Dynamic error control method of precision grinding machine considering grinding wheel diameter

Abstract

The application relates to the technical field of grinding machine error control, in particular to a precision grinding machine dynamic error control method considering the diameter of a grinding wheel, which comprises the following steps: collecting a grinding wheel surface image in a grinding machine and converting the image to a color space; screening non-spark pixel points in the color space according to the color features of sparks generated by the grinding wheel during grinding processing of the grinding machine, to obtain a first pixel point set; removing pixel points belonging to grinding machine parts in the first pixel point set based on the brightness difference of the non-spark pixel points, to obtain a second pixel point set; extracting each circular contour in the second pixel point set through the geometric features of the pixel points in the second pixel point set, obtaining the confidence of each circular contour, and using the confidence to determine the real contour of the grinding wheel in the image; obtaining the diameter of the real contour of the grinding wheel, calculating and adjusting the rotating speed of the grinding wheel, and controlling the dynamic error of the grinding machine. The application improves the control precision of the dynamic error of the grinding machine.

Description

Technical Field

[0001] This application relates to the field of grinding machine error control technology, specifically to a dynamic error control method for precision grinding machines that takes into account the diameter of the grinding wheel. Background Technology

[0002] Bearings, as crucial components in mechanical equipment, have a decisive impact on the coefficient of friction during the movement of rotating mechanical parts due to their machining accuracy and surface quality. Bearing machining typically relies on precision grinding machines. A grinding machine is a machine tool that uses an abrasive wheel to grind the surface of a workpiece, and most grinding machines use high-speed rotating grinding wheels. However, in actual grinding operations, the diameter of the grinding wheel decreases continuously during the grinding process, leading to a decrease in the wheel's linear velocity and consequently, a reduction in grinding accuracy. This makes it difficult for the grinding error of the grinding machine to meet the precision machining requirements of bearings. Therefore, grinding wheel diameter detection is crucial for ensuring a constant grinding wheel linear velocity and achieving dynamic error control in precision grinding.

[0003] With the development of machine vision technology, machine vision-based grinding wheel diameter detection methods have become a commonly used technique, offering advantages such as non-contact operation, real-time performance, and high accuracy. This method extracts the grinding wheel profile from images of the grinding wheel and calculates its diameter. Compared to traditional methods, it is less susceptible to vibrations during grinding in real-time detection of the grinding wheel diameter. However, in actual precision grinding, sparks generated during friction between the grinding wheel and the workpiece, ghosting of the profile caused by high-speed wheel rotation, and obstruction of parts in the camera's view of the grinding machine can all lead to inaccuracies in the detected edge position of the grinding wheel profile from the acquired images. This results in deviations in the grinding wheel diameter detection, affecting the accuracy of dynamic error control during grinding. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a dynamic error control method for precision grinding machines that takes into account the diameter of the grinding wheel, thereby resolving the existing issues.

[0005] The dynamic error control method for precision grinding machines that takes into account the grinding wheel diameter in this application adopts the following technical solution:

[0006] One embodiment of this application provides a dynamic error control method for a precision grinding machine that takes into account the diameter of the grinding wheel. The method includes the following steps:

[0007] Acquire images of the grinding wheel surface in the grinding machine and convert them to a color space;

[0008] Based on the color characteristics of the sparks generated by the grinding wheel during grinding, non-spark pixels in the color space are filtered to obtain the first set of pixels.

[0009] Based on the brightness difference of non-spark pixels, pixels belonging to grinding machine parts are removed from the first pixel set to obtain the second pixel set;

[0010] By extracting the geometric features of the pixels in the second pixel set, the circular contours within the second pixel set are analyzed. The dispersion of the gray values ​​of the pixels on each circular contour and the overall distribution of the brightness values ​​of the pixels are analyzed to obtain the confidence level of each circular contour, which is used to determine the true contour of the grinding wheel in the image.

[0011] Obtain the diameter of the grinding wheel's true profile, calculate and adjust the grinding wheel's rotational speed to control the dynamic error of the grinding machine.

[0012] In one embodiment, the image of the grinding wheel surface is converted to the Lab color space.

[0013] In one embodiment, obtaining the first set of pixels includes:

[0014] Thresholding is performed on the 'a' chromaticity value of all pixels in the grinding wheel surface image. Pixels with 'a' chromaticity values ​​less than the threshold are designated as non-spark pixels and form the first set of pixels.

[0015] In one embodiment, obtaining the second set of pixels includes:

[0016] Thresholding is performed on the L brightness value of all pixels in the first pixel set, and all pixels with L brightness values ​​less than the threshold are grouped into the second pixel set.

[0017] In one embodiment, extracting each circular contour within the second pixel set includes:

[0018] Connectivity analysis is performed on the pixels in the second pixel set, and the circular contours of each circle in the second pixel set are obtained using a circular detection algorithm.

[0019] In one embodiment, obtaining the confidence level of each circular contour includes:

[0020] The degree of dispersion is negatively correlated to obtain the color feature value of each circular contour; the average value of the brightness L of all pixels on each circular contour is used as the brightness feature value of each circular contour.

[0021] The confidence level of each circular contour is determined by combining the brightness feature value and the color feature value.

[0022] In one embodiment, the confidence level of the circular outline is positively correlated with both the brightness feature value and the color feature value.

[0023] In one embodiment, the circular outline with the highest confidence level is taken as the true outline of the grinding wheel in the image.

[0024] In one embodiment, calculating and adjusting the rotational speed of the grinding wheel includes:

[0025] The diameter of the actual contour of the grinding wheel in the image is converted into the actual distance. Combined with the linear velocity of the grinding wheel, the theoretical rotational speed of the grinding wheel is calculated, and the motor speed of the grinding wheel is adjusted to the theoretical rotational speed.

[0026] In one embodiment, calculating the theoretical rotational speed of the grinding wheel includes:

[0027] Calculate the product of pi (π) and the actual distance. The theoretical rotational speed of the grinding wheel is the ratio of the linear velocity of the grinding wheel to the product.

[0028] This application has at least the following beneficial effects:

[0029] This application analyzes the distribution characteristics of the true contour of the grinding wheel, as well as the sparks generated by the grinding wheel, the ghost image of the grinding wheel contour, and the interference of grinding machine parts in the camera image during the grinding process of bearing workpieces on a grinding machine. This analysis yields the true contour of the grinding wheel in the acquired image, and the actual diameter of the grinding wheel is calculated based on the obtained true circular contour. Compared with existing machine vision-based grinding wheel diameter detection methods, this approach effectively reduces the influence of sparks, ghost images of the grinding wheel contour, and interference of grinding machine parts in the camera image on the edge position of the detected grinding wheel contour in the acquired grinding wheel image. This improves the accuracy of the real-time detection results of the grinding wheel diameter, effectively ensuring a constant linear velocity of the grinding wheel in precision grinding and achieving dynamic error control in precision grinding. Attached Figure Description

[0030] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 A flowchart illustrating the steps of the dynamic error control method for precision grinding machines that takes into account the grinding wheel diameter, as provided in this application. Detailed Implementation

[0032] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the dynamic error control method for precision grinding machines considering the grinding wheel diameter proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0034] The following description, in conjunction with the accompanying drawings, details the specific scheme of the dynamic error control method for precision grinding machines that takes into account the diameter of the grinding wheel provided in this application.

[0035] This application provides an embodiment of a dynamic error control method for precision grinding machines that considers the grinding wheel diameter. Specifically, the method is described below. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0036] Step S001: Acquire an image of the grinding wheel surface in the grinding machine and convert it to a color space.

[0037] When grinding bearing workpieces using a precision grinding machine, a CCD camera is used to capture real-time images of the grinding wheel surface by directly facing the grinding wheel in the precision grinding machine. In this embodiment, a white corundum grinding wheel is used for grinding the bearing workpiece, and the image acquisition time interval for the grinding wheel surface is 0.1 seconds. The grinding wheel surface image is an RGB image. The acquisition time interval can be set according to actual conditions.

[0038] Secondly, the acquired grinding wheel surface image is preprocessed. Specifically, this embodiment takes any acquired grinding wheel surface image A as an example. The grinding wheel surface image A is converted into a grayscale image, and the median filtering algorithm is used to denoise the obtained grayscale image. While suppressing noise in the image, the contour details in the image are preserved, resulting in a denoised grinding wheel surface grayscale image B. At the same time, the grinding wheel surface grayscale image B is converted to the Lab color space, and the a, b chromaticity values ​​and L luminance values ​​of each pixel in the grinding wheel surface grayscale image B in the Lab color space are obtained respectively. These values ​​are used to reduce the influence of interference in the grinding wheel surface grayscale image B, namely sparks, contour ghosting, and grinding machine parts, on the subsequent acquisition of the true contour of the grinding wheel. The grayscale image conversion, median filtering algorithm, and Lab color space conversion are all existing known technologies, and the specific process will not be described in detail.

[0039] Step S002: Based on the color characteristics of the sparks generated by the grinding wheel during grinding, filter out non-spark pixels in the color space to obtain the first set of pixels.

[0040] Generally, when precision grinding machines are used to grind bearing workpieces, the sparks generated by the grinding wheel mainly contain red light components, making the sparks appear more reddish in the captured images. The non-spark areas within the grinding wheel region captured by a CCD camera typically do not exhibit this reddish tint. Specifically, white corundum grinding wheels are commonly used in the grinding of bearing steel. White corundum grinding wheels appear grayish-white due to their alumina purity exceeding 99.5%, while the components in precision grinding machines and bearing workpieces are typically silvery-white due to their metallic material. This results in the non-spark areas within the grinding wheel region not exhibiting a significant reddish tint in the captured images taken by a CCD camera.

[0041] Therefore, to avoid the situation where the pixels corresponding to the sparks generated by the grinding wheel during the grinding of bearing workpieces by a precision grinding machine are identified as edge pixels of the grinding wheel contour in the acquired image during subsequent extraction of the grinding wheel contour from the image, the following processing is performed:

[0042] In the Lab color space, the larger the 'a' chromaticity value of a pixel, the more pronounced its red hue, and the greater its likelihood of being a spark pixel. Therefore, taking a grayscale image B of a grinding wheel surface as an example, the 'a' chromaticity values ​​of all pixels in image B are used as input to the Otsu's inter-class variance algorithm, and the 'a' chromaticity threshold is output. The set of all pixels in the grayscale image B whose 'a' chromaticity value is less than the 'a' chromaticity threshold is denoted as the first pixel set G1 of the grayscale image B. This set represents the set of all non-spark pixels remaining after removing spark pixels in the grinding wheel region captured by the CCD camera. The Otsu's inter-class variance algorithm is a well-known technique; implementers can choose other feasible thresholding algorithms.

[0043] Step S003: Based on the brightness difference of non-spark pixels, remove pixels belonging to grinding machine parts from the first pixel set to obtain the second pixel set.

[0044] Because components in precision grinding machines, such as the worktable and spindle, are typically made of metal, they have high light reflectivity. In contrast, white corundum grinding wheels are non-metallic and have high surface roughness, resulting in low light reflectivity. Consequently, the components in precision grinding machines appear brighter in the captured images compared to the grinding wheel itself. Therefore, to prevent pixels corresponding to precision grinding machine components in the grinding wheel area captured by the CCD camera from being identified as edge pixels of the grinding wheel contour in subsequent image extraction, the following processing is performed:

[0045] The L brightness value of all pixels in the first pixel set G1 is used as the input of the Otsu's inter-class variance algorithm, and the output is the brightness threshold. The set of all pixels in the first pixel set G1 whose L brightness value is less than the brightness threshold is denoted as the second pixel set G2 of the grayscale image B of the grinding wheel surface. It is used to characterize the set of all pixels remaining in the grinding wheel region captured by the CCD camera after removing spark pixels and pixels corresponding to precision grinding machine parts.

[0046] Step S004: Extract the circular contours within the second pixel set by using the geometric features of the pixels within the second pixel set, analyze the dispersion of the gray values ​​of the pixels on each circular contour, and the overall distribution of the brightness values ​​of the pixels, and obtain the confidence level of each circular contour to determine the true contour of the grinding wheel in the image.

[0047] Since the CCD camera is directly facing the grinding wheel, the wheel has a complete circular outline in the captured image. Therefore, connected component analysis is performed on all pixels belonging to the second pixel set G2 in the grayscale image B of the grinding wheel surface, and the Hough circle detection algorithm is used to extract the circular outlines of the connected components, and the diameter of each circular outline is obtained. The extracted circular outlines are all suspected to be circles formed by the grinding wheel outline, because the ghost image of the outline formed by the grinding wheel during high-speed rotation also has a complete circular outline in the captured image. Connected component extraction and Hough circle detection are existing known techniques. Implementers can choose other feasible circle detection algorithms, such as circle fitting, deep learning algorithms, etc., and this embodiment does not impose any restrictions on this.

[0048] During the high-speed rotation of the grinding wheel, its various parts move rapidly within the field of view of the image acquisition device. Because of their relatively uniform material and structure, the high-speed rotating grinding wheel exhibits a relatively consistent color distribution in the acquired image. However, the outline ghosting formed by the high-speed rotation is not a direct image of the actual grinding wheel. Instead, it is a blurred image formed by light reflecting off the CCD camera's sensor due to the wheel's rapid movement. It is composed of light reflected from different edges of the grinding wheel at different times, resulting in a lack of consistent color distribution in the acquired image. Furthermore, the brightness of the outline ghosting is usually lower than the brightness of the actual grinding wheel outline.

[0049] Based on the above analysis, taking any extracted circular contour d as an example, the dispersion of gray values ​​of all pixels on the circular contour d is calculated, and the dispersion is negatively correlated and mapped. The result of the negative correlation mapping is used as the color feature value of each circular contour to evaluate whether the circular contour d has a consistent color distribution feature in the grayscale image B of the grinding wheel surface. The larger the color feature value, the more consistent the color distribution feature.

[0050] It should be noted that the negative correlation mapping represents a mathematical relationship, that is, one variable decreases as another variable increases, reflecting the negative correlation between the degree of dispersion and the color feature value. In this embodiment, the reciprocal of the degree of dispersion is used as the color feature value of the circular outline d. In another embodiment, the negative of the degree of dispersion can be used as the color feature value of the circular outline d. During fractional operations, when the denominator is 0, to ensure the calculation result is meaningful, a parameter adjustment factor can be added to the denominator for summation to prevent the denominator from being 0. This parameter adjustment factor is a very small positive number. For example, the value of this parameter adjustment factor can be 0.01. Its specific value can be set by the implementer according to the actual situation, and this application embodiment does not impose specific limitations. Furthermore, the degree of dispersion can be calculated using variance, standard deviation, coefficient of variation, etc. This embodiment uses standard deviation as the calculation method for the degree of dispersion.

[0051] Furthermore, the average L-brightness values ​​of all pixels on the circular contour d are used as the brightness feature value of the circular contour d. The larger the brightness feature value, the higher the brightness of the circular contour d in the grayscale image B of the grinding wheel surface. In another embodiment, the sum of the L-brightness values ​​of all pixels on the circular contour d can be used as the brightness feature value of the circular contour d.

[0052] The color and brightness feature values ​​of all circular contours in the grayscale image B of the grinding wheel surface are normalized using the Min-Max normalization method. Taking circular contour d as an example, the mean of the normalized color feature values ​​and the normalized brightness feature values ​​of circular contour d is used as the confidence score of circular contour d. This score is used to evaluate whether circular contour d is the true contour of the grinding wheel in the grayscale image B. The higher the confidence score, the more likely it is to be the true contour of the grinding wheel. The Min-Max normalization method is a well-known technique, and the specific process will not be described in detail.

[0053] The circular contour with the highest confidence is selected from all circular contours in the grayscale image B of the grinding wheel surface, and is used to represent the true circular contour corresponding to the grinding wheel in the grinding wheel area captured by the CCD camera.

[0054] Step S005: Obtain the diameter of the actual profile of the grinding wheel, calculate and adjust the rotational speed of the grinding wheel to control the dynamic error of the grinding machine.

[0055] The diameter of the circular contour with the highest confidence level is converted into its actual diameter in the real world. This converted actual diameter is denoted as the actual diameter of the grinding wheel at the moment the grayscale image B on the grinding wheel surface is acquired. The method for converting pixel distances in the image into actual distances is a well-known technique, and the specific process will not be elaborated upon.

[0056] Based on the actual diameter of the grinding wheel calculated from the real-time acquired images of the grinding wheel surface and the preset linear velocity of the grinding wheel, the theoretical rotational speed of the grinding wheel is calculated. The specific expression is as follows: Where n represents the theoretical rotational speed of the grinding wheel, in r / min. This indicates the preset linear velocity of the grinding wheel. In this embodiment, V is set to 45 m / s, but the implementer can set it as needed. Pi This indicates the actual diameter of the grinding wheel, in meters (m). The calculation of the grinding wheel rotation speed is a known technique.

[0057] The motor speed of the grinding wheel is adjusted in real time based on the theoretical rotational speed of the grinding wheel obtained by real-time calculation, thereby ensuring that the linear speed of the grinding wheel remains constant during the grinding process of the bearing workpiece by the precision grinding machine, and realizing the dynamic error control of the precision grinding machine.

[0058] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments of this specification have been described above. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0059] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0060] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for controlling the dynamic error of a precision grinding machine considering the diameter of the grinding wheel, characterized in that, The method includes the following steps: Acquire images of the grinding wheel surface in the grinding machine and convert them to the Lab color space; Based on the color characteristics of the sparks generated by the grinding wheel during grinding, non-spark pixels in the color space are filtered to obtain the first set of pixels. Based on the brightness difference of non-spark pixels, pixels belonging to grinding machine parts are removed from the first pixel set to obtain the second pixel set; By extracting the geometric features of the pixels in the second pixel set, the circular contours within the second pixel set are analyzed. The dispersion of the gray values ​​of the pixels on each circular contour and the overall distribution of the brightness values ​​of the pixels are analyzed to obtain the confidence level of each circular contour, which is used to determine the true contour of the grinding wheel in the image. Obtain the diameter of the true profile of the grinding wheel, calculate and adjust the rotational speed of the grinding wheel to control the dynamic error of the grinding machine; The step of obtaining the confidence level of each circular contour includes: performing a negative correlation mapping on the dispersion to obtain the color feature value of each circular contour; taking the average of the brightness values ​​L of all pixels on each circular contour as the brightness feature value of each circular contour; combining the brightness feature value and the color feature value to determine the confidence level of each circular contour; the confidence level of the circular contour is positively correlated with both the brightness feature value and the color feature value.

2. The dynamic error control method for precision grinding machines considering the grinding wheel diameter as described in claim 1, characterized in that, The first set of pixels is obtained, including: Thresholding is performed on the 'a' chromaticity value of all pixels in the grinding wheel surface image. Pixels with 'a' chromaticity values ​​less than the threshold are designated as non-spark pixels and form the first set of pixels.

3. The dynamic error control method for precision grinding machines considering the grinding wheel diameter as described in claim 2, characterized in that, The obtained second pixel set includes: Thresholding is performed on the L brightness value of all pixels in the first pixel set, and all pixels with L brightness values ​​less than the threshold are grouped into the second pixel set.

4. The dynamic error control method for precision grinding machines considering the grinding wheel diameter as described in claim 1, characterized in that, The extraction of each circular contour within the second pixel set includes: Connectivity analysis is performed on the pixels in the second pixel set, and the circular contours of each circle in the second pixel set are obtained using a circular detection algorithm.

5. The dynamic error control method for precision grinding machines considering the grinding wheel diameter as described in claim 1, characterized in that, The circular outline with the highest confidence level is taken as the true outline of the grinding wheel in the image.

6. The dynamic error control method for precision grinding machines considering the grinding wheel diameter as described in claim 5, characterized in that, The calculation and adjustment of the grinding wheel speed includes: The diameter of the actual contour of the grinding wheel in the image is converted into the actual distance. Combined with the linear velocity of the grinding wheel, the theoretical rotational speed of the grinding wheel is calculated, and the motor speed of the grinding wheel is adjusted to the theoretical rotational speed.

7. The dynamic error control method for precision grinding machines considering the grinding wheel diameter as described in claim 6, characterized in that, The calculation of the theoretical rotational speed of the grinding wheel includes: Calculate the product of pi (π) and the actual distance. The theoretical rotational speed of the grinding wheel is the ratio of the linear velocity of the grinding wheel to the product.

Images