Method for quickly correcting axis

Abstract

The invention relates to a method for quickly correcting an axis, which comprises the following steps of: S1, obtaining a ratio of a pixel of movement of a pixel in the center of the axis to an actual movement distance, and taking the ratio as a datum reference of the actual distance; s2, the system records axis center coordinates of different rotation angles at different heights; s3, establishing a regression model, and taking the variation of the regression model by a regression analysis statistical method; and S4, comparing the variation with a set acceptance standard value so as to feed back and display the states of the axis, the cylinder, the motor and the bearing seat in real time. According to the method, the correction camera is used for taking images upwards, axis images of all the suction nozzles are obtained respectively, axis center coordinate information is obtained through all heights and all rotation angles of the axis, then a corresponding model is established through regression analysis, and the corresponding variation is obtained through analysis of a regression equation statistical method; therefore, the state of the axis of the suction nozzle is analyzed to realize rapid correction of the axis.

Description

Technical Field

[0001] The invention relates to the technical field of automated testing, and in particular to a method for rapid axis correction. Background Art

[0002] In the rapid development of industrial automation, the efficient operation of automated equipment can effectively improve production efficiency and product yield. In the automated testing of IC chips, IC chips were usually manually clamped to test equipment or burning equipment for IC chip testing and burning procedures. Because manual operation cannot be done quickly in batches, the efficiency is low and it is prone to errors. Therefore, in order to improve production efficiency and product yield, an automated nozzle handling device was introduced to achieve batch suction and handling of IC chips.

[0003] The automatic nozzle device generally integrates multiple nozzle heads for batch operation, such as Figure 1-1 As shown, the suction nozzle 5 is connected to the cylinder 2 through the axis 3, and the axis 3 is restrained by the bearing seat 4 through the bearing and the bushing. Each suction nozzle is connected to an independent cylinder and is driven to extend and retract by the cylinder. The entire axis, bearing seat, and cylinder module are fixed and formed by the shell structure and connected to the motor screw system to perform synchronous up and down movement of the Z axis. When the cylinder drives the suction nozzle to extend, in an ideal state, the axis extends vertically, thereby driving the suction nozzle to absorb the IC chip in the tray carrier at the specified position. However, there will be certain errors in processing and assembly, such as Figure 1-2 As shown, often during processing and assembly, or after long-term use, the axis, bearing seat and other hardware will shift or tilt. If you are sucking a large IC chip, the impact may not be significant. However, with the high degree of chip integration, the size of some chips is getting smaller and smaller. Small deviations may cause the failure of chip suction, thus affecting production efficiency. Summary of the invention

[0004] The technical problem solved by the present invention is to provide an efficient, fast and accurate method for axis correction, thereby reminding whether the axis is off-axis.

[0005] The technical solution adopted by the present invention to solve its technical problem is:

[0006] The method for rapid axis correction comprises the following steps:

[0007] Step S1: obtaining the ratio of the pixel moved by the axis center pixel to the actual moving distance as a reference of the actual distance;

[0008] Step S2: the system records the axis center coordinates at different rotation angles at different heights;

[0009] Step S3: Establish a regression model and obtain its variance using regression analysis statistical method;

[0010] Step S4: Compare the variation with the set acceptance standard value to display the status of the axis, cylinder, motor and bearing seat in real time.

[0011] Furthermore, in the step S1, the suction nozzle device is imaged by calibrating the camera, calculating the appropriate exposure time, driving the suction nozzle device to move the XY coordinates within the field of view at each Z-axis position and capturing the image, and capturing the image by vertically shooting upward with low-angle annular white light. The algorithm is defined based on the center of the outer ring of the measuring axis to capture the axis center coordinates.

[0012] Preferably, in step S2, the axis center coordinates are recorded with equal Z-axis heights and equal rotation angles.

[0013] Furthermore, in step S2, the following steps are also included:

[0014] Step S101: When the axis is initially not extended, the axis image offset set is obtained and submitted to step S3 for regression analysis to establish a regression model, and compared with the set acceptance standard value to analyze and determine whether it is qualified.

[0015] Furthermore, the regression analysis model uses a nonlinear circular model to find the circle that is closest to the axis image offset set, predict its center and diameter, and compare the diameter with the set acceptance standard value to analyze and determine whether it is qualified.

[0016] Furthermore, the method further comprises the following steps:

[0017] Step S102: Use the XY coordinate values ​​of the center of the circle obtained by the regression analysis in step S101 as compensation values ​​for the original coordinates of each axis, analyze the difference between the original coordinates of each axis and the XY coordinate values ​​of the center of the circle, and compare the difference with the actual nozzle slope to analyze and obtain the status of the axis and the bearing seat.

[0018] Furthermore, the method further comprises the following steps:

[0019] Step S103: driving the cylinder to extend the axes, obtaining the corresponding cylinder image offset, calculating the difference between the cylinder image offset and the original coordinates of the axes, and comparing the difference with the preset acceptance standard to analyze the state of the cylinder.

[0020] Furthermore, the method further comprises the following steps:

[0021] Step S104: Record the third axis image offset at different rotation angles at different heights, and calculate the difference between the third axis image offset and the original coordinates of the axis at different heights and different rotation angles, and compare the difference with the preset acceptance standard to analyze and obtain the status of the axis, the bearing seat, the cylinder, and the motor.

[0022] Furthermore, the XY slopes of the cylinders have similar values ​​and are consistent with the slope of the Z axis. Check whether the calibration camera is properly installed and whether its optical axis is perpendicular to the machine mainboard. After calibrating the calibration camera, re-verify the entire process.

[0023] The beneficial effects of the present invention are:

[0024] The present invention obtains the axis image of each nozzle by calibrating the camera upwards to obtain the axis center coordinate information obtained at each axis height and each rotation angle, and then uses regression analysis to establish a corresponding model, and obtains the corresponding variation through the regression equation statistical method, thereby analyzing the state of the nozzle axis. Then, the software displays the state of the nozzle axis hardware, axis bearing seat hardware, cylinder, motor screw system and other hardware in real time to quickly determine whether the nozzle axis and other hardware are installed normally, thereby realizing the rapid correction of the axis. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is the background technology of the present invention;

[0026] Figure 2 is a system structure diagram of the present invention;

[0027] Figure 3 This is a step diagram of the method for rapid axis correction of the present invention;

[0028] Figure 4 , Figure 5 The regression analysis model adopted by the present invention;

[0029] Figure 6 , Figure 7 and Figure 8 It is a diagram of the technical problems of a specific embodiment of the present invention;

[0030] Fig. 9 It is a measurement diagram of the axis center of the present invention;

[0031] The markings in the figure are:

[0032] 1. Motor screw system, 2. Cylinder, 3. Axis, 4. Bearing seat, 5. Nozzle, 6. Calibration camera, 7. Nozzle device, 8. Axis distribution status; 501. Axis outer ring, 502. Axis center, 601. Camera image interface. DETAILED DESCRIPTION

[0033] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention, so the present invention is not limited by the specific embodiments disclosed below.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0035] like Figure 1-1 As shown, the automated nozzle device 7 generally integrates multiple nozzle heads for batch operation. The nozzle 5 is connected to the cylinder 2 through the axis 3. The axis 3 is restrained by the bearing seat 4 through the bearing and the bushing. Each nozzle 5 is connected to an independent cylinder 2 and is driven to extend and retract by the cylinder. The entire axis 3, the bearing seat 4, and the cylinder 2 module are fixedly mounted by the outer shell structure to form the nozzle device 7, and are connected to the motor screw system 1 to drive the entire device to move up and down along the Z axis. When the cylinder 2 drives the nozzle 5 to extend, ideally, the axis extends vertically, thereby driving the nozzle to absorb the IC chip in the tray carrier at the specified position. However, there will be certain errors in processing and assembly, such as Figure 1-2 As shown, often during processing and assembly, or after long-term use, the cylinder, axis, bearing seat and other hardware will shift or tilt. If you are sucking a large IC chip, the impact may not be significant. However, with the high degree of chip integration, the size of some chips is getting smaller and smaller. Small deviations may cause the failure of chip suction, thus affecting production efficiency.

[0036] To this end, the present invention provides a method for rapid axis calibration, which can quickly identify the status (qualified or abnormal) of hardware such as the axis, bearing seat, cylinder, motor screw system, etc., so as to quickly realize the calibration of the nozzle device. Figure 3 As shown, the following methods are included:

[0037] Step S1: Obtaining the ratio of the actual moving distance of the nozzle to the moving pixels of the axis center in the nozzle image.

[0038] Step S2: The system records the axis center coordinates at different heights and different rotation angles, and calculates the image offset of the actual axis center through the ratio in step S1.

[0039] Step S3: Calculate the variation of the image offset using a statistical method.

[0040] Step S4: Interpretation: comparing and interpreting the variation with the set acceptance standard to display the status of the nozzle axis, cylinder, Z-axis motor, bearing seat and other hardware in real time.

[0041] like Figure 2 and Figure 3 As shown, the present invention arranges or sets a group of upward-facing calibration cameras 6 below the suction nozzle device 7. The calibration cameras 6 are equipped with an image recognition function to quickly obtain the center position of the center of the circle at the end of the suction nozzle (i.e., the axis center position). This position can be represented by the pixel coordinates in the image. By recording the axis center coordinates at different rotation angles at different heights, the recorded coordinate data is calculated to obtain the image offset of the axis center, and the image offset is statistically analyzed to establish an analysis model, and then the variation is obtained. By comparing the variation with the acceptance standard value preset by the system, the specific state of the internal hardware of the feedback suction nozzle device is determined. If the variation is within the acceptance standard, it is qualified, otherwise there is an abnormal state, so that the errors of each hardware component of the suction nozzle device and the distribution between the components can be calculated.

[0042] In step S1, because the imaging images formed by the camera and the lens are different, the size of the image taken is different from the actual size for different cameras, different lenses, and different Z-axis heights. Therefore, it is necessary to calibrate the image module of the system first, so that the actual image offset of the nozzle axis in the image can be obtained. First, the nozzle device 7 is photographed by the calibration camera to take an image, and the exposure time at different Z-axis heights is calculated. Because the brightness of the axis image obtained by the camera module is different at different Z-axis heights of the nozzle device 7, it is necessary to obtain the exposure time at different Z-axis heights first. Then, the nozzle device is driven to move the XY coordinates in the field of view at each Z-axis position and take an image, and the axis center pixel movement coordinates of the nozzle are obtained. Then, the ratio of the actual movement distance of the nozzle to the axis center movement pixels in the image is calculated. This ratio is an important parameter for the subsequent calculation of the nozzle axis center image offset.

[0043] The ratio of the actual nozzle movement distance to the axis center movement pixels in the image is calculated as follows:

[0044] Because the nozzle axis image obtained by the camera below is different at different Z-axis heights, the scale (μm / pixel) obtained in the image at different heights will also be different according to the different Z-axis heights. Therefore, by obtaining the scale at the Z-axis height, the actual offset of the axis center image at different rotation angles obtained at the height can be calculated:

[0045] X-axis scale

[0046] X-axis scale

[0047] Among them, X Scale is the ratio of the actual moving distance of the nozzle in the X-axis direction to the moving distance of the axis center pixel, Xr is the actual moving distance of the nozzle in the X-axis direction, X0 is the original X-axis position coordinate of the nozzle axis image, and X1 is the X-axis position coordinate of the nozzle axis image after movement.

[0048] Y Scale It is the ratio of the actual moving distance of the nozzle in the Y-axis direction to the moving distance of the axis center pixel, Yr is the actual moving distance of the nozzle in the Y-axis direction, Y0 is the original Y-axis position coordinate of the nozzle axis image, and Y1 is the Y-axis position coordinate of the nozzle axis image after movement.

[0049] For example, the nozzle moves 1mm along the XY axis, and the nozzle axis center coordinate in the image moves from the image original coordinate position (105,110) to the image coordinate position (205,210).

[0050]

[0051] Therefore, we can calculate X Scale =10μm / pixel, Y Scale =10μm / pixel. We get the ratio X in the X-axis and Y-axis directions Scale and Y Scale After that, the ratio of the actual moving distance of the nozzle under the Z axis to the axis center moving pixel in the image can be obtained, which is used for the subsequent calculation of the image offset of the axis center pixel.

[0052] In the present invention, Fig. 9 As shown, the calibration camera takes an image by shooting vertically upward and with low-angle annular white light. In the camera image interface 601, when the awakening white light hits the end of the nozzle 5, it will reflect the image at its center to form an outer circle 501. Because the axis 3 is cylindrical, it will form a black axis. Therefore, the software system can quickly obtain the axis outer circle 501, and define the construction algorithm based on the measurement of the center of the axis outer circle 501 to quickly capture the coordinates of its center 502. The system obtains the grayscale value of the nozzle image axis through the calibration camera, and performs binarization processing to quickly obtain the axis region Region. After filtering out noise and trimming the shape through morphology, the outer circle range is filtered out, and then the center of the outer circle, that is, the nozzle image center 502, is found.

[0053] In step S2, the axis center image offsets of several rotation angles at different Z-axis heights are calculated. In the present invention, we take images with equal-proportional Z-axis heights and equal-proportional rotation angles, and calculate the axis center position in the recorded image. The Z-axis heights are 0 mm (axis not extended), 5 mm (axis extended 5 mm, i.e., dropped 5 mm), 10 mm, 15 mm, and 20 mm, respectively. The rotation angle is based on 45°. Samples are taken 8 times at each Z-axis height to obtain the axis center positions at angles of 0°, 45°, 90°, 135°, 180°, 225°, 270°, and 315°, respectively, as shown in the following table:

[0054] Z-axis 0 5 10 15 20 - - - Unit: mm angle 0 45 90 135 180 225 270 315 Unit: °

[0055] Rotate 45° 8 times at 0mm, 5mm, 10mm, 15mm and 20mm respectively to obtain the axis center coordinates at different Z-axis heights with a 45° rotation angle. Then calculate the image offset of the actual axis center through the ratio of the actual distance to the moving pixel in step S1, and then give it to the software system for statistical analysis to establish an analysis model. The variation of the image offset is obtained through statistical analysis, and then the variation is judged against the acceptance standard value to feedback whether it is qualified or abnormal on the software interface.

[0056] In step S3, the variation of the image offset is calculated by a statistical method. In the present invention, we use a nonlinear regression model or a linear model to perform statistical analysis on the data, and perform statistical analysis on the axis center image offset recorded at different angles to find out its variation, and then compare the variation with the acceptable standard value to feedback the hardware status of the nozzle device.

[0057] Further, we will describe the method of the present invention in detail with specific embodiments, such as Figure 4 and Figure 5 As shown, the following steps are also included:

[0058] Step 101: The axis center image offset Rn(w) obtained at 0mm is collected for regression analysis to establish a regression model. The regression analysis model uses a nonlinear circular regression model and finds the closest circle, whose center is Cn and the predicted diameter is Dn. If Dn<Ds (Ds is the acceptable standard value set by the system), it means that the axis is qualified; otherwise, the system will remind abnormalities, including axis and bearing seat abnormalities (the software interface can highlight the font prompt: please replace the axis or bearing seat). This step can quickly analyze and calculate whether there are abnormalities in the installation of the axis and the bearing seat; and whether there is an error offset in the axis during the assembly process, mainly to check whether the axis is bent.

[0059] Wherein, n represents the nozzle number, w represents the rotation angle, if Rn(w) is R1(0), it represents the axis center image offset of nozzle Z1 at 0° (initial angle); if Cn is C1, it represents the circular shape of nozzle Z1; if Dn is D1, it represents the predicted diameter of nozzle Z1.

[0060] The calculation formula of the above axis center image offset Rn(w) is as follows:

[0061] According to the above axis center calculation method, first obtain the axis center position in the image, and compare it with the error amount of the center of the entire image screen, and then use the X-axis direction ratio X calculated above to calculate the axis center position. Scale And the Y axis direction ratio Y Scale The actual image offset Rn(w) at the Z-axis height is obtained by conversion.

[0062] The X coordinate of the center of the image screen ImageX = ImageWidth*0.5;

[0063] The Y coordinate of the center of the image screen ImageY=ImageHeight*0.5;

[0064] Among them, ImageWidth is the width (length) of the image, ImageHeight is the height of the image, and the units of ImageWidth and ImageHeight are both pixels, which are determined by the size of the camera sensor and the image size.

[0065] Furthermore,

[0066] Actual image offset X = (image screen center X coordinate - axis center X coordinate) * X Scale ;

[0067] Actual image offset Y = (image screen center Y coordinate - axis center Y coordinate) * Y Scale ;

[0068] For example, the center coordinates of the image are (640, 480), and the axis center coordinates calculated when the nozzle is 0 mm high and the rotation angle is 0° are (620, 500). The ratio of the actual displacement distance at this height to the axis center moving pixels is X. Scale =10μm / pixel, Y Scale =10μm / pixel, then

[0069] Actual image offset X = (640-620)*10 = 200 μm / pixel;

[0070] The actual image offset Y = (480-500)*10 = -200 μm / pixel.

[0071] When the nozzle is rotated at a Z-axis height of 0 mm and the angle of rotation is 0°, the actual image offset Rn(w) of the axis center is (200, -200). Therefore, an XY coordinate axis can be established to distribute the axis image offset Rn(w) at all rotation angles at this height in the XY coordinate axis, so that the set of actual image offsets of the axis center of nozzle n can be fed back in the form of a chart in the figure.

[0072] like Figure 4 As shown in the figure, they are the sets of the axis center image offsets R1(w) and R2(w) of nozzles Z1 and Z2 respectively. We record the result of rotating the nozzle 45° once at 0mm (axis not extended), and rotate it 8 times in total. As can be seen from the figure, the axis image offset set R1 of nozzle Z1 is more concentrated than the axis image offset set R2 of nozzle Z2. From the result diagram, we can directly see that the offset of nozzle Z1 is small, and we can quickly judge that nozzle 1 meets the standard. However, in the software system, we need to calculate its variation again through statistical analysis, and establish a model analysis to judge whether the nozzle hardware is qualified by comparing the variation with the acceptance standard value preset by the system.

[0073] like Figure 4 As shown in the right figure, we can intuitively see that the axis offset of nozzle 2 is relatively large. At this point, we can build a model through regression analysis. In this embodiment, we use regression analysis to find the closest circle for each offset set, such as Figure 5 As shown, the center of the circle is Cn. We then predict and calculate its diameter Dn (variation), where n represents the nozzle number. If the predicted diameter Dn is less than the acceptable standard value Ds, it is qualified, indicating that it meets the standard. Otherwise, it will indicate an abnormality.

[0074] In this embodiment, the specific rules of statistical analysis are as follows:

[0075] If any of the following conditions are met, the specification is not met and it is judged as abnormal:

[0076] a. Maximum offset in the X direction of the axis - minimum offset in the X direction of the axis > RT;

[0077] b. Maximum offset of axis in Y direction - minimum offset of axis in Y direction > RT;

[0078] c. The standard deviation of the offset in the X direction of the axis is greater than RT / N;

[0079] d. The standard deviation of the axis Y direction offset> RT / N;

[0080] RT is the mechanism tolerance constraint, for example, the system presets RT=0.09.

[0081] We assume that the nominal coordinates (without error) and calculated coordinates (i.e. the measured coordinate centers) of nozzles 1-4 are:

[0082] Nozzle No. Z1 Z2 Z3 Z4 Nominal coordinates Xn1,Yn1 Xn2,Yn2 Xn3,Yn3 Xn4,Yn4 Calculate coordinates Xc1,Yc1 Xc2,Yc2 Xc3,Yc3 Xc4,Yc4

[0083] Among them, when the nominal coordinates of the first nozzle (Z1) are at the origin of the Z axis (0mm), the center of the nozzle is directly opposite to the XY coordinates of the robot arm at the center of the camera image. The nominal coordinates of the second to fourth nozzles are the nominal coordinates of the first nozzle plus the XY offset in the mechanism design. The calculated coordinates are obtained by measuring the coordinates of each angle and then calculating the center of the circle closest to its trajectory.

[0084] Take nozzle Z1 as an example.

[0085]

[0086] Among them, Max(X c1 -X n1 ) represents the calculation of the difference between the axis center coordinates of each image and the nominal coordinates of the nozzle Z1 (rotated 8 times at 45° at 0 mm as described above, and the axis center coordinates of each time are recorded), and the maximum value is taken; Min(X c1 -X n1 ) means calculating the difference between the center coordinates of the image axis of nozzle Z1 and the nominal coordinates each time, and taking the minimum value; Max(X c1 -X n1 )-Min(X c1 -X n1 ) indicates the difference between the maximum and minimum values ​​(i.e., the variation). If this difference is greater than the preset system tolerance limit condition RT (i.e., the preset acceptance standard value), it means that the deviation is too large and does not meet the requirements, and the software system prompts an abnormal state.

[0087] Among them, SQRT(X c1 -X n1 ) represents the calculation (X c1 -X n1 ), that is, the standard deviation of the offset in the X direction of the axis (i.e., the variation). If this deviation is greater than the preset system tolerance limit condition RT / N (i.e., the preset system acceptance standard value, which is RT / 3 in one embodiment), it means that the deviation is too large and does not meet the requirements, and the software system prompts an abnormal state.

[0088] The calculation formula and rules for nozzles Z2, Z3 and Z4 are the same as those for nozzle Z1. If any of the above four conditions a to d is not met, it means that the nozzle axis state is abnormal.

[0089] The judgment formula and function in the software are as follows:

[0090] a.fail if{Max[(Xc1-Xn1),(Xc2-Xn2),(Xc3-Xn3),(Xc4-Xn4)]-Min[(Xc1-Xn1),(Xc2-Xn2),(Xc3-Xn3),(Xc4-Xn4)]}>RT;

[0091] b.fail if{Max[(Yc1-Yn1),(Yc2-Yn2),(Yc3-Yn3),(Yc4-Yn4)]-Min[(Yc1-Yn1),(Yc2-Yn2),(Yc3-Yn3),(Yc4-Yn4)]}>RT;

[0092] c.fail if SQRT[(Xc1-Xn1),(Xc2-Xn2),(Xc3-Xn3),(Xc4-Xn4)]>RT / 3;

[0093] d.fail if SQRT[(Yc1-Yn1),(Yc2-Yn2),(Yc3-Yn3),(Yc4-Yn4)]>RT / 3.

[0094] Further, such as Figure 6 As shown, during the assembly or installation of the nozzle device, the bearing seat and the motor screw system or the entire nozzle device may cause axial displacement or overall displacement due to errors or irregularities in the processing and installation. For this reason, we need to further calculate the corresponding variation, and then compare the variation with the acceptable standard value preset by the system to determine whether there is any abnormality in the bearing seat and the motor screw system.

[0095] like Figure 6 As shown, the axis state distribution Figure 8 The dots in the figure are the projection simulation diagrams of the axis of each nozzle in the XY plane. The axis state distribution diagram of a nozzle device that meets the standard is as follows: Figure 6-1 When the axis, bearing seat or cylinder is offset or there is a large installation error, the axis state distribution Figure 8 like Figure 6-2 As shown, the axis center of individual nozzles will be offset. When the motor screw system 1 of the nozzle device is offset, the entire nozzle device 7 will be offset. Figure 8 like Figure 6-3 Therefore, if the calculated offset of a single nozzle is unqualified, it may be due to the axis, bearing seat hardware, or cylinder machining accuracy; if the slope of the overall offset is similar, it may be due to the motor screw system or the assembly or machining of the entire nozzle device. The software gives different prompts for different situations to remind specific problems and hardware status, so as to facilitate corresponding adjustments and improvements.

[0096] We still assume that the nominal coordinates (without error) and calculated coordinates (i.e., the measured coordinate centers) of nozzles 1-4 are:

[0097] Nozzle No. Z1 Z 2 Z 3 Z 4 Nominal coordinates Xn1,Yn1 Xn2,Yn2 Xn3,Yn3 Xn4,Yn4 Calculate coordinates Xc1,Yc1 Xc2,Yc2 Xc3,Yc3 Xc4,Yc4

[0098] The judgment conditions are as follows:

[0099] a. The X-axis difference of the calculated coordinates of the axis centers of adjacent nozzles in the axis image - the X-axis difference of the nominal coordinates < RT / N;

[0100] b. The Y-axis difference of the calculated coordinates of the axis centers of adjacent nozzles in the axis image - the Y-axis difference of the nominal coordinates < RT / N.

[0101] RT is the tolerance limit of the mechanism (i.e. the acceptance standard value preset by the system).

[0102] Let's take nozzles Z1 and Z2 as examples, and the judgment formula is as follows:

[0103]

[0104] Its meaning is that the coordinate difference between the axis center of nozzle Z1 and the axis center of nozzle Z2 after measurement and calculation needs to be smaller than the specification. In one embodiment, N is 3, that is, RT / 3, and RT is the system preset acceptance standard value (0.09). n2 -X n1 |It can be expressed by the system preset value Pitch(X), which means the X-axis distance between the axes of adjacent nozzles. For example, if the system preset value Pitch(X) is 37mm, it means the distance between adjacent nozzles is 37mm. n2 -Y n1 |It can also be expressed by the system preset value Pitch (Y), which means the Y-axis deviation of the corresponding nozzle axis. The ideal Pitch (Y) is 0mm, which means that the adjacent nozzles are on the same horizontal line. Therefore, the above judgment formula can be converted as follows:

[0105]

[0106] This means that the X-axis and Y-axis coordinates of the axis of nozzle Z1 and the axis of nozzle Z2 after measurement and calculation must be less than the specification, that is, RT / 3.

[0107] like Figure 6-2 As shown, when the bearing seat workpiece 4 of the third nozzle is offset, the corresponding axis 3 will be offset. Therefore, we can quickly determine that the bearing seat is abnormal through the above calculation formula and feedback the abnormal state through software.

[0108] Further, as Figure 6-3 shown, the slopes of the skews in the X and Y directions can be further calculated respectively through linear regression analysis as a reference for hardware improvement. If the offset of an individual nozzle is unqualified, it may be a problem with the hardware processing accuracy; while if the slopes of the offsets are similar, it may be an assembly problem, resulting in the overall offset of the entire nozzle device.

[0109] During the axis assembly and long-term use, since the axis is installed inside the housing, we cannot accurately understand the specific conditions of the internal axis 3 and the Z-axis cylinder 2. Disassembling and inspecting are time-consuming and laborious, and subtle offsets cannot be detected by the naked eye. Therefore, in order to quickly know whether components such as the internal cylinder or bushing are offset, or whether the entire nozzle device is offset, we need to drive the cylinder 2 to extend the axis 3 for further statistical analysis and interpretation.

[0110] Further, it also includes the following steps:

[0111] Through the above steps, we can quickly know the external problem situations such as the positions of the axis, bearing block workpiece, and the entire nozzle device in the initial state. However, we have no way of knowing whether the internal axis 3 and the Z-axis cylinder 2 are offset. Moreover, with long-term use, in order to ensure the accuracy of nozzle grasping, it is necessary to frequently calibrate to obtain the states of workpieces such as the internal axis or Z-axis cylinder of the device. Therefore, further, as Figure 7 and Figure 8 shown, it also includes the following steps:

[0112] Step S103: We drive the cylinder 2 to extend each axis 3 one by one, obtain the corresponding image offset amount Rn(w), where n represents the nozzle number and w represents the rotation angle, and calculate the difference εAn from the original coordinates Nn of the center of each nozzle axis. If εAn < εAs (the set acceptance standard), it is qualified. At the same time, the XY slopes of εA1 → εA2 → εA3 → εA4 are also calculated as a reference for subsequent improvement.

[0113] Step S104: Take images at different Z-axis heights. In the present invention, 0 mm (not extended), 5 mm (extended 5 mm, i.e., lowered 5 mm), 10 mm, 15 mm, and 20 mm are used as measurement cases to obtain the differences εZn-h between Zn-h (h represents the axis height, which are 0, 5, 10, 15, and 20 respectively) and the original coordinates Nn of the axis, and calculate the slopes Tn of each axis εZn-0 → εZn-5 → εZn-10 → εZn-15 → εZn-20. If Tn < Ts (the set acceptance standard), it is qualified; otherwise, an abnormal state is feedback.

[0114] If the XY slopes of the four cylinders εA1→εA2→εA3→εA4 have similar values ​​and meet the slope Tn of the Z axis, it is necessary to check again whether the calibration camera is properly installed and whether the optical axis is still perpendicular to the machine main board. After the calibration camera optical axis is calibrated, the entire verification process should be redone.

[0115] The calculation formula of the above slope Tn is as follows:

[0116] Assume that the measurement coordinates of nozzles Z1 to Z4 at a specified height are as follows:

[0117]

[0118] The slope Tn of the nozzles Z1 to Z4 can be calculated by simple linear regression analysis. The simple linear regression formula is Y=a+bX. In combination with this embodiment, the linear regression formula in this example is as follows:

[0119] Y zn h n =A zn +B zn *X zn h n ;

[0120] Among them, Y zn h n is the Y in the regression formula (known in the above table), X zn h n is the X in the regression formula (known in the above table), such as Y1h0, Y1h5, Y1h10, Y1h15 and Y1h20, and X1h0, X1h5, X1h10, X1h15 and X1h20, Azn is the intercept (to be calculated), and Bzn is the slope (that is, the slope Tn required to be calculated, to be calculated).

[0121] zn represents the nozzle number, such as nozzle Z1, nozzle Z2, nozzle Z3 and nozzle Z4; hn represents different Z-axis heights, where n represents the number of different Z-axis heights, such as H0~H5 in the above table, a total of 5 Z-axis heights, so n is 5, h0 means that the height H is 0mm, and h5 means that the height H is 5mm.

[0122] To this end, further, the linear regression formula of the four nozzles is as follows:

[0123]

[0124] Furthermore, the average value of the X coordinate and the average value of the Y coordinate of each nozzle measurement coordinate are taken, that is:

[0125]

[0126] Where, zn represents the nozzle number, X zn Avg represents the average value of the X coordinate. zn h5 represents the X-axis coordinate of the axis center measured at a height of 5 mm; Y zn Avg represents the average value of the Y coordinate. zn h5 indicates the Y-axis coordinate of the axis center measured at a height H of 5 mm.

[0127] Furthermore, the above average values ​​are brought into the above linear regression formula to calculate the slope Bn and intercept An. The calculation formula is as follows:

[0128]

[0129] The present invention obtains the axis image of each nozzle by correcting the camera upwards to obtain the axis center coordinate information obtained at each axis height and each rotation angle, and then uses regression analysis to establish a corresponding model, and obtains the corresponding variation (image offset) through the regression equation statistical method, thereby analyzing the state of the nozzle axis. Then, the software displays the state of the nozzle axis hardware, axis bearing seat hardware, cylinder, motor screw system and other hardware in real time to quickly determine whether the nozzle axis and other hardware are installed normally, thereby realizing rapid correction of the axis.

[0130] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for rapid axis correction, characterized in that: The following steps are involved: Step S1: Obtaining the ratio of the actual moving distance of the nozzle to the moving pixel of the axis center in the nozzle image; Step S2: The system records the axis center coordinates at different heights and different rotation angles, and calculates the image offset of the actual axis center through the ratio in step S1; Step S3: Calculate the variation of the image offset using a statistical method; Step S4: Interpretation: comparing and interpreting the variation with the set acceptance standard to display the status of the hardware such as the nozzle axis, cylinder, Z-axis motor, and bearing seat in real time.

2. The method for rapid axis correction according to claim 1, characterized in that: In step S1, the nozzle device is imaged by calibrating the camera, and the appropriate exposure time is calculated. The nozzle device is driven to move the XY coordinates within the field of view at each Z-axis position and image is captured. The image is captured by vertically upward shooting and low-angle annular white light. The algorithm is defined based on the center of the outer ring of the measuring axis to capture the axis center coordinates.

3. The method for rapid axis correction according to claim 1, characterized in that: In the step S2, the axis center coordinates are recorded with equal Z-axis heights and equal rotation angles.

4. The method for rapid axis correction according to claim 1, characterized in that: In step S2, the following steps are also included: Step S101: When the axis is initially not extended, the axis image offset set is obtained and submitted to step S3 for regression analysis to establish a regression model, and compared with the set acceptance standard value to analyze and determine whether it is qualified.

5. The method for rapid axis correction as claimed in claim 4, characterized in that: The regression analysis model adopts a nonlinear circular model to find the circle closest to the axis image offset set, predict its center and diameter, and compare the diameter with the set acceptance standard value to analyze and determine whether it is qualified.

6. The method for rapid axis correction according to claim 5, characterized in that: The following steps are also included: Step S102: Use the XY coordinate values ​​of the center of the circle obtained by the regression analysis in step S101 as compensation values ​​for the original coordinates of each axis, analyze the difference between the original coordinates of each axis and the XY coordinate values ​​of the center of the circle, and compare the difference with the actual nozzle slope to analyze and obtain the status of the axis and the bearing seat.

7. The method for rapid axis correction according to claim 6, characterized in that: The following steps are also included: Step S103: driving the cylinder to extend the axes, obtaining the corresponding cylinder image offset, calculating the difference between the cylinder image offset and the original coordinates of the axes, and comparing the difference with the preset acceptance standard to analyze the state of the cylinder.

8. The method for rapid axis correction according to claim 7, characterized in that: The following steps are also included: Step S104: Record the third axis image offset at different rotation angles at different heights, and calculate the difference between the third axis image offset and the original coordinates of the axis at different heights and different rotation angles, and compare the difference with the preset acceptance standard to analyze and obtain the status of the axis, the bearing seat, the cylinder, and the motor.

9. The method for rapid axis correction according to claim 7 or 8, characterized in that: The XY slopes of the cylinders have similar values ​​and are consistent with the slope of the Z axis. Check whether the calibration camera is properly installed and whether its optical axis is perpendicular to the machine mainboard. After calibrating the calibration camera, re-verify the entire process.

Images