An error analysis method for five-axis machine tool precision detection

Abstract

The present invention discloses an error analysis method for precision detection of five-axis machine tools, which belongs to the technical field of machine tool error detection and is characterized in that it includes the following steps: a. operating the machine tool to perform forward specimen cutting detection or sequential non-cutting no-load motion detection; b. operating the machine tool to perform reverse specimen cutting detection or reverse non-cutting no-load motion detection; c. matching the errors of the same corresponding position in the forward and reverse detections; d. obtaining dynamic error and static error through cancellation calculation; e. calculating the proportion of static error factors and the proportion of dynamic error factors. The present invention does not need to use a variety of different detection schemes to detect different error source types separately. It only needs two repeated detections in opposite directions and a simple superposition cancellation calculation to separate the static error and dynamic error that are mutually coupled during the detection process, thereby improving detection efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of machine tool error detection, and in particular to an error analysis method for precision detection of five-axis machine tools. Background Art

[0002] Five-axis CNC machine tools, specialized equipment for machining complex curved surfaces, are widely used in machining complex curved structures such as aircraft structural parts, turbine blades, and automotive molds. They play a pivotal role in the aerospace and automotive manufacturing industries. Machining accuracy is a core metric for evaluating the performance and technical level of five-axis CNC machine tools. Testing this accuracy is crucial for the use and improvement of these tools.

[0003] The machining accuracy testing of five-axis CNC machine tools is mainly divided into two categories. These are static testing and comprehensive testing that simulates real machining conditions. Static testing refers to the use of instruments such as laser interferometers and contact-type precision side heads to perform static fixed-point measurements of the accuracy of various positions of the machine tool. This type of method can accurately reflect the positioning accuracy information of five-axis CNC machine tools and is therefore widely used. However, the error sources that affect the machining accuracy of five-axis CNC machine tools include static positioning errors caused by machining errors and assembly errors of machine tool components, as well as dynamic errors that occur during machining motion due to factors such as servo system tracking delays. Static testing methods cannot reflect the impact of dynamic errors on machining accuracy. Therefore, comprehensive testing that can simulate real machining conditions is also widely used to guide the accuracy testing and error compensation of five-axis CNC machine tools.

[0004] A Chinese patent document with publication number CN112518422A, published on March 19, 2021, discloses a method for modeling and separating geometric errors of a five-axis AC rotary head gantry machine tool, characterized by comprising the following steps:

[0005] Step 1) Based on the structure and transmission form of the five-axis AC swing head gantry machine tool, the inter-axis relative motion conversion matrix is obtained according to the multi-body kinematics theory and homogeneous coordinate transformation theory, and the kinematic model of the machine tool from the tool coordinate system to the workpiece coordinate system is established;

[0006] Step 2) First, the physical meaning of the 21 geometric errors of the machine tool's linear axes and the 20 geometric errors of the rotary axes are defined. Then, based on the different motion combinations of the machine tool, tool tip spatial error models are established that consider only the linear axis, only the rotary axis, and both the linear and rotary axes' geometric errors.

[0007] Step 3) Within the maximum travel range of the machine tool, according to the principle of having the same number of measurement points in the X, Y, and Z directions, a spatial measurement trajectory of the linear axis geometric errors is generated. A laser tracker is used to measure the actual spatial coordinates of the tool tip point at each measurement point. A single-base station multi-measurement method is used, and the installation positions of the base station and target ball are changed six times. Based on the difference between the actual distance and the theoretical distance from the tool tip point to the base station at each measurement position, 21 linear axis geometric errors are calculated separately.

[0008] Step 4) Based on a tool tip spatial error model that considers only the linear axis geometric error, the ballbar length variation caused by the linear axis geometric error is calculated. The theoretical value of the length variation caused by the linear axis geometric error is removed from the experimentally measured length variation to obtain the length variation caused only by the rotational axis geometric error. The theoretical coordinates of the centers of the balls at both ends of the ballbar under each A- and C-axis measurement mode are used to obtain a calculation formula for the theoretical value of the length variation caused by the rotational axis geometric error under the corresponding measurement mode.

[0009] Step 5) Design the measurement motion trajectories for the A and C axes, respectively. One end of the ballbar is fixed to the workbench and the other end is fixed to the tool tip. The A axis is required to rotate from -90° to 90°, and the C axis is required to rotate from 0° to 360°. The A and C axes and the ballbar must meet the axial, radial, and tangential mounting relationships, respectively. The theoretical coordinates of the centers of the stylus balls at both ends of the ballbar are obtained under six different measurement modes.

[0010] Step 6) The motion transformation matrix of the tool tip side ball in the machine tool coordinate system is established based on the relative motion relationship between the rotary axis and the linear axis. The ballbar length change corresponding to different measurement angles when each PIGE error acts alone is calculated based on the coordinates of the ball centers at both ends of the ballbar. The sensitive directions of the eight PIGE errors are thus determined. The separation of the various PIGE errors is achieved based on the expressions of the ballbar length changes corresponding to the sensitive directions of the eight PIGE errors.

[0011] Step 7) The relationship between the 12 PDGE errors and the ballbar length change is established using the rotation axis geometric error model. Expressions for the length changes caused by the six PDGE errors on the A and C axes are derived. By measuring the length changes of the three ballbars at different initial installation coordinates, simultaneous equations are established to derive expressions for the six PDGE errors on the A and C axes. Substituting the measured length changes at different rotation angles into the equations, the PDGE errors for the A and C axes at different angles throughout the entire travel range are obtained, thus achieving separation of the 12 PDGE errors.

[0012] The geometric error modeling and separation method for a five-axis AC gantry machine tool disclosed in this patent document has good versatility. However, it cannot effectively separate the static and dynamic errors that are intertwined and coupled during the detection process, affecting detection efficiency. Summary of the Invention

[0013] In order to overcome the defects of the above-mentioned prior art, the present invention provides an error analysis method for the precision detection of five-axis machine tools. The present invention does not need to use multiple different detection schemes to detect different error source types separately. It only needs two repeated detections in opposite directions and a simple superposition cancellation calculation to separate the static errors and dynamic errors that are mixed and coupled with each other during the detection process, thereby improving the detection efficiency.

[0014] The present invention is achieved through the following technical solutions:

[0015] An error analysis method for precision detection of a five-axis machine tool, characterized by comprising the following steps:

[0016] a. Operate the machine tool to perform forward specimen cutting test or sequential non-cutting no-load motion test;

[0017] b. Operate the machine tool to perform reverse specimen cutting test or reverse non-cutting no-load motion test;

[0018] c. Match the errors at the same corresponding position in the forward and reverse tests. If the error data is the machining error data of the specimen, match the machining errors at the same measuring point of the specimen for the forward cutting test piece and the reverse cutting test piece. If the error data is the position error data obtained by non-cutting no-load motion testing, arrange the position error data obtained by the reverse test in reverse chronological order and match them with the position error data obtained by the sequential test in chronological order.

[0019] d. By canceling the calculation, the dynamic error and static error are obtained, and the error detection data of the five-axis machine tool is decomposed into dynamic error components and static error components;

[0020] e. Calculate the proportion of static error factors and the proportion of dynamic error factors.

[0021] In step a, the specimen cutting test refers to measuring the machining errors of the surface points of the specimen using a three-coordinate machine.

[0022] In step a, the non-cutting no-load motion detection refers to obtaining the position error by a ballbar or calculating the position error by collecting internal data of the machine tool.

[0023] In step d, the static error refers to a design defect, a machine tool component error or an assembly error.

[0024] In step d, the dynamic error refers to the tracking error of the servo control system of each motion axis.

[0025] In step d, the static error is calculated by formula 1;

[0026] Formula 1;

[0027] Where, is the static error, is the positive specimen cutting error, The error of the reverse specimen cutting test is shown in Figure 2.

[0028] In step d, the dynamic error is calculated by formula 2;

[0029] Formula 2;

[0030] Where, is the dynamic error, is the positive specimen cutting error, The error of the reverse specimen cutting test is shown in Figure 2.

[0031] In step e, the static error factor ratio is calculated by formula 3;

[0032] Formula 3;

[0033] Where, is the proportion of static error factors, is the static error, is the dynamic error.

[0034] In step e, the dynamic error factor ratio is calculated by formula 4;

[0035] Formula 4;

[0036] Where, is the proportion of dynamic error factors, is the dynamic error, is the static error.

[0037] The beneficial effects of the present invention are mainly manifested in the following aspects:

[0038] 1. The present invention comprises the following steps: a. operating a machine tool to perform a forward specimen cutting test or a sequential non-cutting no-load motion test; b. operating a machine tool to perform a reverse specimen cutting test or a reverse non-cutting no-load motion test; c. matching the errors at the same corresponding position in the forward and reverse tests; if the error data is the machining error data of the specimen, then for the forward cutting specimen and the reverse cutting specimen, matching the machining errors at the same measuring point of the specimen; if the error data is the position error data obtained by the non-cutting no-load motion test, then arranging the position error data obtained by the reverse test in reverse chronological order and comparing them with the position error data obtained by the sequential test. The obtained position error data is mapped according to the time series; d. Through cancellation calculation, the dynamic error and static error are obtained, and the error detection data of the five-axis machine tool is decomposed into dynamic error component and static error component; e. The proportion of static error factors and the proportion of dynamic error factors are calculated. As a complete technical solution, compared with the existing technology, there is no need to use multiple different detection schemes to detect different error source types separately. Only two repeated detections in opposite directions and simple superposition cancellation calculations are required to separate the static errors and dynamic errors that are mixed and coupled with each other during the detection process, thereby improving the detection efficiency.

[0039] 2. The present invention can effectively increase the machine tool accuracy information contained in the detection results and improve the versatility of the accuracy detection results.

[0040] 3. The present invention can avoid designing detection schemes for different error types, thereby improving detection efficiency and reducing detection costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments:

[0042] Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION

[0043] Example 1

[0044] See also Figure 1 , an error analysis method for five-axis machine tool precision detection, comprising the following steps:

[0045] a. Operate the machine tool to perform forward specimen cutting test or sequential non-cutting no-load motion test;

[0046] b. Operate the machine tool to perform reverse specimen cutting test or reverse non-cutting no-load motion test;

[0047] c. Match the errors at the same corresponding position in the forward and reverse tests. If the error data is the machining error data of the specimen, match the machining errors at the same measuring point of the specimen for the forward cutting test piece and the reverse cutting test piece. If the error data is the position error data obtained by non-cutting no-load motion testing, arrange the position error data obtained by the reverse test in reverse chronological order and match them with the position error data obtained by the sequential test in chronological order.

[0048] d. By canceling the calculation, the dynamic error and static error are obtained, and the error detection data of the five-axis machine tool is decomposed into dynamic error components and static error components;

[0049] e. Calculate the proportion of static error factors and the proportion of dynamic error factors.

[0050] This embodiment is the most basic implementation method. As a complete technical solution, compared with the existing technology, there is no need to use multiple different detection schemes to detect different error source types separately. Only two repeated detections in opposite directions and simple superposition cancellation calculations are required to separate the static errors and dynamic errors that are mixed and coupled with each other during the detection process, thereby improving detection efficiency.

[0051] Example 2

[0052] See also Figure 1 , an error analysis method for five-axis machine tool precision detection, comprising the following steps:

[0053] a. Operate the machine tool to perform forward specimen cutting test or sequential non-cutting no-load motion test;

[0054] b. Operate the machine tool to perform reverse specimen cutting test or reverse non-cutting no-load motion test;

[0055] c. Match the errors at the same corresponding position in the forward and reverse tests. If the error data is the machining error data of the specimen, match the machining errors at the same measuring point of the specimen for the forward cutting test piece and the reverse cutting test piece. If the error data is the position error data obtained by non-cutting no-load motion testing, arrange the position error data obtained by the reverse test in reverse chronological order and match them with the position error data obtained by the sequential test in chronological order.

[0056] d. By canceling the calculation, the dynamic error and static error are obtained, and the error detection data of the five-axis machine tool is decomposed into dynamic error components and static error components;

[0057] e. Calculate the proportion of static error factors and the proportion of dynamic error factors.

[0058] Preferably, in step a, the specimen cutting inspection refers to measuring the machining errors of the surface points of the specimen using a three-coordinate machine.

[0059] In step a, the non-cutting no-load motion detection refers to obtaining the position error by a ballbar or calculating the position error by collecting internal data of the machine tool.

[0060] In step d, the static error refers to a design defect, a machine tool component error or an assembly error.

[0061] In step d, the dynamic error refers to the tracking error of the servo control system of each motion axis.

[0062] This embodiment is a preferred implementation method, which can effectively increase the machine tool accuracy information contained in the detection results and improve the versatility of the accuracy detection results.

[0063] Example 3

[0064] See also Figure 1 , an error analysis method for five-axis machine tool precision detection, comprising the following steps:

[0065] a. Operate the machine tool to perform forward specimen cutting test or sequential non-cutting no-load motion test;

[0066] b. Operate the machine tool to perform reverse specimen cutting test or reverse non-cutting no-load motion test;

[0067] c. Match the errors at the same corresponding position in the forward and reverse tests. If the error data is the machining error data of the specimen, match the machining errors at the same measuring point of the specimen for the forward cutting test piece and the reverse cutting test piece. If the error data is the position error data obtained by non-cutting no-load motion testing, arrange the position error data obtained by the reverse test in reverse chronological order and match them with the position error data obtained by the sequential test in chronological order.

[0068] d. By canceling the calculation, the dynamic error and static error are obtained, and the error detection data of the five-axis machine tool is decomposed into dynamic error components and static error components;

[0069] e. Calculate the proportion of static error factors and the proportion of dynamic error factors.

[0070] In step a, the specimen cutting test refers to measuring the machining errors of the surface points of the specimen using a three-coordinate machine.

[0071] In step a, the non-cutting no-load motion detection refers to obtaining the position error by a ballbar or calculating the position error by collecting internal data of the machine tool.

[0072] In step d, the static error refers to a design defect, a machine tool component error or an assembly error.

[0073] In step d, the dynamic error refers to the tracking error of the servo control system of each motion axis.

[0074] More preferably, in step d, the static error is calculated by formula 1;

[0075] Formula 1;

[0076] Where, is the static error, is the positive specimen cutting error, The error of the reverse specimen cutting test is shown in Figure 2.

[0077] In step d, the dynamic error is calculated by formula 2;

[0078] Formula 2;

[0079] Where, is the dynamic error, is the positive specimen cutting error, The error of the reverse specimen cutting test is shown in Figure 2.

[0080] In step e, the static error factor ratio is calculated by formula 3;

[0081] Formula 3;

[0082] Where, is the proportion of static error factors, is the static error, is the dynamic error.

[0083] In step e, the dynamic error factor ratio is calculated by formula 4;

[0084] Formula 4;

[0085] Where, is the proportion of dynamic error factors, is the dynamic error, is the static error.

[0086] This embodiment is the best implementation method, which can avoid designing detection solutions for different error types, improve detection efficiency, and reduce detection costs.

[0087] The non-cutting no-load motion detection process is as follows:

[0088] Operate the machine tool to perform no-load motion and use motion error detection instruments to collect error data.

[0089] When using a ballbar to collect error data, the machine tool motion is circular interpolation;

[0090] When using a rotary axis analyzer to collect error data, the machine tool motion is a fixed-point motion with the tool tip fixed and the rotary axis moving, and the error data is obtained. , where t is the time series within the effective data collection range;

[0091] Reversely arrange all the instruction items in the no-load motion CNC program to form a reverse detection program, and use the motion error detection instrument to collect error data to obtain error data. , where t is the time series within the effective data collection range;

[0092] Match the no-load motion test results in the forward and reverse directions according to the specimen sequence. and That is, the error vector of forward cutting and reverse no-load motion at the same position;

[0093] Due to the detection error of the same corresponding position in the forward and reverse direction detection 、 and dynamic error , static error The following relationship exists:

[0094] Formula 5;

[0095] Therefore, you can use 、 Perform cancellation calculations:

[0096] Formula 6.

Claims

1. An error analysis method for five-axis machine tool precision detection, characterized in that: The following steps are involved: a. Operate the machine tool to perform forward specimen cutting test or sequential non-cutting no-load motion test; b. Operate the machine tool to perform reverse specimen cutting test or reverse non-cutting no-load motion test; c. Match the errors at the same corresponding position in the forward and reverse tests. If the error data is the machining error data of the specimen, match the machining errors at the same measuring point of the specimen for the forward cutting test piece and the reverse cutting test piece. If the error data is the position error data obtained by non-cutting no-load motion testing, arrange the position error data obtained by the reverse test in reverse chronological order and match them with the position error data obtained by the sequential test in chronological order. d. By canceling the calculation, the dynamic error and static error are obtained, and the error detection data of the five-axis machine tool is decomposed into dynamic error components and static error components; e. Calculate the proportion of static error factors and the proportion of dynamic error factors.

2. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step a, the specimen cutting test refers to measuring the machining errors of the surface points of the specimen using a three-coordinate machine.

3. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step a, the non-cutting no-load motion detection refers to obtaining the position error by a ballbar or calculating the position error by collecting internal data of the machine tool.

4. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step d, the static error refers to a design defect, a machine tool component error or an assembly error.

5. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step d, the dynamic error refers to the tracking error of the servo control system of each motion axis.

6. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step d, the static error is calculated by formula 1; Formula 1; Where, is the static error, is the positive specimen cutting error, The error of the reverse specimen cutting test is shown in Figure 2.

7. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step d, the dynamic error is calculated by formula 2; Formula 2; Where, is the dynamic error, is the positive specimen cutting error, The error of the reverse specimen cutting test is shown in Figure 2.

8. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step e, the static error factor ratio is calculated by formula 3; Formula 3; Where, is the proportion of static error factors, is the static error, is the dynamic error.

9. The error analysis method for five-axis machine tool precision detection according to claim 1, characterized in that: In step e, the dynamic error factor ratio is calculated by formula 4; Formula 4; Where, is the proportion of dynamic error factors, is the dynamic error, is the static error.

Images