Intelligent rotor shaft detection method and system

Abstract

The invention relates to the field of rotor shaft image processing, in particular to a rotor shaft intelligent detection method and system. The method comprises the following steps: acquiring images of a plurality of continuous moments in the cutting process of a rotor shaft; the edge, in contact with the cutter, of the rotor shaft in the image is a cutting edge, and the edge, symmetrical about the central axis of the rotor shaft, of the cutting edge is a detection edge; a target image is screened out from the images at the multiple continuous moments, and when the target image meets at least one condition, the rotor shaft is unqualified; wherein the first condition is that the smoothness index of the detection edge is not in a preset smoothness index range; the second condition is that the variance between the coordinate of each pixel point of the detected edge and the preset coordinate of each pixel point is not in a preset variance range; the third condition is that the coordinates of the edge pixel points are detected not to be within the preset pixel point coordinate range and the number of the pixel points exceeds the preset number. The problem that unqualified rotor shafts cannot be timely and accurately screened is effectively solved.

Description

Technical Field

[0001] The present invention relates to the field of rotor shaft image processing, and more particularly to a rotor shaft intelligent detection method and system. Background Art

[0002] Ensuring machining accuracy is crucial during the rotor shaft cutting process, but various factors can contribute to inaccurate cutting. Over time, the cutting edge of the tool gradually dulls, increasing cutting forces, increasing surface roughness, and even causing dimensional deviations. Insufficient machine tool rigidity can also lead to even minor deformations during high-speed cutting or when machining hard materials, potentially affecting machining accuracy. Unstable tool clamping can cause offset or vibration during cutting, while insecure workpiece clamping can cause displacement or deformation. Improper cutting parameters can lead to inaccurate cutting and may also cause tool breakage, poor workpiece surface quality, or dimensional errors. Improper tool selection, such as tool material and geometry mismatched with the workpiece, can also compromise cutting results and reduce machining accuracy. Furthermore, improper use of cutting fluid is crucial. Inappropriate cutting fluids can cause excessive temperatures in the cutting area, affecting the thermal stability of the tool and workpiece, and consequently, compromising rotor shaft machining accuracy. Reduced machine tool accuracy, such as wear on key components like guideways, lead screws, and bearings, can reduce positioning and repeatability, impacting rotor shaft cutting accuracy.

[0003] At present, the patent application document with publication number CN117697536A and titled "A Gas Generator Set Rotor Shaft Combined Processing Device" discloses a gas generator set rotor shaft combined processing device, which uses a laser scanner to scan and detect the surface of the rotor shaft body, detects the processing accuracy of the rotor shaft body, and transmits the detection data to the controller. The controller automatically adjusts the No. 1 servo frame, No. 2 servo frame and No. 3 servo frame, and re-processes the positions with unqualified accuracy through the processing host to make the processing accuracy of the rotor shaft body meet the requirements, avoiding the need to remove the rotor shaft body for detection and re-positioning processing, thereby improving the accuracy and efficiency of the rotor shaft body processing.

[0004] Since the rotor shaft rotates at an extremely high speed during processing, the laser scanning method cannot process large amounts of data in a timely manner, and therefore cannot accurately screen out unqualified rotor shafts in a timely manner. Summary of the Invention

[0005] In order to solve the above-mentioned problem of being unable to timely and accurately screen out unqualified rotor shafts, the present invention provides solutions in the following aspects.

[0006] In a first aspect, the present invention provides an intelligent detection method for a rotor shaft, comprising: collecting images of a plurality of consecutive moments during the cutting process of the rotor shaft; the edge of the rotor shaft in contact with the cutting blade in the image is the cutting edge, and the edge of the cutting edge that is symmetrical about the central axis of the rotor shaft is the detection edge; screening out a target image from the images of the plurality of consecutive moments, and when the target image satisfies at least one condition, the rotor shaft is unqualified; wherein the first condition is: the smoothness index of the detection edge is not within a preset smoothness index range; the second condition is: the variance between the coordinates of each pixel point of the detection edge and the coordinates of each preset pixel point is not within a preset variance range; the third condition is: the coordinates of the pixel point of the detection edge are not within a preset pixel point coordinate range and the number of pixel points exceeds a preset number; the target image is the image corresponding to the minimum image recognition coefficient; the image recognition coefficient ,in, The vertical coordinate of the pixel point where the cutting edge of the first image contacts the cutting blade tip in the images at the plurality of consecutive moments, represents the ordinate of the rotor axis in the mth image, Indicates the edge detection of the mth image The horizontal coordinates of corresponding pixels are the same as the vertical coordinates of the pixels.

[0007] By calculating the image recognition coefficient for each image and screening out target images, the method can select the most appropriate image for further analysis based on the minimum image recognition coefficient, effectively improving image recognition accuracy. Based on the image's smoothness index, pixel coordinate variance, and pixel coordinate range, it can accurately detect whether the rotor shaft meets predetermined standards during machining. This multi-dimensional detection based on images and angular velocity comprehensively reflects the machining accuracy and quality of the rotor shaft. In summary, this method enables intelligent, precise, and rapid detection and screening of unqualified rotor shafts.

[0008] Preferably, the smoothness index of the detected edge is specifically: ,in, is the smoothness index of the detected edge, The number of pixels for edge detection in the target image, Indicates the target image edge detection The coordinates of the pixel points and The distance between the coordinates of the pixel points, Indicates the target image edge detection The coordinates of the pixel points and The distance between the coordinates of the pixel points.

[0009] By calculating the distance between each pixel and taking into account its distribution, we can more accurately describe the smoothness of the detected edge. Edge smoothness directly impacts the quality of rotor shaft machining, and quantifying the smoothness index helps accurately determine whether the rotor shaft has machining defects or irregularities.

[0010] Preferably, the variance between the coordinates of each pixel point of the detected edge and the coordinates of each preset pixel point is specifically: , where p is the variance between the coordinates of each pixel point of the detected edge and the coordinates of each preset pixel point, The number of pixels for edge detection in the target image, Indicates the horizontal coordinate of the ath pixel point on the edge of the target image detection. Indicates the vertical coordinate of the ath pixel point on the edge of the target image detection. Indicates the horizontal coordinate of the a-th pixel point of the preset detection edge, Indicates the vertical coordinate of the ath pixel of the preset detected edge.

[0011] By calculating the variance between the pixel coordinates of the target image's detected edge and the preset edge coordinates, the accuracy and consistency of the detected edge can be effectively quantified. A smaller variance indicates a closer match between the target image and the preset edge; a smaller variance may indicate deviation or irregular cutting during the machining process. Variance calculation is based on the coordinate information of each pixel point, reflecting changes in edge morphology from multiple perspectives. This multi-dimensional detection allows for more comprehensive identification of potential issues during rotor shaft machining, avoiding missed or misidentified errors.

[0012] Preferably, the preset smoothness index range includes: collecting the smoothness indices of the target image detection edges corresponding to multiple qualified rotor shafts in history after being cut at the time corresponding to the first image, and determining the preset smoothness index range based on the maximum and minimum values ​​therein.

[0013] By analyzing the smoothness index of the target image detected by multiple qualified rotor shafts after cutting at the time corresponding to the first image, we can determine the range of rotor shaft smoothness indexes that meet quality standards during actual machining. This historical data-based approach better reflects actual production conditions, avoids theoretical inaccuracies, and is more practical.

[0014] Preferably, the preset variance range includes: collecting the variances of the coordinates of each pixel point on the edge of the target image corresponding to multiple qualified rotor shafts cut at the time corresponding to the first image and the preset coordinates of each pixel point, and determining the preset variance range based on the maximum and minimum values ​​therein.

[0015] Preferably, the preset pixel point coordinate range includes: collecting the coordinates of each pixel point on the edge of the target image corresponding to multiple qualified rotor shafts after being cut at the time corresponding to the first image in history, and determining the preset pixel point coordinate range based on the maximum and minimum values ​​therein.

[0016] Preferably, the method of selecting a target image from images at multiple consecutive moments includes: presetting an image selection time interval as ,from Filter out the target image from the corresponding image, and The calculation method is as follows: the moment corresponding to the first image is used as the starting moment to collect the rotor shaft angular velocity, and the collection period is T. , , s represents the number of times the angular velocity is collected when the sum of the product of the angular velocity and the collection period is greater than or equal to π for the first time.

[0017] The angular velocity of the rotor shaft is constantly changing during machining. This method can minimize the time interval containing the target image, thereby significantly reducing the amount of calculation.

[0018] In a second aspect, the present invention further provides a rotor shaft intelligent detection system, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-mentioned rotor shaft intelligent detection method.

[0019] The beneficial effects of the present invention are as follows: by calculating the image recognition coefficient for each image and screening out the target image, the most appropriate image can be selected for further analysis based on the minimum value of the image recognition coefficient, which can effectively improve the accuracy of image recognition. Based on the image smoothness index, the variance of the pixel coordinates, and the pixel coordinate range, it is possible to accurately detect whether the rotor shaft meets the predetermined standards during the processing process. This multi-dimensional detection based on images and angular velocity can comprehensively reflect the processing accuracy and quality of the rotor shaft. In summary, this method can intelligently, accurately, and quickly detect and screen out unqualified rotor shafts. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present invention are shown in an illustrative and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 This is a flow chart of a rotor shaft intelligent detection method provided by an embodiment of the present invention; Figure 2 This is a block diagram of a rotor shaft intelligent detection system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] 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 them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0022] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 This is a flow chart of a rotor shaft intelligent detection method provided by an embodiment of the present invention, comprising the following steps: S101. Collect images of the rotor shaft at multiple consecutive moments during the cutting process; the edge of the rotor shaft in the image that contacts the cutting blade is the cutting edge, and the edge of the cutting edge that is symmetrical about the central axis of the rotor shaft is the detection edge.

[0024] In the first image, the rotor shaft is cut by the cutting blade at the cutting edge. Because the cutting edge is in contact with the cutting blade, the image captured at this time does not reflect the actual condition of the rotor shaft after cutting. Therefore, the rotor shaft's true condition needs to be analyzed by detecting the rotor shaft edge in the image. The cutting edge of the rotor shaft is generally the top edge. Therefore, an image acquisition device can be installed at a suitable location on the side of the rotor shaft, aligned with the rotor axis, to capture the rotor shaft image. The captured image must include the entire rotor shaft.

[0025] S102. Filter out a target image from images at multiple consecutive moments. When the target image meets at least one condition, the rotor shaft is unqualified; wherein the first condition is that the smoothness index of the detected edge is not within a preset smoothness index range; the second condition is that the variance between the coordinates of each pixel point of the detected edge and the coordinates of each preset pixel point is not within a preset variance range; and the third condition is that the coordinates of the pixel point of the detected edge are not within a preset pixel point coordinate range and the number of pixels exceeds a preset number.

[0026] During the rotor shaft machining process, the cutting edge of the rotor shaft is cut by the cutting tool in the first image. To detect whether the rotor shaft is qualified after this cutting, it is necessary to collect an image of the detected edge after the rotor shaft is rotated 180° for analysis. The image collected at this time is the target image. Therefore, a method is needed to filter out the target image from the multiple collected images. In some embodiments, the target image is the image corresponding to the minimum image recognition coefficient. The image recognition coefficient is ,in, The vertical coordinate of the pixel point where the cutting edge of the first image contacts the cutting blade tip in the images at the plurality of consecutive moments, represents the ordinate of the rotor axis in the mth image, Indicates the edge detection of the mth image The horizontal coordinates of corresponding pixels are the same as the vertical coordinates of the pixels. It shows that in the mth image, and The degree of symmetry about the rotor shaft centerline, The smaller it is, the higher the degree of symmetry is, and the more likely the mth image is the image of the rotor shaft rotated 180° after the cutting edge is cut by the cutting knife in the first image.

[0027] In the above method, the target image screening needs to calculate the image recognition coefficient of a large number of collected images, which is computationally intensive and slow. Therefore, in some embodiments, the image screening time interval can be preset as ,from Filter out the target image from the corresponding image, and The calculation method is as follows: the moment corresponding to the first image is used as the starting moment to collect the rotor shaft angular velocity, and the collection period is T. , , s represents the number of times the angular velocity is collected when the sum of the product of the angular velocity and the collection period is greater than or equal to π for the first time. For example, to detect whether the rotor shaft is qualified after the cutting edge is cut by the cutting knife in the first image, the angular velocity of the rotor shaft is collected starting at the time corresponding to the first image, and the collection period is T. The angular velocities collected in chronological order are 、 、 、 , , ,therefore , .

[0028] After obtaining the target image, it is necessary to analyze the detected edges in the target image to determine whether the rotor shaft is qualified. In some embodiments, the qualification of the rotor shaft can be determined by calculating whether the smoothness index of the detected edges of the rotor shaft in the target image is within a preset smoothness index range. The smoothness index of the detected edges of the rotor shaft in the target image is specifically: ,in, The smoothness index of the edge detected by the rotor axis of the target image, The number of pixels for edge detection in the target image, Indicates the target image edge detection The coordinates of the pixel points and The distance between the coordinates of the pixel points, Indicates the target image edge detection The coordinates of the pixel points and The preset smoothness index range is generally determined by collecting the smoothness indices of the target image edge detected by multiple qualified rotor shafts after being cut at the time corresponding to the first image, and the preset smoothness index range is determined based on the maximum and minimum values ​​among them.

[0029] In addition, whether the rotor shaft is qualified can also be determined by calculating whether the variance between the coordinates of each pixel point of the rotor shaft detection edge of the target image and the coordinates of each preset pixel point is within a preset variance range, specifically: , The variance between the coordinates of each pixel point on the detected edge and the coordinates of each preset pixel point, wherein, The number of pixels for edge detection in the target image, Indicates the horizontal coordinate of the ath pixel point on the edge of the target image detection. Indicates the vertical coordinate of the ath pixel point on the edge of the target image detection. Indicates the horizontal coordinate of the a-th pixel point of the preset detection edge, Represents the ordinate of the ath pixel of the preset detection edge. The preset variance range is generally the variance between the coordinates of each pixel of the target image detection edge corresponding to the time corresponding to the first image of multiple qualified rotor shafts cut in the history and the preset coordinates of each pixel. The preset variance range is determined based on the maximum and minimum values. By calculating the variance between the pixel coordinates of the target image detection edge and the preset detection edge coordinates, the accuracy and consistency of the detection edge can be effectively quantified. The smaller the variance, the better the match between the target image and the preset edge. Conversely, it may indicate deviation or irregular cutting during the processing process.

[0030] In addition, whether the rotor shaft is qualified can be determined by calculating whether the number of pixel points whose coordinates of the target image detection edge pixel points are not within the preset pixel point coordinate range exceeds a preset number. The preset pixel point coordinate range is generally the coordinates of each pixel point of the target image detection edge corresponding to multiple qualified rotor shafts in history after being cut at the time corresponding to the first image. The preset pixel point coordinate range is determined based on the maximum and minimum values ​​therein.

[0031] The present invention calculates the image recognition coefficient for each image and selects the target image. Based on the image's smoothness index, the variance of pixel coordinates, and the pixel coordinate range, it accurately detects whether the rotor shaft meets predetermined standards during machining. This multi-dimensional detection based on images and angular velocity comprehensively reflects the machining accuracy and quality of the rotor shaft. In summary, this method allows for intelligent, precise, and rapid detection and screening of unqualified rotor shafts. Furthermore, screening unqualified rotor shafts during machining allows for timely cessation of machining, reducing energy waste and machine wear.

[0032] The present invention also provides a rotor shaft intelligent detection system. Figure 2 As shown, the system includes a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, the rotor shaft intelligent detection method described in the present invention is implemented.

[0033] The system further includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and thus will not be described in detail here.

[0034] In the present invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, the computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium can be part of, accessible to, or connectable to a device. Any application or module described in the present invention can be implemented using computer-readable / executable instructions that can be stored or otherwise maintained by such a computer-readable medium.

[0035] In the description of this specification, "multiple" and "several" mean at least two, such as two, three or more, etc., unless otherwise clearly defined.

[0036] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A rotor shaft intelligent detection method, characterized in that: include: Collect images of the rotor shaft at multiple consecutive moments during the cutting process; the edge of the rotor shaft in contact with the cutting blade in the image is the cutting edge, and the edge of the cutting edge symmetrical to the central axis of the rotor shaft is the detection edge; Screening out a target image from images at multiple consecutive moments, when the target image satisfies at least one condition, the rotor shaft is unqualified; The first condition is that the smoothness index of the detected edge is not within the preset smoothness index range; the second condition is that the variance between the coordinates of each pixel point of the detected edge and the preset coordinates of each pixel point is not within the preset variance range; the third condition is that the coordinates of the pixel point of the detected edge are not within the preset pixel point coordinate range and the number of pixels exceeds the preset number; The target image is the image corresponding to the minimum image recognition coefficient; The image recognition coefficient ,in, The ordinate of the pixel point where the cutting edge of the first image among the images at the plurality of consecutive moments contacts the cutting blade tip is represented, represents the ordinate of the rotor axis in the mth image, Indicates the edge detection of the mth image The corresponding pixel points have the same vertical coordinate as the horizontal coordinate of the pixel point.

2. The rotor shaft intelligent detection method according to claim 1, characterized in that: The smoothness index of the detected edge is specifically: , where q is the smoothness index of the detected edge, The number of pixels for edge detection in the target image, Indicates the target image edge detection The coordinates of the pixels and The distance between the coordinates of the pixel points, Indicates the target image edge detection The coordinates of the pixels and The distance between the coordinates of the pixel points.

3. The rotor shaft intelligent detection method according to claim 1, characterized in that: The variance of the coordinates of each pixel point of the detected edge and the coordinates of each preset pixel point is specifically: ,in, is the variance between the coordinates of each pixel point of the detected edge and the coordinates of each preset pixel point, The number of pixels for edge detection in the target image, Represents the horizontal coordinate of the ath pixel point on the edge of the target image detection. Indicates the ordinate of the ath pixel point on the edge of the target image detection. Indicates the horizontal coordinate of the ath pixel of the preset detected edge. Indicates the ordinate of the ath pixel of the preset detected edge.

4. The intelligent detection method for rotor shaft according to claim 1, characterized in that: The preset smoothing index range includes: The smoothness index of the target image detection edge corresponding to a plurality of qualified rotor shafts in history after being cut at the time corresponding to the first image is collected, and a preset smoothness index range is determined according to the maximum and minimum values ​​thereof.

5. The rotor shaft intelligent detection method according to claim 1, characterized in that: The preset variance range includes: The variances of the coordinates of each pixel point on the edge of the target image corresponding to multiple qualified rotor shafts cut at the time corresponding to the first image and the preset coordinates of each pixel point are collected, and the preset variance range is determined according to the maximum and minimum values ​​therein.

6. The intelligent detection method for rotor shaft according to claim 1, characterized in that: The preset pixel point coordinate range includes: The coordinates of each pixel point on the edge of the target image corresponding to the cutting of multiple qualified rotor shafts at the time corresponding to the first image are collected, and the preset pixel point coordinate range is determined according to the maximum and minimum values ​​therein.

7. The rotor shaft intelligent detection method according to claim 1, characterized in that: The method of selecting a target image from images at multiple consecutive moments includes: presetting an image selection time interval as ,from Filter out the target image from the corresponding image. and The calculation method is as follows: the rotor shaft angular velocity is collected at the time corresponding to the first image as the starting time, and the collection period is T. , , s represents the number of times the angular velocity is collected when the sum of the product of the angular velocity and the collection period is greater than or equal to π for the first time.

8. A rotor shaft intelligent detection system, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the rotor shaft intelligent detection method according to any one of claims 1 to 7.

Images