Evolutionary compensation method based on in-situ timing measurement and for spindle rotation errors of computer numerical control grinding machine
By collecting the trajectory images of the spindle gyro error characteristic points in real time on the CNC grinder and establishing and updating the compensation model, the problems of measurement hysteresis and low accuracy in the existing technology are solved, and high-precision real-time compensation of the spindle gyro error of the CNC grinder is achieved to ensure that Continuous optimization of machining accuracy.
Patent Information
- Application Number
- PCT/CN2023/139878
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-06-01
- Filing Date
- 2023-12-19
- Publication Date
- 2024-12-05
AI Technical Summary
The existing CNC grinder spindle dynamic gyro error measurement methods have problems such as measurement hysteresis, difficulty in ensuring accuracy and unreal-time compensation, which affects the processing accuracy.
The evolutionary compensation method of spindle gyro error based on timing in-situ measurement is adopted. By determining feature points at the edge of the central hole of the spindle, the camera collects trajectory images, establishes compensation models, and continuously updates to achieve real-time compensation.
The spindle slewing error measurement accuracy and compensation accuracy are improved, and the real-time compensation requirements are met, and the problems of measurement hysteresis and low accuracy are overcome, ensuring continuous optimization of machining accuracy.
Smart Images

Figure CN2023139878_05122024_PF_FP_ABST
Abstract
Description
Evolutionary compensation method for spindle rotation error of CNC grinding machine based on in-situ timing measurement
[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on June 1, 2023, with application number 202310640942.5 and invention name “Evolutionary compensation method for spindle rotation error of CNC grinding machine based on in-situ measurement of timing”, the entire contents of which are incorporated into this application by reference. Technical Field
[0002] The invention relates to a method for compensating a spindle rotation error of a numerically controlled grinding machine, and in particular to a method for compensating a spindle rotation error of a numerically controlled grinding machine based on timing in-situ measurement. Background Art
[0003] Precision parts are widely used in aviation, aerospace, national defense and other fields. CNC grinding machines provide a strong guarantee for the high-precision and high-quality processing of such parts. The spindle unit, as the core functional component of the CNC grinding machine, directly affects the processing quality of the processed parts. For a long time, domestic CNC machine tool manufacturers have usually used the static rotation accuracy of the machine tool when it leaves the factory as the measurement indicator of the spindle rotation accuracy. However, with the continuous improvement of the manufacturing level, CNC grinding machines are mostly used for precision and ultra-precision processing. The extremely slight rotation error of the grinding machine spindle may have a serious impact on the processing accuracy. It is difficult to guarantee the processing quality of the parts by compensating only for the static rotation error. Therefore, accurately compensating for the rotation error based on the measurement results of the dynamic rotation accuracy of the spindle is of great significance to ensuring the processing accuracy of CNC grinding machines.
[0004] The existing methods for measuring the dynamic rotation error of the spindle of a CNC grinder are mainly:
[0005] (1) Displacement sensor measurement method: A standard workpiece is clamped on the grinding machine spindle, and multiple displacement sensors are arranged in the axial and radial directions of the spindle to indirectly measure the spindle's rotation error. The rotation error is determined by roundness error separation technology. This method has problems such as difficulty in separating roundness errors and ensuring accuracy, and difficulty in arranging displacement sensors.
[0006] (2) Machine vision measurement method: The feature point trajectory image is collected and processed during the CNC grinder spindle rotation process, and the spindle rotation error is analyzed and determined. This method has problems such as the inability to compensate on the machine, the complex feature point trajectory image processing process, and the obvious lag in the measurement results.
[0007] During the actual machining process of a CNC grinder, factors such as spindle speed, load, operating conditions, lubrication, and thermal deformation all affect the spindle's rotational accuracy. Real-time, in-situ measurement of rotational errors provides data support for the compensation process, significantly impacting the spindle's rotational accuracy and, ultimately, the machining accuracy of the CNC grinder. Existing research methods are unable to achieve axial and radial spindle error compensation for CNC grinders based on real-time, in-situ measurement.
[0008] Summary of the Invention
[0009] In order to solve the problems existing in the background technology, the present invention proposes an evolutionary compensation method for the spindle rotation error of a CNC grinder based on timing in-situ measurement, which is used to achieve timing in-situ compensation of the axial and radial rotation errors of the spindle of the CNC grinder. The present invention can overcome the shortcomings of existing methods. During the operation of the CNC grinder, the spindle rotation error is determined by an in-situ measurement device for the spindle rotation error of the CNC grinder, and a CNC grinder spindle rotation error compensation model is established by integrating the spindle rotation error characteristics, thereby compensating for the measurement lag problem of the measuring device; the measurement process is repeated every m minutes, and the compensation model is continuously improved and optimized. The rotation error compensation value is determined based on the timing in-situ measurement, thereby improving the compensation accuracy of the spindle rotation error of the CNC grinder.
[0010] To achieve the above object, the technical solution of the present invention is as follows:
[0011] S1: Fix the CNC grinding machine spindle rotation error in-situ measurement device on the CNC grinding machine, determine multiple feature points on the edge of the spindle center hole, and align the CNC grinding machine spindle rotation error in-situ measurement device with the edge of the spindle center hole;
[0012] S2: The CNC grinder is in operation, and the CNC grinder spindle rotation error in-situ measurement device is used to acquire a trajectory image of the characteristic points on the edge of the spindle center hole;
[0013] S3: The spindle rotation error measurement value is obtained by calculating the trajectory image of the feature points on the edge of the spindle center hole;
[0014] S4: The spindle rotation error characteristics and spindle rotation error measurement values of the CNC grinder are integrated to establish a spindle rotation error compensation model for the CNC grinder. The compensation model outputs a spindle rotation error compensation value, and the spindle rotation error of the CNC grinder is compensated according to the spindle rotation error compensation value.
[0015] S5: Repeat S2-S3 to calculate and obtain the spindle rotation error measurement value, use the current spindle rotation error measurement value to calibrate the current CNC grinder spindle rotation error compensation model and update the compensation model, thereby obtaining the latest spindle rotation error compensation value;
[0016] S6: Repeat S5, continuously calibrate and update the CNC grinder spindle rotation error compensation model to achieve continuous optimization compensation of the CNC grinder spindle rotation error.
[0017] Compared with the prior art, the present invention has the following beneficial effects:
[0018] 1) The spindle rotation error measuring device for a CNC grinding machine of the present invention is not affected by the internal structural arrangement of the grinding machine, is easy to install, and can simultaneously obtain the axial and radial rotation errors of the spindle of the CNC grinding machine.
[0019] 2) The spindle rotation error measurement result of the present invention does not include the spindle roundness error, and the measurement accuracy is high.
[0020] 3) The CNC grinding machine spindle rotation error evolution compensation model established by the present invention meets the real-time requirements of rotation error compensation, the compensation capability is continuously improved and optimized, and the rotation error compensation accuracy is high.
[0021] Figures in the specification
[0022] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0023] FIG1 is a flow chart of the method of the present invention;
[0024] FIG2 is a schematic diagram of the overall structure of an in-situ measurement device for the spindle rotation error of a CNC grinding machine;
[0025] FIG3 is a schematic diagram of the structural decomposition of an in-situ measurement device for the spindle rotation error of a CNC grinding machine;
[0026] FIG4 is a schematic diagram of the structural decomposition of a calibration light source in an in-situ measurement device for the spindle rotation error of a CNC grinding machine;
[0027] FIG5 is a schematic diagram of an in-situ measurement device for the spindle rotation error of a CNC grinding machine.
[0028] Explanation of symbols: vertical guide rail frame 1, horizontal guide rail 2, slider 3, front and rear guide rails 4, sliding table 5, pitch guide rail 6, calibration light source 7, right camera vertical moving guide rail 8, right industrial camera 9, right camera rotation table 10, left camera vertical moving guide rail 11, left industrial camera 12, left camera rotation table 13, rotary aperture adjustment ring 71, silicon-based target piece 72, support 73, LED light source 74. DETAILED DESCRIPTION
[0029] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0030] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0031] In the specific implementation, as shown in Figures 2 and 3, the in-situ measurement device for the spindle rotation error of a CNC grinding machine includes a vertical guide rail frame 1, a horizontal guide rail 2, a slider 3, front and rear guide rails 4, a sliding table 5, a pitch guide rail 6, a calibration light source 7, a left camera assembly and a right camera assembly.
[0032] The vertical guide rail frame 1 is fixedly installed inside the movable door of the CNC grinder through the slide groove structure behind it. The horizontal guide rail 2 is installed in the vertical guide rail frame 1 through the slider 3, and the horizontal guide rail 2 slides up and down in the vertical guide rail frame 1; the front and rear guide rails 4 are installed on the horizontal guide rail 2 through the slider 3, and the front and rear guide rails 4 slide horizontally in the horizontal guide rail 2; the sliding table 5 is installed on the front and rear guide rails 4, and the sliding table 5 slides back and forth in the front and rear guide rails 4; a calibration light source 7 is fixedly installed on the side of the sliding table 5, and a pitch guide rail 6 is installed on the sliding table 5, and the left camera assembly and the right camera assembly are installed in the pitch guide rail 6, and the left camera assembly and the right camera assembly are arranged at intervals.
[0033] The right camera assembly includes a right camera vertical moving guide rail 8, a right industrial camera 9 and a right camera rotating platform 10. The right camera rotating platform 10 is installed in the sliding platform 5. The right camera vertical moving guide rail 8 is fixedly installed on the right camera rotating platform 10. The right industrial camera 9 is installed on the side of the right camera vertical moving guide rail 8. The right industrial camera 9 has displacement freedom in the Y and Z directions and rotation freedom on the Z axis relative to the pitch guide rail 6. The right industrial camera 9 has displacement freedom in the Y direction relative to the left industrial camera 12.
[0034] The left camera assembly includes a left camera vertical moving guide rail 11, a left industrial camera 12 and a left camera rotating platform 13. The left camera rotating platform 13 is installed in the sliding platform 5. The left camera vertical moving guide rail 11 is fixedly installed on the left camera rotating platform 13. The left industrial camera 12 is installed on the side of the left camera vertical moving guide rail 11. The left industrial camera 12 has displacement freedom in the Y and Z directions and rotation freedom in the Z axis relative to the pitch guide rail 6.
[0035] As shown in FIG4 , the calibration light source 7 includes a rotatable aperture adjustment ring 71 , a silicon-based target piece 72 , a support 73 and an LED light source 74 .
[0036] The support 73 is fixedly mounted on the LED light source 74, the rotary aperture adjustment ring 71 is mounted on the support 73, the silicon-based target piece 72 is mounted between the rotary aperture adjustment ring 71 and the support 73, a through hole is opened in the middle of the support 73, and a light-transmitting ring hole is opened in the middle of the silicon-based target piece 72. The light beam emitted by the LED light source 74 passes through the through hole and the light-transmitting ring hole of the support 73 in turn and then emerges from the rotary aperture adjustment ring 71. The emergent light beam is aligned with the edge of the main shaft center hole, and the rotary aperture adjustment ring 71 is used to adjust the intensity of the emergent light beam.
[0037] As shown in FIG1 , a surface grinder according to an embodiment of the present invention is described as an example, and includes the following steps:
[0038] S1: As shown in Figure 5, the CNC grinding machine spindle rotation error in-situ measurement device is fixed to the CNC grinding machine. Multiple feature points are determined on the edge of the spindle center hole. The calibration light source 7 of the CNC grinding machine spindle rotation error in-situ measurement device is aligned with the edge of the spindle center hole. The feature points are set to a size of 5 μm to 20 μm. In the specific implementation, three feature points are set to a size of 10 μm.
[0039] S2: Start the CNC grinder, which is in operation. Use the CNC grinder spindle rotation error in-situ measurement device to acquire a trajectory image of the feature points on the edge of the spindle center hole.
[0040] In S2, the camera shutter time of the CNC grinder's spindle rotation error in-situ measurement device is the same as the spindle rotation period. Every m minutes, the trajectories of the spindle center hole edge feature points as the spindle rotates N times are captured to obtain an initial feature point image. After image enhancement and edge extraction, an image of the spindle center hole edge feature point trajectory is obtained. The choice of m depends on the actual machining conditions and is typically 10; the value of N depends on the spindle speed n and is typically N = 2n. Image enhancement and edge extraction are performed on the initial feature point image. Specifically, the trajectories of all spindle center hole edge feature points in the initial feature point image as the spindle rotates are processed cycle by cycle. Image enhancement is used to increase the contrast of the original image. Edge extraction is performed using the Canny operator to obtain the trajectory coordinate set of the feature points.
[0041] S3: The spindle rotation error measurement value is obtained based on the trajectory image of the feature points on the edge of the spindle center hole.
[0042] S3 specifically:
[0043] According to the trajectory images of the edge feature points of the spindle center hole captured by the right industrial camera 9 and the left industrial camera 12 respectively, the axis coordinates of the spindle in the camera coordinate systems corresponding to the right industrial camera 9 and the left industrial camera 12 are determined respectively. Then, the axis coordinates of the spindle in the two camera coordinate systems are fused and analyzed using the dual-target positioning method to obtain the world coordinates of the spindle axis of the CNC grinder. Finally, the spindle axial rotation error and radial rotation error are determined according to the world coordinates of the spindle axis of the CNC grinder and recorded as the spindle rotation error measurement value.
[0044] Specifically:
[0045] According to the target trajectory coordinates of each center hole edge feature point obtained in S2, the goal is to minimize the total deviation of all coordinate points to the fitting circle: Goal = min a,b [(x i -a) 2 +(y i -b) 2 -r 2 ]
[0046] Among them, Goal represents the total minimum deviation of all coordinate points to the fitting circle, x i ,y i are the two coordinate values of the target trajectory of the feature point, a and b are the two coordinate values of the center of the fitting circle, and r is the radius of the fitting circle. The nonlinear target problem is transformed into a linear problem, that is,
[0047] Among them, X i =x i 2 , Y i =y i 2 , A=a 2 , B=b 2 , R=r 2 , X i ,Y i are the two coordinate values of the transformed feature point target trajectory, A and B are the two coordinate values of the transformed fitting circle center, and R is the transformed fitting circle radius. Therefore, the optimization goal becomes:
[0048] Among them, i is the first relaxation factor, ζ i * represents the second relaxation factor, C represents the penalty factor, w represents the target weight, |||| represents the two-norm operation, n represents the total number of relaxation factors, and i represents the relaxation factor index. Given the KKT constraints, the Lagrange multiplier method is used to establish a fitting circle for the target trajectory of the center hole edge feature point and determine the center position.
[0049] The grayscale value corresponding to the center of each feature point's target trajectory on the image is calculated circle by circle, and the coordinate point corresponding to the grayscale average is used as the spindle axis coordinate in the camera coordinate system. At this point, the spindle axis coordinates for N revolutions are obtained in each camera coordinate system. This is then combined with the dual-target positioning method to construct the world coordinates of the CNC grinder's spindle axis.
[0050] The obtained world coordinates of the spindle axis are compared with the world coordinates of the axis at the initial moment, and the axial and radial rotation errors of the spindle are determined according to the axial and radial separation.
[0051] S4: Taking into account the complexity of the trajectory image processing process of the feature points and the lag in the calculation results of the spindle rotation error measurement value, the present invention integrates the spindle rotation error characteristics and the spindle rotation error measurement value of the CNC grinder to establish a spindle rotation error compensation model for the CNC grinder. The compensation model outputs the spindle rotation error compensation value, and the spindle rotation error of the CNC grinder is compensated according to the spindle rotation error compensation value; the spindle rotation error characteristics of the CNC grinder include axial rotation error characteristics and radial rotation error characteristics. The axial rotation error characteristic is specifically the temperature of the temperature sensitive point of the CNC grinder, and the radial rotation error characteristic is specifically the change in the spindle speed of the CNC grinder. In the specific implementation, the spindle rotation error compensation model of the CNC grinder is a gate loop unit neural network. The temperature of the temperature-sensitive point of the CNC grinder and the spindle speed changes are used as inputs of the gate loop unit neural network. The spindle rotation error measurement value is used to calibrate the gate loop unit neural network for the first time. The compensation model continuously outputs the spindle rotation error compensation value, and compensates for the spindle rotation error of the CNC grinder according to the spindle rotation error compensation value. Specifically, the coordinate origin offset method is used to compensate for the axial and radial rotation errors of the CNC grinder spindle. In the specific implementation, a value equal to the compensated spindle rotation error and opposite in direction is input into the CNC controller through the I / O interface, and the spindle rotation error of the CNC grinder is calculated by using the external mechanical coordinate system origin translation function of the CNC system.
[0052] S5: Repeat S2-S3 to calculate and obtain the spindle rotation error measurement value, use the current spindle rotation error measurement value to calibrate the current CNC grinder spindle rotation error compensation model and update the compensation model, and then obtain the latest spindle rotation error compensation value, thereby compensating the CNC grinder spindle rotation error.
[0053] In S5, considering that the accuracy of the compensation model is difficult to maintain for a long time, the present invention calculates the spindle rotation error measurement value based on each trajectory image and regularly calibrates it. At the moment of acquisition of the current trajectory image, the spindle rotation error compensation value is output by the CNC grinder spindle rotation error compensation model to achieve parameter optimization of the compensation model, and then update the compensation model, thereby improving the model accuracy.
[0054] S6: Repeat S5, continuously calibrate and update the CNC grinder spindle rotation error compensation model to achieve continuous optimization compensation of the CNC grinder spindle rotation error.
[0055] The above embodiments are used to illustrate the present invention rather than to limit the present invention. Any modifications and changes made to the present invention within the spirit of the present invention and the protection scope of the claims shall fall within the protection scope of the present invention.
Claims
1. A CNC grinder spindle rotation error evolution compensation method based on timing in-situ measurement, characterized in that: The following steps are involved: S1: Fix the CNC grinder spindle rotation error in-situ measuring device on the CNC grinder, determine multiple feature points on the edge of the spindle center hole, and align the CNC grinder spindle rotation error in-situ measuring device with the edge of the spindle center hole; S2: The CNC grinder is in operation, and the CNC grinder spindle rotation error in-situ measurement device is used to acquire the trajectory image of the feature points on the edge of the spindle center hole; S3: Obtain the spindle rotation error measurement value based on the trajectory image of the feature points on the edge of the spindle center hole; S4: The spindle rotation error characteristics of the CNC grinder and the spindle rotation error measurement value are integrated to establish a spindle rotation error compensation model for the CNC grinder. The compensation model outputs a spindle rotation error compensation value, and the spindle rotation error of the CNC grinder is compensated according to the spindle rotation error compensation value. S5: repeat S2-S3, calculate and obtain the spindle rotation error measurement value, use the current spindle rotation error measurement value to calibrate the current CNC grinder spindle rotation error compensation model and update the compensation model, so as to obtain the latest spindle rotation error compensation value; S6: Repeat S5, continuously calibrate and update the CNC grinder spindle rotation error compensation model, and realize continuous optimization compensation of the CNC grinder spindle rotation error.
2. The method for compensating the spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1 is characterized in that: In S1, the size of the feature point is set to 5 μm-20 μm.
3. The method for compensating the spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1 is characterized in that: In S2, the camera shutter time of the CNC grinder spindle rotation error in-situ measuring device is the same as the spindle rotation period, and the trajectory of the feature points of the edge of the spindle center hole as the spindle rotates N times is collected every m minutes to obtain an initial feature point image. After image enhancement and edge extraction processing are performed on the initial feature point image, a trajectory image of the feature points in the edge of the spindle center hole is obtained.
4. The method for compensating the spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1 is characterized in that: The S3 is specifically: According to the trajectory images of the edge feature points of the spindle center hole acquired by the right industrial camera (9) and the left industrial camera (12), the right industrial camera (9) and the left industrial camera (12) are respectively determined. The axis coordinates of the spindle in the corresponding camera coordinate system are then integrated and analyzed using the dual-target positioning method to obtain the world coordinates of the spindle axis of the CNC grinder. Finally, the axial rotation error and radial rotation error of the spindle are determined according to the world coordinates of the spindle axis of the CNC grinder and recorded as the spindle rotation error measurement value.
5. The method for compensating spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1, characterized in that: In S4, the CNC grinder spindle rotation error characteristics include axial rotation error characteristics and radial rotation error characteristics. The axial rotation error characteristics are specifically the temperature of the temperature sensitive point of the CNC grinder, and the radial rotation error characteristics are specifically the change in spindle speed of the CNC grinder.
6. The method for compensating the spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1, characterized in that: In S5, the spindle rotation error measurement value is calculated and calibrated according to each trajectory image. At the time of acquisition of the current trajectory image, the spindle rotation error compensation value is output by the CNC grinder spindle rotation error compensation model to optimize the parameters of the compensation model and then update the compensation model.
7. The method for compensating spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1, characterized in that: In S1, the CNC grinding machine spindle rotation error in-situ measurement device comprises a vertical guide rail frame (1), a horizontal guide rail (2), a front and rear guide rail (4), a sliding table (5), a pitch guide rail (6), a calibration light source (7), a left camera assembly and a right camera assembly; The vertical guide rail frame (1) is fixedly installed inside the CNC grinding machine, the horizontal guide rail (2) is installed in the vertical guide rail frame (1), and the horizontal guide rail (2) slides up and down in the vertical guide rail frame (1); the front and rear guide rails (4) are installed on the horizontal guide rails (2), and the front and rear guide rails (4) slide horizontally in the horizontal guide rails (2); the sliding table (5) is installed on the front and rear guide rails (4), and the sliding table (5) slides forward and backward in the front and rear guide rails (4); a calibration light source (7) is fixedly installed on the side of the sliding table (5), a pitch guide rail (6) is installed on the sliding table (5), and a left camera assembly and a right camera assembly are installed in the pitch guide rail (6).
8. The method for compensating spindle rotation error evolution of a CNC grinder based on timing in-situ measurement according to claim 1, characterized in that: The calibration light source (7) comprises a rotating aperture adjustment ring (71), a silicon-based target sheet (72), a support (73) and an LED light source (74); The support (73) is fixedly mounted on the LED light source (74), the rotating aperture adjustment ring (71) is mounted on the support (73), and the silicon-based target sheet (72) is mounted on the rotating aperture adjustment ring (71). A through hole is provided in the middle of the support (73), a light-transmitting ring hole is provided in the middle of the silicon-based target sheet (72), and a light beam emitted by the LED light source (74) passes through the through hole of the support (73) and the light-transmitting ring hole in sequence and then emerges from the rotating aperture adjustment ring (71).
Citation Information
Patent Citations
Trajectory regeneration compensation method of numerical control machine error
CN101943896A
Standard for measurement of main shaft radial turning errors
CN107036553A
Method for measuring radial rotation error of main shaft based on target trajectory tracking
CN107101593A
Main shaft radial runout measurement method based on machine vision
CN114166117A
Method and system for compensating eccentric error of main shaft of circular grating encoder
CN115682917A
Cited By
Compensation prediction method and system for numerically-controlled machine tool spindle state diagnosis
CN119937451A
A compensation prediction method and system for CNC machine tool spindle status diagnosis
CN119937451B
Rotation center calibration method, pose adjusting mechanism and wafer testing device
CN120593625A
