A Method and System for Real-time Monitoring of Production Equipment Data in a Digital Factory

Abstract

The present invention relates to the technical field of image data processing, and particularly relates to a method and system for real-time monitoring of production equipment data in a digital factory, including: collecting and clustering thermal images of castings to obtain several regions of the castings; obtaining the edge fluctuation degree and temperature difference degree of each junction pixel point; through the edge line where the junction pixel point is located, obtaining two different thickness regions corresponding to each junction pixel point, obtaining the temperature tomography degree of the pixel points in the two different thickness regions corresponding to each junction pixel point, and combining the temperature difference degree of each junction pixel point to obtain the temperature anomaly manifestation degree of each junction pixel point; based on the temperature anomaly manifestation degree and the edge fluctuation degree, obtaining the temperature anomaly degree of each junction pixel point and performing threshold judgment to obtain the temperature anomaly junction pixel points and determine the abnormal temperature region. The purpose of the present invention is to improve the accuracy of detecting small bubble defects in the junction zone of different thicknesses.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and particularly relates to a method and system for real-time monitoring of production equipment data in a digital factory. Background Art

[0002] The rise of digital factories marks the entry of the manufacturing industry into a new production mode. As an important production equipment, die-casting machines have significantly improved production efficiency and quality management levels through the application of intelligent and data-driven technologies. Die-casting machines mainly rely on the experience of operators and manual adjustments to control the production process. With the mature application of the Internet of Things, big data analysis, and artificial intelligence, die-casting machines have been comprehensively upgraded and integrated in digital factories.

[0003] Traditional die-casting machine temperature detection is to detect the temperature distribution on the body and mold surface of the die-casting machine through an infrared thermal imager to ensure uniform overall temperature distribution, thereby guaranteeing the quality of castings and the stability of molding. During the die-casting process, there may be thickness variations in different parts of the casting. Such thickness variations will result in different cooling rates and metal fluidities in different parts. Gas cannot be fully discharged during the solidification of the metal, forming bubbles. It is difficult to detect small bubble defects in the junction areas of different thicknesses by detecting temperature differences at different positions. Summary of the Invention

[0004] The present invention provides a method and system for real-time monitoring of production equipment data in a digital factory to solve existing problems.

[0005] The method and system for real-time monitoring of production equipment data in a digital factory of the present invention adopt the following technical solutions:

[0006] An embodiment of the present invention provides a method for real-time monitoring of production equipment data in a digital factory, the method comprising the following steps:

[0007] Collect thermal images of the casting, perform clustering on the thermal images to obtain several regions of the casting;

[0008] Obtain the edge lines where each boundary pixel point is located in the junction areas of different regions of the casting, and obtain the edge fluctuation degree of each boundary pixel point according to the temperature differences of the pixel points on the edge lines where each boundary pixel point is located;

[0009] Obtain the temperature difference degree of each junction pixel according to the temperature changes on both sides of the edge line where each junction pixel is located; through the edge line where the junction pixel is located, obtain two different thickness regions corresponding to each junction pixel, analyze the temperature changes of the pixels in the two different thickness regions, and obtain the temperature fault degree of the pixels in the two different thickness regions corresponding to each junction pixel; according to the temperature difference degree of each junction pixel and the temperature fault degree of the pixels in the two corresponding different thickness regions, obtain the temperature anomaly manifestation degree of each junction pixel;

[0010] Obtain the temperature anomaly degree of each junction pixel according to the temperature anomaly manifestation degree and the edge fluctuation degree of each junction pixel;

[0011] Perform a threshold judgment on the temperature anomaly degree of each junction pixel, obtain the junction pixels with temperature anomalies, and determine the abnormal temperature region.

[0012] Further, the specific method for obtaining the edge line where each junction pixel is located is as follows:

[0013] In the junction area between any two different regions, which is the common edge of the two different regions, select the th junction pixel, and record this junction area as the edge line where the th junction pixel is located.

[0014] Further, the obtaining of the edge fluctuation degree of each junction pixel includes:

[0015] Record all the pixels on the edge line where the th junction pixel is located except the th junction pixel as the adjacent pixels on the edge line where the th junction pixel is located; the specific calculation formula for the edge fluctuation degree of the th junction pixel is:

[0016]

[0017] Among them, represents the edge fluctuation degree of the th junction pixel; represents the temperature of the th junction pixel; represents the temperature of the th adjacent pixel on the edge line where the th junction pixel is located; represents the number of adjacent pixels on the edge line where the th junction pixel is located; represents the linear normalization function.

[0018] Further, the specific method for obtaining the temperature difference degree of each junction pixel is as follows:

[0019] Based on the edge line where each junction pixel is located, a window is constructed for each junction pixel respectively, obtaining the first part and the second part in the window of each junction pixel, and several adjacent pixels within the window of each junction pixel; the specific calculation formula for the temperature difference degree of the th junction pixel is:

[0020]

[0021] Wherein, represents the temperature difference degree of the th junction pixel; represents the temperature of the th junction pixel; represents the temperature of the th adjacent pixel within the window of the th junction pixel; represents the number of pixels in the window except the central pixel; represents the th junction pixel, and the absolute value of the difference between the temperature of the th pixel in the first part and the temperature of the th pixel in the second part in the window of the represents the number of pixels in the first part and the second part of the window; represents the linear normalization function; represents the absolute value function.

[0022] Further, the constructing a window for each junction pixel respectively, obtaining the first part and the second part in the window of each junction pixel, and several adjacent pixels within the window of each junction pixel includes:

[0023] Taking the th junction pixel as the central pixel, in the vertical direction of the tangent line of the edge line where the th junction pixel is located at the th junction pixel, a window with a size of is constructed. Through the tangent line of the edge line where the th junction pixel is located at the th junction pixel, the window is divided into the first part and the second part, and all the pixels in the window are recorded as the adjacent pixels of the window of the th junction pixel.

[0024] Further, obtaining two different thickness regions corresponding to each junction pixel point through the edge line where the junction pixel point is located, and analyzing the temperature changes of the pixel points in the two different thickness regions, to obtain the temperature tomography degree of the pixel points in the two different thickness regions corresponding to each junction pixel point includes:

[0025] Denote the two different regions corresponding to the junction area where the th junction pixel point is located as the two different thickness regions corresponding to the th junction pixel point; in the two different thickness regions corresponding to the th junction pixel point, denote the thickness region with the largest average pixel point temperature as the thicker region, and the region with the smallest average pixel point temperature as the thinner region. Taking the left to right of the th junction pixel point as the traversal order, for each pixel point on the edge line where the th junction pixel point is located, respectively select the pixel points closest to each pixel point in the thicker region and the thinner region, and use them as the pixel points corresponding to the positions of each pixel point on the edge line in the thicker region and the thinner region;

[0026] Denote the absolute value of the difference between the temperature of the pixel point at the th position in the thinner region corresponding to the th junction pixel point and the temperature of the pixel point at the th position in the thicker region as the temperature change amount of the pixel points at the th position in the two different thickness regions corresponding to the th junction pixel point;

[0027] Obtain the temperature change amounts of all the pixel points at all positions on the edge line where the th junction pixel point is located in the two different thickness regions corresponding to the th junction pixel point, and form a temperature sequence according to the traversal order;

[0028] Obtain the curvature of the temperature sequence according to the finite difference method, normalize the curvature of the temperature sequence, and denote the normalized result as the temperature tomography degree of the pixel points in the two different thickness regions corresponding to the th junction pixel point.

[0029] Further, obtaining the temperature anomaly manifestation degree of each junction pixel point includes:

[0030] Denote the product of the temperature difference degree of the th junction pixel point and the temperature tomography degree of the pixel points in the two different thickness regions corresponding to the th junction pixel point as the temperature anomaly manifestation degree of the th junction pixel point.

[0031] Further, the obtaining of the temperature anomaly degree of each junction pixel point includes:

[0032] Multiply the temperature, the temperature anomaly manifestation degree, and the edge fluctuation degree of the -th junction pixel point, and denote it as the temperature anomaly degree of the -th junction pixel point.

[0033] Further, the obtaining of the temperature anomaly junction pixel points includes:

[0034] When the temperature anomaly degree of the -th junction pixel point is greater than or equal to a preset anomaly threshold, mark the -th junction pixel point as a temperature anomaly junction pixel point;

[0035] When the temperature anomaly degree of the -th junction pixel point is less than the preset anomaly threshold, mark the -th junction pixel point as a temperature normal junction pixel point.

[0036] The present invention also provides a real-time monitoring system for digital factory production equipment data, including a memory, a processor, and a computer program stored in the memory and running on the processor. The processor executes the computer program stored in the memory to implement the steps of the foregoing real-time monitoring method for digital factory production equipment data.

[0037] The beneficial effects of the technical solution of the present invention are as follows: Collecting thermal images of the casting, clustering the thermal images to obtain several regions of the casting, which helps to identify the temperature distribution characteristics of different parts; obtaining the edge lines where each intersection pixel in the boundary area between different regions of the casting is located, and obtaining the edge fluctuation degree of each intersection pixel, which is beneficial to determining the local characteristics of temperature change on the boundary line and helps to identify the regions where the temperature changes violently on the boundary line; obtaining the temperature difference degree of each intersection pixel according to the temperature changes on both sides of the edge line where each intersection pixel is located, determining the temperature dividing line between different regions, and providing a basis for anomaly detection; obtaining two different thickness regions corresponding to each intersection pixel through the edge line where the intersection pixel is located, analyzing the temperature changes of the pixels in the two different thickness regions, and obtaining the temperature fault degree of the pixels in the two different thickness regions corresponding to each intersection pixel, revealing the temperature change characteristics of the internal structure of the casting and helping to understand the heat conduction situation inside the casting, especially the heat distribution difference in different thickness regions; obtaining the temperature anomaly manifestation degree of each intersection pixel according to the temperature difference degree of each intersection pixel and the temperature fault degree of the pixels in the two different thickness regions corresponding thereto; obtaining the temperature anomaly degree of each intersection pixel according to the temperature anomaly manifestation degree and the edge fluctuation degree of each intersection pixel, which is beneficial to accurately identifying the regions with temperature anomalies; performing a threshold judgment on the temperature anomaly degree of each intersection pixel, obtaining the intersection pixels with temperature anomalies and determining the abnormal temperature regions, and improving the detection accuracy of small bubble defects in the boundary area with different thicknesses. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0039] Figure 1 It is a flowchart of the steps of a method for real-time monitoring of data of production equipment in a digital factory according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific implementation manners, structures, features and effects of a method and system for real-time monitoring of data of production equipment in a digital factory according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this invention belongs.

[0042] The following specifically describes the specific solutions of a method and system for real-time monitoring of digital factory production equipment data provided by the present invention in conjunction with the accompanying drawings.

[0043] Please refer to Figure 1 , which shows a flowchart of the steps of a method for real-time monitoring of digital factory production equipment data provided by an embodiment of the present invention. The method includes the following steps:

[0044] Step S001: Collect thermal images of the casting, perform clustering on the thermal images, and obtain several regions of the casting.

[0045] Collect thermal images of the casting during the production process of the die-casting machine through an infrared thermal imager, obtain the temperature of each pixel point in the thermal image of the casting, and use the DBSCAN clustering algorithm to cluster all pixel points in the thermal image of the casting according to the temperature of each pixel point. The distance metric uses the absolute value of the difference between the temperatures of pixel points, and several clusters are obtained. Each cluster corresponds to several regions, so several regions of the casting are obtained.

[0046] Step S002: Obtain the edge line where each boundary pixel point is located in the boundary area between different regions of the casting, and obtain the edge fluctuation degree of each boundary pixel point according to the temperature difference of the pixel points on the edge line where each boundary pixel point is located.

[0047] It should be noted that during the die-casting process of the die-casting machine, due to the uneven thickness of different regions of the casting, it affects the solidification speed and thermal conductivity of the metal, and further affects the release of gas and the uniformity of metal fluidity during the solidification process of the metal. In the region with a larger thickness, due to slower heat conduction, the solidification speed is relatively slow, and the gas in the region cannot be effectively discharged and the metal flow speed is not sufficient to push out the gas, resulting in the formation of bubbles during the solidification process; the thickness difference of some regions of the casting is relatively large, resulting in uneven metal fluidity, and thus bubbles will be generated in the boundary area between different regions of the casting. The bubbles in the boundary area between different regions of the casting will change the thermal conductivity of the metal, resulting in uneven temperature distribution around the boundary area, causing abnormal temperature changes, that is, local temperature fluctuations. Temperature anomalies will cause large fluctuations in the edges of different clustered regions.

[0048] Specifically, in the boundary area between any two different regions, this boundary area is the common edge of the two different regions. Select the th boundary pixel point, and record this boundary area as the edge line where the th boundary pixel point is located; the All the pixel points on the edge where the th junction pixel point is located, except the th junction pixel point, are denoted as the adjacent pixel points on the edge where the th junction pixel point is located. According to the temperature difference of the pixel points on the edge where the th junction pixel point is located, the edge fluctuation degree of the

[0049] th junction pixel point is obtained. Furthermore, the specific calculation formula for the edge fluctuation degree of the

[0050]

[0051] th junction pixel point is as follows: where, represents the edge fluctuation degree of the th junction pixel point; represents the temperature of the th junction pixel point; represents the temperature of the th adjacent pixel point on the edge where the th junction pixel point is located; represents the number of adjacent pixel points on the edge where the th junction pixel point is located; represents the linear normalization function, and the normalization object is all junction pixel points'

[0052] It should be noted that represents the average temperature of all adjacent pixel points on the edge where the th junction pixel point is located, and represents the absolute value of the difference between the temperature of the th junction pixel point and the average temperature of all adjacent pixel points on the edge where it is located, which measures the magnitude of the average temperature difference between the temperature of the th junction pixel point and other pixel points in the direction of temperature boundary change. The larger this value is, the more uneven the temperature change around the represents the relative magnitude of the temperature difference between the th junction pixel point and its surrounding pixel points. Relative to the temperature of the th junction pixel point, the larger this value is, the more significant the influence of the temperature difference between the th junction pixel point and its surrounding environment on its own temperature. The larger the value of is, the greater the edge fluctuation degree of the

[0053] Step S003: Obtain the temperature difference degree of each intersection pixel according to the temperature changes on both sides of the edge line where each intersection pixel is located; obtain two different thickness regions corresponding to each intersection pixel through the edge line where the intersection pixel is located, analyze the temperature changes of the pixels in the two different thickness regions, and obtain the temperature tomography degree of the pixels in the two different thickness regions corresponding to each intersection pixel; obtain the temperature anomaly manifestation degree of each intersection pixel according to the temperature difference degree of each intersection pixel and the temperature tomography degree of the pixels in the two corresponding different thickness regions.

[0054] It should be noted that after obtaining the edge fluctuation degree of the th intersection pixel by analyzing the temperature difference between the th intersection pixel and the adjacent pixels on the edge line, since simply analyzing the temperature difference between the th intersection pixel and the adjacent pixels on the edge line cannot comprehensively reflect the temperature anomaly of the pixels in the junction area of different regions, it is necessary to further analyze by considering the temperature difference between the two sides of the intersection pixel on the edge line where it is located and the temperature difference of the pixels in different thickness regions.

[0055] It should be noted that when there are no bubbles generated in the junction area of different regions, the temperature difference between the two sides of the junction area is large, and the bubbles generated in the junction area of different regions have an obvious impact on the temperature distribution of the surrounding environment, that is, there are temperature changes on both sides of any intersection pixel in the junction area on the edge line where it is located. When the temperature changes on both sides of the intersection pixel on the edge line where it is located are large and the temperature difference between the corresponding pixels on both sides is small, the temperature of this intersection pixel may show anomalies.

[0056] Specifically, in the junction area, with the th intersection pixel as the central pixel, in the vertical direction of the tangent line of the edge line where the th intersection pixel is located at the th intersection pixel, construct a window with a size of . Through the tangent line of the edge line where the th intersection pixel is located at the th intersection pixel, divide the window into a first part and a second part, and record all the pixels in the window as the adjacent pixels in the window of the th intersection pixel; in this embodiment, , are used to construct the window. If the intersection pixel is close to the boundary of the thermal image and a complete window cannot be obtained, quadratic linear interpolation is used to fill the window.

[0057] It should be noted that the greater the temperature difference between the corresponding pixel points in the first part and the second part, the smaller the possibility that the temperature of the th junction pixel is abnormal. According to the temperature difference between the corresponding pixel points in the first part and the second part and the change of the temperature data of the th junction pixel and all adjacent pixel points within the window, the temperature difference degree of the th junction pixel is determined.

[0058] Specifically, the specific calculation formula for the temperature difference degree of the th junction pixel is as follows:

[0059]

[0060] Among them, represents the temperature difference degree of the th junction pixel; represents the temperature of the th junction pixel; represents the temperature of the th adjacent pixel within the window of the th junction pixel; represents the number of pixels within the window except the central pixel; represents the th junction pixel, and the absolute value of the difference between the temperature of the th pixel in the first part and the temperature of the th pixel in the second part in the window of the represents the number of pixels in the first part and the second part within the window; represents the linear normalization function, and the normalization object is the of all junction pixels; represents the absolute value function.

[0061] It should be noted that represents the average value of the temperatures of all adjacent pixels within the window, represents the degree of deviation of the temperature of the th junction pixel from the average value of the temperatures of all adjacent pixels within the window. The greater the degree of deviation, the greater the temperature difference between the th junction pixel and its two adjacent pixels, and the corresponding temperature difference degree is greater. represents the average value of the absolute values of the differences between the temperatures of the corresponding pixels in the first part and the second part in the window of the th junction pixel. The larger this value, the more normal the temperature difference on both sides of the th junction pixel, and the The smaller the possibility that the temperature of the junction pixel is abnormal, the greater the temperature difference of the

[0062] junction pixel.

[0063] Furthermore, it should be noted that the thicknesses of different regions of the thermal image of the casting are different, and there are significant temperature differences. Since the amount of metal in the thicker region is large and the heat distribution is average, the temperature is higher and evenly distributed, without obvious temperature anomalies and changes; since the metal thickness in the thinner region is thin, the heat conduction speed is fast, the temperature changes rapidly, and there are obvious temperature anomalies or change characteristics. When there is no bubble region at the junction of different thickness regions, due to the heat conduction performance of the metal, the temperature of the thicker region will quickly conduct to the thinner region, making the temperature difference of the pixel points in different thickness regions small; when the bubbles generated at the junction of different thickness regions form a heat bridge or thermal resistance, affecting the heat conduction, the temperature of the thicker region cannot quickly conduct to the thinner region, bypassing the bubbles and conducting to the thinner region from both sides of the edge of the bubbles, making the temperature of the pixel points on both sides of the thinner region higher and the temperature of the pixel points in the middle lower, and the temperature difference of the pixel points in different thickness regions is large, and the temperature shows a fault phenomenon.

[0063] Furthermore, the two different regions corresponding to the junction area where the junction pixel is located are denoted as the two different thickness regions corresponding to the junction pixel; in order to determine the temperature fault degree of the pixel points in the two different thickness regions corresponding to the junction pixel, the temperatures of the pixel points in the two different thickness regions corresponding to the junction pixel are quantified. In the two different thickness regions corresponding to the junction pixel, the thickness region with the largest average pixel point temperature is denoted as the thicker region, and the region with the smallest average pixel point temperature is denoted as the thinner region. Taking the left to right of the junction pixel as the traversal order, for each pixel point on the edge line where the junction pixel is located, the pixel points closest to each pixel point in the thicker region and the thinner region are respectively selected, that is, for each pixel point in the traversal order, the adjacent pixel points in the two different thickness regions are obtained and used as the pixel points corresponding to the positions of each pixel point on the edge line in the thicker region and the thinner region.

[0064] Furthermore, the specific calculation formula for the temperature change amount of the pixel points at the junction pixel corresponding to the two different thickness regions at the position is:

[0065]

[0066] Among them, represents the temperature change of the pixels at the -th position corresponding to the two different thickness regions of the -th junction pixel point; represents the temperature of the pixel at the -th position in the thinner region corresponding to the -th junction pixel point; represents the temperature of the pixel at the -th position in the thicker region corresponding to the -th junction pixel point; represents the absolute value function.

[0067] Furthermore, according to the above method, obtain the temperature change of the pixels at all positions on the edge line where the -th junction pixel point is located, corresponding to the two different thickness regions of the -th junction pixel point, and form a temperature sequence in the traversal order ; obtain the curvature of the temperature sequence according to the finite difference method, normalize the curvature of the temperature sequence, and record the normalized result as the temperature fault degree of the pixels corresponding to the two different thickness regions of the -th junction pixel point; the greater the curvature, the greater the fault degree. Among them, the finite difference method is a well-known technology, and the specific method will not be introduced here.

[0068] It should be noted that when the temperature difference degree of the -th junction pixel point is greater, and the temperature fault degree of the pixels in the two different thickness regions corresponding to the -th junction pixel point is greater, it indicates that the temperature of the -th junction pixel point may show abnormality.

[0069] Specifically, the specific calculation formula for the temperature abnormality degree of the -th junction pixel point is:

[0070]

[0071] Among them, represents the temperature abnormality degree of the -th junction pixel point; represents the temperature difference degree of the -th junction pixel point; represents the temperature fault degree of the pixels in the two different thickness regions corresponding to the -th junction pixel point.

[0072] It should be noted that the larger the value of, the greater the indication of the The temperature performance of the boundary pixel points is more abnormal.

[0073] Step S004: Obtain the temperature abnormality degree of each boundary pixel point according to the temperature abnormality performance degree and the edge fluctuation degree of each boundary pixel point.

[0074] It should be noted that the temperature abnormality performance degree of the th boundary pixel point and the edge fluctuation degree of the th boundary pixel point both show the connection with the temperature abnormality of the th boundary pixel point, comprehensively reflecting the temperature abnormality situation of the th boundary pixel point in the boundary area of different regions. The greater the temperature abnormality performance degree of the th boundary pixel point and the greater the edge fluctuation degree of the th boundary pixel point, the greater the possibility that the temperature of the th boundary pixel point is abnormal. Take the temperature abnormality performance degree of the th boundary pixel point and the edge fluctuation degree of the th boundary pixel point as the weight of the temperature of the th boundary pixel point to judge the temperature abnormality degree of the

[0075] The specific calculation formula for the temperature abnormality degree of the

[0076]

[0077] Among them, represents the temperature abnormality degree of the th boundary pixel point; represents the temperature abnormality performance degree of the th boundary pixel point; represents the edge fluctuation degree of the th boundary pixel point; represents the temperature of the th boundary pixel point.

[0078] It should be noted that the larger the value of , the greater the temperature abnormality degree of the

[0079] Similarly, obtain the temperature abnormality degree of each boundary pixel point according to the above steps.

[0080] Step S005: Perform a threshold judgment on the temperature abnormality degree of each boundary pixel point to obtain the boundary pixel points with temperature abnormality and determine the abnormal temperature area.

[0081] Threshold judgment is performed on the temperature anomaly degree of each obtained junction pixel. The anomaly threshold preset in this embodiment is taken as an example for description. In other embodiments, it can be set to other values, which are not limited in this embodiment. When the temperature anomaly degree of the th junction pixel is greater than or equal to the anomaly threshold , the th junction pixel is recorded as a temperature-anomaly junction pixel; when the temperature anomaly degree of the th junction pixel is less than the anomaly threshold , the

[0082] th junction pixel is recorded as a temperature-normal junction pixel.

[0083] Based on the temperature-anomaly junction pixels, several anomaly temperature regions are obtained through the region growing algorithm. Each anomaly temperature region is the detected small bubble defect in the junction zone with different thicknesses, completing the real-time monitoring of the data of the die-casting machine in the digital factory production equipment. Among them, the region growing algorithm is a well-known technology, and the specific method is not introduced here.

[0084] This embodiment is completed here.

[0085] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for real-time monitoring of production equipment data in a digital factory, characterized in that, The method includes the following steps: Collect thermal images of the casting, perform clustering on the thermal images, and obtain several regions of the casting; Obtain the edge line where each junction pixel in the junction area of different regions of the casting is located, and obtain the edge fluctuation degree of each junction pixel according to the temperature difference of the pixels on the edge line where each junction pixel is located; Obtain the temperature difference degree of each junction pixel according to the temperature changes on both sides of the edge line where each junction pixel is located; obtain two different thickness regions corresponding to each junction pixel through the edge line where the junction pixel is located, analyze the temperature changes of the pixels in the two different thickness regions, and obtain the temperature fault degree of the pixels in the two different thickness regions corresponding to each junction pixel; obtain the temperature anomaly manifestation degree of each junction pixel according to the temperature difference degree of each junction pixel and the temperature fault degree of the pixels in the two different thickness regions corresponding thereto; Obtain the temperature anomaly degree of each junction pixel according to the temperature anomaly manifestation degree and the edge fluctuation degree of each junction pixel; Perform threshold judgment on the temperature anomaly degree of each junction pixel, obtain the temperature anomaly junction pixels, and determine the abnormal temperature region.

2. The real-time monitoring method for the production equipment data of a digital factory according to claim 1, characterized in that The specific method for obtaining the edge line where each junction pixel is located is: At the boundary between any two different regions, which is the common edge of the two different regions, select the th boundary pixel, and denote the boundary region as the edge line where the th boundary pixel is located.

3. The real-time monitoring method for data of a digital factory production device according to claim 1, characterized in that The obtaining of the edge fluctuation degree of each junction pixel includes: The edge line where the th junction pixel point is located, except for the th junction pixel point, are denoted as the adjacent pixel points on the edge line where the th junction pixel point is located; The specific calculation formula for the edge fluctuation degree of the th junction pixel point is: Among them, represents the edge fluctuation degree of the th junction pixel point; represents the temperature of the th junction pixel point; represents the temperature of the th adjacent pixel point on the edge line where the th junction pixel point is located; represents the number of adjacent pixel points on the edge line where the th junction pixel point is located; represents the linear normalization function.

4. The real-time monitoring method for the data of a digital factory production device according to claim 1, characterized in that, The specific method for obtaining the temperature difference degree of each junction pixel is: Based on the edge line where each junction pixel is located, a window is constructed for each junction pixel respectively, obtaining the first part and the second part in the window of each junction pixel, as well as several adjacent pixels within the window of each junction pixel; the specific calculation formula for the temperature difference degree of the nth junction pixel is as follows: Among them, represents the temperature difference degree of the th junction pixel point; represents the temperature of the th junction pixel point; represents the temperature of the th adjacent pixel point within the window of the th junction pixel point; represents the number of pixel points in the window except the central pixel point; represents the absolute value of the difference in temperature between the th pixel point in the first part and the th pixel point in the second part in the window of the th junction pixel point; represents the number of pixel points in the first part and the second part of the window; represents the linear normalization function; represents the absolute value function.

5. The real-time monitoring method for the data of a digital factory production device according to claim 4, characterized in that The constructing of a window for each junction pixel respectively to obtain the first part and the second part in the window of each junction pixel, and several adjacent pixels in the window of each junction pixel includes: Taking the th junction pixel as the central pixel, in the vertical direction of the tangent line at the th junction pixel on the edge line where the th junction pixel is located, construct a window with a size of . Through the tangent line of the edge line where the th junction pixel is located at the th junction pixel, divide the window into a first part and a second part, and denote all the pixels in the window as the adjacent pixels of the window of the th junction pixel.

6. The real-time monitoring method for the data of a digital factory production device according to claim 1, characterized in that, The obtaining of two different thickness regions corresponding to each junction pixel through the edge line where the junction pixel is located, analyzing the temperature changes of the pixels in the two different thickness regions, and obtaining the temperature fault degree of the pixels in the two different thickness regions corresponding to each junction pixel includes: Denote the two different regions corresponding to the boundary zone where the th boundary pixel is located as the two different thickness regions corresponding to the th boundary pixel; In the two different thickness regions corresponding to the th boundary pixel, denote the thickness region with the largest average pixel temperature as the thicker region, and the region with the smallest average pixel temperature as the thinner region. Taking the left to right traversal order of the th boundary pixel, for each pixel on the edge line where the th boundary pixel is located, select the pixel closest to each pixel in the thicker region and the thinner region respectively, and use them as the pixels corresponding to each pixel on the edge line in the thicker region and the thinner region; Take the absolute value of the difference between the temperature of the pixel at the th position in the thinner region corresponding to the th junction pixel point and the temperature of the pixel at the th position in the thicker region, and denote it as the temperature change of the pixels at the th junction pixel point in two different thickness regions at the th position; Obtain the temperature change amounts of the pixels at all positions on the edge line where the two different thickness regions corresponding to the th junction pixel point are located, and form a temperature sequence in accordance with the traversal order; Obtain the curvature of the temperature sequence according to the finite difference method, normalize the curvature of the temperature sequence, and denote the normalized result as the temperature tomography degree of the pixels in two different thickness regions corresponding to the th junction pixel point.

7. The real-time monitoring method for the data of a digital factory production device according to claim 1, wherein The obtaining of the temperature anomaly manifestation degree of each junction pixel includes: Multiply the temperature difference degree of the th boundary pixel by the temperature tomography degree of the pixels in the two different thickness regions corresponding to the th boundary pixel, and denote it as the temperature anomaly manifestation degree of the th boundary pixel.

8. The real-time monitoring method for the data of a digital factory production device according to claim 1, wherein The obtaining of the temperature anomaly degree of each junction pixel includes: Multiply the temperature, temperature anomaly manifestation degree, and edge fluctuation degree of the th boundary pixel point, and denote it as the temperature anomaly degree of the th boundary pixel point.

9. The real-time monitoring method for the data of the production equipment in a digital factory according to claim 1, characterized in that, The obtaining of the temperature anomaly junction pixels includes: When the temperature anomaly degree of the th boundary pixel point is greater than or equal to a preset anomaly threshold, the th boundary pixel point is recorded as a temperature anomaly boundary pixel point; When the temperature anomaly degree of the th boundary pixel point is less than the preset anomaly threshold, the th boundary pixel point is recorded as a temperature-normal boundary pixel point.

10. A real-time monitoring system for production equipment data in a digital factory, comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, When the computer program is executed by a processor, it implements the steps of a method for real-time monitoring of digital factory production equipment data as described in any one of claims 1-9.

Images