Numerical control machine tool machining domain error field T spline model optimization method based on regular term
By using the T-spline model of regular terms for local refinement and optimization in the CNC machine tool processing domain error field model, the problems of overfitting and underfitting of NURBS models are solved, and higher fitting accuracy and smoothness are achieved , improves the machining accuracy of CNC machine tools.
Patent Information
- Application Number
- CN202411306391.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-19
AI Technical Summary
The existing CNC machine tool processing domain error field model uses NURBS and least squares method to overfit and underfit, and lack local refinement capabilities, resulting in high computational complexity and poor fitting results on the test set.
A T-spline model based on regular terms is adopted to generate a smoother error field model through local refinement and punishment force parameter optimization to prevent overfitting and improve fitting effect.
It significantly improves the smoothness of the fitted surface, removes unnecessary features, improves the fitting level on the test set and the overall data set, avoids overfitting, and improves the accuracy of the error field model of CNC machine tools.
Smart Images

Figure CN119087911A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the research field of machining errors of precision CNC machine tools, and in particular relates to a T-spline model optimization method of a CNC machine tool machining domain error field based on a regularization term. Background Art
[0002] With the rapid development of modern manufacturing, CNC machine tools are widely used in different fields such as aerospace, automobile, and medical. Higher processing accuracy and smaller errors are our common goals. The processing accuracy of CNC machine tools also largely determines the level of manufacturing. At present, there are two main solutions to the problem of improving the processing accuracy of CNC machine tools: error prevention method and error compensation method. Error prevention method refers to the use of more advanced technology and more suitable materials in the manufacture of CNC machine tools, so that the errors generated by CNC machine tools during operation are smaller, and the thermal deformation caused by running heat is smaller. However, this method is limited by the current manufacturing level of CNC machine tools and cannot meet the expected requirements. The more popular method is the error compensation method. The error compensation method refers to fitting the measured discrete error data into a unified mathematical model under the premise that a certain machine tool has a processing error, and adding a certain compensation algorithm to compensate for the error before the machine tool is processed, so that the machine tool can achieve higher processing accuracy.
[0003] Studying how to establish a unified CNC machine tool processing domain error field model plays a vital role in the subsequent CNC machine tool processing domain error field compensation algorithm. At present, the main tool for CNC machine tool processing domain error field modeling is NURBS (non-uniform rational B-spline), and the main fitting method is the least squares method. The main steps are as follows: first, collect discrete data; second, design the NURBS grid; finally, use the least squares method to solve the NURBS grid control points to obtain a unified processing domain error field surface model.
[0004] The current method of using NURBS and least squares method to calculate the error field in the machining domain of CNC machine tools still has the following problems:
[0005] 1. NURBS requires control points to be distributed in a grid-like manner. It does not have the ability of true local refinement and flexible topological structure. When inserting control points, it is necessary to insert a whole row and a whole column of control points at the same time, resulting in the emergence of a large number of redundant control points, which greatly increases the complexity of calculation and the complexity of the final model.
[0006] Second, the error of the measured data is not taken into account. Due to the measurement error, when there are many control points, the final fitting surface will have many unnecessary features, that is, overfitting occurs, while when there are fewer control points, underfitting occurs, and the error of a single data point may have a greater impact on the entire surface. Overall, the fitting results on the training set are better, while the fitting results on the test set are worse.
[0007] like Figure 3 As shown in Figure 2, the existence of measurement errors is inevitable in the actual sampling of error fields in CNC machine tool processing. Figure 3 a is the assumed error surface in the CNC machine tool processing domain, Figure 3 b is the ideal error-free sampling data distribution, Figure 3 c is the data distribution with errors obtained from actual sampling. Summary of the invention
[0008] The purpose of the present invention is to solve the problems existing in the prior art and to provide a T-spline model optimization method for the error field in the machining domain of a CNC machine tool based on a regularization term, the purpose of which is to take the measurement error of the error field data in the machining domain of a CNC machine tool into consideration, to prevent the overfitting phenomenon of the general NURBS model caused by the existence of the measurement error, and to provide a method for improving the fitting level of the model and the smoothness of the fitting surface by adding a regularization term to the optimization function to constrain its smoothness.
[0009] In order to achieve the above-mentioned invention object, the present invention specifically adopts the following technical solutions:
[0010] In a first aspect, the present invention provides a T-spline model optimization method for a CNC machine tool machining domain error field based on a regularization term, comprising the following steps:
[0011] S1. Obtain discrete data of the error field in the CNC machine tool processing domain and perform preprocessing, and divide the preprocessed discrete data into two parts: a training set and a test set; wherein the discrete data is the difference between each data point in the real processing surface and each data point in the actual processing surface, the real processing surface is the surface formed by the theoretical motion trajectory during the CNC machine tool processing of parts, and the actual processing surface is the surface formed by the actual motion trajectory during the CNC machine tool processing of parts;
[0012] S2, parameterizing the preprocessed discrete data into a two-dimensional square parameter domain as a T-spline surface mesh, locally refining the area in the T-spline surface mesh that needs to be locally refined, setting nodes in two directions parallel to the T-spline surface mesh lines on the mesh of each control point in the T-spline surface mesh, and generating node vectors corresponding to two nodes of each control point according to the T-spline vector node generation rule; wherein the control point is a set of data that determines the T-spline surface shape of the error field of the CNC machine tool when the T-spline surface mesh is known; and the T-spline surface is a surface obtained by fitting the discrete data using T-spline;
[0013] S3, determining the position of each data point in the training set mapped on the T-spline surface grid, and assigning nodes in two directions parallel to the T-spline surface grid lines to each data point in the training set;
[0014] S4, designing a regularization term function of the T-spline surface optimization function, taking a penalty intensity parameter as a weight, and adding the T-spline surface errors of all data points in the training set to the weighted regularization term function to obtain a T-spline surface optimization function with a regularization term;
[0015] S5, setting the initial value of the coordinates of each control point in the T-spline surface grid;
[0016] S6. Adjust the penalty intensity parameter value within a preset range, obtain a T-spline surface under each penalty intensity parameter value, calculate the error of each generated T-spline surface on the test set, take the penalty intensity parameter with the smallest error as the optimal penalty intensity parameter, take the T-spline surface with the smallest error as the final optimized CNC machine tool error field model, and realize the T-spline model optimization of the CNC machine tool processing domain error field.
[0017] Based on the above scheme, each step can be implemented in the following preferred specific manner.
[0018] Preferably, in step S4, for a T-spline surface mesh with n control points, the regularization term function g(P) is in the form of:
[0019]
[0020] Among them, Z x ,Z xx are the first-order partial derivative and second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface with respect to the x direction; Z y ,Z yy are the first-order partial derivative and second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface in the y direction; u i2 ,v i2Respectively represent the i-th control point P i The node vector U i and V i The third component in .
[0021] Preferably, in step S4, the kth data point Q in the training set k The T-spline surface error is the k-th data point Q in the training set k Its T-spline surface at the corresponding point P(u k ,v k ), then the T-spline surface error of all data points in the training set is:
[0022]
[0023] Wherein, m represents the number of data points in the training set.
[0024] Preferably, in step S4, for a T-spline surface mesh with n control points, the T-spline surface optimization function is in the form of:
[0025]
[0026] Among them, λ represents a penalty intensity parameter greater than 0.
[0027] As a preference, for the kth data point Q in the training set k The corresponding point P(u k ,v k ), which is calculated according to the following T-spline surface formula:
[0028]
[0029] Where P(u,v) represents the T-spline surface; n represents the number of control points; w i represents the i-th control point P i The weight of x i ,y i ,z i Respectively represent the x, y, and z coordinate components of the three-dimensional coordinates of the i-th control point; B i (u,v) represents the T-spline basis function of the i-th control point; Indicates that the node vector U i Cubic B-spline basis functions on ; Indicates the node vector V i Cubic B-spline basis functions on .
[0030] As a preference, for a node vector U i and V i The i-th control point Pi , the initial value of its coordinates is set as follows: Get the maximum values X and Y of the x and y components of the three-dimensional coordinates of the data point in the training set, then set the i-th control point P i The initial value of the coordinate is [u i2 *X,v i2 *Y,0].
[0031] In the second aspect, the present invention provides an error compensation method for the error field in the processing domain of a CNC machine tool based on regularization terms, and the specific steps are as follows: first, discrete data to be error compensated in the error field in the processing domain of a CNC machine tool is obtained and preprocessed, the preprocessed discrete data is input into the CNC machine error field model optimized by the T-spline model optimization method for the error field in the processing domain of a CNC machine tool based on regularization terms as described in the first aspect above, to obtain a T-spline surface of the CNC machine error field, an error compensation algorithm is used on the T-spline surface of the CNC machine error field, an error compensation result is obtained and fed back to the control system of the processing domain of the CNC machine tool, and finally an optimized processing surface is output; wherein, the optimized processing surface is a surface formed by the actual motion trajectory after error compensation during the process of processing parts by the CNC machine tool.
[0032] Compared with the prior art, the present invention has the following beneficial effects:
[0033] The main problem with the currently used NURBS least squares method for solving the error field model of CNC machine tools is that due to the existence of data errors, a higher level of fitting for the training set will lead to overfitting and a poor fitting level on the test set. After adding a regularization term to constrain the smoothness of the surface, the present invention allows for a small error in the data, and comprehensively considers the surface fitting error and the surface smoothness to obtain a CNC machine tool error field model with a better fitting effect.
[0034] Compared with the prior art, the present invention greatly improves the smoothness of the fitting surface, removes unnecessary surface features, effectively prevents overfitting of the error field model of CNC machine tools, and significantly improves the fitting level on the test set and the overall data set. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 is a flow chart of the steps of the present invention;
[0036] Figure 2 Schematic diagram of the difference between T-spline mesh and NURBS mesh refinement; (a) is a schematic diagram of the initial mesh that needs to be locally refined on the surface F; (b) is a schematic diagram of NURBS mesh refinement; (c) is a schematic diagram of T-spline mesh refinement;
[0037] Figure 3Schematic diagram of the difference between ideal sampling and actual sampling under a certain ideal surface; among them, (a) is a schematic diagram of an ideal error surface; (b) is a schematic diagram of ideal error-free sampling data; and (c) is a schematic diagram of actual data with errors. DETAILED DESCRIPTION
[0038] In order to make the above-mentioned purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation mode of the present invention is described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. The technical features in each embodiment of the present invention can be combined accordingly without conflicting with each other.
[0039] like Figure 1 As shown, in a preferred implementation of the present invention, the above-mentioned T-spline model optimization method for the error field in the machining domain of a CNC machine tool based on regularization term includes the following steps S1 to S6. The specific implementation process is described in detail below.
[0040] S1. Obtain discrete data of the error field in the CNC machine tool processing domain and perform preprocessing, and divide the preprocessed discrete data into two parts: a training set and a test set; wherein the discrete data is the difference between each data point in the real processing surface and each data point in the actual processing surface, the real processing surface is the surface formed by the theoretical motion trajectory during the CNC machine tool processing of parts, and the actual processing surface is the surface formed by the actual motion trajectory during the CNC machine tool processing of parts.
[0041] S2, parameterize the preprocessed discrete data into a two-dimensional square parameter domain as a T-spline surface mesh, locally refine the area that needs to be locally refined in the T-spline surface mesh, and set the i-th control point P in the T-spline surface mesh. i Nodes u on the grid in two directions parallel to the T-spline surface grid lines i2 and v i2 , generating node vectors corresponding to the two nodes of each control point according to T-mesh knot interval rules. The control points are a set of data that determine the shape of the T-spline surface of the error field of the CNC machine tool when the T-spline surface mesh is known; the T-spline surface is a surface obtained by fitting the discrete data using T-spline.
[0042] In step S2 of the present invention, the i-th control point P i For example, the node vector in the U direction is denoted as Ui =[u i0 ,u i1 ,u i2 ,u i3 ,u i4 ],u i0 ,u i1 ,u i2 ,u i3 ,u i4 Represent the node vector U i The 1st, 2nd, 3rd, 4th, and 5th elements in the V direction are denoted by V i =[v i0 ,v i1 ,v i2 ,v i3 ,v i4 ]v i0 ,v i1 ,v i2 ,v i3 ,v i4 Represent the node vector V i The 1st, 2nd, 3rd, 4th, and 5th elements in .
[0043] It should also be noted that when the difference between two discrete data is large, the area corresponding to the two discrete data in the T-spline surface mesh is regarded as an area that needs to be locally refined, and then the local refinement is performed.
[0044] S3, determine the position of each data point in the training set mapped on the T-spline surface grid, and for the kth data point Q in the training set k Nodes u are assigned in two directions parallel to the T-spline surface grid lines. k and v k .
[0045] S4. Design a regularization term function of the T-spline surface optimization function, use the penalty intensity parameter as a weight, add the T-spline surface errors of all data points in the training set to the weighted regularization term function to obtain a T-spline surface optimization function with a regularization term.
[0046] It should be noted that, in the present invention, for a T-spline surface mesh with n control points, the regularization term function is in the form of:
[0047]
[0048] Among them, g(P) represents the regularization function of the T-spline surface optimization function, which is used to measure the smoothness of the T-spline surface; Z x ,Z xxare the first-order partial derivative and second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface with respect to the x direction; Z y ,Z yy are the first-order partial derivative and second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface in the y direction; u i2 ,v i2 Respectively represent the i-th control point P i The node vector U i =[u i0 ,u i1 ,u i2 ,u i3 ,u i4 ] and V i =[v i0 ,v i1 ,v i2 ,v i3 ,v i4 ] is used to represent the i-th control point P i The location of the map on the T-spline surface mesh.
[0049] In the embodiment of the present invention, when calculating the first-order partial derivative and the second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface in the x and y directions, it is necessary to write the T-spline surface P(u, v) as the following parametric surface for easy calculation:
[0050]
[0051] Among them, x′, y′, z′ represent the x, y, z coordinate components of the three-dimensional coordinates of a point on the T-spline surface respectively; x(u,v), y(u,v), z(u,v) represent the functions of x′, y′, z′ with respect to u and v respectively; u and v represent the nodes in two directions parallel to the T-spline surface grid lines corresponding to a point on the T-spline surface grid.
[0052] It should be noted that, in the present invention, the kth data point Q in the training set k The T-spline surface error is the k-th data point Q in the training set k Its T-spline surface at the corresponding point P(u k ,v k ), then the T-spline surface error for m data points in the training set is:
[0053]
[0054] It should be noted that, in the present invention, for a T-spline surface mesh with n control points, the form of the T-spline surface optimization function is:
[0055]
[0056] Among them, λ represents a penalty intensity parameter greater than 0.
[0057] For the kth data point Q in the training set k The corresponding point P(u k ,v k ), which can be calculated according to the T-spline surface formula:
[0058]
[0059] Where P(u,v) represents the T-spline surface; n represents the number of control points; w i represents the i-th control point P i The weight of x i ,y i ,z i Respectively represent the x, y, and z coordinate components of the three-dimensional coordinates of the i-th control point; B i (u,v) represents the T-spline basis function of the i-th control point; Indicates that the node vector U i Cubic B-spline basis functions on ; Indicates the node vector V i Cubic B-spline basis functions on ;
[0060] In the present invention, the i-th control point P i The weight w i It can be set according to the actual situation. In this embodiment, the i-th control point P i The weight w i The default setting is 1.
[0061] In addition, when the T-spline surface mesh, each control point and its node vector are determined, the T-spline surface is uniquely determined by the above T-spline surface formula.
[0062] S5. Set the initial value of the coordinates of each control point in the T-spline surface mesh.
[0063] It should be noted that, in the present invention, for the node vector U i =[u i0 ,u i1 ,u i2 ,u i3 ,u i4 ] and V i =[v i0 ,v i1 ,v i2 ,v i3 ,v i4 ]’s ith control point P i, the initial value of its coordinates is set as follows: Get the maximum values X and Y of the x and y components of the three-dimensional coordinates of the data point in the training set, then set the i-th control point P i The initial value of the coordinate is [u i2 *X,v i2 *Y,0].
[0064] S6. Adjust the value of the penalty intensity parameter λ within a preset range, obtain a T-spline surface under each penalty intensity parameter value, calculate the error of each generated T-spline surface on the above test set, take the penalty intensity parameter λ with the smallest error as the optimal penalty intensity parameter, and take the T-spline surface with the smallest error as the final optimized CNC machine tool error field model to achieve T-spline model optimization of the error field in the CNC machine tool processing domain.
[0065] It should be noted that in the embodiment of the present invention, the preset range is [0.01, 1.00], and then the penalty intensity parameter λ is adjusted at intervals of 0.01 (0.01, 0.02, ..., 1.00) within this preset range, the purpose is to find a penalty intensity parameter that is as optimal as possible, and the penalty intensity parameter is greater than zero. Under each penalty intensity parameter λ value, a corresponding set of control point coordinates is obtained, and each set of control point coordinates can uniquely obtain a smooth T-spline surface. Subsequently, the error of each generated T-spline surface is verified on the test set, and it can be found that as the penalty intensity parameter λ increases, the error of the generated T-spline surface on the test set first decreases and then increases. That is because as the penalty intensity parameter changes from small to large, the generated T-spline surface changes from overfitting to a good fitting level and then to underfitting. Finally, the penalty intensity parameter selected when the error is minimum is used as the required optimal penalty intensity parameter, and the final T-spline surface is generated by the control point coordinates with the minimum error, and the final T-spline surface is used as the final optimized CNC machine tool error field model.
[0066] Figure 2 The difference between the T-spline used in the present invention and the traditional NURBS is explained. Since the T-spline allows the existence of T-shaped nodes in the grid, the method of the present invention can achieve local refinement in a real sense. The method of the present invention can greatly reduce redundant control points and reduce model complexity.
[0067] It should also be noted that the final optimized CNC machine error field model obtained by the present invention can be used for subsequent error compensation. Therefore, an error compensation method for a CNC machine error field based on a regularization term is also provided in an embodiment of the present invention, and the specific steps are as follows: obtaining discrete data to be error compensated in the CNC machine error field in the processing domain and preprocessing it, inputting the preprocessed discrete data into the CNC machine error field model optimized by the T-spline model optimization method for the CNC machine error field in the processing domain based on the regularization term described in the above embodiment, obtaining a T-spline surface of the CNC machine error field, applying an error compensation algorithm to the T-spline surface of the CNC machine error field, obtaining an error compensation result and feeding it back to the control system of the CNC machine processing domain, and finally outputting an optimized processing surface; wherein, the optimized processing surface is a surface formed by the actual motion trajectory after error compensation during the CNC machine processing of parts.
[0068] It should also be noted that the error compensation algorithm may adopt a mirror compensation algorithm and other existing methods, and their implementation methods all belong to the existing technology and will not be described in detail.
[0069] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.
Claims
1. A T-spline model optimization method for error field in machining domain of CNC machine tools based on regularization term, characterized in that: The following steps are involved: S1. Obtain discrete data of the error field in the CNC machine tool processing domain and perform preprocessing, and divide the preprocessed discrete data into two parts: a training set and a test set; wherein the discrete data is the difference between each data point in the real processing surface and each data point in the actual processing surface, the real processing surface is the surface formed by the theoretical motion trajectory during the CNC machine tool processing of parts, and the actual processing surface is the surface formed by the actual motion trajectory during the CNC machine tool processing of parts; S2, parameterizing the preprocessed discrete data into a two-dimensional square parameter domain as a T-spline surface mesh, locally refining the area in the T-spline surface mesh that needs to be locally refined, setting nodes in two directions parallel to the T-spline surface mesh lines on the mesh of each control point in the T-spline surface mesh, and generating node vectors corresponding to two nodes of each control point according to the T-spline vector node generation rule; wherein the control point is a set of data that determines the T-spline surface shape of the error field of the CNC machine tool when the T-spline surface mesh is known; and the T-spline surface is a surface obtained by fitting the discrete data using T-spline; S3, determining the position of each data point in the training set mapped on the T-spline surface grid, and assigning nodes in two directions parallel to the T-spline surface grid lines to each data point in the training set; S4, designing a regularization term function of the T-spline surface optimization function, taking the penalty intensity parameter as a weight, and adding the T-spline surface errors of all data points in the training set to the weighted regularization term function to obtain a T-spline surface optimization function with a regularization term; S5, setting the initial value of the coordinates of each control point in the T-spline surface grid; S6. Adjust the penalty intensity parameter value within a preset range, obtain a T-spline surface under each penalty intensity parameter value, calculate the error of each generated T-spline surface on the test set, take the penalty intensity parameter with the smallest error as the optimal penalty intensity parameter, take the T-spline surface with the smallest error as the final optimized CNC machine tool error field model, and realize the T-spline model optimization of the CNC machine tool processing domain error field.
2. The T-spline model optimization method for the error field in the machining domain of a CNC machine tool based on a regularization term as claimed in claim 1, characterized in that: In step S4, for a T-spline surface mesh with n control points, the regularization term function g(P) is in the form of: Among them, Z x , Z xx are the first-order partial derivative and second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface with respect to the x direction; Z y , Z yy are the first-order partial derivative and second-order partial derivative of the z-coordinate component z′ of the three-dimensional coordinates of a point in the T-spline surface in the y direction; u i2 , v i2 Respectively represent the i-th control point P i The node vector U i and V i The third component in .
3. The T-spline model optimization method for the error field in the machining domain of a CNC machine tool based on a regularization term as claimed in claim 2, characterized in that: In step S4, the kth data point Q in the training set k The T-spline surface error is the k-th data point Q in the training set k Its T-spline surface at the corresponding point P(u k , v k ), then the T-spline surface error of all data points in the training set is: Wherein, m represents the number of data points in the training set.
4. The T-spline model optimization method for the error field in the machining domain of a CNC machine tool based on a regularization term as claimed in claim 3, characterized in that: In step S4, for a T-spline surface mesh with n control points, the T-spline surface optimization function is in the form of: Among them, λ represents a penalty intensity parameter greater than 0.
5. The method for optimizing the T-spline model of the error field in the machining domain of a CNC machine tool based on regularization term according to claim 4, characterized in that: For the kth data point Q in the training set k The corresponding point P(u k , v k ), which is calculated according to the following T-spline surface formula: Where P(u, v) represents the T-spline surface; n represents the number of control points; w i represents the i-th control point P i The weight of x i ,y i , z i Respectively represent the x, y, and z coordinate components of the three-dimensional coordinates of the i-th control point; B i (u, v) represents the T-spline basis function of the i-th control point; Indicates that the node vector U i Cubic B-spline basis functions on ; Indicates the node vector V i Cubic B-spline basis functions on .
6. The T-spline model optimization method for the error field in the machining domain of a CNC machine tool based on a regularization term as claimed in claim 2, characterized in that: For the node vector U i and V i The i-th control point P i , the initial value of its coordinates is set as follows: Get the maximum values X and Y of the x and y components of the three-dimensional coordinates of the data point in the training set, then set the i-th control point P i The initial value of the coordinate is [u i2 *X,v i2 *Y,0].
7. An error compensation method for a CNC machine tool machining domain error field based on a regularization term, characterized in that: The specific steps are as follows: obtaining discrete data to be error compensated in the error field of the CNC machine tool processing domain and preprocessing the data, inputting the preprocessed discrete data into the CNC machine tool error field model optimized by the T-spline model optimization method of the CNC machine tool processing domain error field based on the regularization term according to claims 1 to 6, obtaining a T-spline surface of the CNC machine tool error field, applying an error compensation algorithm to the T-spline surface of the CNC machine tool error field, obtaining an error compensation result and feeding it back to the control system of the CNC machine tool processing domain, and finally outputting an optimized processing surface; wherein, the optimized processing surface is a surface formed by the actual motion trajectory after error compensation during the CNC machine tool processing of parts.
Citation Information
Patent Citations
Machining error rapid detection and compensation method based on NURBS curved surface
CN103777570A
Regular triangular mesh spiral machining path generation method
CN105045204A
Offset calculation method for T-spline curved surface
CN112231901A
Method for compensating geometric errors of three-coordinate measuring machine based on elastic network algorithm
CN113932707A
Internet of Things cloud equipment data acquisition system based on VPN (Virtual Private Network)
CN114253242A
