A method for dynamically adjusting the accuracy of the guide rails of a large CNC gantry milling machine bed

Through computer simulation technology and pseudo-inverse algorithm, dynamically adjusting the joint angle and execution speed of the guide rail of the CNC gantry milling machine, the complexity of the guide rail accuracy adjustment is solved, and the machining accuracy and the flexibility and stability of the machine tool are improved.

CN118768620BActive Publication Date: 2024-11-29启东大同电机有限公司
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411269673.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-11
Publication Date
2024-11-29
Estimated Expiration
2044-09-11

AI Technical Summary

Technical Problem

In the prior art, the guide rail accuracy adjustment cannot be achieved by adjusting the joint angle of the moving parts and the execution speed of the end effector, resulting in low machining accuracy, increased complexity, and lack of effective solutions.

Method used

Using computer simulation technology to establish a finite element model of CNC gantry milling machine, using pseudo-inverse algorithm and motion state simulation, dynamically adjust the joint angle of moving parts and the speed of the end effector, and optimize the parallelism and straight lines of the guide rails by comparing error values ​​and expected values. , achieve accuracy adjustment.

Benefits of technology

Improve the flexibility and reliability of the machine tool, accurately obtain the guide rail accuracy deviation value, deeply understand the relationship between joint angle, execution speed and guide rail accuracy, enhance the stability and adaptability of the machine tool in different environments, and ensure that the guide rails are in various places. Efficient and accurate operation under various conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118768620B_ABST
    Figure CN118768620B_ABST
Patent Text Reader

Abstract

The invention discloses a method for dynamically adjusting the accuracy of a bed guide rail of a large-scale CNC gantry milling machine, and relates to the field of dynamic adjustment of guide rail accuracy. The method comprises the following steps: using computer simulation technology to establish a finite element model of a CNC gantry milling machine, and using a collimator as a measurement reference in the finite element model to simulate and obtain the parallelism and straightness of the bed guide rail of the milling machine when working; comparing the parallelism and straightness of the guide rails with the set expected values ​​to obtain error values, and based on the error values, reversely judging the joint angles of the moving parts and the speed of the end effector in the milling machine, and taking the joint angles of the moving parts and the speed of the end effector as the motion state of the milling machine; simulating the change of the error value under set conditions based on the motion state and the pseudo-inverse algorithm, and designing a time-varying function according to the simulation results to perform motion planning adjustment on the moving parts and the end effector. The invention achieves the purpose of dynamically adjusting the accuracy of the guide rails, and improves the flexibility and reliability of the machine tool.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of dynamic adjustment of guide rail precision, and in particular to a method for dynamic adjustment of guide rail precision of a bed of a large-scale CNC gantry milling machine. Background Art

[0002] A CNC gantry milling machine is a large machine tool controlled by a CNC system and used to process large workpieces. It is usually composed of components such as a beam, a column, a workbench, and a CNC system. A CNC gantry milling machine has the ability to perform high-precision processing and can achieve complex processing of large workpieces to ensure processing accuracy and surface quality. A CNC gantry milling machine has a variety of processing functions and can perform milling, drilling, tapping and other processing operations. It is suitable for the processing needs of complex workpieces. The CNC system can flexibly adjust processing parameters and processing paths according to processing requirements to adapt to the processing requirements of different workpieces, and has strong flexibility and applicability.

[0003] The guide rails of a milling machine bed refer to the guide rail structure on the bed of the milling machine, which is used to support and guide the movement of the worktable or beam. The design and quality of the guide rails directly affect the machining accuracy and stability of the milling machine. Common guide rails for milling machines include sliding guides, rolling guides, inclined guides and linear guides. The accuracy of the guide rails directly affects the machining accuracy of mechanical equipment. In the manufacturing industry, the machining requirements of many workpieces are very high, and they need to be precisely machined on a millimeter or even sub-millimeter scale. If the guide rail accuracy is inaccurate, it will lead to deviations in the machining size of the workpiece, unsatisfactory surface quality, and even failure to meet the design requirements, affecting product quality and performance.

[0004] However, in the prior art, the dynamic adjustment technology of guide rail accuracy cannot achieve the purpose of guide rail accuracy adjustment by adjusting the joint angle of the moving parts and the execution speed of the end effector. The adjustment accuracy is mostly achieved by adjusting the structure of the guide rail, which increases the complexity of the adjustment.

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

[0006] In order to solve the above problems, the present invention proposes a method for dynamically adjusting the accuracy of the guide rails of a large CNC gantry milling machine bed, which can dynamically adjust the joint angles of moving parts and the speed of the end effector to achieve the purpose of dynamically adjusting the guide rail accuracy.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] The present invention provides a method for dynamically adjusting the accuracy of a bed guide rail of a large-scale CNC gantry milling machine, and the method comprises the following steps:

[0009] S1. Use computer simulation technology to establish a finite element model of a CNC gantry milling machine, and use a collimator as a measurement reference in the finite element model to simulate and obtain the parallelism and straightness of the guide rails of the milling machine bed during operation;

[0010] S2, comparing the parallelism and straightness of the guide rail with the set expected values ​​to obtain error values, and based on the error values, reversely judging the joint angle of the moving parts and the speed of the end effector in the milling machine, and taking the joint angle of the moving parts and the speed of the end effector as the motion state of the milling machine;

[0011] S3, based on the motion state and pseudo-inverse algorithm, simulate the change of error value under set conditions, and design time-varying functions according to the simulation results to adjust the motion planning of the moving parts and the end effector;

[0012] S4. Apply motion planning to the finite element model to obtain the adjusted guide rail parallelism and straightness, and compare the guide rail parallelism and straightness with the set expected values ​​to determine the degree of optimization of the guide rail accuracy;

[0013] S5. If the comparison result is less than the expected value, return to step S3 to repeat the simulation optimization. If the comparison result is greater than the expected value, the motion planning is applied as an adjustment method to the CNC gantry milling machine controller to adjust the guide rail accuracy.

[0014] Furthermore, the guide rail parallelism and straightness are compared with the set expected values ​​to obtain error values, and the joint angle of the moving part and the speed of the end effector in the milling machine are reversely judged based on the error values. The joint angle of the moving part and the speed of the end effector are used as the motion state of the milling machine, which includes the following steps:

[0015] S21. Conduct independence test on the expected value based on collinearity diagnosis and correlation analysis, and use the information method to determine the information value of the expected value and implement weight quantification;

[0016] S22. Building an error comparison model based on the weighted quantitative results and linear regression technology, and inputting the guide rail parallelism and straightness into the error comparison model to obtain error values ​​as guide rail accuracy deviation values;

[0017] S23. Combining the error value with the reverse engineering method, it is inferred that when the guide rail has an accuracy deviation, the joint angle of the moving parts in the milling machine and the execution speed of the end effector during operation, and the joint angle and the execution speed of the end effector are used as the motion state of the milling machine.

[0018] Furthermore, based on collinearity diagnosis and correlation analysis, the independence test of the set expected value is carried out, and the information value of the expected value is determined by the information method. The implementation of weight quantification includes the following steps:

[0019] S211. Analyze the linear correlation between variables within the set expected value, and use tolerance and variance inflation factor based on the correlation results to perform multicollinearity diagnosis on the set expected value, where the variables include the set parallelism value and straightness value;

[0020] S212. Determine the independence of variables based on the collinearity diagnosis results, evaluate the information value of the variables based on the independence results and the information method, and sum the information values ​​of the variables as a comparison index;

[0021] S213. Construct a judgment matrix according to the analysis method to implement weight allocation for the comparison indicators, and define the information value based on the weight allocation result.

[0022] Furthermore, the error value is combined with the reverse engineering method to infer the joint angle of the moving parts in the milling machine and the execution speed of the end effector when the guide rail has an accuracy deviation. The joint angle and the execution speed of the end effector are used as the motion state of the milling machine, which includes the following steps:

[0023] S231, extracting three-dimensional information of the bed guide rail, the milling machine moving parts and the end effector in the finite element model and converting them into component data points;

[0024] S232, using the adjacency matrix and the working principle of the milling machine to describe the intrinsic relationship between the joint angle of the moving part, the execution speed of the end effector and the accuracy of the guide rail to create a three-dimensional model of the part;

[0025] S233, performing a local polynomial surface operation on the component data points, determining the projection residual of the component data points on the surface to obtain a feature point index, and combining the feature point index with a broken line growth technique to extract a feature line of the component three-dimensional model;

[0026] S234. Based on the feature lines, the three-dimensional model of the component is transformed into a voxel matrix to describe the differences between the three-dimensional features when the guide rail has parallelism and straightness errors, and the convolutional neural network is trained through two-dimensional typical views. After forward and backward propagation operations, the joint angle and execution speed recognition results are output, and the joint angle and execution speed are used as the motion state of the milling machine.

[0027] Further, the component data points are subjected to a local polynomial surface operation, the projection residual of the component data points in the surface is determined to obtain the feature point index, and the feature point index is combined with the broken line growth technology to extract the feature line of the component three-dimensional model, including the following steps:

[0028] S2331, fitting each data point in the three-dimensional model of the part into a local polynomial surface in its adjacent range, and selecting any component data point in the three-dimensional model of the component to analyze whether it meets the constraint conditions;

[0029] S2332, deriving the projection residuals of all neighborhood points of the component data point, and selecting the maximum value as the projection residual of the component data point, and comparing the projection residual with a preset critical value, where the preset critical value represents the residual mean of all component data points;

[0030] S2333, if the projection residual is greater than the preset critical value, the component data point is considered to be a feature point; if the projection residual is less than the preset critical value, return to step S2332 to reselect and compare;

[0031] S2334, creating a sequence for storing feature points, using component analysis to calculate the correlation index between feature points, obtaining the correlation index of all feature points and sorting them in order of high and low, and saving the sorting result in the sequence;

[0032] S2335. Add an evaluation direction in the component three-dimensional model as an operation termination condition, and select several component data points in the sequence as seed points to simulate broken line growth in the component three-dimensional model to extract feature lines of the component three-dimensional model.

[0033] Furthermore, the calculation formula of the correlation index is:

[0034] ;

[0035] In the formula, or Represents the relevant index value; P Represents a sequence for saving feature points; , , They respectively represent the eigenvalues ​​of the feature points in their neighborhood points.

[0036] Furthermore, based on the feature lines, the three-dimensional model of the component is transformed into a voxel matrix to describe the difference between the three-dimensional features when the guide rail has parallelism and straightness errors, and the convolutional neural network is trained through two-dimensional typical views. After forward and backward propagation operations, the joint angle and execution speed recognition results are output, and the joint angle and execution speed are used as the motion state of the milling machine, which includes the following steps:

[0037] S2341, selecting the highest and lowest values ​​of feature point coordinates in the component three-dimensional model based on the feature line, and combining the feature point coordinates with a preset voxel unit coordinate system to set the coordinate axis direction to be parallel to the initial coordinate axis;

[0038] S2342, align the axis bounding box of the component 3D model and x , y , z The edges parallel to the axis are divided into equal distances in turn, and the model is divided along the x , y , zThe number of rows, columns and layers of the axis divided is represented as a voxel matrix to complete the voxelization process;

[0039] S2343, based on the voxel matrix describing the difference between the three-dimensional features when the guide rail has parallelism and straightness errors, randomly expanding the voxelized three-dimensional model of the component to obtain a voxel enhancement data set of a set multiple;

[0040] S2344. The convolutional neural network is trained through two-dimensional typical views combined with forward and backward propagation. The parameter values ​​of the three-dimensional model of the component are output after the forward and backward propagation training to obtain the joint angle and execution speed recognition results, and the joint angle and execution speed are used as the motion state of the milling machine.

[0041] Furthermore, based on the motion state and the pseudo-inverse algorithm, the variation of the error value under the set conditions is simulated, and the time-varying function is designed according to the simulation results to adjust the motion planning of the moving parts and the end effector, including the following steps:

[0042] S31, setting a plurality of sets of joint angle values ​​and actuator speed values ​​in the component three-dimensional model to simulate the motion process of the moving component and the end effector;

[0043] S32, combining the pseudo-inverse algorithm with the simulation results to obtain the change trend of the guide rail accuracy error value, and describing the correlation between the joint angle, execution speed and guide rail accuracy according to the change trend;

[0044] S33. Based on the influence relationship and simulation results, a time-varying function is designed to obtain an error threshold, and a motion planning adjustment scheme is designed according to the error threshold to correct the joint angle of the moving parts and the execution speed of the end effector.

[0045] Furthermore, the pseudo-inverse algorithm is combined with the simulation results to obtain the change trend of the guide rail accuracy error value, and the correlation between the joint angle, execution speed and guide rail accuracy is described according to the change trend, including the following steps:

[0046] S321, performing redundancy analysis on the motion process based on the pseudo-inverse algorithm and the simulation results, and judging the trend of the error value changing with the joint angle and the execution speed according to the analysis results;

[0047] S322, fitting the guide rail accuracy error value change function according to the judgment result to obtain the change of the joint angle and the execution speed;

[0048] S323. Analyze the influence relationship between the joint angle, execution speed and guide rail accuracy based on the change situation to obtain the correlation result.

[0049] Further, the motion planning adjustment scheme includes:

[0050] When the guide rail accuracy error |e| Greater than the error thresholdt When , the adjustment range of the moving part joint angle is set to △ i = k θ . e ,in k θ is the joint angle adjustment coefficient, k θ The value range is [0.1-0.5], and the end effector execution speed is adjusted to △ v = k v . e ,in k v is the joint angle adjustment coefficient, k v The value range of is [0.1-0.5];

[0051] When the guide rail accuracy error |e| Less than the error threshold t When , the adjustment range of the moving part joint angle is set to △ i = k ' θ . e ,in k ' θ is the joint angle adjustment coefficient, k ' θ The value range is [0.001-0.1], and the end effector execution speed is adjusted to △ v = k ' v . e ,in k ' v is the joint angle adjustment coefficient, k ' v The value range of is [0.001-0.1];

[0052] When the guide rail accuracy error |e| Equal to the error threshold t , the original joint angle and execution speed remain unchanged.

[0053] The beneficial effects of the present invention are:

[0054] 1. The present invention establishes a finite element model by utilizing computer simulation technology. It can accurately simulate the parallelism and straightness of the guide rails of the milling machine bed without actually operating the milling machine, so as to facilitate the later evaluation and optimization of performance. At the same time, by comparing the simulated guide rail parallelism and straightness with the expected values, the current guide rail accuracy state can be understood. By utilizing the pseudo-inverse algorithm and the simulation of the motion state, the joint angles of the moving parts and the speed of the end effector can be dynamically adjusted to achieve the purpose of dynamically adjusting the guide rail accuracy, thereby improving the flexibility and reliability of the machine tool.

[0055] 2. The present invention compares the parallelism and straightness of the guide rails with the set expected values ​​and reversely judges the joint angles of the moving parts and the speed of the end effector in the milling machine based on the error values, so as to accurately obtain the deviation value of the guide rail accuracy, and provide an accurate basis for subsequent adjustments. By performing an independence test on the expected values ​​and using the information method for weight quantification, it can be ensured that in error evaluation and subsequent processing, the error comparison model constructed using linear regression technology can accurately describe the relationship between the parallelism and straightness of the guide rails and the expected values, and provide a scientific method for error quantification. At the same time, the error value is combined with the reverse engineering method to infer the specific motion state of the moving parts, including the joint angles and the speed of the end effector, when the guide rail has an accuracy deviation. This helps to identify the root cause of the accuracy deviation and provide a basis for adjustment.

[0056] 3. The present invention simulates the change of error values ​​under set conditions based on motion state and pseudo-inverse algorithm, and performs motion planning adjustment on moving parts and end effectors according to the simulation results. It can accurately obtain the changing trend of guide rail accuracy error values, deeply understand the complex correlation between joint angles, execution speeds and guide rail accuracy, and thus provide a theoretical basis for precise adjustment. At the same time, by designing time-varying functions and motion planning adjustment schemes according to error thresholds, the machine tool can dynamically adapt to different working conditions and requirements, enhance the stability and adaptability of the machine tool in different environments, and optimize the motion performance of the machine tool by simulating different joint angle values ​​and actuator speed values, ensuring that the guide rail can maintain efficient and precise operation under various working conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0058] Figure 1 The present invention is a flowchart of a method for dynamically adjusting the accuracy of a bed guide rail of a large CNC gantry milling machine according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0060] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0061] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0062] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0063] See also Figure 1 The present invention provides a method for dynamically adjusting the accuracy of a bed guide rail of a large CNC gantry milling machine, and the method comprises the following steps:

[0064] S1. Use computer simulation technology to establish a finite element model of a CNC gantry milling machine, and use the collimator as the measurement reference in the finite element model to simulate and obtain the parallelism and straightness of the milling machine bed guide rails during operation.

[0065] In this embodiment, a finite element model of a CNC gantry milling machine is established by using computer simulation technology, and a collimator is used as a measurement reference in the finite element model to simulate and obtain the parallelism and straightness of the guide rails of the milling machine bed during operation, which generally includes the following steps:

[0066] The 3D model of the CNC gantry milling machine is established using computer-aided design software, including drawing the gantry, worktable, beam and other main structural components, assembling them into a complete machine tool model, and importing the model into the finite element analysis (FEA) software to establish a mathematical model.

[0067] In the finite element model, it is necessary to define boundary conditions, including the constraints and loading conditions of each machine tool component. After the finite element model is established and the boundary conditions are defined, numerical simulation analysis is performed, which includes solving the static or dynamic equations of the model and calculating the stress, deformation and other physical quantities of the guide rail under different working conditions. The deformation of the guide rail can be obtained through simulation analysis, and then its parallelism and straightness can be evaluated.

[0068] In the finite element simulation, the collimator is used as the measurement reference to simulate the parallelism and straightness of the guide rail during operation.

[0069] S2. Compare the parallelism and straightness of the guide rail with the set expected values ​​to obtain error values, and reversely judge the joint angles of the moving parts and the speed of the end effector in the milling machine based on the error values, and use the joint angles of the moving parts and the speed of the end effector as the motion state of the milling machine.

[0070] In this embodiment, the guide rail parallelism and straightness are compared with the set expected values ​​to obtain error values, and the joint angle of the moving part and the speed of the end effector in the milling machine are reversely judged based on the error values. The joint angle of the moving part and the speed of the end effector are used as the motion state of the milling machine, which includes the following steps:

[0071] S21. Based on collinearity diagnosis and correlation analysis, the independence test of the set expected value is carried out, and the information amount method is used to determine the information amount of the expected value to implement weight quantification.

[0072] Specifically, the independence test of the set expected value is carried out based on collinearity diagnosis and correlation analysis, and the information value of the expected value is determined by the information method to implement weight quantification, which includes the following steps:

[0073] S211. Analyze the linear correlation between variables within the set expected value, and use tolerance and variance inflation factor based on the correlation results to perform multicollinearity diagnosis on the set expected value, where the variables include the set parallelism value and straightness value;

[0074] S212. Determine the independence of variables based on the collinearity diagnosis results, evaluate the information value of the variables based on the independence results and the information method, and sum the information values ​​of the variables as a comparison index;

[0075] S213. Construct a judgment matrix according to the analysis method to implement weight allocation for the comparison indicators, and define the information value based on the weight allocation result.

[0076] S22. An error comparison model is constructed based on the weighted quantitative results and linear regression technology, and the guide rail parallelism and straightness are input into the error comparison model to obtain error values ​​as guide rail accuracy deviation values.

[0077] Specifically, the error comparison model is constructed based on the weighted quantitative results and the linear regression technology, and the guide rail parallelism and straightness are input into the error comparison model to obtain the error value as the guide rail accuracy deviation value, including the following steps:

[0078] Determine the weights of rail parallelism and straightness in error comparison. The weights can be set according to the importance and influence of rail accuracy. They can be determined based on expert experience or data analysis, and the collected rail parallelism and straightness data can be preprocessed.

[0079] The error comparison model was constructed using linear regression technology. The guide rail parallelism and straightness were used as independent variables, and the error value was used as the dependent variable to establish a linear regression model. The constructed linear regression model was verified and evaluated. The new guide rail parallelism and straightness data were input into the error comparison model, and the corresponding error value was obtained through model calculation.

[0080] S23. Combining the error value with the reverse engineering method, it is inferred that when the guide rail has an accuracy deviation, the joint angle of the moving parts in the milling machine and the execution speed of the end effector during operation, and the joint angle and the execution speed of the end effector are used as the motion state of the milling machine.

[0081] Specifically, the error value is combined with the reverse engineering method to infer the joint angle of the moving parts in the milling machine and the execution speed of the end effector when the guide rail has an accuracy deviation. The joint angle and the execution speed of the end effector are used as the motion state of the milling machine, which includes the following steps:

[0082] S231, extracting three-dimensional information of the bed guide rail, the milling machine moving parts and the end effector in the finite element model and converting them into component data points;

[0083] S232, using the adjacency matrix and the working principle of the milling machine to describe the intrinsic relationship between the joint angle of the moving part, the execution speed of the end effector and the accuracy of the guide rail to create a three-dimensional model of the part;

[0084] S233, perform a local polynomial surface operation on the component data points, determine the projection residual of the component data points on the surface to obtain the feature point index, and combine the feature point index with the broken line growth technology to extract the feature line of the component three-dimensional model.

[0085] Among them, performing a local polynomial surface operation on the component data points, determining the projection residual of the component data points on the surface to obtain the feature point index, and combining the feature point index with the broken line growth technology to extract the feature line of the component three-dimensional model includes the following steps:

[0086] S2331, fitting each data point in the three-dimensional model of the part into a local polynomial surface in its adjacent range, and selecting any component data point in the three-dimensional model of the component to analyze whether it meets the constraint conditions;

[0087] S2332, deriving the projection residuals of all neighborhood points of the component data point, and selecting the maximum value as the projection residual of the component data point, and comparing the projection residual with a preset critical value, where the preset critical value represents the residual mean of all component data points;

[0088] S2333, if the projection residual is greater than the preset critical value, the component data point is considered to be a feature point; if the projection residual is less than the preset critical value, return to step S2332 to reselect and compare;

[0089] S2334, creating a sequence for storing feature points, using component analysis to calculate the correlation index between feature points, obtaining the correlation index of all feature points and sorting them in order of high and low, and saving the sorting result in the sequence;

[0090] The calculation formula of the correlation index is:

[0091] ;

[0092] In the formula, or Represents the relevant index value; P Represents a sequence for saving feature points; , , They respectively represent the eigenvalues ​​of the feature points in their neighborhood points.

[0093] S2335. Add an evaluation direction in the component three-dimensional model as an operation termination condition, and select several component data points in the sequence as seed points to simulate broken line growth in the component three-dimensional model to extract feature lines of the component three-dimensional model.

[0094] The termination condition represents the average of the force exerted on the guide rail by the joint angle of the moving part and the execution speed of the end effector.

[0095] S234. Based on the feature lines, the three-dimensional model of the component is transformed into a voxel matrix to describe the differences between the three-dimensional features when the guide rail has parallelism and straightness errors, and the convolutional neural network is trained through two-dimensional typical views. After forward and backward propagation operations, the joint angle and execution speed recognition results are output, and the joint angle and execution speed are used as the motion state of the milling machine.

[0096] Among them, based on the feature line, the three-dimensional model of the component is transformed into a voxel matrix to describe the difference between the three-dimensional features when the guide rail has parallelism and straightness errors, and the convolutional neural network is trained through two-dimensional typical views. After forward and backward propagation operations, the joint angle and execution speed recognition results are output, and the joint angle and execution speed are used as the motion state of the milling machine. The following steps are included:

[0097] S2341, selecting the highest and lowest values ​​of feature point coordinates in the component three-dimensional model based on the feature line, and combining the feature point coordinates with a preset voxel unit coordinate system to set the coordinate axis direction to be parallel to the initial coordinate axis;

[0098] S2342, align the axis bounding box of the component 3D model and x , y , z The edges parallel to the axis are divided into equal distances in turn, and the model is divided along the x , y , z The number of rows, columns and layers of the axis divided is represented as a voxel matrix to complete the voxelization process;

[0099] S2343, based on the voxel matrix describing the difference between the three-dimensional features when the guide rail has parallelism and straightness errors, randomly expanding the voxelized three-dimensional model of the component to obtain a voxel enhancement data set of a set multiple;

[0100] S2344. The convolutional neural network is trained through two-dimensional typical views combined with forward and backward propagation. The parameter values ​​of the three-dimensional model of the component are output after the forward and backward propagation training to obtain the joint angle and execution speed recognition results, and the joint angle and execution speed are used as the motion state of the milling machine.

[0101] Therefore, by comparing the parallelism and straightness of the guide rails with the set expected values ​​and reversely judging the joint angles of the moving parts and the speed of the end effector in the milling machine based on the error values, the deviation value of the guide rail accuracy can be accurately obtained, providing an accurate basis for subsequent adjustments. By conducting an independence test on the expected values ​​and using the information method for weight quantification, it can be ensured that in error evaluation and subsequent processing, the error comparison model constructed using linear regression technology can accurately describe the relationship between the parallelism and straightness of the guide rails and the expected values.

[0102] S3. Based on the motion state and pseudo-inverse algorithm, simulate the change of error value under set conditions, and design time-varying functions according to the simulation results to adjust the motion planning of the moving parts and the end effector.

[0103] In this embodiment, based on the motion state and the pseudo-inverse algorithm, simulating the change of the error value under the set conditions, and designing a time-varying function according to the simulation results to adjust the motion planning of the moving part and the end effector includes the following steps:

[0104] S31. Setting a plurality of sets of joint angle values ​​and actuator speed values ​​in the three-dimensional model of the component to simulate the motion process of the moving component and the end effector.

[0105] Specifically, several groups of joint angle values ​​of moving parts are set in the three-dimensional model of the parts. The joint angle values ​​can represent the position and posture of the moving parts under different working conditions. At the same time, the speed values ​​of the corresponding end effectors are set. The actuator speed values ​​can represent the movement speed of the end effector during the processing. The established three-dimensional model is imported into the corresponding simulation software to simulate the movement process and interaction of machine tool parts.

[0106] S32. Combining the pseudo-inverse algorithm with the simulation results to obtain the changing trend of the guide rail accuracy error value, and describing the correlation between the joint angle, execution speed and guide rail accuracy based on the changing trend.

[0107] Specifically, the pseudo-inverse algorithm is combined with the simulation results to obtain the change trend of the guide rail accuracy error value, and the correlation between the joint angle, execution speed and guide rail accuracy is described according to the change trend, including the following steps:

[0108] S321, performing redundancy analysis on the motion process based on the pseudo-inverse algorithm and the simulation results, and judging the trend of the error value changing with the joint angle and the execution speed according to the analysis results;

[0109] S322, fitting the guide rail accuracy error value change function according to the judgment result to obtain the change of the joint angle and the execution speed;

[0110] S323. Analyze the influence relationship between the joint angle, execution speed and guide rail accuracy based on the change situation to obtain the correlation result.

[0111] It should be explained that the change of joint angle will directly affect the position and posture of the end effector, and then affect the precision error of the guide rail. The change of execution speed will also affect the dynamic response and stability of the CNC gantry milling machine, thereby indirectly affecting the precision error of the guide rail. Therefore, the relationship between joint angle, execution speed and guide rail precision mainly includes the following aspects:

[0112] Relationship between joint angle and guide rail accuracy: Changes in joint angles will directly affect the position and posture of the moving parts, thereby affecting the accuracy error of the guide rail. Under normal circumstances, large changes in joint angles may cause an increase in the position deviation of the end effector, which in turn leads to an increase in the guide rail accuracy error.

[0113] Relationship between execution speed and guide rail accuracy: Changes in execution speed will affect the dynamic response and stability of the end effector, and thus affect the accuracy error of the guide rail. High-speed movement may cause vibration and inertia effects, thereby increasing the guide rail accuracy error, while low-speed movement may reduce the dynamic response of the end effector and may also lead to an increase in accuracy error.

[0114] Relationship between joint angle and execution speed: Changes in joint angle usually affect the selection and adjustment of execution speed. For example, in some cases, in order to ensure the stability and accuracy of the CNC gantry milling machine, it may be necessary to reduce the execution speed to reduce the impact of vibration and inertia effects.

[0115] S33. Based on the influence relationship and simulation results, a time-varying function is designed to obtain an error threshold, and a motion planning adjustment scheme is designed according to the error threshold to correct the joint angle of the moving parts and the execution speed of the end effector.

[0116] Movement planning adjustments include:

[0117] When the guide rail accuracy error |e| Greater than the error threshold t When , the adjustment range of the moving part joint angle is set to △ i = k θ . e ,in k θ is the joint angle adjustment coefficient, k θ The value range is [0.1-0.5]. For example, k θ =0.1, and when the error value is 10, the joint angle is adjusted to 1°.

[0118] The end effector execution speed is adjusted to △ v = k v . e ,in k v is the joint angle adjustment coefficient, k v The value range is [0.1-0.5]. For example, k v =0.5, and the error value is 10, the execution speed is adjusted to 5rad / s.

[0119] When the guide rail accuracy error |e| Less than the error threshold t When , the adjustment range of the moving part joint angle is set to △ i = k ' θ . e ,in k ' θ is the joint angle adjustment coefficient, k ' θ The value range is [0.001-0.1]. For example,k θ =0.001, and when the error value is 10, the joint angle is adjusted to 0.01°.

[0120] The end effector execution speed is adjusted to △ v = k ' v . e ,in k ' v is the joint angle adjustment coefficient, k ' v The value range is [0.001-0.1]. For example, k v =0.005, and when the error value is 10, the execution speed is adjusted to 0.05rad / s.

[0121] When the guide rail accuracy error |e| Equal to the error threshold t , the original joint angle and execution speed remain unchanged.

[0122] Therefore, by simulating the change of error values ​​under set conditions based on the motion state and pseudo-inverse algorithm, and adjusting the motion planning of the moving parts and the end effector according to the simulation results, it is possible to accurately obtain the changing trend of the guide rail accuracy error value, and deeply understand the complex correlation between the joint angle, execution speed and guide rail accuracy, thereby providing a theoretical basis for precise adjustment. At the same time, by designing time-varying functions and designing motion planning adjustment schemes according to error thresholds, the machine tool can dynamically adapt to different working conditions and requirements, enhance the stability and adaptability of the machine tool in different environments, and optimize the motion performance of the machine tool by simulating different joint angle values ​​and actuator speed values, ensuring that the guide rail can maintain efficient and precise operation under various working conditions.

[0123] S4. Apply motion planning to the finite element model to obtain the adjusted guide rail parallelism and straightness, and compare the guide rail parallelism and straightness with the set expected values ​​to determine the degree of optimization of the guide rail accuracy.

[0124] In this embodiment, applying motion planning to the finite element model to obtain the adjusted guide rail parallelism and straightness, and comparing the guide rail parallelism and straightness with the set expected values ​​to determine the degree of guide rail precision optimization includes:

[0125] Apply the motion-planned parameters to the finite element model, perform simulation or analysis within the finite element model, and obtain the adjusted guide rail parallelism and straightness data to reflect the actual accuracy of the guide rail after the motion planning is applied. Set the expected values ​​of guide rail parallelism and straightness according to design requirements and actual needs.

[0126] Compare the obtained guide rail parallelism and straightness data with the set expected values, and judge whether the guide rail accuracy has reached the expected value through comparison analysis. If it reaches or is close to the expected value, it means that the guide rail accuracy has been optimized. If there is a large deviation from the expected value, further optimization is required.

[0127] Based on the comparison results, the degree of optimization of the guide rail accuracy is determined. If the guide rail accuracy has met the requirements, no further optimization is required. If there is a large deviation, further optimization measures need to be considered, such as adjusting motion planning parameters, optimizing structural design, etc.

[0128] S5. If the comparison result is less than the expected value, return to step S3 to repeat the simulation optimization. If the comparison result is greater than the expected value, the motion planning is applied as an adjustment method to the CNC gantry milling machine controller to adjust the guide rail accuracy.

[0129] Specifically, a performance evaluation is conducted based on the experimental results to evaluate the optimization effect of motion planning as an adjustment method and the degree of improvement in guide rail accuracy. If the effect is achieved, the adjustment scheme can be formally applied to the CNC gantry milling machine controller for guide rail accuracy adjustment in the actual machining process. Applying motion planning as an adjustment method to the CNC gantry milling machine controller can achieve guide rail accuracy adjustment.

[0130] To sum up, with the help of the above-mentioned technical scheme, the present invention establishes a finite element model by utilizing computer simulation technology. It can accurately simulate the parallelism and straightness of the guide rails of the milling machine bed without actually operating the milling machine, so as to facilitate the later evaluation and optimization of performance. At the same time, by comparing the simulated guide rail parallelism and straightness with the expected values, the current guide rail accuracy state can be understood, and the pseudo-inverse algorithm and motion state simulation can be used to dynamically adjust the joint angles of the moving parts and the speed of the end effector to achieve the purpose of dynamically adjusting the guide rail accuracy, thereby improving the flexibility and reliability of the machine tool. The present invention aims at the comparison between the parallelism and straightness of the guide rails and the set expected values, and reversely judges the joint angles of the moving parts and the speed of the end effector in the milling machine based on the error values, so as to accurately obtain the deviation value of the guide rail accuracy, and provide an accurate basis for subsequent adjustments. By performing an independence test on the expected values ​​and using the information quantity method for weight quantification, it can be ensured that in error evaluation and subsequent processing, the error comparison model constructed by the linear regression technology can accurately describe the relationship between the parallelism and straightness of the guide rails and the expected values, and provide a scientific method for error quantification. At the same time, the error value is combined with the reverse engineering method, so as to infer the specific motion state of the moving parts, including the joint angles and the speed of the end effector, when the guide rail has an accuracy deviation, which helps to identify the root cause of the accuracy deviation and provide a basis for adjustment. The present invention simulates the change of error values ​​under set conditions based on motion states and pseudo-inverse algorithms, and performs motion planning adjustments on moving parts and end effectors according to the simulation results. It can accurately obtain the changing trend of guide rail accuracy error values, and deeply understand the complex correlation between joint angles, execution speeds and guide rail accuracy, thereby providing a theoretical basis for precise adjustment. At the same time, by designing time-varying functions and designing motion planning adjustment schemes according to error thresholds, the machine tool can dynamically adapt to different working conditions and requirements, enhance the stability and adaptability of the machine tool in different environments, and optimize the motion performance of the machine tool by simulating different joint angle values ​​and actuator speed values, ensuring that the guide rail can maintain efficient and precise operation under various working conditions.

[0131] Those skilled in the art will appreciate that the units, i.e., algorithm steps, of the various examples described in conjunction with this embodiment can be implemented in electronic hardware or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0132] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A method for dynamically adjusting the accuracy of a bed guide rail of a large CNC gantry milling machine, characterized in that: The dynamic precision adjustment method comprises the following steps: S1. Use computer simulation technology to establish a finite element model of a CNC gantry milling machine, and use a collimator as a measurement reference in the finite element model to simulate and obtain the parallelism and straightness of the guide rails of the milling machine bed during operation; S2, comparing the parallelism and straightness of the guide rail with the set expected values ​​to obtain error values, and based on the error values, reversely judging the joint angle of the moving parts and the speed of the end effector in the milling machine, and taking the joint angle of the moving parts and the speed of the end effector as the motion state of the milling machine; S3, based on the motion state and pseudo-inverse algorithm, simulate the change of error value under set conditions, and design time-varying functions according to the simulation results to adjust the motion planning of the moving parts and the end effector; S4. Apply motion planning to the finite element model to obtain the adjusted guide rail parallelism and straightness, and compare the guide rail parallelism and straightness with the set expected values ​​to determine the degree of optimization of the guide rail accuracy; S5. If the comparison result is less than the expected value, return to step S3 to repeat the simulation optimization. If the comparison result is greater than the expected value, apply the motion planning as an adjustment method to the CNC gantry milling machine controller to adjust the guide rail accuracy. The method of comparing the parallelism and straightness of the guide rails with the set expected values ​​to obtain error values, and reversely judging the joint angle of the moving parts and the speed of the end effector in the milling machine based on the error values, and taking the joint angle of the moving parts and the speed of the end effector as the motion state of the milling machine includes the following steps: S21. Conduct independence test on the expected value based on collinearity diagnosis and correlation analysis, and use the information method to determine the information value of the expected value and implement weight quantification; S22. Building an error comparison model based on the weighted quantitative results and linear regression technology, and inputting the guide rail parallelism and straightness into the error comparison model to obtain error values ​​as guide rail accuracy deviation values; S23, combining the error value with the reverse engineering method to infer the joint angle of the moving parts in the milling machine and the execution speed of the end effector when the guide rail has an accuracy deviation, and taking the joint angle and the execution speed of the end effector as the motion state of the milling machine; The independence test of the set expected value based on collinearity diagnosis and correlation analysis, and the weight quantification by using the information method to determine the information value of the expected value include the following steps: S211, analyzing the linear correlation between variables within the set expected value, and performing multicollinearity diagnosis on the set expected value using tolerance and variance inflation factor based on the correlation result, wherein the variables include the set parallelism value and straightness value; S212. Determine the independence of variables based on the collinearity diagnosis results, evaluate the information value of the variables based on the independence results and the information method, and sum the information values ​​of the variables as a comparison index; S213. Construct a judgment matrix according to the analysis method to implement weight allocation for the comparison indicators, and define the information value based on the weight allocation result.

2. The method for dynamically adjusting the accuracy of the guide rails of a large CNC gantry milling machine bed according to claim 1 is characterized in that: The method of combining the error value with the reverse engineering method to infer the joint angle of the moving parts in the milling machine and the execution speed of the end effector when the guide rail has an accuracy deviation, and using the joint angle and the execution speed of the end effector as the motion state of the milling machine includes the following steps: S231, extracting three-dimensional information of the bed guide rail, the milling machine moving parts and the end effector in the finite element model and converting them into component data points; S232, using the adjacency matrix and the working principle of the milling machine to describe the intrinsic relationship between the joint angle of the moving part, the execution speed of the end effector and the accuracy of the guide rail to create a three-dimensional model of the part; S233, performing a local polynomial surface operation on the component data points, determining the projection residual of the component data points on the surface to obtain a feature point index, and combining the feature point index with a broken line growth technique to extract a feature line of the component three-dimensional model; S234. Based on the feature lines, the three-dimensional model of the component is transformed into a voxel matrix to describe the differences between the three-dimensional features when the guide rail has parallelism and straightness errors, and the convolutional neural network is trained through two-dimensional typical views. After forward and backward propagation operations, the joint angle and execution speed recognition results are output, and the joint angle and execution speed are used as the motion state of the milling machine.

3. The method for dynamically adjusting the accuracy of the guide rails of a large CNC gantry milling machine bed according to claim 2 is characterized in that: The method of performing a local polynomial surface operation on the component data points, determining the projection residual of the component data points on the surface to obtain the feature point index, and combining the feature point index with the broken line growth technology to extract the feature line of the component three-dimensional model comprises the following steps: S2331, fitting each data point in the three-dimensional model of the part into a local polynomial surface in its adjacent range, and selecting any component data point in the three-dimensional model of the component to analyze whether it meets the constraint conditions; S2332, deriving the projection residuals of all neighborhood points of the component data point, and selecting the maximum value as the projection residual of the component data point, and comparing the projection residual with a preset critical value, wherein the preset critical value represents the residual mean of all component data points; S2333, if the projection residual is greater than the preset critical value, the component data point is considered to be a feature point; if the projection residual is less than the preset critical value, return to step S2332 to reselect and compare; S2334, creating a sequence for storing feature points, using component analysis to calculate the correlation index between feature points, obtaining the correlation index of all feature points and sorting them in order of high and low, and saving the sorting result in the sequence; S2335. Add an evaluation direction in the component three-dimensional model as an operation termination condition, and select several component data points in the sequence as seed points to simulate broken line growth in the component three-dimensional model to extract feature lines of the component three-dimensional model.

4. The method for dynamically adjusting the accuracy of the guide rails of a large CNC gantry milling machine bed according to claim 3 is characterized in that: The calculation formula of the correlation index is: ; In the formula, η Represents the relevant index value; P Represents a sequence for saving feature points; , , They respectively represent the eigenvalues ​​of the feature points in their neighborhood points.

5. A method for dynamically adjusting the accuracy of a bed guide rail of a large CNC gantry milling machine according to claim 4, characterized in that: The method transforms the component three-dimensional model into a voxel matrix based on the feature line to describe the difference between the three-dimensional features when the guide rail has parallelism and straightness errors, and trains the convolutional neural network through two-dimensional typical views, outputs the joint angle and execution speed recognition results after forward and reverse propagation operations, and uses the joint angle and execution speed as the motion state of the milling machine, including the following steps: S2341, selecting the highest and lowest values ​​of feature point coordinates in the component three-dimensional model based on the feature line, and combining the feature point coordinates with a preset voxel unit coordinate system to set the coordinate axis direction to be parallel to the initial coordinate axis; S2342, align the axis bounding box of the component 3D model and x , y , z The edges parallel to the axis are divided into equal distances in turn, and the model is divided along the x , y , z The number of rows, columns and layers of the axis divided is represented as a voxel matrix to complete the voxelization process; S2343, based on the voxel matrix describing the difference between the three-dimensional features when the guide rail has parallelism and straightness errors, randomly expanding the voxelized three-dimensional model of the component to obtain a voxel enhancement data set of a set multiple; S2344. The convolutional neural network is trained through two-dimensional typical views combined with forward and backward propagation. The parameter values ​​of the three-dimensional model of the component are output after the forward and backward propagation training to obtain the joint angle and execution speed recognition results, and the joint angle and execution speed are used as the motion state of the milling machine.

6. A method for dynamically adjusting the accuracy of a bed guide rail of a large CNC gantry milling machine according to claim 5, characterized in that: The method of simulating the change of the error value under set conditions based on the motion state and the pseudo-inverse algorithm, and adjusting the motion planning of the moving part and the end effector by designing a time-varying function according to the simulation results includes the following steps: S31, setting a plurality of sets of joint angle values ​​and actuator speed values ​​in the component three-dimensional model to simulate the motion process of the moving component and the end effector; S32, combining the pseudo-inverse algorithm with the simulation results to obtain the change trend of the guide rail accuracy error value, and describing the correlation between the joint angle, execution speed and guide rail accuracy according to the change trend; S33. Based on the influence relationship and simulation results, a time-varying function is designed to obtain an error threshold, and a motion planning adjustment scheme is designed according to the error threshold to correct the joint angle of the moving parts and the execution speed of the end effector.

7. A method for dynamically adjusting the accuracy of a bed guide rail of a large CNC gantry milling machine according to claim 6, characterized in that: The method of combining the pseudo-inverse algorithm with the simulation results to obtain the variation trend of the guide rail accuracy error value and describing the correlation between the joint angle, the execution speed and the guide rail accuracy according to the variation trend comprises the following steps: S321, performing redundancy analysis on the motion process based on the pseudo-inverse algorithm and the simulation results, and judging the trend of the error value changing with the joint angle and the execution speed according to the analysis results; S322, fitting the guide rail accuracy error value change function according to the judgment result to obtain the change of the joint angle and the execution speed; S323. Analyze the influence relationship between the joint angle, execution speed and guide rail accuracy based on the change situation to obtain the correlation result.

8. The method for dynamically adjusting the accuracy of the guide rails of a large CNC gantry milling machine bed according to claim 7, characterized in that: The motion planning adjustment scheme includes: When the guide rail accuracy error |e| Greater than the error threshold t When , the adjustment range of the moving part joint angle is set to △ θ = k θ . e ,in k θ is the joint angle adjustment coefficient, k θ The value range is [0.1-0.5], and the end effector execution speed is adjusted to △ v = k v . e ,in k v is the joint angle adjustment coefficient, k v The value range of is [0.1-0.5]; When the guide rail accuracy error |e| Less than the error threshold t When , the adjustment range of the moving part joint angle is set to △ θ = k ' θ . e ,in k ' θ is the joint angle adjustment coefficient, k ' θ The value range is [0.001-0.1], and the end effector execution speed is adjusted to △ v = k ' v . e ,in k ' v is the joint angle adjustment coefficient, k ' v The value range of is [0.001-0.1]; When the guide rail accuracy error |e| Equal to the error threshold t , the original joint angle and execution speed remain unchanged.

Citation Information

Patent Citations

  • Butt joint leveling method of long lathe body

    CN103810349A

  • Method for detecting and correcting parallelism of guide rails of portal machine tool

    CN109489589A