Automatic Input System and Method for Geometric Parameters of Numerical Control Tools

Abstract

The present invention relates to the field of numerical control machining technology, and particularly to an automatic input system and method for the geometric parameters of a numerical control tool. The method includes the following steps: obtaining the point cloud data and material properties of the tool, removing abnormal points to clean the data, extracting the edge contour of the tool, performing geometric parameter mapping through the edge contour, determining the geometric features of the tool and reconstructing the tool framework, obtaining a numerical control machining task, performing cutting load simulation on the reconstructed tool, analyzing the initial wear data, predicting the thermal effect based on the cutting load, calculating the degree of tool thermal deformation, compensating for deformation of the initial wear data, generating geometric deformation data, and plotting a deformation-time curve, and performing sequential tracking parameter compensation according to the tool change process to achieve real-time input and update. The present invention improves the management efficiency and machining accuracy of numerical control tools as a whole, and enhances the stability and predictability of the production process.

Description

Technical Field

[0001] The present invention relates to the technical field of numerical control machining, and in particular to a system and method for automatically inputting geometric parameters of numerical control tools. Background Art

[0002] Traditional methods for entering tool geometry parameters often rely on manual operation and static measurement, resulting in insufficient accuracy and real-time performance of data entry. Manual operation is not only prone to human errors, but also unable to promptly reflect changes in tool status. This problem is particularly prominent in high-precision machining and large-scale production environments. Traditional methods are relatively weak in data processing and analysis capabilities, and it is difficult to meet the needs of modern manufacturing for intelligence and automation. Tool wear and deformation are key factors affecting machining quality and efficiency. Existing technologies for monitoring tool wear rely more on regular inspections and fail to achieve real-time monitoring and dynamic adjustment. Especially under complex machining conditions, the thermal deformation and wear characteristics of the tool change rapidly. Traditional methods are unable to capture these changes in a timely manner, resulting in delayed evaluation of tool performance, which in turn affects the stability and reliability of the overall machining process. The lack of an effective compensation mechanism makes it impossible to effectively guarantee the tool life and machining quality. Summary of the invention

[0003] Based on this, it is necessary to provide a system and method for automatically entering the geometric parameters of CNC tools to solve at least one of the above technical problems.

[0004] To achieve the above purpose, the method for automatically entering the geometric parameters of a CNC tool comprises the following steps:

[0005] Step S1: Acquire tool point cloud data and tool material properties; remove abnormal points in the tool point cloud data to obtain cleaning tool point cloud data; extract tool edge contour based on the cleaning tool point cloud data;

[0006] Step S2: mapping geometric parameters through the tool edge profile to determine the tool geometric features; reconstructing the tool frame based on the tool geometric features;

[0007] Step S3: Acquire a numerical control machining task; simulate the machining cutting load of the reconstructed tool frame based on the numerical control machining task to obtain simulated machining cutting load data; and analyze the initial tool wear data according to the simulated machining cutting load data;

[0008] Step S4: Predicting the thermal effect based on the simulated machining cutting load data to obtain the predicted machining thermal effect; calculating the thermal deformation degree of the tool according to the tool material characteristics and the predicted machining thermal effect;

[0009] Step S5: Perform deformation compensation on the initial tool wear data based on the degree of tool thermal deformation to generate tool geometric deformation data; calibrate the corresponding deformation timestamps for the tool geometric deformation data, plot the deformation-time curve, and determine the tool change process based on the deformation-time curve;

[0010] Step S6: Perform sequential tracking parameter compensation according to the tool change process to generate tracked compensated tool parameters; perform real-time entry and update based on the tracked compensated tool parameters.

[0011] By obtaining the tool point cloud data and tool material characteristics, the present invention ensures the accuracy and integrity of the data. The elimination of outliers improves the data quality, enabling the more accurate extraction of the tool edge contour from the cleaned point cloud data, thereby ensuring the accuracy of geometric parameter mapping. The process of reconstructing the tool framework enhances the visualization and practicality of tool design, ensuring the stability and reliability of the tool during machining. The acquisition of the numerical control machining task in combination with the cutting load simulation provides an in-depth analysis of tool wear, providing a scientific basis for tool performance evaluation. The simulated machining cutting load data can reflect the force condition of the tool during actual machining, laying a foundation for subsequent wear analysis. The combination of thermal effect prediction and material characteristics can accurately calculate the degree of tool thermal deformation during machining, forming a comprehensive understanding of the tool thermal stress. The prediction of thermal deformation provides a basis for optimizing tool material selection and machining plans. The geometric deformation data generated by deformation compensation provides a reliable reference for the actual use of the tool, effectively reducing the machining errors caused by thermal deformation. The addition of timestamps and the plotting of the deformation-time curve enable the real-time monitoring of tool state changes. The analysis of the deformation-time curve helps technicians better understand the performance of the tool during machining. The tracked compensated tool parameters generated by sequential tracking parameter compensation ensure the real-time update and accuracy of tool parameters. The implementation of tracked compensation improves the machining accuracy of the tool, reduces production fluctuations caused by tool parameter changes, overall improves the management efficiency and machining accuracy of numerical control tools, promotes the development and application of intelligent manufacturing, and ultimately realizes the optimization of tool life cycle management, making the production process more efficient and economical, enhancing the competitiveness of enterprises. Especially in the context of the increasing demand for high-precision manufacturing, this method provides a new solution for the intelligent management of numerical control tools, promotes the transformation of the manufacturing industry towards digitalization and intelligence, ensures the tool performance in complex machining environments, improves the stability and predictability of the production process, and provides strong technical support and guarantee for the development of modern manufacturing.

[0012] The present invention also provides an automatic entry system for the geometric parameters of a numerical control tool, which is used to execute the automatic entry method for the geometric parameters of the numerical control tool as described above. The automatic entry system for the geometric parameters of the numerical control tool includes:

[0013] A point cloud acquisition module, configured to acquire tool point cloud data and tool material characteristics; eliminate abnormal points in the tool point cloud data to obtain cleaned tool point cloud data; extract the tool edge contour based on the cleaned tool point cloud data.

[0014] A geometric mapping module, configured to perform geometric parameter mapping through the tool edge contour to determine tool geometric characteristics; reconstruct the tool framework based on the tool geometric characteristics.

[0015] A cutting simulation module, configured to obtain a numerical control machining task; perform machining cutting load simulation on the reconstructed tool framework based on the numerical control machining task to obtain simulated machining cutting load data; analyze the initial tool wear data based on the simulated machining cutting load data.

[0016] A thermal effect prediction module, configured to perform thermal effect prediction based on the simulated machining cutting load data to obtain predicted machining thermal effects; calculate the tool thermal deformation degree based on the tool material characteristics and the predicted machining thermal effects.

[0017] A deformation compensation module, configured to perform deformation compensation on the initial tool wear data based on the tool thermal deformation degree to generate tool geometric deformation data; calibrate a corresponding deformation timestamp for the tool geometric deformation data, plot a deformation-time curve, and determine the tool change process based on the deformation-time curve.

[0018] A parameter tracking module, configured to perform sequential tracking parameter compensation according to the tool change process to generate tracked compensated tool parameters; perform real-time entry and update based on the tracked compensated tool parameters.

[0019] Through the application of the point cloud acquisition module, the present invention realizes the comprehensive acquisition of the tool point cloud data and material characteristics, eliminates abnormal points to improve the accuracy of the data, and the cleaned point cloud data ensures the accurate extraction of the tool edge contour. The introduction of the geometric mapping module makes the determination of the tool geometric characteristics more reliable. The ability to reconstruct the tool framework enhances the visualization and practicality of tool design. The cutting simulation module combines with the numerical control machining task to be able to perform real machining cutting load simulation, providing in-depth analysis of tool wear. The simulation results provide a scientific basis for tool performance evaluation. The thermal effect prediction module can accurately predict the thermal effect of the tool during the working process by analyzing the machining cutting load data, and calculate the degree of tool thermal deformation, enabling the selection of tool materials and the optimization of machining plans to have theoretical support. The deformation compensation module compensates the initial tool wear data based on the degree of thermal deformation, and the generated geometric deformation data provides a reliable basis for the actual use of the tool. The addition of timestamps and the plotting of the deformation-time curve facilitate real-time monitoring of tool state changes. The analysis of the deformation-time curve helps technicians understand the performance of the tool during machining. The parameter tracking module can perform sequential tracking according to the tool change process, generating tracking compensation tool parameters to ensure the real-time update and accuracy of tool parameters. The implementation of tracking compensation improves the machining accuracy of the tool and reduces production fluctuations caused by tool parameter changes. The entire system improves the management efficiency and machining accuracy of numerical control tools, promotes the development and application of intelligent manufacturing, and finally realizes the optimization of tool life cycle management, making the production process more efficient and economical, enhancing the competitiveness of enterprises. Especially in the context of the increasing demand for high-precision manufacturing, it provides a new solution for the intelligent management of numerical control tools, promotes the process of the manufacturing industry towards digitalization and intelligentization, ensures the stability and predictability of tool performance in complex machining environments, provides strong technical support and guarantee for the development of modern manufacturing, promotes the scientific and systematic management of tools, and improves production efficiency. Brief Description of the Drawings

[0020] Figure 1 It is a schematic diagram of the step flow of an automatic input method for the geometric parameters of a numerical control tool;

[0021] Figure 2 It is a schematic diagram of the detailed implementation step flow of step S2;

[0022] The realization, functional features and advantages of the object of the present invention will be further described in conjunction with the embodiments with reference to the drawings. Detailed Embodiment

[0023] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0024] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0025] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0026] To achieve this, please refer to Figures 1 to 2 , a method for automatically entering geometric parameters of a numerical control tool, comprising the following steps:

[0027] Step S1: Acquire tool point cloud data and tool material properties; remove abnormal points in the tool point cloud data to obtain cleaning tool point cloud data; extract tool edge contour based on the cleaning tool point cloud data;

[0028] Step S2: mapping geometric parameters through the tool edge profile to determine the tool geometric features; reconstructing the tool frame based on the tool geometric features;

[0029] Step S3: Acquire a numerical control machining task; simulate the machining cutting load of the reconstructed tool frame based on the numerical control machining task to obtain simulated machining cutting load data; and analyze the initial tool wear data according to the simulated machining cutting load data;

[0030] Step S4: Predicting the thermal effect based on the simulated machining cutting load data to obtain the predicted machining thermal effect; calculating the thermal deformation degree of the tool according to the tool material characteristics and the predicted machining thermal effect;

[0031] Step S5: Perform deformation compensation on the initial tool wear data based on the degree of tool thermal deformation to generate tool geometric deformation data; calibrate the corresponding deformation timestamps for the tool geometric deformation data, draw the deformation-time curve, and determine the tool change process based on the deformation-time curve;

[0032] Step S6: Perform sequential tracking parameter compensation according to the tool change process to generate tracking-compensated tool parameters; perform real-time entry and update based on the tracking-compensated tool parameters.

[0033] The present invention ensures the accuracy and integrity of data by obtaining tool point cloud data and tool material characteristics. The elimination of outliers improves the data quality, enabling the more accurate extraction of the tool edge contour from the cleaned point cloud data, thereby ensuring the accuracy of geometric parameter mapping. The process of reconstructing the tool framework enhances the visualization and practicality of tool design, ensuring the stability and reliability of the tool during machining. The acquisition of CNC machining tasks combined with cutting load simulation provides in-depth analysis of tool wear, providing a scientific basis for tool performance evaluation. The simulated machining cutting load data can reflect the force condition of the tool during actual machining, laying a foundation for subsequent wear analysis. The prediction of thermal effects combined with material characteristics can accurately calculate the degree of tool thermal deformation during machining, forming a comprehensive understanding of tool thermal stress. The prediction of thermal deformation provides a basis for optimizing tool material selection and machining plans. The geometric deformation data generated by deformation compensation provides a reliable reference for the actual use of the tool, effectively reducing machining errors caused by thermal deformation. The addition of timestamps and the drawing of the deformation-time curve enable the real-time monitoring of tool state changes. The analysis of the deformation-time curve helps technicians better understand the performance of the tool during machining. The tracking-compensated tool parameters generated by sequential tracking parameter compensation ensure the real-time update and accuracy of tool parameters. The implementation of tracking compensation improves the machining accuracy of the tool, reduces production fluctuations caused by tool parameter changes, overall improves the management efficiency and machining accuracy of CNC tools, promotes the development and application of intelligent manufacturing, and ultimately realizes the optimization of tool life cycle management, making the production process more efficient and economical, enhancing the competitiveness of the enterprise. Especially in the context of the increasing demand for high-precision manufacturing, this method provides a new solution for the intelligent management of CNC tools, promotes the transformation of the manufacturing industry towards digitalization and intelligentization, ensures tool performance in complex machining environments, improves the stability and predictability of the production process, and provides strong technical support and guarantee for the development of modern manufacturing.

[0034] In the embodiment of the present invention, the automatic entry method for the geometric parameters of the CNC tool includes the following steps:

[0035] Step S1: Acquire tool point cloud data and tool material properties; remove abnormal points in the tool point cloud data to obtain cleaning tool point cloud data; extract tool edge contour based on the cleaning tool point cloud data;

[0036] In this embodiment, during the tool point cloud data acquisition stage, a three-dimensional laser scanner Leica AbsoluteScanner model LAS-20 is used, the scanning head resolution is set to 0.01 mm, the scanning frequency is 500 Hz, and the scanning distance is set within 400 mm. The standard carbide turning tool is fully scanned to obtain the tool surface point cloud data. The data format is uniformly converted into an xyz coordinate format. The tool material properties are indexed by the scan number and the preset material table in the database, and the tool material recorded by the corresponding number is extracted. The recorded attributes include material type, density, thermal expansion coefficient, Young's modulus and hardness. The material density is limited to 14.5 grams per cubic centimeter, and the thermal expansion coefficient is 4.8×10⁻ 6 Per degree Celsius, Young's modulus is set to 590GPa, hardness range is HRC85 to HRC90. In the stage of eliminating abnormal points, based on the statistical filtering method, the distance threshold is set to 0.3 mm, and the average distance of 50 neighboring points with a radius of 0.5 mm in the neighborhood of each point in the point cloud is calculated. Points exceeding the average value ± 2 times the standard deviation are eliminated as abnormal points. The cleaned tool point cloud data is subjected to edge detection. The edge detection algorithm based on the normal vector change rate is used to detect the position where the normal vector angle is greater than 45° as the edge point. The edge point cloud is extracted and fitted into a continuous contour segment, and the tool edge contour data is output.

[0037] Step S2: mapping geometric parameters through the tool edge profile to determine the tool geometric features; reconstructing the tool frame based on the tool geometric features;

[0038] In this embodiment, the edge contour of the tool is extracted and the Bezier curve interpolation method is used to fit the discrete point sequence of the edge contour into a continuous curve. The Bezier curve order is set to fourth order, and the interpolation interval is set to 0.05 mm. The geometric characteristics of the tool, such as the front angle, back angle, tool tip arc radius, cutting edge length and main deflection angle, are calculated based on the fitting curve. The front angle range is limited to 5° to 15°, the back angle is 8° to 20°, and the tool tip arc radius is set to 0.2 mm to 0.8 mm. Based on the above geometric characteristics, the parametric modeling method is used to reconstruct the three-dimensional geometric structure of the tool. The reconstruction uses the SolidWorks modeling module, and the input parameter value setting is strictly consistent with the extracted characteristic value, and the complete tool three-dimensional model STL format data file is output.

[0039] Step S3: Obtain the numerical control machining task; perform a machining cutting load simulation on the reconstructed tool framework based on the numerical control machining task to obtain simulation machining cutting load data; analyze the initial tool wear data according to the simulation machining cutting load data;

[0040] In this embodiment, in the numerical control machining task data acquisition stage, call the machining task database in the MES system (Task Manager), extract the workpiece material, machining depth, machining feed rate, spindle speed, tool type, and machining environment temperature under the current machining task number. The workpiece material is 45# steel, the machining depth is 5 mm, the feed rate is 300 mm per minute, and the spindle speed is set to 2500 revolutions per minute. Based on the machining task parameters, import the reconstructed tool framework into the Deform-3D simulation platform, construct a tool-workpiece-environment simulation model, set the interface friction factor to 0.12, the simulation step size to 0.001 s, the simulation cutting path to be consistent with the task path, simulate and output the three-axis components of the cutting force Fx, Fy, and Fz. After the simulation ends, extract the average cutting force data at all cutting step sizes, calculate the tool wear rate based on the average cutting force distribution and in combination with the wear experience model, use the criterion that the cutting force peak exceeds 700 N to determine the wear critical point, and generate the initial tool wear data table.

[0041] Step S4: Perform a thermal effect prediction based on the simulation machining cutting load data to obtain the predicted machining thermal effect; calculate the tool thermal deformation degree according to the tool material characteristics and the predicted machining thermal effect;

[0042] In this embodiment, based on the cutting load data obtained from the simulation, import it into the COMSOL Multiphysics simulation platform, construct a tool thermal-mechanical coupling model, set the initial tool temperature to 25 °C, convert the workpiece contact surface heat flux density according to the simulation cutting force and friction power, and set the heat flux density to 5×10 5 W / m², the simulation time is 5 s, the step size is 0.01 s, output the temperature field distribution, calculate the linear expansion amount of the tool in the maximum heat area according to the tool material thermal expansion coefficient of 4.8×10⁻ 6 per °C. The expansion length ΔL is equal to the thermal expansion coefficient multiplied by the length multiplied by the temperature rise. The length is selected as the length from the tool tip to the tool shank, which is 50 mm, and the temperature rise is taken as the highest temperature obtained from the simulation, 320 °C, minus the initial temperature of 25 °C, calculate the tool thermal deformation degree, and output the thermal deformation vector field data of each part of the tool.

[0043] Step S5: Perform deformation compensation on the initial tool wear data based on the tool thermal deformation degree to generate tool geometric deformation data; calibrate the corresponding deformation time stamp for the tool geometric deformation data, draw a deformation-time curve, and at the same time determine the tool change process based on the deformation-time curve;

[0044] In this embodiment, the degree of tool thermal deformation is applied to the initial tool wear data. Based on the finite element node interpolation method, the thermal deformation vector is superimposed on the corresponding wear position coordinates to correct the wear size, and the tool geometric deformation data after deformation compensation is obtained. The data format uses a csv table, which includes three columns: node coordinates, deformation values, and wear values after compensation. All data is added with a timestamp, the timestamp accuracy is set to 0.01 seconds, the starting value is set to the machining start time, a deformation-time curve is plotted, the abscissa is time, the ordinate is the deformation of each key node, the Matplotlib graphics library is used, the line chart mode is set, the line width is 2 pixels, the peak points of each stage are marked, and according to the curve change trend, the tool change process is segmented and marked. The change process is divided into a linear growth segment, a slow growth segment, and a sharp rise segment, and a complete tool change process data file is output.

[0045] Step S6: Perform timing tracking parameter compensation according to the tool change process to generate tracking compensation tool parameters; perform real-time entry and update based on the tracking compensation tool parameters.

[0046] In this embodiment, according to the tool change process data, a timing tracking compensation rule is set. If the tool deformation at a certain moment is greater than 0.15 mm, it is determined that the tool wear is critical. According to the compensation rule, the tool geometric parameters at the current moment are adjusted to the state with the minimum deformation, and all geometric parameters affected by the deformation are corrected synchronously. The corrected parameter values are written into the tracking compensation tool parameter table, which includes the current moment, tool type, corrected geometric parameter values, and wear status marks, and are real-time entered into the numerical control tool parameter library. The entry period is set to 0.01 seconds, and the entry tool uses a real-time data writing interface based on SQL Server. The interface cache queue depth is 128, ensuring that all tracking compensation tool parameters are accurately updated to the tool parameter database in chronological order.

[0047] Preferably, step S1 includes the following steps:

[0048] Step S11: Perform multi-angle laser scanning on the surface of the numerical control tool to obtain tool point cloud data. Among them, the scanning angle range is set to 15° to 75°, and the single-scan point distance is set to 0.01 mm to 0.05 mm;

[0049] Step S12: Calibrate the spatial coordinate system of the tool point cloud data to generate calibrated tool point cloud data; monitor the outliers of the calibrated tool point cloud data to obtain marked abnormal point position data;

[0050] Step S13: Perform abnormal correction on the calibrated tool point cloud data based on the marked abnormal point position data to obtain cleaned tool point cloud data. Among them, the correction range is limited to ±0.05 mm;

[0051] Step S14: Perform 3D surface fitting on the cleaned tool point cloud data and conduct edge detection to generate the tool edge contour, where the surface fitting residual is controlled within 0.01 mm.

[0052] In this embodiment, when performing surface multi-angle laser scanning on a numerical control tool, a three-dimensional laser scanning device of model FARO QuantumS is used, and it is paired with a FARO LLP (Laser Line Probe) to perform high-density point cloud acquisition. The tool is fixed at the center of a five-axis numerical control turntable. The tool is rotated around the Y-axis through the turntable, and the rotation angle range is limited to 15° to 75°, the stepping angle is set to 5°, the emission wavelength of the scanning light source is set to 632.8 nm, the laser line width is limited to 0.02 mm, the single-scan point distance is set at 0.02 mm, and the point cloud density reaches more than 2,500 points per square millimeter. The point distance is jointly controlled by the acquisition rate of the laser head and the rotation speed of the turntable, and the rotation speed is limited to 5° / s. The scanning path adopts a spiral progressive layer-by-layer trajectory. The obtained point cloud data is recorded in the form of three-dimensional coordinates, with the unit of millimeters, and in the.ply file format, including XYZ coordinates and reflection intensity. Ensure that there is no shadow area on the tool surface during each scan. Perform a spatial coordinate calibration operation on the acquired tool point cloud data. Use a standard positioning ball group with a fixed spherical center radius of 10 mm as a reference. The positioning balls are distributed around the scanning area, and the three-dimensional coordinate origin is set at the center of any one of the positioning balls. Use the least squares method to fit the point cloud on the surface of the positioning ball, extract the spherical center coordinates, calculate the rotation matrix and translation vector according to the position of the fitted spherical center and the theoretical standard coordinates, perform coordinate transformation, and complete the unification of the spatial coordinate system. The calibrated tool point cloud data is recorded with five decimal places of precision, and the coordinate unit is millimeters. For the calibrated point cloud data, use a statistical outlier detection method, set the Z-score threshold to 3, calculate the mean value of each point and its 50 neighboring points, calculate the deviation value. If the deviation value exceeds 3 times the standard deviation, it is determined as an abnormal point, and the abnormal point data is marked in the additional marking column of the point cloud file. The abnormal point value is recorded as 1, and the normal point value is recorded as 0, forming a point cloud data set containing abnormal point position information. When performing abnormal correction on the calibrated tool point cloud data based on the marked abnormal point position data, use a weighted neighborhood average interpolation algorithm, set the search radius to 0.1 mm, extract no less than 30 normal point positions within the search radius, and obtain the weighted average value of the neighborhood points according to the inverse distance weighting to replace the abnormal point position coordinates. The corrected value is limited within the range of ±0.05 mm above and below the original point position value. If the corrected calculation result exceeds this range, the upper and lower limit values are directly taken as the interval boundary values to complete the correction operation. After the correction is completed, all abnormal point marks are cleared, and the abnormal point position marking column is uniformly set to 0. The corrected cleaned tool point cloud data is retained to five decimal places of precision. The corrected point cloud file adopts...PLY format, including XYZ coordinates, reflection intensity, and columns of marked abnormal points after clearing. When performing three-dimensional surface fitting on the cleaned tool point cloud data, the rational B-spline surface modeling method is adopted. The point cloud is divided into several sub-regions according to the fitting area. The size of each sub-region is limited to 1mm×1mm, and the number of points in a single sub-region shall not be less than 500. The number of fitting control points is fixed at 10×10. Global surface fitting is performed using Rational B-Spline, and the fitting residual is controlled within 0.01mm. The residual value is calculated from the position deviation value between the fitting points and the original point cloud points. After fitting, an edge detection operation is performed. The Canny operator is used for edge detection. The upper limit of the Canny algorithm threshold is set to 0.3, and the lower limit is set to 0.1. The set of detected edge points is stored in the form of two-dimensional projection coordinates in millimeters. The output file format is set to.csv, and the file contains X coordinates, Y coordinates, and edge intensity values. Finally, complete and accurate tool edge contour data is obtained.

[0053] Preferably, step S2 includes the following steps:

[0054] Step S21: Perform Fourier descriptor transformation on the tool edge contour to obtain the frequency domain characteristics of the tool contour; perform principal component analysis on the frequency domain characteristics of the tool contour to generate the geometric parameters of the tool spindle;

[0055] Step S22: Determine the tool geometric parameters based on the geometric parameters of the tool spindle and the tool contour parameters; extract the tool geometric features according to the tool geometric parameters;

[0056] Step S23: Perform surface envelope fitting according to the tool geometric features to generate the tool outer envelope; perform axial cross-section tomography on the tool outer envelope and dissect it into equal cross-section slices;

[0057] Step S24: Perform geometric projection fusion based on the equal cross-section slices to obtain the contour feature fusion data; reconstruct the tool framework according to the contour feature fusion data and the tool geometric features.

[0058] In this embodiment, when performing Fourier Descriptor transformation on the tool edge profile data, the two-dimensional contour projection data in.csv format obtained in the previous step is used as the input. The X and Y coordinates are extracted to form a complex number point sequence. The X coordinate is used as the real part and the Y coordinate is used as the imaginary part. The total number of data points is not less than 2048. The Fast Fourier Transform (FFT) algorithm is used to perform Fourier transform on the complex number point sequence, and the complex coefficients in the transformation result are extracted as the frequency domain features of the tool profile. The first 64 complex coefficients are retained as the frequency domain components mainly describing the contour shape, and the high-frequency coefficients after the 64th order are discarded. The components with the amplitude of the high-frequency coefficients lower than 0.001 mm are directly set to zero. The frequency domain feature data is recorded in the form of a two-column array. The first column is the coefficient modulus value with the unit of millimeter, and the second column is the coefficient phase with the unit of radian. Principal Component Analysis (PCA) is performed on the obtained 64-order frequency domain feature data, and the first 3 principal components with the variance contribution rate exceeding 95% are extracted. The eigenvectors of these 3 principal components are multiplied by the original frequency domain feature matrix to obtain a low-dimensional tool profile feature vector. This feature vector corresponds to the geometric parameters in the tool spindle direction, and the number of parameters is fixed at 3, which respectively represent the spindle diameter, eccentricity, and yaw angle, and the numerical units are unified as millimeters. Based on the above tool spindle geometric parameters and the original tool profile frequency domain features, the complete tool geometric parameters are determined. First, the 64-order frequency domain feature data is inverse-transformed back to the spatial contour curve to restore the contour two-dimensional coordinate points. The number of restored points is the same as the original number of points, and the point distance error does not exceed 0.005 mm. The maximum contour diameter value is extracted in combination with the spindle diameter parameter, the eccentricity parameter determines the offset position of the contour center, and the yaw angle parameter defines the angle between the contour spindle and the Z axis of the coordinate system. Combining the above data, the outer diameter, center offset, spindle tilt direction angle, end face thickness, and edge radius of the tool are determined. In the extraction method, quantitative geometric measurement is used for each parameter. The outer diameter is directly taken as the maximum diameter, the end face thickness is calculated according to the difference between the maximum and minimum values of the contour point projection along the Z axis, and the edge radius is obtained by fitting the edge arc segment and calculating the radius value. The number of points for fitting the arc segment is not less than 128, and the fitting residual does not exceed 0.005 mm. When extracting the tool geometric features according to the complete tool geometric parameters, based on parameters such as the outer diameter, spindle tilt direction angle, and eccentricity, according to the envelope surface generation rule, the outer circle part of the tool is defined as the reference surface, the end face part is defined as the normal surface, and the edge radius area is defined as the transition surface. The surface joint points are generated based on the contact points of the three. A multi-segment Bezier curve is used to fit the envelope contour. The number of control points for each segment of the Bezier curve is 4, the length of the curve segment does not exceed 0.5 mm, and the difference between the fitting residuals of adjacent curves does not exceed 0.0.003 mm. The fitting curve records coordinates with five - decimal - place precision. All control points and curvature continuity of the fitting curve are recorded in the geometric feature database, generating a tool geometric feature code. The code includes the outer - circle diameter, eccentricity, tilt - direction angle, end - face thickness, edge - radius, positions of curvature extreme points, and corresponding curvature values. According to the above - mentioned tool geometric features, a tool - shape envelope is generated using a two - way surface - envelope fitting method. The reference outer - circle curve and end - face curve are used as fitting boundary conditions, and two - way Bezier surface fitting is utilized. The surface order is set to the third order, and the control - point array is set to 16×16. The fitting residual is controlled within 0.01 mm. After the envelope fitting is completed, an axial - section tomography operation is performed. Along the Z - axis direction of the tool axis, the surface is cut at a step of 0.1 mm to generate equal - cross - section slices. The number of contour points in each slice is not less than 512, and the recording precision of the contour - point coordinates is 0.001 mm. Each slice is stored as a separate two - dimensional contour data in.csv format, including the X - coordinate, Y - coordinate, and Z - axis coordinate where the slice is located. Based on all equal - cross - section slices, geometric - projection fusion is performed. All slice contour points are projected onto the XZ plane and YZ plane along the Z - axis direction, respectively generating the XZ - projection contour and YZ - projection contour. Edge detection is performed on the projection contours using the Sobel operator with a threshold set to 0.2. The contour - boundary coordinate points are extracted, and the average width, maximum width, minimum width, and edge - curvature distribution of each projection contour are calculated. The contour - characteristic data are recorded in the form of an array with the unit of millimeters. According to all projection - contour characteristic data and the aforementioned tool geometric features, a tool - space framework is reconstructed using a two - way Bezier surface. The number of surface control points is set to 32×32, and the reconstruction - surface residual is controlled within 0.015 mm. The surface control points correspond to the original geometric - feature points, and the coordinate precision of the control points is maintained to five decimal places, including complete tool - geometric - space - structure data.

[0059] Preferably, step S3 includes the following steps:

[0060] Step S31: Obtain a numerical - control machining task; extract task - process parameters according to the numerical - control machining task;

[0061] Step S32: Predict the cutting - force distribution based on the task - process parameters; perform a cutting - force response simulation on the reconstructed tool framework according to the predicted cutting - force distribution, generating simulation - cutting - response data;

[0062] Step S33: Perform a cutting - load mapping according to the simulation - cutting - response data, generating simulation - machining - cutting - load data;

[0063] Step S34: Calculate the cumulative - load sequence based on the simulation - machining - cutting - load data; analyze the theoretical wear rate according to the cumulative - load sequence;

[0064] Step S35: Estimate the tool wear of the reconstructed tool framework according to the theoretical wear rate to generate initial tool wear data.

[0065] In this embodiment, when obtaining a numerical control machining task, the task scheduling module in the machine tool control system is called to extract the current machining work order to be executed from the machining task scheduling table built into the numerical control machine (NC machine). Through parsing the G-code program file stored in the machine tool controller, process parameter information such as the machining process number, feed rate, spindle speed, machining path trajectory coordinates, feed direction, tool number, and cutting depth is extracted. All process parameters adopt a unified structured format and are extracted in sequence according to the machining order. Among them, the set range of the feed rate is from 80 mm / min to 3000 mm / min, the set range of the spindle speed is from 500 rpm to 20000 rpm, and the cutting depth range is from 0.1 mm to 5 mm. After all parameter extractions are completed, they are organized into a standardized JSON data format and stored in the intermediate buffer for subsequent calls in the link of predicting the cutting force distribution. When predicting the cutting force distribution based on the task process parameters, a cutting force prediction model constructed based on the BP neural network (Back Propagation Neural Network) is used. The input parameters include the spindle speed, feed rate, cutting depth, tool geometric parameters, and material mechanical property parameters. The tool geometric parameters include the rake angle, clearance angle, helix angle, and nose radius. The material mechanical property parameters include Young's modulus, Poisson's ratio, and yield strength. The specific value ranges are as follows: the rake angle is from 6° to 20°, the clearance angle is from 6° to 15°, the helix angle is from 30° to 60°, the nose radius is from 0.2 mm to 1.2 mm, Young's modulus is from 160 GPa to 210 GPa, Poisson's ratio is from 0.28 to 0.34, and the yield strength is from 400 MPa to 900 MPa. The BP neural network adopts a three-layer structure, with 12 nodes in the input layer, 24 nodes in the hidden layer, and 3 nodes in the output layer, corresponding to the predicted values of the cutting force distribution in the X, Y, and Z directions. The neural network training samples are sourced from the measured cutting force data under existing machining conditions. The mean square error minimization is used as the error function, and the number of iterations is 5000. After the prediction is completed, the cutting force distribution field data in a three-dimensional coordinate system is generated. According to the predicted cutting force distribution, a cutting power response simulation is carried out on the reconstructed tool structure. Using the ANSYS Mechanical module of the finite element analysis software, the tool structure is established as a three-dimensional solid model and divided into tetrahedral elements, with the side length of each element not greater than 0.2 mm. The predicted cutting force distribution data is applied as an external load to the main cutting edge area of the tool, and the fixed constraint condition of the tool handle is set. The tool material property is defined as high-speed steel, with a density of 7800 kg / m³, Young's modulus of 210 GPa, and Poisson's ratio of 0.3. The static analysis module is used to solve the displacement, stress, and strain responses of each node of the tool, and an output simulation cutting response data file is recorded, including the node number, coordinate position, displacement vector, maximum principal stress, and equivalent strain value. The simulation step size is 0.0.01 seconds, cumulative simulation time of 1 second. Perform cutting load mapping based on the simulated cutting response data. Using grid mapping technology, project the node load results of the simulation model to the corresponding machining path position points. The path position points are generated according to the machining path G-code, with a path step distance of 0.5 mm. During the mapping process, based on the nearest neighbor interpolation method, assign the cutting load values of each simulation node to the machining path points, and record the simulated machining cutting load data after mapping. The data format is CSV, including path point number, three-dimensional coordinates, and load values in the X, Y, and Z directions. Calculate the cumulative load sequence based on the simulated machining cutting load data. Using the cumulative superposition method, sequentially add the loads in the X, Y, and Z directions point by point along the path from the starting point to the ending point of the machining path, and record the cumulative values in the corresponding path point numbers. The update step of the cumulative sequence is 0.5 mm. The generated cumulative load sequence has a unit of N·mm, and the sequence length is equal to the number of machining path points. Analyze the theoretical wear rate based on the cumulative load sequence. Using the Archard wear model, the input parameters are the cumulative normal force, cutting path length, tool hardness, and wear coefficient. The tool hardness is taken as 62 HRC, and the wear coefficient is taken as 0.0000025 mm³ / N·mm. Calculate the wear rate per unit length of each path point and summarize to obtain the theoretical wear rate data. Estimate the tool wear of the reconstructed tool framework based on the theoretical wear rate. Apply the corresponding wear rate to each cutting path point position based on the three-dimensional solid model. Using the geometric reduction method, reduce the corresponding wear depth along the cutting contact normal direction according to the position of the path point. The wear depth calculation method is the wear rate multiplied by the path step distance. Generate the initial tool wear data file for the updated tool three-dimensional model, recording the node numbers, coordinate positions, and wear depths of each surface of the tool.

[0066] Particularly importantly, step S32 includes:

[0067] Execute the force field load mapping conversion on the task process parameter data to obtain the cutting load mapping data;

[0068] Subdivide the contact area action weights of the cutting load mapping data to generate the regional load distribution data;

[0069] Perform load gradient matrix fitting based on the regional load distribution data to obtain the gradient load matrix;

[0070] Execute the tool framework load mapping coupling on the gradient load matrix data to obtain the tool load action data;

[0071] Perform the cutting contact surface effect mapping based on the tool load action data to obtain the cutting response mapping data;

[0072] Execute the response trajectory reconstruction simulation on the cutting response mapping data to obtain the simulated cutting response data.

[0073] In this embodiment, when performing a force field load mapping conversion on task process parameter data, a cutting force calculation engine is called, and task process parameter data including spindle speed, feed rate, cutting depth, rake angle, clearance angle, helix angle, nose radius, Young's modulus of the material, yield strength, and Poisson's ratio is input. This data is used as the input quantity of the cutting force empirical prediction model. The prediction model adopts a method of superimposing a multivariable regression model and a physical model. The coefficients of the regression model are fitted based on existing cutting force measurement data. The physical model is calculated based on the relationship between the main cutting edge length, cutting area, and load per unit area. The load per unit area ranges from 500 N / mm² to 3500 N / mm². Combining the G-code trajectory information of the machining path, the prediction results are mapped to the corresponding coordinates of the path points in the form of three-dimensional vectors. The load components are calculated separately in the X, Y, and Z directions, and the cutting load mapping data is output and stored as a three-column vector format data table. The fields of the data table include path point number, three-dimensional coordinates, X-direction load, Y-direction load, and Z-direction load. When dividing the action weights of the contact areas of the dissected cutting load mapping data, according to the position of the path points on the machining path and the principle of dividing the contact areas on the cutting edge surface of the tool, the cutting area is divided into three types of areas: the rake face contact area, the flank face contact area, and the nose transition area. The division of the contact areas is determined by the nearest neighbor relationship between the mesh node positions of the tool model and the machining path points. The mesh node density is set to 30 nodes per millimeter. According to the cutting contact theory, the action weight of the rake face contact area is set to 0.5, the action weight of the flank face contact area is set to 0.35, and the action weight of the nose transition area is set to 0.15. Multiply each path point load value by the corresponding action weight, and output the regional load distribution data. The data table structure maintains four fields: path point number, regional type, X-direction load, Y-direction load, and Z-direction load. When fitting the load gradient matrix based on the regional load distribution data, a three-dimensional quadratic polynomial surface fitting method is adopted. The fitting equation form is the quadratic polynomial relationship of the regional load distribution in the X, Y, and Z directions with respect to the machining path coordinate points X, Y, and Z. The least squares method is used for fitting. All fitting sample points are provided by the regional load distribution data table. The fitting result outputs a three-dimensional gradient matrix. The matrix dimension is set to the number of points of the machining path length / step distance. Each matrix element records the fitted gradient value. The unit of the gradient value is N / mm. Perform maximum, minimum, and mean statistics on the three-dimensional gradient matrix respectively, and record them in the load gradient matrix data table. The data table contains four columns: matrix number, three-dimensional coordinate position, X gradient, Y gradient, and Z gradient. When performing the tool frame load mapping coupling on the gradient load matrix data, import the tool three-dimensional finite element mesh model, and use the ANSYS APDL script module. Input the gradient load matrix data as the external load and apply it to the tool model mesh nodes. The node matching method is based on the nearest neighbor interpolation method of the path point coordinates and the tool mesh node positions. Each tool model node applies the X, Y, and Z three-direction coupling loads according to the interpolation results of the gradient loads of adjacent path points. The load value is calculated by linear interpolation. The boundary condition of the tool model is set to fix the tool handle and release other degrees of freedom. After the load application is completed, output the tool load action data. The data format is the ANSYS standard RST result file, which records the force condition, displacement, and stress response values of each node. The node number corresponds one-to-one with the load application point number. When performing the cutting contact surface effect mapping based on the tool load action data, call the ANSYS Workbench Mechanical module, read the RST result file, and extract the stress, strain, and displacement response values at the surface mesh nodes of the tool model. According to the definition of the tool-workpiece contact area, only retain the surface node response data located in the rake face, flank face, and tool tip area of the tool. The node screening condition is that the angle between the surface normal and the cutting direction is less than 30°. After screening, map the stress, strain, and displacement values of each surface node to the corresponding machining path point. The mapping method uses the Euclidean distance weighted average method. Each path point takes the response values of the 5 surface nodes closest to it, with weight coefficients of 0.4, 0.3, 0.15, 0.1, 0.05 Perform weighted summation, and the calculated cutting contact surface response data is used as cutting response mapping data, which is saved as a data table with seven columns of path point number, X, Y, Z coordinates, normal stress, normal strain, and normal displacement. When performing response trajectory reconstruction simulation on the cutting response mapping data, based on the ABAQUS Explicit dynamics solver, establish a discrete point trajectory model of the machining path, use the cutting response mapping data as the external force input of the path point, define the normal stress, normal strain, and normal displacement of the path point as loading conditions, with a step spacing of 0.5 mm for the trajectory point, a total path length of 100 mm, a time step of 0.0001 s, a total simulation time of 0.1 s, and the applied external force is loaded in a linearly increasing manner with time. The simulation outputs the displacement trajectory, velocity change, and path deformation response of the path point, records the simulation cutting response data, and saves it as a TXT format file. The file content includes five fields: time step number, path point number, displacement vector, velocity vector, and deformation amount.

[0074] Preferably, the thermal effect prediction based on the simulated machining cutting load data in step S4 includes:

[0075] Perform power structure separation based on the simulated machining cutting load data to obtain the power data in each direction;

[0076] Perform cutting plastic zone power mapping based on the power data in each direction and the tool material characteristics to generate cutting deformation power;

[0077] Determine the tool-workpiece contact unit according to the reconstructed tool framework; perform contact zone thermally induced response simulation based on the power data in each direction and the tool-workpiece contact unit to obtain the simulated contact zone response;

[0078] Perform thermal power backtracking mapping based on the simulated contact zone response to generate contact friction power;

[0079] Perform thermal conversion rate conversion on the cutting deformation power and contact friction power to generate deformation heat power data and friction heat power data respectively;

[0080] Perform thermal accumulation processing on the deformation heat power data and friction heat power data according to the simulated machining cutting load data to obtain the accumulated deformation heat and accumulated friction heat respectively;

[0081] Perform heat source spatial mapping based on the accumulated deformation heat and accumulated friction heat to generate heat source distribution data;

[0082] Perform heat source superposition coupling on the accumulated deformation heat and accumulated friction heat according to the heat source distribution data, and perform thermal diffusion response prediction to obtain the predicted machining thermal effect.

[0083] In this embodiment, power structure separation is performed based on the cutting load data of simulation machining. First, the cutting process is simulated through simulation machining software (such as ANSYS, Abaqus, etc.) to obtain the cutting load data, including the mechanical responses in different cutting stages. According to the simulation results, the cutting load is separated into power in different cutting directions. Specifically, by decomposing the cutting force data, the cutting force is respectively converted into axial power, cutting force power, and radial power data according to different cutting directions. Based on mechanical calculations, the power data range in each direction is set. Usually, the power data is separated according to the X-axis (axial), Y-axis (radial), and Z-axis (cutting force). The power data in each direction will be processed separately through a mechanical calculation model and normalized according to cutting parameters such as cutting speed, feed rate, and cutting depth to ensure accurate calculation of the power data in different directions. Finally, the power data in each direction is obtained. Based on the power data in each direction and the tool material characteristics, power mapping of the cutting plastic zone is performed. First, the characteristics of the tool material, such as hardness and plasticity, are analyzed. The material characteristics of the tool will affect the formation and power consumption of the plastic zone during the cutting process. After obtaining the power data in each direction, based on the characteristics of the tool material, power mapping of the cutting plastic zone is performed. The tool material models used in the mapping process include the shear strength model and the plastic deformation model. Assuming that the contact area between the tool and the workpiece is the cutting plastic zone, the power of the cutting plastic zone is calculated through the correlation between the cutting force and the cutting angle. This power is manifested as the generation and consumption of heat during the plastic deformation process. Combining the actual process parameters, the value of the cutting deformation power is obtained. The tool-workpiece contact unit is determined according to the reconstructed tool framework. When reconstructing the tool framework, based on the geometric shape of the tool and the initial state of the workpiece, the contact surface between the tool and the workpiece is calculated through numerical simulation. First, it is necessary to model through the contact situation between the cutting edge of the tool and the workpiece surface. Using the Finite Element Analysis (FEA) method, the contact area between the tool and the workpiece is divided into multiple small units, and the cutting edge of the tool is used as the boundary of the contact unit. By analyzing the stress and strain in the contact area, the tool-workpiece contact unit is determined. On this basis, further calculations are performed using the contact model to ensure the accurate modeling and reconstruction of the contact unit. Based on the power data in each direction and the tool-workpiece contact unit, thermal response simulation of the contact area is performed. First, based on the power data in each direction obtained previously, the power data is used as the input condition for thermal response simulation. During the thermal response simulation, considering the influence of different power directions on the temperature change in the contact area, the heat change in the contact area is calculated through the heat conduction and heat convection models. The heat conduction equation and the temperature field distribution in the contact unit are used for simulation. In the simulation, parameters such as cutting temperature, material thermal conductivity, and contact pressure need to be input. Finally, the simulated response data of the contact area is obtained, where the response data includes temperature distribution, heat generation, etc.During the simulation process, ensure the matching of the heat source and power to obtain an accurate thermally induced response in the contact zone. Based on the simulated contact zone response, perform a backtracking mapping of the thermal power. First, use the temperature data obtained from the simulation of the thermally induced response in the contact zone and combine it with the power data for backtracking analysis. During the analysis process, use a thermodynamic backtracking model to compare the response in the contact zone with the initial input power, and inversely calculate the thermal power generation of each contact unit. The process of backtracking mapping of the thermal power needs to be adjusted according to the thermal physical properties of the material and the thermal effects during the cutting process. In particular, the influence of the contact force between the tool and the workpiece needs to be taken into account. Finally, obtain the contact friction power, which is the heat generated by friction during the contact between the tool and the workpiece. Perform a thermal conversion rate conversion on the cutting deformation power and the contact friction power. First, combine the material properties of the tool and the workpiece to calculate the thermal conversion rate of the cutting deformation power and the contact friction power. Specifically, parameters such as the friction coefficient between the tool and the workpiece, the specific heat capacity and thermal conductivity of the material will affect the thermal conversion rate. Based on these thermal physical properties, use a thermal conversion rate model for conversion, and calculate the thermal conversion rates of the cutting deformation power and the contact friction power respectively. During this process, the thermal conversion rate is usually set within a fixed range and adjusted according to different material properties to ensure that the converted data accurately reflects the heat distribution during the machining process. Perform a thermal accumulation process on the deformation heat power data and the friction heat power data according to the simulated machining cutting load data. During the accumulation process, accumulate the thermal power data for each time period to obtain the accumulated heat. The time range of the accumulation process is set according to the machining time, usually with a processing time span of 30 seconds to 5 minutes. At each time step, perform heat accumulation based on the change of the thermal power data. The accumulation process of the cutting deformation power data and the friction heat power data will perform data fusion based on the corresponding temperature change curves to ensure that the finally obtained accumulated deformation heat and friction heat data can accurately represent the thermal energy consumption during the machining process. Perform a heat source spatial mapping based on the accumulated deformation heat and the accumulated friction heat. First, use the accumulated deformation heat and friction heat data for spatial mapping. Use a heat source distribution model to distribute the accumulated heat in the contact area between the tool and the workpiece. During the mapping, considering the spatial distribution characteristics and temperature gradient of the heat source, use the finite element method to numerically simulate the heat source and distribute the accumulated heat according to the spatial coordinates. Finally, obtain the heat source distribution data, which describes the distribution of thermal energy between the tool and the workpiece during the machining process. Perform a heat source superposition coupling on the accumulated deformation heat and the accumulated friction heat according to the heat source distribution data, and predict the thermal diffusion response. First, couple the heat source distribution data with the accumulated heat, and combine the contact situation between the tool and the workpiece in the actual machining process to perform a superposition process on the accumulated heat and the heat source. Through a thermal diffusion model, simulate the diffusion of heat in the workpiece and the tool. The coupled data will be input into the thermal diffusion equation for calculation, and predict the propagation and diffusion of heat through a heat conduction model. Finally,Obtain the predicted machining thermal effect during the machining process. The predicted data includes the distribution of heat, the diffusion path, and the temperature change, etc., to ensure that the final prediction of the thermal effect conforms to the actual cutting conditions.

[0084] Preferably, the calculation of the tool thermal deformation degree according to the tool material characteristics and the predicted machining thermal effect in step S4 includes:

[0085] Decompose the tool material characteristics into material groups and perform thermal elastic coefficient mapping to generate material thermodynamics parameters;

[0086] Determine the tool thermal response characteristics based on the material thermodynamics parameters;

[0087] Perform tool surface heat load distribution mapping based on the predicted machining thermal effect to generate tool surface heat load data;

[0088] Perform thermally induced strain driving processing according to the tool surface heat load data and the tool thermal response characteristics to obtain the local strain response of the tool;

[0089] Infer the geometric deformation of the edge contact based on the local strain response of the tool; perform non-contact edge thermal expansion simulation of the tool according to the local strain response of the tool to generate non-edge thermal expansion geometric deformation;

[0090] Perform geometric shape migration and reconstruction according to the edge contact geometric deformation and the non-edge thermal expansion geometric deformation to obtain the thermal deformation geometric shape; construct the tool thermal deformation field based on the thermal deformation geometric shape;

[0091] Infer the tool thermal deformation through the tool thermal deformation field for the reconstructed tool framework and evaluate the tool thermal deformation degree.

[0092] In this embodiment, the material characteristics of the cutting tool are decomposed into material groups and the thermoelastic coefficient is mapped. First, by obtaining the chemical composition data of the cutting tool material, the proportion of each component in the cutting tool material is analyzed. For example, the proportion of alloying elements such as carbon, tungsten, cobalt, etc. The microstructure of the cutting tool material is analyzed by scanning electron microscopy (SEM) or X-ray diffraction (XRD). Combining with the standard material mechanics database, the initial thermoelastic coefficient of the cutting tool material is determined. For each component, according to its thermal expansion coefficient, specific heat capacity, thermal conductivity and other thermal performance parameters, the material thermodynamics model is used for calculation to obtain the thermodynamics parameters of the cutting tool material. Usually, these parameters include the thermal diffusivity, thermal stress constant, etc. of the material. Through the thermal strain test at different temperatures, the mapping relationship between temperature and thermoelastic coefficient is established, so as to obtain the detailed thermoelastic characteristic data of the cutting tool material. Based on the material thermodynamics parameters, the thermal response characteristics of the cutting tool are determined. First, according to the obtained thermodynamics parameters of the cutting tool material, combined with the temperature distribution of the cutting tool during the cutting process, the thermal response characteristics of the cutting tool are simulated. During the simulation, it is assumed that the heat source during the cutting process is the main influencing factor, and the thermodynamics model is used to calculate the thermal response of the cutting tool. The specific process is to distribute the heat generated by cutting to each area of the cutting tool according to different material characteristics, such as the cutting edge, flank face, etc. of the cutting tool, and calculate the distribution of temperature gradient and thermal stress according to the thermophysical characteristics of the cutting tool. Through the numerical simulation of the temperature field, the thermal response characteristic data of the cutting tool are obtained. This data includes the temperature change, thermal stress value and thermal strain distribution of the cutting tool surface, cutting edge and back surface, etc., and finally forms the thermal response characteristic model of the cutting tool. Based on the prediction of the machining thermal effect, the thermal load distribution on the cutting tool surface is mapped. First, according to the prediction of the cutting force and heat distribution during the machining process, the thermal load distribution model is used to map the thermal load on the cutting tool surface. The finite element analysis (FEA) is used to simulate the heat flow situation of the cutting tool during the cutting process. Specifically, the cutting force data is combined with the heat source distribution to calculate the thermal load generated on the cutting tool surface during the cutting process. Further, according to the position and size of the heat source, the thermal load density of different areas on the cutting tool surface is determined. In the thermal load distribution mapping, the thermal conductivity of the cutting tool material and the dynamic changes during the cutting process need to be considered, and finally the accurate distribution of the thermal load on the cutting tool surface during the machining process is obtained. According to the thermal load data on the cutting tool surface and the thermal response characteristics of the cutting tool, the thermally induced strain driving process is carried out to obtain the local strain response of the cutting tool. First, the thermal load data on the cutting tool surface obtained in the previous step is used as the input, and it is combined with the thermal response characteristics of the cutting tool to calculate the thermally induced strain. In this process, it is assumed that when the cutting tool surface is driven by the thermal load, the strain of the material will respond to the change of the thermal load. The thermal strain model is used to perform thermally induced strain analysis on the cutting tool surface in combination with the thermodynamics parameters. During the specific operation, by setting a certain time step, the thermal strain of each part of the cutting tool is calculated step by step. Finally, the local strain response of the cutting tool surface is obtained. This strain response reflects the geometric deformation of the cutting tool caused by the thermal load.Infer the geometric deformation of the cutting edge contact based on the local strain response of the tool. Combine the tool surface heat load and thermally induced strain response data to infer the geometric deformation of the cutting edge contact area of the tool. First, determine the thermal strain value of the cutting edge contact area based on the geometric shape of the tool. Analyze the geometric deformation of the cutting edge according to the strain distribution. Usually, calculate the temperature distribution and thermal strain value of the contact area to estimate the geometric deformation of this area. For example, the cutting edge of the tool may undergo deformations such as bending and expansion. Combine the plastic properties of the material and use the finite element method to simulate the geometric deformation of the cutting edge, and finally obtain the deformation form and deformation amount of this area. Conduct non-contact cutting edge thermal expansion simulation of the tool based on the local strain response of the tool to generate non-cutting-edge thermal expansion geometric deformation. First, according to the local strain response data of the tool obtained in the previous step, conduct thermal expansion simulation for the non-contact cutting edge of the tool. In this process, the geometric deformation of the non-contact cutting edge is mainly caused by the thermal expansion effect. The non-contact cutting edge of the tool expands to a greater extent because it is heated differently from the contact cutting edge. Through numerical simulation, use the thermal expansion model to conduct thermal expansion analysis of the non-contact cutting edge, specifically analyze the relationship between the temperature change and expansion coefficient of the non-cutting edge, and thus calculate the geometric deformation of the non-cutting edge, and finally obtain the non-cutting-edge thermal expansion geometric deformation result. Conduct geometric shape migration and reconstruction based on the cutting edge contact geometric deformation and non-cutting-edge thermal expansion geometric deformation to obtain the thermally deformed geometric shape. First, merge the data of the cutting edge contact geometric deformation and non-cutting-edge thermal expansion geometric deformation to form a thermally deformed geometric shape model of the whole tool. In specific operations, use the geometric shape migration algorithm to combine the geometric deformation data of the two to conduct the overall thermal deformation reconstruction of the tool. During the reconstruction process, based on the thermal expansion coefficient and strain data of each part of the tool, estimate the overall deformation result of the tool. Finally, obtain the thermally deformed geometric shape data of the tool, which reflects the overall geometric deformation of the tool due to temperature changes during the machining process. Construct the tool thermally deformed field based on the thermally deformed geometric shape. First, based on the thermally deformed geometric shape data obtained in the previous step, combine the temperature distribution data of the tool to construct the tool thermally deformed field. The specific steps are as follows: Analyze the thermal deformation degree of each part of the tool during the cutting process through the thermodynamic model. Based on the heat and strain data of different regions, use the numerical simulation method to construct the tool thermally deformed field. Usually, the establishment of the thermally deformed field needs to consider factors such as the material properties, temperature distribution, and thermally induced strain response of the tool. Finally, obtain the thermally deformed field data of each part of the tool during the machining process, which provides a complete picture of the deformation process experienced by the tool during the cutting process. Conduct tool thermal deformation inference on the reconstructed tool structure through the tool thermally deformed field and evaluate the degree of tool thermal deformation. First, according to the constructed tool thermally deformed field, conduct thermal deformation inference on the tool. Combine the geometric structure of the tool and the thermally deformed field data, and use the finite element analysis (FEA) method to analyze the thermal deformation of each part of the tool and estimate the degree of tool thermal deformation. During the inference process,It is necessary to combine the stresses, strains and heat source effects of each part of the tool to evaluate the degree of thermal deformation of the tool during actual cutting, and finally obtain the thermal deformation amount generated by the tool during the machining process. The inferred results provide a basis for tool design optimization and machining accuracy evaluation.

[0093] Especially importantly, perform thermally induced strain-driven processing based on the tool surface thermal load data and the tool thermal response characteristics to obtain the local strain response of the tool, including:

[0094] Perform zoning reduction of the thermal load density on the tool surface thermal load data to obtain the zoned thermal load density;

[0095] Combine the zoned thermal load density data with the tool thermal response characteristics for thermally coupled zoning coupling mapping to obtain the thermal zone coupling response data;

[0096] Perform thermal gradient flow fitting based on the thermal zone coupling response data to obtain the gradient thermal stress distribution;

[0097] Perform thermal effect hierarchical superposition reconstruction according to the gradient thermal stress distribution data to obtain the thermal stress superposition field;

[0098] Perform strain-driven function conversion on the thermal stress superposition field to obtain the driving strain response;

[0099] Determine the local strain response of the tool based on the driving strain response.

[0100] In this embodiment, when performing the reduction of the heat load density in the heat load data on the tool surface, the tool surface in the numerical control machining area is divided into rectangular units of 5 mm × 5 mm. The surface temperature values of each unit are collected in real time by a thermocouple array measurement device. The measurement device uses K-type thermocouples with a resolution of 0.1 °C and a sampling frequency set to 1000 Hz. According to the surface temperature values, the heat load density of each unit is calculated through the established heat conduction model. In the model, the thermal conductivity of the tool material is taken as 42 W / (m·K). The heat load density values of all units are divided into 5×5 sub-blocks, and the average heat load density in each sub-region is calculated to obtain the sub-region heat load density data. All the sub-region heat load density data are stored in a 64×64 two-dimensional array matrix with the unit of W / m². When performing the thermal coupling partition coupling mapping operation on the sub-region heat load density data in combination with the tool heat response characteristics, a preset tool heat response library is called. This library contains the thermoelastic parameter data obtained by decomposing different material groups. The thermoelastic parameters include the coefficient of thermal expansion set to 13.5×10^-6 1 / K, the Young's modulus set to 215 GPa, and the Poisson's ratio set to 0.29. The data in the heat response library are from the test results of a high-temperature dynamic mechanical analyzer DMA. The heat load density value of each sub-region is calculated point by point with the corresponding heat response parameters to obtain the heat zone coupling response data. The coupling method is to multiply the average heat load density in the sub-region by the combined weight value of the coefficient of thermal expansion and the thermal conductivity. Finally, a heat zone coupling response matrix is formed, and the matrix dimension is consistent with the heat load density matrix. When performing the heat gradient flow fitting based on the heat zone coupling response data, the finite difference method is used to calculate the gradient of adjacent unit points in the heat zone coupling response matrix. The gradient direction is determined by the temperature difference between adjacent points. The gradient value is obtained by taking the difference between the coupling response values of adjacent two points divided by the distance using the central difference method, and the distance is fixed at 5 mm. All the heat gradient values in the sub-regions are mapped to a two-dimensional coordinate heat stress distribution diagram. Each pixel point in the diagram represents a 5 mm × 5 mm area, and the color intensity corresponds to the gradient heat stress value with the unit of MPa. The heat stress value is obtained by multiplying the heat gradient by the thermoelastic parameters. The heat stress distribution diagram is smoothed by the kernel density estimation method to improve the continuity of the heat stress gradient. When performing the heat effect hierarchical stacking and reconstruction operation based on the gradient heat stress distribution data, the heat stress distribution diagram is divided along the tool radius direction into 0.A 5-mm equidistant annular region is defined. The average thermal stress value of all elements within each annular region is calculated, and these values are successively superimposed to form a thermal stress superposition field. During the superposition, the sum of the thermal stress value of each layer and the upper layer is taken as the cumulative thermal stress value of that layer. The cumulative value is calculated using a recursive iterative accumulation method. The data of the superposition field is stored in the form of a three-dimensional matrix, with dimensions of the number of layers in the radial direction × the number of angular partitions × the thermal stress value. The unit of the thermal stress value is MPa. After the superposition is completed, a thermal stress superposition field map is generated, where the color intensity in the map corresponds to the cumulative thermal stress intensity value of each region. When performing a strain-driven function transformation on the thermal stress superposition field, a linear thermal strain relationship is selected. The thermal stress value of each element in the superposition field is divided by the Young's modulus of the corresponding region of the tool, with the Young's modulus taken as 215 GPa, to obtain the thermal-induced strain value of that element, with the unit being dimensionless. The strain value of each element is associated with its spatial coordinates and stored in a three-dimensional strain response matrix, whose dimensions are the same as those of the superposition field matrix. When determining the local strain response of the tool based on the driving strain response, the strain values in the three-dimensional strain response matrix are clustered according to their spatial positions. The clustering method uses the density-based spatial clustering algorithm DBSCAN, with the neighborhood radius set to 5 mm and the minimum number of points set to 10. After clustering, the high-density strain response region is extracted as the local strain response region of the tool, and the average strain, maximum strain, and coordinate distribution information within this region are output. All output results are recorded in a table, with the table fields including the partition number, average strain, strain extreme value, and position coordinates. Among them, the strain response unit is dimensionless, and the position coordinate unit is mm.

[0101] Preferably, the deformation compensation of the initial tool wear data based on the tool thermal deformation degree in step S5 includes:

[0102] Performing a mapping of the heat flux density distribution on the initial tool wear data to obtain the tool thermal gradient distribution;

[0103] Constructing a three-dimensional deformation vector field of the tool based on the tool thermal gradient distribution;

[0104] Optimizing the boundary constraint conditions of the three-dimensional deformation vector field of the tool according to the tool thermal deformation degree and constructing an actual thermal deformation model of the tool;

[0105] Determining the geometric deformation data of the tool based on the actual thermal deformation model of the tool.

[0106] In this embodiment, a heat flux density distribution mapping is performed on the initial tool wear data. First, a high-precision infrared thermal imager is used to measure the surface temperature distribution of the tool. The temperature measurement accuracy of the instrument is 0.1°C, and the measurement range covers the temperature range from 100°C to 1000°C. The temperature distribution data of the tool under different cutting conditions is obtained. Based on the temperature distribution data, the heat flux density on the surface and inside of the tool is calculated. A heat flux density model based on thermal conductivity and cutting force is adopted. The thermal conductivity parameter is selected according to the tool material. For example, the thermal conductivity of a high-speed steel tool is 25 W / (m·K), and the thermal conductivity of a cemented carbide tool is 150 W / (m·K). The distribution data of the tool heat flux density is obtained. Further, this data is used to construct the thermal gradient distribution on the tool surface. The thermal gradient reflects the rate of change of the tool temperature and provides basic data for the subsequent deformation model. The unit of the thermal gradient distribution is usually K / m. By combining the high-precision heat flux density distribution with the temperature gradient, the accuracy of the thermal gradient data is ensured to reach 0.1 K / m. A three-dimensional deformation vector field of the tool is constructed based on the tool thermal gradient distribution. First, according to the thermal gradient data and in combination with the geometric model of the tool, a three-dimensional space grid of the tool is established. The size of the grid is set according to the geometric complexity of the tool, and the grid size is between 0.1 mm and 1 mm. A refined grid is selected to ensure the deformation accuracy. The deformation vector field of the tool is numerically calculated by the finite element method (FEM). The calculation process is based on the thermal expansion coefficient of the tool, the stress-strain relationship, and the temperature field and force field during the cutting process. The range of the thermal expansion coefficient generally ranges from 5×10⁻ 6 1 / °C to 15×10⁻ 61 / °C, select a specific value according to the tool material. Using a computer simulation tool, construct a three-dimensional deformation vector field of the tool under different cutting states. This vector field describes the deformation of each grid node of the tool, with an accuracy of up to 0.001 mm. Optimize the boundary constraint conditions of the tool's three-dimensional deformation vector field according to the degree of tool thermal deformation, and construct an actual thermal deformation model of the tool. First, set the boundary conditions according to the contact conditions between the tool and the workpiece. The contact force is calculated using the data from the force sensor, and the range of the contact force is from 10 N to 500 N. The specific value is determined according to the actual loading situation during the cutting process. The boundary conditions include the fixing method of the tool and the fixture, the contact surface between the tool and the workpiece, and the temperature distribution data of the tool itself. The optimization algorithm uses the least squares method or the genetic algorithm. During the optimization process, correct and adjust the thermal deformation model to ensure that the model more accurately reflects the actual thermal deformation of the tool. Finally, obtain the actual thermal deformation model of the tool. This model provides the geometric change information of each point of the tool, and the calculation accuracy is within the range of 0.001 mm. Determine the geometric deformation data of the tool based on the actual thermal deformation model of the tool. Use a three-dimensional measuring instrument (such as a coordinate measuring machine) to accurately measure the geometric shape of the tool, with a measurement accuracy of 0.001 mm. By comparing the thermal deformation model of the tool in the actual machining process with the actual geometric shape, extract the geometric deformation data of the tool. The deformation data of the tool usually includes the wear of the tool edge, the deformed tool shape, and the geometric changes of each surface and cutting edge of the tool. During the measurement process, combined with the actual machining conditions, record the tool shape data before and after cutting, and determine the deformation amount through calculation. This deformation amount is used to adjust the tool geometric parameters in the numerical control system to ensure the geometric compensation effect of the tool. Finally, obtain the accurate geometric data of the tool.

[0107] Preferably, in step S5, calibrate the corresponding deformation timestamp for the tool geometric deformation data, and draw a deformation-time curve. At the same time, the process of determining the tool change based on the deformation-time curve includes:

[0108] Perform cutting time sequence synchronization processing on the tool geometric deformation data to generate timestamp-associated deformation data;

[0109] Map the timestamp-associated deformation data to a deformation-time discrete point set;

[0110] Fit the deformation-time discrete point set to a deformation-time curve;

[0111] Detect the curve inflection points of the deformation-time curve and mark them as deformation key nodes;

[0112] Perform an analysis of the slope change at the inflection point on the deformation-time curve based on the deformation key nodes and map it to a deformation acceleration feature;

[0113] Synchronously project deformation nodes according to the deformation acceleration characteristics and tool geometric deformation data to generate deformation synchronization data;

[0114] Determine the tool change process based on the deformation synchronization data.

[0115] In this embodiment, the tool geometric deformation data is processed for cutting time sequence synchronization. First, the CNC system records the start time and end time of each cutting process with a precision requirement reaching the millisecond level. The recorded data includes the tool position, tool rotation speed, feed speed, cutting force, etc. during cutting. Through the synchronous timestamp, all the tool geometric deformation data is sorted and matched according to the time sequence during the cutting process. The timestamp is obtained through a high-speed data acquisition system with a sampling frequency of 1 kHz. Within each sampling period, the geometric shape and wear condition of the tool are updated once. The corresponding timestamp is recorded and matched with the geometric deformation data, and each position of the tool is marked in detail to ensure that each data point can correspond to the accurate cutting time. The timestamp-associated deformation data is mapped into a deformation-time discrete point set. The data interpolation method is used to interpolate the tool geometric deformation data at different timestamps. The specific interpolation method uses cubic spline interpolation to ensure the smoothness and accuracy of the data. The interpolated data is displayed through a graphical tool (such as the Matlab or Matplotlib toolbox of Python) to form a deformation-time discrete point set. Each discrete point represents the relationship between a tool deformation data and the corresponding timestamp. The abscissa of the discrete point set is time in seconds, and the ordinate is the tool deformation amount in millimeters or micrometers. The number of discrete points in the set depends on the duration of the cutting process and the sampling frequency. Among them, the number of sample points can reach several thousand, and the distribution of points shows a regular change with the change of cutting time. The deformation-time discrete point set is fitted into a deformation-time curve. First, the least squares method is used to fit the discrete point set to generate a smooth curve of deformation and time, which can better reflect the geometric deformation change trend of the tool during the cutting process. During the fitting process, a suitable function model is first set. Commonly used models include polynomial models, exponential models, etc. The specific selection depends on the distribution of the data. The fitting accuracy requirement is 0.001 mm, and the fitting error range of the curve is controlled within 5%. Through the fitting of the curve, the change law of the tool deformation is obtained, and a continuous deformation-time curve is formed to ensure that the fitted curve can truly reproduce the geometric deformation process of the tool. Detect the inflection points of the deformation-time curve and mark them as deformation key nodes. Based on the deformation-time curve, a numerical analysis tool is used to detect the inflection points of the curve. The condition of the first derivative being zero is used to determine the inflection point position. By detecting the change trend of the deformation curve, the key turning points during the tool deformation process are determined. The inflection points correspond to important stages of the tool geometric shape change. The change passing through the inflection point usually means that the tool wear intensifies or the deformation speed changes significantly. The detection of the inflection points requires setting a threshold to ensure its accuracy. When the threshold is set that the change amplitude of the deformation amount exceeds 0.01 mm, it is regarded as a key node. The processed inflection points are marked as deformation key nodes in the deformation-time curve.Analyze the slope change at the inflection point of the deformation-time curve based on the key deformation nodes, and map it to the deformation acceleration feature. First, calculate the tangent slope of the deformation data before and after each key node. The slope represents the rate of deformation. By calculating the change in slope at the inflection point, the acceleration feature of the tool deformation is obtained. The specific calculation method is to select several time points near the inflection point and calculate the rate of change of the slope between adjacent points to obtain the acceleration data. The deformation acceleration of the tool is usually in units of millimeters per second squared. The region with a large slope change usually corresponds to the stage of severe tool wear. The deformation acceleration feature is of great significance for judging the remaining life of the tool. The calculated deformation acceleration feature will be combined with the specific geometric deformation data of the tool for further analysis of the tool change process. Perform deformation node synchronous projection based on the deformation acceleration feature and the tool geometric deformation data to generate deformation synchronous data. Based on the previously obtained deformation acceleration feature, through time series synchronous analysis, map the acceleration data of each deformation node to its corresponding tool geometric deformation data, establish a one-to-one correspondence between the acceleration feature and the time stamp, and generate deformation synchronous data. The deformation synchronous data can truly reflect the geometric changes and acceleration features of the tool in each cutting stage. The time accuracy of the deformation synchronous data is 1 ms, and the number of data points is determined by the duration of the cutting process and the synchronous accuracy. The generated synchronous data provides accurate geometric data for subsequent tool life prediction and automatic input. Determine the tool change process based on the deformation synchronous data. Finally, through the processing of the deformation synchronous data, using data analysis methods, draw a graph of the tool change process, display the deformation of the tool during the entire cutting process through the graph. The change process graph comprehensively presents data such as time, deformation amount, and acceleration. Through this graph, the relationship between tool wear and cutting time can be intuitively understood, and the tool life cycle can be further estimated. The tool change process graph can provide accurate tool status information for the numerical control system, and then automatically input the tool geometric parameters.

[0116] Preferably, step S6 includes the following steps:

[0117] Step S61: Segment the tool change process in time series to obtain tool change time series data, where the single-segment time length is set within the range of 5 s to 30 s;

[0118] Step S62: Measure the parameter drift of the tool change time series data to obtain the tool drift parameter, where the parameter drift amplitude threshold is limited within ±0.05 mm;

[0119] Step S63: Perform compensation fitting processing based on the tool drift parameter data to generate tracking compensation tool parameters, where the compensation coefficient range is set to 0.8 to 1.2;

[0120] Step S64: Based on the tracking compensation tool parameters, perform real-time input and update to obtain real-time updated tool parameters.

[0121] In this embodiment, the tool change process is segmented in time series. First, the deformation data of the tool during the machining process is collected, and the data is segmented according to the time sequence. The time length of each segment is between 5 seconds and 30 seconds. The specific time segment length is adjusted according to the cutting process and machining characteristics of the tool. Usually, the tool will experience a relatively stable working state during this time period. When segmenting, it is set according to the data volatility. If the data changes violently, a shorter time period may be selected for analysis. The key point of data segmentation is to flexibly set the cutting segment according to the real-time collected data fluctuation and cutting parameter change. Use signal processing tools or algorithms (such as Fast Fourier Transform (FFT)) to analyze the frequency characteristics of the tool deformation data, and measure the parameter drift of the tool change time series data. First, analyze the data of each time period and calculate the offset (drift amount) in the tool change process. This drift amount is defined as the difference between the geometric deformation of the tool in each time period and the theoretical value. When measuring the drift, first select the reference tool deformation value, which can be obtained from the initial geometric shape of the tool or the most recent precision calibration. The drift amplitude of each segment of data needs to be compared with a threshold of ±0.05 mm. If the offset exceeds this threshold, further adjustment is required. The calculation of the drift parameter is completed by the difference method. Each data point is compared with the adjacent data points before and after to obtain the change in the drift amount. If the drift amount continues to increase, it indicates that the tool may have suffered wear or other deformation. The drift amplitude threshold is limited to ±0.05 mm. Perform compensation fitting processing based on the tool drift parameter data. The compensation processing mainly corrects the tool deformation through the compensation coefficient. The compensation coefficient range is set between 0.8 and 1.2. A suitable coefficient is set to fit the drift data of each segment. The selection of the compensation coefficient is based on the stability of the drift parameter and the size of the offset. If the drift amount is large, the compensation coefficient is close to 1.2. If the drift amount is small, it is close to 0.8. When performing compensation fitting, first calculate the compensation curve according to the drift data and the theoretical value. The compensation coefficient is obtained through curve fitting. When performing curve fitting, methods such as polynomial regression or exponential regression can be selected. The compensated data will be mapped back to the tool geometric parameters to obtain the real-time compensated tool parameters, ensuring a high degree of matching between the tool geometry and the actual cutting state. Based on the tracked compensated tool parameters, perform real-time entry and update. First, synchronize the compensated tool parameters with the numerical control system in real time through the data input module, and update the geometric parameters of the tool in real time, including the radius of the tool, the edge angle, the length of the tool, etc. Each time the compensated parameters will be automatically entered into the numerical control system to ensure that the tool always maintains the best working state during the machining process. During the entry process, the accuracy requirement for parameter update is 0.001 mm. For each compensated tool geometric parameter, the time accuracy of real-time update is 1 ms. In the numerical control system, parameter synchronization is carried out through data interfaces (such as RS-232, Ethernet, etc.), so that the geometric data of the tool can be immediately fed back to the machining tool, and any subtle changes in the tool can be compensated by means of real-time update to ensure that the change process of the tool can be reflected and corrected in real time.

[0122] The present invention also provides an automatic input system for the geometric parameters of a numerical control tool, which is used to execute the automatic input method for the geometric parameters of the numerical control tool as described above. The automatic input system for the geometric parameters of the numerical control tool includes:

[0123] A point cloud acquisition module, which is used to acquire tool point cloud data and tool material characteristics; remove abnormal points in the tool point cloud data to obtain cleaned tool point cloud data; extract the tool edge contour according to the cleaned tool point cloud data;

[0124] A geometric mapping module, which is used to perform geometric parameter mapping through the tool edge contour to determine the geometric characteristics of the tool; reconstruct the tool framework based on the geometric characteristics of the tool;

[0125] A cutting simulation module, which is used to obtain a numerical control machining task; perform machining cutting load simulation on the reconstructed tool framework based on the numerical control machining task to obtain simulation machining cutting load data; analyze the initial tool wear data according to the simulation machining cutting load data;

[0126] A thermal effect prediction module, which is used to perform thermal effect prediction based on the simulation machining cutting load data to obtain predicted machining thermal effects; calculate the tool thermal deformation degree according to the tool material characteristics and the predicted machining thermal effects;

[0127] A deformation compensation module, which is used to perform deformation compensation on the initial tool wear data based on the tool thermal deformation degree to generate tool geometric deformation data; calibrate the corresponding deformation timestamp for the tool geometric deformation data, draw a deformation-time curve, and at the same time determine the tool change process based on the deformation-time curve;

[0128] A parameter tracking module, which is used to perform timing tracking parameter compensation according to the tool change process to generate tracked compensated tool parameters; perform real-time input and update based on the tracked compensated tool parameters.

[0129] Through the application of the point cloud acquisition module, the present invention realizes the comprehensive acquisition of the tool point cloud data and material characteristics, eliminates abnormal points to improve the accuracy of the data, and the cleaned point cloud data ensures the accurate extraction of the tool edge contour. The introduction of the geometric mapping module makes the determination of the tool geometric characteristics more reliable. The ability to reconstruct the tool framework enhances the visualization and practicality of tool design. The cutting simulation module combines with the numerical control machining task to perform real machining cutting load simulation, providing in-depth analysis of tool wear. The simulation results provide a scientific basis for tool performance evaluation. The thermal effect prediction module can accurately predict the thermal effect of the tool during the working process by analyzing the machining cutting load data, and calculate the degree of tool thermal deformation, providing theoretical support for the selection of tool materials and the optimization of machining plans. The deformation compensation module compensates the initial tool wear data based on the degree of thermal deformation, and the generated geometric deformation data provides a reliable basis for the actual use of the tool. The addition of timestamps and the drawing of the deformation-time curve facilitate the real-time monitoring of tool state changes. The analysis of the deformation-time curve helps technicians understand the performance of the tool during machining. The parameter tracking module can perform sequential tracking according to the tool change process, generate tracking compensation tool parameters, ensuring the real-time update and accuracy of tool parameters. The implementation of tracking compensation improves the machining accuracy of the tool and reduces production fluctuations caused by tool parameter changes. The entire system improves the management efficiency and machining accuracy of CNC tools, promotes the development and application of intelligent manufacturing, and finally realizes the optimization of tool life cycle management, making the production process more efficient and economical, enhancing the competitiveness of enterprises. Especially in the context of the increasing demand for high-precision manufacturing, it provides a new solution for the intelligent management of CNC tools, promotes the process of the transformation of the manufacturing industry to digital and intelligent, ensures the stability and predictability of tool performance in complex machining environments, provides strong technical support and guarantee for the development of modern manufacturing, promotes the scientific and systematic management of tools, and improves production efficiency.

[0130] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to cover all changes falling within the meaning and scope of the equivalent elements of the application documents within the present invention.

[0131] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. An automatic input method for the geometric parameters of a numerical control tool, characterized in that, The following steps are involved: Step S1: Acquire tool point cloud data and tool material properties; remove abnormal points in the tool point cloud data to obtain cleaning tool point cloud data; Extract tool edge contour based on cleaning tool point cloud data; Step S2: mapping geometric parameters through the tool edge profile to determine the tool geometric features; reconstructing the tool frame based on the tool geometric features; Step S3: Acquire a numerical control machining task; simulate the machining cutting load of the reconstructed tool frame based on the numerical control machining task to obtain simulated machining cutting load data; and analyze the initial tool wear data according to the simulated machining cutting load data; Step S4: predicting the thermal effect based on the simulated machining cutting load data to obtain the predicted machining thermal effect; calculating the thermal deformation degree of the tool according to the tool material characteristics and the predicted machining thermal effect; wherein predicting the thermal effect based on the simulated machining cutting load data includes: Power structure separation is performed based on the simulated machining cutting load data to obtain power data in each direction; Based on the power data in each direction and the material characteristics of the tool, the power mapping of the cutting plastic zone is performed to generate the cutting deformation power; Determine the tool-workpiece contact unit according to the reconstructed tool frame; simulate the thermal response of the contact area based on the power data in each direction and the tool-workpiece contact unit to obtain the simulated contact area response; Perform thermal power back-mapping based on simulated contact zone response to generate contact friction power; Perform heat conversion rate conversion on cutting deformation power and contact friction power to generate deformation heat power data and friction heat power data respectively; According to the simulated machining cutting load data, the deformation heat power data and the friction heat power data are heat accumulated to obtain the accumulated deformation heat and accumulated friction heat respectively; Based on the accumulated deformation heat and accumulated friction heat, heat source space mapping is performed to generate heat source distribution data; According to the heat source distribution data, the accumulated deformation heat and the accumulated friction heat are coupled with heat sources, and the thermal diffusion response is predicted to obtain the predicted processing thermal effect; Step S5: performing deformation compensation on the initial tool wear data based on the degree of thermal deformation of the tool to generate tool geometric deformation data; calibrating the corresponding deformation timestamp for the tool geometric deformation data, and drawing a deformation-time curve, and determining the tool change process based on the deformation-time curve; Step S6: Perform time-series tracking parameter compensation according to the tool change process to generate tracking compensation tool parameters; and perform real-time input and update based on the tracking compensation tool parameters.

2. The automatic input method for the geometric parameters of a numerically controlled cutting tool according to claim 1, characterized in that, Step S1 includes the following steps: Step S11: performing multi-angle laser scanning on the surface of the CNC tool to obtain tool point cloud data, wherein the scanning angle range is set to 15°~75°, and the single scanning point distance is set to 0.01mm~0.05mm; Step S12: calibrating the tool point cloud data in a spatial coordinate system to generate corrected tool point cloud data; performing outlier monitoring on the corrected tool point cloud data to obtain marked abnormal point data; Step S13: performing abnormal correction on the calibration tool point cloud data based on the marked abnormal point data to obtain cleaning tool point cloud data, wherein the correction range is limited to ±0.05 mm; Step S14: Perform 3D surface fitting on the cleaned tool point cloud data and conduct edge detection to generate the tool edge contour, where the surface fitting residual is controlled within 0.01 mm.

3. The automatic input method for the geometric parameters of a numerically controlled cutting tool according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: Perform Fourier descriptor transformation on the tool edge contour to obtain the frequency domain characteristics of the tool contour; conduct principal component analysis on the frequency domain characteristics of the tool contour to generate the geometric parameters of the tool spindle. Step S22: Determine the tool geometric parameters based on the geometric parameters of the tool spindle and the tool contour parameters; extract the tool geometric features according to the tool geometric parameters. Step S23: Perform surface envelope fitting according to the tool geometric features to generate the tool shape envelope; conduct axial cross-section tomography on the tool shape envelope and dissect it into equal cross-section slices. Step S24: Conduct geometric projection fusion based on the equal cross-section slices to obtain the contour feature fusion data; reconstruct the tool framework according to the contour feature fusion data and the tool geometric features.

4. The automatic input method for the geometric parameters of a numerically controlled tool according to claim 1, characterized in that, Step S3 includes the following steps: Step S31: Obtain the numerical control machining task; extract the task process parameters according to the numerical control machining task. Step S32: Predict the cutting force distribution based on the task process parameters; conduct cutting dynamic response simulation on the reconstructed tool framework according to the predicted cutting force distribution to generate the simulation cutting response data. Step S33: Conduct cutting load mapping according to the simulation cutting response data to generate the simulation machining cutting load data. Step S34: Calculate the cumulative load sequence based on the simulation machining cutting load data; analyze the theoretical wear rate according to the cumulative load sequence. Step S35: Conduct tool wear estimation on the reconstructed tool framework according to the theoretical wear rate to generate the initial tool wear data.

5. The automatic input method for the geometric parameters of a numerically controlled tool according to claim 1, characterized in that, The deduction of the tool thermal deformation degree according to the tool material characteristics and the predicted machining thermal effect in Step S4 includes: Conduct material group decomposition on the tool material characteristics and perform thermal elastic coefficient mapping to generate the material thermodynamics parameters. Determine the tool thermal response characteristics based on the material thermodynamics parameters. Perform tool surface heat load distribution mapping based on the predicted machining thermal effect to generate the tool surface heat load data. Conduct thermally induced strain driving processing according to the tool surface heat load data and the tool thermal response characteristics to obtain the tool local strain response. Infer the edge contact geometric deformation based on the tool local strain response; conduct non-edge thermal expansion simulation of the tool according to the tool local strain response to generate the non-edge thermal expansion geometric deformation. Conduct geometric shape migration reconstruction according to the edge contact geometric deformation and the non-edge thermal expansion geometric deformation to obtain the thermal deformation geometric shape; construct the tool thermal deformation field based on the thermal deformation geometric shape. Infer the tool thermal deformation on the reconstructed tool framework through the tool thermal deformation field and evaluate the tool thermal deformation degree.

6. The automatic input method for the geometric parameters of a numerically controlled tool according to claim 1, characterized in that The deformation compensation for the initial tool wear data based on the tool thermal deformation degree in Step S5 includes: Perform heat flux density distribution mapping on the initial tool wear data to obtain the tool thermal gradient distribution. Construct the tool three-dimensional deformation vector field based on the tool thermal gradient distribution. Optimize the boundary constraint conditions of the tool three-dimensional deformation vector field according to the tool thermal deformation degree and construct the tool actual thermal deformation model. Determine the tool geometric deformation data based on the tool actual thermal deformation model.

7. The automatic input method for the geometric parameters of a numerically controlled tool according to claim 1, characterized in that Step S5 of calibrating the corresponding deformation timestamp for the tool geometric deformation data and drawing the deformation-time curve, and determining the tool change process based on the deformation-time curve includes: Perform cutting timing synchronization processing on tool geometry deformation data to generate timestamp-associated deformation data; Mapping timestamp-associated deformation data into a deformation-time discrete point set; Fitting the deformation-time discrete point set into a deformation-time curve; Detect the inflection points of the deformation-time curve and mark them as key deformation nodes; Based on the deformation key nodes, the slope change of the deformation-time curve at the inflection point is analyzed and mapped into the deformation acceleration characteristics; Perform deformation node synchronization projection according to deformation acceleration characteristics and tool geometry deformation data to generate deformation synchronization data; The tool change process is determined based on the deformation synchronization data.

8. The automatic input method for the geometric parameters of a numerically controlled tool according to claim 1, characterized in that, Step S6 includes the following steps: Step S61: segmenting the tool change process into time series to obtain tool change time series data, wherein the length of a single segment is set within a range of 5s to 30s; Step S62: measuring the parameter drift of the tool change time sequence data to obtain the tool drift parameter, wherein the parameter drift amplitude threshold is limited to within ±0.05 mm; Step S63: performing compensation fitting processing according to the tool drift parameter data to generate tracking compensation tool parameters, wherein the compensation coefficient range is set to 0.8-1.2; Step S64: Based on the tracking compensation tool parameters, real-time input and update are performed to obtain real-time updated tool parameters.

9. An automatic input system for the geometric parameters of a numerical control tool, characterized in that, Used to execute the automatic input method of the geometric parameters of the numerical control tool as claimed in claim 1, the automatic input system of the geometric parameters of the numerical control tool comprises: The point cloud acquisition module is used to obtain the tool point cloud data and tool material characteristics; remove abnormal points in the tool point cloud data to obtain the cleaning tool point cloud data; extract the tool edge contour based on the cleaning tool point cloud data; A geometric mapping module is used to map geometric parameters through the tool edge profile to determine the tool geometric features; and reconstruct the tool frame based on the tool geometric features; A cutting simulation module is used to obtain a numerical control machining task; based on the numerical control machining task, a machining cutting load simulation is performed on the reconstructed tool frame to obtain simulated machining cutting load data; and initial tool wear data is analyzed based on the simulated machining cutting load data; The thermal effect prediction module is used to predict the thermal effect based on the simulated machining cutting load data to obtain the predicted machining thermal effect; the thermal deformation degree of the tool is calculated according to the tool material characteristics and the predicted machining thermal effect; The deformation compensation module is used to perform deformation compensation on the initial tool wear data based on the degree of thermal deformation of the tool to generate tool geometric deformation data; calibrate the corresponding deformation timestamp for the tool geometric deformation data, draw the deformation-time curve, and determine the tool change process based on the deformation-time curve; The parameter tracking module is used to perform time-series tracking parameter compensation according to the tool change process, generate tracking compensation tool parameters, and perform real-time input and update based on the tracking compensation tool parameters.

Images