controlling costs.

[0003] In the design of cutting tools, grooves are an important structure. They work closely with lathes and other equipment to cut holes or surfaces in workpieces to form the required channels or paths. There are many types of cutting tool grooves, and each tool has a unique shape, size and properties to meet different processing requirements. The shape, size and position of the groove directly affect the cutting effect and processing quality; therefore, it is very important to carefully inspect the grooves. Through inspection, the quality and performance of the cutting tool can be evaluated to ensure that it can meet specific processing requirements.

[0004] In the traditional cutting tool groove detection process, manual visual inspection or simple measuring tools are usually used. This method not only consumes a lot of time and manpower, but is also easily affected by the operator's subjective judgment, resulting in low detection accuracy. In addition, for complex or tiny groove structures, manual detection is more difficult and more likely to lead to error accumulation, making it difficult to meet the needs of modern manufacturing for high-precision and high-efficiency detection.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In view of the problems in the related art, the present invention proposes a cutting tool groove detection method and system based on artificial intelligence to overcome the above-mentioned technical problems existing in the existing related art.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to a first aspect of the present invention, there is provided a cutting tool groove detection method based on artificial intelligence, the cutting tool groove detection method based on artificial intelligence comprising:

[0009] S1, obtaining groove parameters of a cutting tool;

[0010] S2. acquiring point cloud data of groove parameters based on point cloud technology, and analyzing the groove shape according to the point cloud data of the groove parameters to obtain a classification result of the groove shape;

[0011] S3. Build a groove three-dimensional model based on the classification results of the groove shape and perform a visual display.

[0012] Furthermore, obtaining the groove parameters of the cutting tool includes:

[0013] S11, synchronously detecting the cutting tool through a preset ultrasonic sensor to obtain ultrasonic signal data of the groove;

[0014] S12, preprocessing the ultrasonic signal data, and obtaining the distance difference between the groove and each ultrasonic sensor through an ultrasonic time difference algorithm;

[0015] S13, calculating by an ultrasonic attenuation algorithm to obtain a distance ratio between the groove and each ultrasonic sensor;

[0016] S14. Using the distance difference and the distance ratio as observation values ​​for data fusion, an iterative expansion filtering algorithm is used to iteratively calculate the groove position to obtain groove parameters.

[0017] Furthermore, the ultrasonic signal data is preprocessed, and the distance difference between the groove and each ultrasonic sensor is obtained by an ultrasonic time difference algorithm, including:

[0018] S121, based on the preprocessed ultrasonic signal data, identifying and marking the arrival point of the reflected wave signal data through edge detection;

[0019] S122, calculating the time difference of each ultrasonic sensor receiving reflected wave signal data according to a plurality of ultrasonic sensors;

[0020] S123, combining the preset sound speed with the obtained time difference, and using the distance difference formula to calculate the distance difference between the groove and each sensor.

[0021] Furthermore, the distance ratios between the groove and each ultrasonic sensor are calculated by using an ultrasonic attenuation algorithm, including:

[0022] S131, using an ultrasonic signal decomposition algorithm, decomposing the preprocessed ultrasonic signal data into a number of intrinsic mode function components;

[0023] S132, performing correlation analysis on each intrinsic modal function component to obtain a target intrinsic modal function component associated with each intrinsic modal function component and ultrasonic reflection signal data;

[0024] S133, calculating the energy entropy values ​​of several target intrinsic modal function components, and combining with the energy attenuation model, obtaining the distance between the groove and each ultrasonic sensor;

[0025] S134 . Compare the distances between the groove and each ultrasonic sensor to obtain a distance ratio between the groove and each ultrasonic sensor.

[0026] Further, performing correlation analysis on each intrinsic modal function component to obtain a target intrinsic modal function component associated with each intrinsic modal function component and the ultrasonic reflection signal data includes:

[0027] S1321, by calculating the autocorrelation function of each intrinsic mode function component, identifying the position and height of the peak in the autocorrelation function;

[0028] S1322, by calculating the correlation coefficient between each intrinsic mode function component and combining the correlation coefficient with the peak position and height in the autocorrelation function, comprehensively analyzing the correlation between each intrinsic mode function component and the ultrasonic reflection signal data;

[0029] S1323. Based on the comprehensive analysis results, the correlation between each intrinsic modal function component and the ultrasonic reflection signal data is sorted, and the first N intrinsic modal function components are selected as the target intrinsic modal function components.

[0030] Furthermore, the calculation formula of energy entropy is:

[0031]

[0032] In the formula, H represents the energy entropy value;

[0033] m Indicates time difference;

[0034] k represents the target intrinsic mode function component;

[0035] i represents a natural number;

[0036] p k Indicates the percentage of the energy of the kth target intrinsic mode function component of the ultrasonic signal data to the total energy.

[0037] Furthermore, based on the point cloud technology, point cloud data of the groove parameters are obtained, and the groove shape is analyzed according to the point cloud data of the groove parameters, and the classification results of the groove shape are obtained, including:

[0038] S21, scanning the acquired groove position information by a laser scanner to obtain groove point cloud data, and preprocessing the groove point cloud data;

[0039] S22, extracting feature data from the preprocessed point cloud data to obtain a feature data set;

[0040] S23, using a decision tree algorithm to perform shape annotation on the groove point cloud data, and classifying the groove shapes into regular shapes and irregular shapes according to a preset classification standard;

[0041] S24, generating a label file according to the feature data set and the classification result;

[0042] S25. Based on the label file, a groove shape classification model is established using a machine learning algorithm;

[0043] S26, inputting the groove point cloud data into the groove shape classification model to obtain the groove shape classification result.

[0044] Further, the classification criteria include the aspect ratio of the groove, angle measurement, edge line and area to perimeter ratio.

[0045] According to a second aspect of the present invention, there is also provided a cutting tool groove detection system based on artificial intelligence, the cutting tool groove detection system based on artificial intelligence comprising:

[0046] A parameter acquisition module, used to acquire groove parameters of a cutting tool;

[0047] A point cloud analysis module is used to obtain point cloud data of groove parameters based on point cloud technology, and analyze the groove shape according to the point cloud data of the groove parameters to obtain a classification result of the groove shape;

[0048] A classification and display module is used to construct a groove three-dimensional model based on the classification results of the groove shape and to perform visual display;

[0049] The parameter acquisition module is connected with the classification display module through the point cloud analysis module.

[0050] According to a third aspect of the present invention, a computer-readable storage medium is provided.

[0051] Furthermore, a computer program is stored on the computer-readable storage medium, and the steps of the above method are implemented when the computer program is executed by the processor.

[0052] The beneficial effects of the present invention are:

[0053] 1. The present invention uses ultrasonic sensor technology for synchronous detection, which not only significantly improves the accuracy of cutting tool groove detection, but also realizes the precise capture and measurement of groove parameters, improves the reliability of the cutting tool groove detection process, and ensures the high quality standards of cutting tools in industrial production; at the same time, it promotes a significant improvement in production efficiency and provides strong technical support for the sustainable development and quality control of the manufacturing industry.

[0054] 2. The present invention significantly improves the accuracy of cutting tool groove detection and effectively reduces human misjudgment by combining the groove positioning algorithm with point cloud technology; at the same time, by analyzing the shape of the groove, the groove can be accurately classified, ensuring the stability and reliability of the detection process, providing strong data support for subsequent precise processing and quality control, making the identification and classification of different grooves more efficient, thereby promoting the continuous improvement and enhancement of cutting tool quality management.

[0055] 3. By utilizing three-dimensional model visualization technology, the present invention can obtain more comprehensive and in-depth groove information, improve the ability to quickly identify and accurately classify groove positions and categories during cutting tool groove detection, which not only improves the efficiency of the production quality control process and shortens the detection cycle, but also ensures that product quality standards are more strictly complied with, thereby enhancing the market competitiveness of the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 is a flow chart of a cutting tool groove detection method based on artificial intelligence according to an embodiment of the present invention;

[0058] Figure 2 It is a principle block diagram of a cutting tool groove detection system based on artificial intelligence according to an embodiment of the present invention.

[0059] In the figure:

[0060] 1. Parameter acquisition module; 2. Point cloud analysis module; 3. Classification display module. DETAILED DESCRIPTION

[0061] To further illustrate each embodiment, the present invention provides drawings, which are part of the disclosure of the present invention and are mainly used to illustrate the embodiments. They can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, ordinary technicians in this field should be able to understand other possible implementation methods and advantages of the present invention.

[0062] According to an embodiment of the present invention, a cutting tool groove detection method and system based on artificial intelligence are provided.

[0063] The present invention is further described with reference to the accompanying drawings and specific embodiments. Figure 1 As shown, according to an embodiment of the present invention, a cutting tool groove detection method based on artificial intelligence includes:

[0064] S1. Obtaining groove parameters of a cutting tool.

[0065] In the description of the present invention, obtaining the groove parameters of the cutting tool includes:

[0066] S11. Synchronously detect the cutting tool through a preset ultrasonic sensor to obtain ultrasonic signal data of the groove.

[0067] It should be noted that, since several ultrasonic sensors are inspected synchronously, a time synchronization algorithm needs to be used to calibrate the time difference between the ultrasonic sensors to ensure the synchronization of ultrasonic signal data in time; wherein, calibrating the time difference between the ultrasonic sensors by the time synchronization algorithm includes the following steps:

[0068] An ultrasonic sensor is randomly selected as a reference sensor, and a synchronization signal (such as a pulse signal of a specific frequency) is sent from the reference sensor to other ultrasonic sensors. Other ultrasonic sensors receive the synchronization signal and record the receiving timestamp. The timestamp of the signal sent by the reference sensor is compared with the timestamp of the signal received by each ultrasonic sensor, and the time difference of each ultrasonic sensor relative to the reference sensor is calculated. According to the calculated time difference, the time of each ultrasonic sensor is adjusted to achieve time synchronization. The above steps are repeated for multiple rounds of time synchronization to optimize the time synchronization accuracy between each ultrasonic sensor.

[0069] S12, pre-processing the ultrasonic signal data, and obtaining the distance difference between the groove and each ultrasonic sensor through an ultrasonic time difference algorithm.

[0070] It should be noted that the acquired ultrasonic signal data is used as the input data of the adaptive filter, and the input data is framed in combination with the preset reference data; the framed data is iteratively processed through the adaptive filter, wherein the iterative processing includes calculating the error data and updating the weight vector; the above steps are repeated to update the weights of the framed data until the adaptive filter reaches the preset number of iterations; the output data of the adaptive filter is the denoised ultrasonic signal data.

[0071] The denoised ultrasonic signal data is calculated using the Z-score standardization calculation formula to convert the ultrasonic signal data into a unified format with similar scale and range. The Z-score standardization calculation formula is:

[0072]

[0073] In the formula, Z Expressed as standard scores;

[0074] X It is expressed as the value of a specific data in the ultrasonic signal data;

[0075] It is expressed as the mean value of ultrasonic signal data;

[0076] Expressed as the standard deviation of the ultrasonic signal data.

[0077] In the description of the present invention, preprocessing the ultrasonic signal data and obtaining the distance difference between the groove and each ultrasonic sensor by using the ultrasonic time difference algorithm includes:

[0078] S121. Based on the preprocessed ultrasonic signal data, identify and mark the arrival point of the reflected wave signal data through edge detection.

[0079] It should be noted that the preprocessed ultrasonic signal data is converted into an image format, wherein the time series data can be mapped to the horizontal axis (horizontal direction) of the image, and the signal amplitude can be mapped to the vertical axis (vertical direction) of the image; the Sobel operator is used to calculate the gradient and direction of each pixel in the image; at each pixel, the gradient values ​​of adjacent pixels are compared along the gradient direction, combined with the preset high threshold and low threshold, if the gradient value is greater than the high threshold, the pixel is marked as an edge point, if the gradient value is between the high and low thresholds, check whether the pixel is connected to a pixel with a gradient value higher than the high threshold, if so, mark the pixel as an edge point; by finding continuous edge lines in the image, connect the edge points to form a complete edge; by finding the local maximum on the edge line, identify the arrival point of the reflected wave; map the edge points in the image back to the time axis of the preprocessed ultrasonic signal data, and mark the corresponding time or position.

[0080] S122. Calculate, based on a plurality of ultrasonic sensors, the time difference between each ultrasonic sensor receiving reflected wave signal data.

[0081] It should be noted that several ultrasonic sensors are used to send ultrasonic signal data simultaneously and record the timestamp of the sent signal data; when the ultrasonic signal data is reflected by the groove, each ultrasonic sensor receives the reflected wave signal and records the timestamp of the received signal; for any two ultrasonic sensors, the time difference is obtained by the difference between the timestamps of their received signals.

[0082] S123, combining the preset sound speed with the obtained time difference, and using the distance difference formula to calculate the distance difference between the groove and each sensor.

[0083] It should be noted that the propagation speed of the ultrasonic wave in the cutting tool, that is, the speed of sound, is determined according to the material of the cutting tool; the distance difference between the groove and each sensor is calculated by the distance difference formula in combination with the obtained time difference, where the distance difference formula is calculated as follows:

[0084]

[0085] In the formula, d is the distance difference, i.e., the distance difference from the groove to the two sensors;

[0086] c is the propagation speed of ultrasonic waves in the medium, that is, the speed of sound;

[0087] Δt It is the time difference between the two sensors receiving the reflected waves.

[0088] S13. Calculate using an ultrasonic attenuation algorithm to obtain a distance ratio between the groove and each ultrasonic sensor.

[0089] In the description of the present invention, the distance ratio between the groove and each ultrasonic sensor is calculated by an ultrasonic attenuation algorithm, including:

[0090] S131. Decompose the preprocessed ultrasonic signal data into a number of intrinsic mode function components using an ultrasonic signal decomposition algorithm.

[0091] It should be noted that the ultrasonic signal decomposition algorithm is an empirical mode decomposition algorithm, which can decompose complex signals into a series of intrinsic mode functions.

[0092] Using the ultrasonic signal decomposition algorithm, decomposing the preprocessed ultrasonic signal data into several intrinsic mode function components includes the following steps:

[0093] Step 1: Filter out the first intrinsic mode function component from the preprocessed ultrasonic signal data.

[0094] Identify all local maximum and minimum points of the preprocessed ultrasonic signal data, and construct an upper envelope through the local maximum points and a lower envelope through the local minimum points by using cubic spline interpolation technology; calculate the average values ​​of the upper envelope and the lower envelope to obtain the local mean of the ultrasonic signal data; subtract the local mean from the preprocessed ultrasonic signal data to obtain a new ultrasonic signal data; repeat the above process until the average value of the envelope of the new ultrasonic signal data is zero or small enough, indicating that the first intrinsic mode function component has been extracted.

[0095] Step 2: Separate the first intrinsic modal function component extracted in step 1 from the preprocessed ultrasonic signal data, and repeat step 1 for the remaining ultrasonic signal data until no new intrinsic modal function components can be extracted, that is, stop when the remaining ultrasonic signal data can no longer be decomposed into components that meet the intrinsic modal function component conditions.

[0096] S132, performing correlation analysis on each intrinsic modal function component to obtain a target intrinsic modal function component associated with each intrinsic modal function component and the ultrasonic reflection signal data.

[0097] In the description of the present invention, performing correlation analysis on each intrinsic modal function component to obtain a target intrinsic modal function component associated with each intrinsic modal function component and ultrasonic reflection signal data includes:

[0098] S1321. Calculate the autocorrelation function of each intrinsic mode function component to identify the position and height of the peak in the autocorrelation function.

[0099] It should be noted that the calculation formula of the autocorrelation function is:

[0100]

[0101] In the formula, R represents the autocorrelation function;

[0102] N Indicates the length of the ultrasonic reflection signal;

[0103] τ Indicates the delay time;

[0104] t Indicates time;

[0105] X represents the intrinsic mode function component;

[0106] Represents the mean value of the ultrasonic reflection signal.

[0107] Identifying the location and height of the peak in the autocorrelation function involves the following steps:

[0108] Traverse the autocorrelation function and obtain the local maximum point by comparing the autocorrelation value of each point with the points before and after it; take the position of the local maximum point as the peak position of the autocorrelation function and correspond it to the specific delay time in the ultrasonic reflection signal data; record the function value of each peak point and use these function values ​​as the peak height of the autocorrelation function.

[0109] In addition, the delay time corresponding to the peak position can be used to identify the periodicity in the ultrasonic reflection signal data. If the corresponding time interval is an integer multiple of the ultrasonic reflection signal data period, it indicates that the ultrasonic reflection signal data has a frequency component of that period. The peak height indicates the intensity or energy of the ultrasonic reflection signal data in the corresponding period. The higher the height, the greater the proportion of the period in the ultrasonic reflection signal data.

[0110] S1322. By calculating the correlation coefficient between each intrinsic mode function component and combining the correlation coefficient with the peak position and height in the autocorrelation function, the correlation between each intrinsic mode function component and the ultrasonic reflection signal data is comprehensively analyzed.

[0111] It should be noted that by calculating the correlation coefficient between each intrinsic mode function component and combining the correlation coefficient with the peak position and height in the autocorrelation function, the comprehensive analysis of the correlation between each intrinsic mode function component and the ultrasonic reflection signal data includes the following steps:

[0112] The Pearson correlation coefficient formula is used to calculate the correlation coefficient between each intrinsic mode function component, where the Pearson correlation coefficient formula is:

[0113]

[0114] In the formula, r represents the Pearson correlation coefficient;

[0115] n represents the number of intrinsic mode function components;

[0116] i Indicates the index quantity;

[0117] X i represents the observed value of the random intrinsic mode function component;

[0118] represents the mean of the observed values ​​of the random intrinsic mode function components;

[0119] Y i represents the observed value of the random intrinsic mode function component;

[0120] Represents the mean of the observations of the random intrinsic mode function components.

[0121] According to the obtained correlation coefficient, a correlation coefficient matrix is ​​formed, and the intrinsic mode function component pairs with higher absolute values ​​of correlation coefficients are identified in the correlation coefficient matrix; the peak position and height in the autocorrelation function are combined with the correlation coefficient matrix to analyze the autocorrelation peak characteristics of the intrinsic mode function component pairs with higher absolute values; based on the analysis results, a correlation model is constructed to link the autocorrelation peak of the intrinsic mode function component with the specific characteristics of the ultrasonic reflection signal data.

[0122] S1323. Based on the comprehensive analysis results, the correlation between each intrinsic modal function component and the ultrasonic reflection signal data is sorted, and the first N intrinsic modal function components are selected as the target intrinsic modal function components.

[0123] S133, calculating the energy entropy values ​​of several target intrinsic modal function components, and combining with the energy attenuation model to obtain the distance between the groove and each ultrasonic sensor.

[0124] In the description of the present invention, the calculation formula of energy entropy value is:

[0125]

[0126] In the formula, H represents the energy entropy value;

[0127] m Indicates time difference;

[0128] k represents the target intrinsic mode function component;

[0129] i represents a natural number;

[0130] p k Indicates the percentage of the energy of the kth target intrinsic mode function component of the ultrasonic signal data to the total energy.

[0131] It should be supplemented that the geometric relationship between the ultrasonic sensor and the groove is analyzed, and an energy attenuation model is established based on the principle of sound wave propagation. The energy attenuation model and the energy entropy value of the target inherent modal function component are combined with the measured ultrasonic signal energy to calculate the distance between the groove and each ultrasonic sensor. The calculated distance is compared with the actual measured distance, and the energy attenuation model parameters are adjusted according to the comparison results to optimize the performance of the energy attenuation model.

[0132] S134 . Compare the distances between the groove and each ultrasonic sensor to obtain a distance ratio between the groove and each ultrasonic sensor.

[0133] S14. Using the distance difference and the distance ratio as observation values ​​for data fusion, an iterative expansion filtering algorithm is used to iteratively calculate the groove position to obtain groove parameters.

[0134] It should be noted that the state transfer equation and the observation equation are established, where the state transfer equation is:

[0135]

[0136] In the formula, r(k+1) represents the state transition function;

[0137] I 3 represents the unit matrix;

[0138] r(k) represents the state vector function;

[0139] represents the process noise function;

[0140] k Indicates time.

[0141] The observation equation is:

[0142]

[0143] In the formula, y(k) represents the observation function;

[0144] represents a nonlinear observation function;

[0145] v(k) represents the observation noise function;

[0146] k Indicates time.

[0147] Use the state transfer equation to predict the state vector and covariance matrix at the next moment, and use the observed data to update the state vector and covariance matrix; repeat the above steps to determine whether the iteration converges, where the convergence condition is that the change in the covariance for multiple consecutive iterations is less than the preset threshold; when the iteration converges, output the final groove position information.

[0148] In addition, the covariance matrix at the next moment is:

[0149]

[0150] In the formula, P represents the covariance matrix;

[0151] F k represents the state transfer Jacobian matrix;

[0152] represents the transpose of the state transfer Jacobian matrix;

[0153] Q k represents the process noise covariance;

[0154] k Indicates time;

[0155] T Represents the transpose of a matrix.

[0156] The updated covariance matrix is:

[0157]

[0158] In the formula, P represents the covariance matrix;

[0159] I 3 represents the unit matrix;

[0160] K k represents the Kalman gain;

[0161] H k represents the observation Jacobian matrix;

[0162] k Indicates time.

[0163] S2. Obtaining point cloud data of groove parameters based on point cloud technology, and analyzing the groove shape according to the point cloud data of the groove parameters to obtain a classification result of the groove shape.

[0164] In the description of the present invention, point cloud data of groove parameters are obtained based on point cloud technology, and the groove shape is analyzed according to the point cloud data of the groove parameters, and the classification results of the groove shape are obtained, including:

[0165] S21, scanning the acquired groove position information by means of a laser scanner to obtain groove point cloud data, and preprocessing the groove point cloud data.

[0166] It should be noted that the preprocessing of the groove point cloud data includes filtering. Specifically, the acquired groove point cloud data is traversed to find all neighboring points in the neighborhood of each point (usually a sphere or cube neighborhood); for each point in the neighborhood, the Gaussian weight is calculated according to its distance from the center point; the calculated weight is used to perform weighted averaging on the points in the neighborhood to obtain the filtered point coordinates; the iterative filtering process is repeated to further smooth the groove point cloud data.

[0167] The calculation formula of Gaussian weight is:

[0168]

[0169] In the formula, w represents Gaussian weight;

[0170] d Indicates the distance between the field point and the central store;

[0171] σ Represents the standard deviation of the Gaussian distribution.

[0172] The calculated weights are used to perform weighted averaging on the points in the neighborhood, and the calculation formula for the filtered point coordinates is:

[0173]

[0174]

[0175]

[0176] In the formula, Represents the horizontal coordinate of the point after filtering;

[0177] Represents the ordinate of the point after filtering;

[0178] Indicates the vertical coordinate of the point after filtering;

[0179] xRepresents the horizontal coordinate of each point in the neighborhood;

[0180] y Represents the ordinate of each point in the neighborhood;

[0181] z Represents the vertical coordinates of each point in the neighborhood;

[0182] n Represents the total number of neighborhood points;

[0183] i represents a natural number;

[0184] w represents Gaussian weights.

[0185] S22. Extract feature data from the preprocessed point cloud data to obtain a feature data set.

[0186] It should be noted that the feature data includes position features, geometric features and shape features;

[0187] Among them, position features include coordinates, distance, etc.; geometric features include curvature, normal, main direction, etc.; shape features include diameter, area, volume, etc.

[0188] S23. Use a decision tree algorithm to perform shape annotation on the groove point cloud data, and classify the groove shapes into regular shapes and irregular shapes according to a preset classification standard.

[0189] In the description of the present invention, the classification criteria include the aspect ratio of the groove, the angle measurement, the edge line and the ratio of the area to the perimeter.

[0190] It should be noted that if the aspect ratio of the groove is a constant, that is, the length-to-width ratio is a constant, the shape of the groove is a regular shape. Conversely, if the aspect ratio of the groove is not a constant, the shape of the groove is an irregular shape.

[0191] If the angle measurement of the groove includes common angles such as 90°, 45° and 30°, the groove shape is regular. Conversely, if the angle measurement of the groove does not include these common angles, the groove shape is irregular.

[0192] If the edge line of the groove is smooth and continuous without sharp corners or discontinuities, the shape of the groove is regular; otherwise, if there are sharp corners or discontinuities, the shape of the groove is irregular;

[0193] If the ratio of the area of ​​the groove to the perimeter is a constant, the shape of the groove is a regular shape. Conversely, if the ratio of the area of ​​the groove to the perimeter is not a constant, the shape of the groove is an irregular shape.

[0194] According to the above standards, if all conditions are met, the groove shape is a regular shape, otherwise, the groove shape is an irregular shape.

[0195] S24. Generate a label file based on the feature data set and the classification result.

[0196] It should be noted that the format of the label file should be determined, including the label name, data index, classification results and other information; the classification results should be matched with the feature data set to ensure that each feature data has a corresponding classification label; according to the determined label file format, the label file can be generated in common data formats such as CSV, JSON, etc.; the generated label file should be verified to ensure the consistency and accuracy of the label and feature data set.

[0197] S25. Based on the label file, a groove shape classification model is established using a machine learning algorithm.

[0198] It should be noted that the label file is analyzed and the category is marked to obtain the groove shape category; according to the groove shape category, a groove shape classification model is established based on the support vector machine; the labeled data is divided into a training set, a validation set and a test set; the training set is used to train the groove shape classification model to learn the relationship between the characteristics of the groove shape and the category. During the training process, the hyperparameters of the model are adjusted according to the performance of the validation set to improve the generalization ability of the model; the test set is used to evaluate the performance of the trained groove shape classification model, including accuracy, precision, recall rate, F1 value and other indicators to determine the classification effect of the model; the trained groove shape classification model is applied to new point cloud data to realize automatic classification and recognition of groove shapes.

[0199] S26, inputting the groove point cloud data into the groove shape classification model to obtain the groove shape classification result.

[0200] S3. Build a groove three-dimensional model based on the classification results of the groove shape and perform a visual display.

[0201] It should be noted that, based on the classification results of the groove position information and the groove shape, constructing the groove three-dimensional model and performing a visual display includes the following steps:

[0202] According to the position information of the groove, the coordinate system of the three-dimensional space is converted to ensure that the groove is in the correct position; a scene is created in the three-dimensional modeling software, and the position information of the groove is applied to the scene to ensure that the groove is in the appropriate position; according to the classification results of the groove shape, the appropriate modeling tool or method is selected for modeling. For regular-shaped grooves, the parametric modeling function or programming tool of the three-dimensional modeling software is used to generate the groove shape. For irregular-shaped grooves, the surface modeling tool is used, such as the surface modeling function of Blender to generate the groove shape; the constructed groove three-dimensional model is exported to common three-dimensional model file formats, such as OBJ, STL, etc., for subsequent visualization and application.

[0203] like Figure 2 As shown, according to a second embodiment of the present invention, there is also provided a cutting tool groove detection system based on artificial intelligence, the cutting tool groove detection system based on artificial intelligence comprising:

[0204] Parameter acquisition module 1, used to acquire groove parameters of the cutting tool;

[0205] Point cloud analysis module 2, used for acquiring point cloud data of groove parameters based on point cloud technology, and analyzing the groove shape according to the point cloud data of the groove parameters to obtain a classification result of the groove shape;

[0206] The classification display module 3 is used to construct a groove three-dimensional model based on the classification result of the groove shape and perform visual display;

[0207] The parameter acquisition module 1 is connected to the classification display module 3 through the point cloud analysis module 2.

[0208] According to a third embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0209] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0210] In summary, with the help of the above technical solution of the present invention, synchronous detection through ultrasonic sensor technology not only significantly improves the accuracy of cutting tool groove detection, but also realizes the accurate capture and measurement of groove parameters, improves the reliability of cutting tool groove detection process, and ensures the high quality standards of cutting tools in industrial production; at the same time, it promotes a significant improvement in production efficiency and provides strong technical support for the sustainable development and quality control of the manufacturing industry. By combining the groove positioning algorithm with point cloud technology, the accuracy of cutting tool groove detection is significantly improved, and human misjudgment is effectively reduced; at the same time, by analyzing the shape of the groove, the groove can be accurately classified, ensuring the stability and reliability of the detection process, providing strong data support for subsequent precise processing and quality control, making the identification and classification of different grooves more efficient, thereby promoting the continuous improvement and improvement of cutting tool quality management. By using three-dimensional model visualization technology, more comprehensive and in-depth groove information can be obtained, which improves the ability to quickly identify and accurately classify the groove position and category during the cutting tool groove detection process, which not only improves the efficiency of the production quality control process and shortens the detection cycle, but also ensures that the product quality standards are more strictly followed, thereby improving the market competitiveness of the product.

[0211] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A cutting tool groove detection method based on artificial intelligence, characterized in that: The cutting tool groove detection method based on artificial intelligence includes: S1, obtaining groove parameters of a cutting tool; S2. acquiring point cloud data of groove parameters based on point cloud technology, and analyzing the groove shape according to the point cloud data of the groove parameters to obtain a classification result of the groove shape; S3, constructing a groove three-dimensional model based on the classification results of the groove shape and performing a visual display; The obtaining of groove parameters of the cutting tool comprises: S11, synchronously detecting the cutting tool through a preset ultrasonic sensor to obtain ultrasonic signal data of the groove; S12, preprocessing the ultrasonic signal data, and obtaining the distance difference between the groove and each ultrasonic sensor through an ultrasonic time difference algorithm; S13, calculating by an ultrasonic attenuation algorithm to obtain a distance ratio between the groove and each ultrasonic sensor; S14, using the distance difference and the distance ratio as observation values ​​for data fusion, and using an iterative expansion filtering algorithm to iteratively calculate the groove position to obtain groove parameters; The preprocessing of the ultrasonic signal data and obtaining the distance difference between the groove and each ultrasonic sensor by using the ultrasonic time difference algorithm includes: S121, converting the preprocessed ultrasonic signal data into an image format, wherein the time series data is mapped to the horizontal axis of the image, and the signal amplitude is mapped to the vertical axis of the image; identifying and marking the arrival point of the reflected wave signal data through edge detection; S122, calculating the time difference of each ultrasonic sensor receiving reflected wave signal data according to a plurality of ultrasonic sensors; S123, combining the preset sound speed with the obtained time difference, and using the distance difference formula to calculate the distance difference between the groove and each sensor; The distance ratio between the groove and each ultrasonic sensor obtained by calculating through the ultrasonic attenuation algorithm includes: S131, using an ultrasonic signal decomposition algorithm, decomposing the preprocessed ultrasonic signal data into a number of intrinsic mode function components; S132, performing correlation analysis on each intrinsic modal function component to obtain a target intrinsic modal function component associated with each intrinsic modal function component and ultrasonic reflection signal data; S133, calculating the energy entropy values ​​of several target intrinsic modal function components, and combining with the energy attenuation model, obtaining the distance between the groove and each ultrasonic sensor; S134 . Compare the distances between the groove and each ultrasonic sensor to obtain a distance ratio between the groove and each ultrasonic sensor.

2. The cutting tool groove detection method based on artificial intelligence according to claim 1 is characterized in that: The performing correlation analysis on each intrinsic modal function component to obtain a target intrinsic modal function component associated with each intrinsic modal function component and the ultrasonic reflection signal data comprises: S1321, by calculating the autocorrelation function of each intrinsic mode function component, identifying the position and height of the peak in the autocorrelation function; S1322, by calculating the correlation coefficient between each intrinsic mode function component and combining the correlation coefficient with the peak position and height in the autocorrelation function, comprehensively analyzing the correlation between each intrinsic mode function component and the ultrasonic reflection signal data; S1323. Based on the comprehensive analysis results, the correlation between each intrinsic modal function component and the ultrasonic reflection signal data is sorted, and the first N intrinsic modal function components are selected as the target intrinsic modal function components.

3. The cutting tool groove detection method based on artificial intelligence according to claim 2 is characterized in that: The calculation formula of the energy entropy value is: In the formula, H represents the energy entropy value; m Indicates time difference; k represents the target intrinsic mode function component; i represents a natural number; p k Indicates the percentage of the energy of the kth target intrinsic mode function component of the ultrasonic signal data to the total energy.

4. The cutting tool groove detection method based on artificial intelligence according to claim 1, characterized in that: The point cloud data of the groove parameters are obtained based on the point cloud technology, and the groove shape is analyzed according to the point cloud data of the groove parameters to obtain the classification result of the groove shape, including: S21, scanning the acquired groove position information by a laser scanner to obtain groove point cloud data, and preprocessing the groove point cloud data; S22, extracting feature data from the preprocessed point cloud data to obtain a feature data set; S23, using a decision tree algorithm to perform shape annotation on the groove point cloud data, and classifying the groove shapes into regular shapes and irregular shapes according to a preset classification standard; S24, generating a label file according to the feature data set and the classification result; S25. Based on the label file, a groove shape classification model is established using a machine learning algorithm; S26, inputting the groove point cloud data into the groove shape classification model to obtain the groove shape classification result.

5. The cutting tool groove detection method based on artificial intelligence according to claim 4 is characterized in that: The classification criteria include the aspect ratio of the groove, angle measurement, edge line and area to perimeter ratio.

6. An artificial intelligence-based cutting tool groove detection system, used to implement the artificial intelligence-based cutting tool groove detection method according to any one of claims 1 to 5, characterized in that: The AI-based cutting tool groove detection system includes: A parameter acquisition module, used to acquire groove parameters of a cutting tool; A point cloud analysis module is used to obtain point cloud data of groove parameters based on point cloud technology, and analyze the groove shape according to the point cloud data of the groove parameters to obtain a classification result of the groove shape; A classification and display module is used to construct a groove three-dimensional model based on the classification results of the groove shape and to perform visual display; The parameter acquisition module is connected to the classification display module through the point cloud analysis module.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Citation Information

Patent Citations

  • System and method for extracting tool parameter

    CN101246506A

  • Material parameter self-learning method based on ultrasonic sensor array

    CN118130630A