Crystal cooling process monitoring system and method based on multi-source data fusion

Abstract

The present invention discloses a crystal cooling process monitoring system and method based on multi-source data fusion, which belongs to the general control system field. The present invention imports the image abnormality analysis change results within a set period into a cooling gradient abnormality analysis model to perform cooling gradient abnormality analysis, imports the obtained cooling gradient abnormality analysis results into an actual cooling gradient analysis model to obtain an actual cooling gradient, adjusts the cooling gradient during the cooling process according to the obtained actual cooling gradient, and analyzes the crystal quality change with temperature during the cooling process by performing abnormality evaluation on the crystal image during the cooling process. The required cooling gradient is accurately calculated and analyzed by comprehensively analyzing the crystal quality during the cooling process and the cooling safety, thereby improving the crystal quality during the cooling process and ensuring the safety during the cooling process.

Description

Technical Field

[0001] The present invention belongs to the general field of control systems, and more specifically relates to a crystal cooling process monitoring system and method based on multi-source data fusion. Background Art

[0002] The crystal cooling process in the prior art mainly utilizes the advantages of data fusion technology, integrates and analyzes multi-source data in the crystal cooling process to improve the accuracy and efficiency of the monitoring system, extracts key features from the fused data, such as cooling rate, temperature gradient, etc., and establishes a mathematical model of the crystal cooling process to predict and analyze the cooling behavior. According to the prediction results of the model and real-time monitoring data, the crystal cooling process is monitored in real time. When the cooling gradient deviates from the preset value, the parameters of the cooling system are adjusted by the control unit, such as adjusting the flow rate and temperature of the cooling medium. Through continuous data analysis and system feedback, the data fusion algorithm and monitoring strategy are optimized, so that the system can adapt to different crystal materials and process requirements, and improve the generalization ability of the monitoring system. Through this monitoring system based on multi-source data fusion, the control accuracy of the crystal cooling process can be effectively improved, and the heat treatment process can be optimized;

[0003] However, the prior art ignores the effect of the cooling gradient on the crystal quality during the crystal cooling process, and is unable to analyze the change of the crystal quality with temperature during the cooling process by performing abnormal evaluation on the crystal image during the cooling process, which results in the inability to accurately calculate and analyze the required cooling gradient based on the crystal quality and cooling safety during the cooling process, thereby causing the crystal quality to be reduced during the cooling process. Most of the prior art has the above problems, such as in the Chinese patents with application numbers CN112255964A and CN113759996A;

[0004] The above patents all have the problems raised by this background technology. In order to solve the problems raised by this background technology, this application designs a crystal cooling process monitoring system and method based on multi-source data fusion. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes a crystal cooling process monitoring system and method based on multi-source data fusion. The present invention obtains the crystal image in the cooling process in real time, and simultaneously obtains the temperature change gradient of the crystal cooling process, obtains the crystal image within a set time period, imports the obtained crystal image into an image abnormality analysis model for image abnormality analysis, imports the image abnormality analysis change result within a set period into a cooling gradient abnormality analysis model for cooling gradient abnormality analysis, imports the obtained cooling gradient abnormality analysis result into an actual cooling gradient analysis model to obtain the actual cooling gradient, adjusts the cooling gradient during the cooling process according to the obtained actual cooling gradient, performs abnormal evaluation on the crystal image during the cooling process, thereby analyzing the change of crystal quality with temperature during the cooling process, accurately calculates and analyzes the required cooling gradient by comprehensively considering the crystal quality during the cooling process and the cooling safety, thereby improving the crystal quality during the cooling process while ensuring the safety during the cooling process.

[0006] To achieve the above object, the present invention provides the following technical solution: a crystal cooling process monitoring method based on multi-source data fusion, which comprises the following specific steps:

[0007] Acquire the crystal image in the cooling process in real time, and simultaneously acquire the temperature gradient of the crystal in the cooling process;

[0008] Acquire crystal images within a set time period, and import the acquired crystal images into an image anomaly analysis model for image anomaly analysis;

[0009] Importing the image abnormality analysis change results within a set period into the cooling gradient abnormality analysis model to perform cooling gradient abnormality analysis;

[0010] According to the obtained abnormal analysis results of the cooling gradient, the actual cooling gradient is imported into the actual cooling gradient analysis model to obtain the actual cooling gradient;

[0011] The temperature reduction gradient is adjusted during the temperature reduction process according to the actual temperature reduction gradient obtained.

[0012] It should be noted here that as a preferred technical solution of the crystal cooling process monitoring method based on multi-source data fusion, the specific steps of obtaining the crystal image in the cooling process in real time and simultaneously obtaining the temperature change gradient of the crystal cooling process are:

[0013] S101, obtaining a crystal image during a cooling process with a set temperature change gradient, and simultaneously obtaining a set real-time temperature change gradient;

[0014] S102, pre-processing the acquired crystal image to obtain a clear crystal image, obtain crack data on the surface of the crystal image and pixel data of each pixel point, wherein the surface crack data includes surface crack length, maximum width and position data, and simultaneously obtain crystal safety pixel range data and store it in a storage component.

[0015] It should be noted that, as a preferred technical solution of the crystal cooling process monitoring method based on multi-source data fusion, the acquisition of crystal images within a set time period and importing the acquired crystal images into the image anomaly analysis model for image anomaly analysis includes the following specific steps:

[0016] S201, obtaining all crystal images within a set time period from the current moment, obtaining crystal images at the time nodes at the end of the set time period, and importing the crystal images at the time nodes at the end of the set time period into the image change abnormal value calculation formula to calculate the image change abnormal value, wherein the image change abnormal value calculation formula is: , where n is the number of crystal image pixels, xiw is the pixel value of the i-th pixel of the crystal image at the end of the set time period, xiwm is the standard pixel value of the i-th pixel of the crystal image at the end of the set time period, xiq is the pixel value of the i-th pixel of the crystal image at the start of the set time period, xiqm is the standard pixel value of the i-th pixel of the crystal image at the start of the set time period, T is the length of the set time period, xit is the pixel value of the i-th pixel of the crystal image at the time t of the set time period, dt is the time integral, xitm is the standard pixel value of the i-th pixel of the crystal image at the time t of the set time period, and a is the first accounting coefficient;

[0017] S202, obtaining a crystal image at a time node at the end of a set time period, obtaining crack data on the surface of the crystal image, and importing the crack data on the surface of the crystal image into an image crack change abnormal value calculation formula to calculate the image crack change abnormal value, wherein the image crack change abnormal value calculation formula is: , where m is the number of cracks on the crystal image at the end of the set time period, kj is the length of the jth crack on the crystal image at the end of the set time period, zj is the maximum width of the jth crack on the crystal image at the end of the set time period, M is the number of cracks on the crystal image at the beginning of the set time period, lc is the length of the cth crack on the crystal image at the beginning of the set time period, fc is the width of the cth crack on the crystal image at the beginning of the set time period, and exp() is the power of the natural constant e.

[0018] It should be noted here that as a preferred technical solution for the crystal cooling process monitoring method based on multi-source data fusion, the method of importing the image abnormality analysis change results within a set period into the cooling gradient abnormality analysis model for cooling gradient abnormality analysis includes the following specific steps:

[0019] S301, obtaining the calculated image change abnormal value and image crack change abnormal value, and substituting the calculated image change abnormal value and image crack change abnormal value into the image abnormality analysis change abnormality coefficient calculation formula to calculate the image abnormality analysis change abnormality coefficient, wherein the image abnormality analysis change abnormality coefficient calculation formula is: ,in, is the proportion of abnormal values ​​in image changes;

[0020] S302. Compare the image abnormality analysis change abnormality coefficient with the set image abnormality analysis change abnormality threshold. If the image abnormality analysis change abnormality coefficient is greater than or equal to the set image abnormality analysis change abnormality threshold, it is judged that the cooling gradient is abnormal and the cooling gradient needs to be adjusted. If the image abnormality analysis change abnormality coefficient is less than the set image abnormality analysis change abnormality threshold, it is judged that the cooling gradient is normal and no cooling gradient adjustment is needed.

[0021] It should be noted here that as a preferred technical solution for the crystal cooling process monitoring method based on multi-source data fusion, the obtained cooling gradient abnormality analysis result is imported into the actual cooling gradient analysis model to obtain the actual cooling gradient: the calculated image abnormality analysis change abnormality coefficient, image abnormality analysis change abnormality threshold and cooling gradient are obtained and imported into the actual cooling gradient calculation formula to calculate the actual cooling gradient, wherein the actual cooling gradient calculation formula is: , where Tm is the cooling gradient, ln() is the logarithm with the natural constant e as the base, and K is the abnormal threshold of image abnormality analysis change;

[0022] The calculated actual cooling gradient is obtained and compared with the set safe cooling gradient range. If the actual cooling gradient is within the set safe cooling gradient range, the cooling gradient is adjusted to the actual cooling gradient. If the actual cooling gradient is not within the set safe cooling gradient range, the safe cooling gradient with the smallest difference from the actual cooling gradient within the safe cooling gradient range is selected and set as the second actual cooling gradient, and the cooling gradient is adjusted to the second actual cooling gradient.

[0023] A crystal cooling process monitoring system based on multi-source data fusion, which is implemented based on the above-mentioned crystal cooling process monitoring method based on multi-source data fusion, and specifically includes a data acquisition module, an image abnormality analysis module, a cooling gradient abnormality analysis module, an actual cooling gradient acquisition module, a cooling gradient adjustment module and a control module, wherein the data acquisition module is used to acquire the crystal image in the cooling process in real time, and simultaneously acquire the temperature change gradient in the cooling process of the crystal;

[0024] The image anomaly analysis module is used to obtain crystal images within a set time period, and import the obtained crystal images into the image anomaly analysis model to perform image anomaly analysis;

[0025] The cooling gradient anomaly analysis module is used to import the image anomaly analysis change results within a set period into the cooling gradient anomaly analysis model to perform cooling gradient anomaly analysis;

[0026] The actual cooling gradient acquisition module is used to import the obtained cooling gradient anomaly analysis results into the actual cooling gradient analysis model to obtain the actual cooling gradient;

[0027] The temperature-lowering gradient adjustment module is used to adjust the temperature-lowering gradient during the temperature-lowering process according to the actual temperature-lowering gradient obtained;

[0028] The control module is used for controlling the operation of the data acquisition module, the image anomaly analysis module, the cooling gradient anomaly analysis module, the actual cooling gradient acquisition module and the cooling gradient adjustment module.

[0029] An electronic device comprises: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0030] The processor executes the above-mentioned crystal cooling process monitoring method based on multi-source data fusion by calling the computer program stored in the memory.

[0031] A computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the crystal cooling process monitoring method based on multi-source data fusion as described above.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] The present invention obtains the crystal image in the cooling process in real time, obtains the temperature change gradient of the crystal in the cooling process, obtains the crystal image within a set time period, imports the obtained crystal image into an image anomaly analysis model for image anomaly analysis, imports the image anomaly analysis change result within a set period into a cooling gradient anomaly analysis model for cooling gradient anomaly analysis, imports the obtained cooling gradient anomaly analysis result into an actual cooling gradient analysis model to obtain the actual cooling gradient, adjusts the cooling gradient during the cooling process according to the obtained actual cooling gradient, conducts an abnormality evaluation on the crystal image during the cooling process, thereby analyzing the change of crystal quality with temperature during the cooling process, accurately calculates and analyzes the required cooling gradient by comprehensively considering the crystal quality during the cooling process and the cooling safety, thereby improving the crystal quality during the cooling process and ensuring the safety during the cooling process. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic diagram of the overall flow of the crystal cooling process monitoring method based on multi-source data fusion of the present invention;

[0035] Figure 2 It is a schematic diagram of the steps of the actual temperature reduction gradient analysis model in the crystal temperature reduction process monitoring method based on multi-source data fusion of the present invention;

[0036] Figure 3 It is a schematic diagram of the overall framework of the crystal cooling process monitoring system based on multi-source data fusion of the present invention;

[0037] Figure 4 This is a schematic diagram of data transmission according to the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present application and its application or use.

[0039] Example 1

[0040] To solve the technical problem raised in the background technology: the prior art ignores the influence of the cooling gradient on the crystal quality during the crystal cooling process, and is unable to analyze the crystal quality changes with temperature during the cooling process by performing abnormal evaluation on the crystal image during the cooling process, which leads to the inability to accurately calculate and analyze the required cooling gradient based on the crystal quality and cooling safety during the cooling process, thereby causing the crystal quality to be reduced during the cooling process; the present invention provides a preferred embodiment: Figure 1-Figure 2As shown, the crystal cooling process monitoring method based on multi-source data fusion includes the following specific steps:

[0041] S1, obtaining a crystal image in the cooling process in real time, and simultaneously obtaining a temperature gradient of the crystal in the cooling process;

[0042] S2, acquiring crystal images within a set time period, and importing the acquired crystal images into an image anomaly analysis model for image anomaly analysis;

[0043] S3, importing the image abnormality analysis change results within the set period into the cooling gradient abnormality analysis model to perform cooling gradient abnormality analysis;

[0044] S4, importing the obtained abnormal analysis result of the cooling gradient into the actual cooling gradient analysis model to obtain the actual cooling gradient;

[0045] S5, adjusting the temperature reduction gradient during the temperature reduction process according to the actual temperature reduction gradient obtained;

[0046] In this embodiment, the specific steps of acquiring the crystal image in the cooling process in real time and acquiring the temperature gradient of the crystal in the cooling process are:

[0047] S101, obtaining a crystal image during a cooling process with a set temperature change gradient, and simultaneously obtaining a set real-time temperature change gradient;

[0048] S102, performing a pre-clearing process on the acquired crystal image to obtain a clear crystal image, obtaining crack data on the surface of the crystal image and pixel data of each pixel point, wherein the surface crack data includes surface crack length, maximum width and position data, and simultaneously obtaining crystal safety pixel range data and storing it in a storage component;

[0049] The specific steps are as follows: S1021, image acquisition:

[0050] Acquire images of the crystals using appropriate imaging equipment (e.g., microscope, high-resolution camera, etc.).

[0051] S1022, Image import:

[0052] Import the acquired images into image processing software or programming environments, such as the OpenCV library in Python or MATLAB.

[0053] S1023, image clarity preprocessing:

[0054] Denoising: Use filters (such as Gaussian filtering, median filtering) to remove image noise.

[0055] Contrast Enhancement: Adjust the image contrast to make the image clearer.

[0056] Sharpening: Apply sharpening filters (such as Sobel filtering, Laplacian filtering, Unsharp Masking) to enhance image edges;

[0057] The sample code is as follows: import cv2

[0058] import numpy as np

[0059] # Read the image

[0060] image = cv2.imread('crystal_image.jpg')

[0061] # Gaussian denoising

[0062] blurred = cv2.GaussianBlur(image, (5, 5), 0)

[0063] # Contrast enhancement

[0064] hist_equalized = cv2.equalizeHist(blurred)

[0065] # Sharpening

[0066] sharpened = cv2.addWeighted(hist_equalized, 1.5, cv2.Laplacian(hist_equalized, -1), 0, 0)

[0067] # Save the cleared image

[0068] cv2.imwrite('sharpened_crystal_image.jpg', sharpened);

[0069] S1024, crack detection

[0070] Use an edge detection algorithm (such as Canny edge detection) to identify cracks.

[0071] Apply morphological operations (such as dilation and erosion) to connect crack regions;

[0072] The sample code is as follows:# Canny edge detection

[0073] edges = cv2.Canny(sharpened, 100, 200)

[0074] Morphological operations

[0075] kernel = np.ones((3,3), np.uint8)

[0076] dilated = cv2.dilate(edges, kernel, iterations=1)

[0077] # Find crack outlines

[0078] contours, _ = cv2.findContours(dilated, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

[0079] # Draw the cracks

[0080] for contour in contours:

[0081] cv2.drawContours(sharpened, [contour], -1, (0, 255, 0), 2);

[0082] In this embodiment, obtaining a crystal image within a set time period and importing the obtained crystal image into an image anomaly analysis model for image anomaly analysis includes the following specific steps:

[0083] S201, obtaining all crystal images within a set time period from the current moment, obtaining crystal images at the time nodes at the end of the set time period, and importing the crystal images at the time nodes at the end of the set time period into the image change abnormal value calculation formula to calculate the image change abnormal value, wherein the image change abnormal value calculation formula is: , where n is the number of crystal image pixels, xiw is the pixel value of the i-th pixel of the crystal image at the end of the set time period, xiwm is the standard pixel value of the i-th pixel of the crystal image at the end of the set time period, xiq is the pixel value of the i-th pixel of the crystal image at the start of the set time period, xiqm is the standard pixel value of the i-th pixel of the crystal image at the start of the set time period, T is the length of the set time period, xit is the pixel value of the i-th pixel of the crystal image at the time t of the set time period, dt is the time integral, xitm is the standard pixel value of the i-th pixel of the crystal image at the time t of the set time period, and a is the first accounting coefficient;

[0084] S202, obtaining a crystal image at a time node at the end of a set time period, obtaining crack data on the surface of the crystal image, and importing the crack data on the surface of the crystal image into an image crack change abnormal value calculation formula to calculate the image crack change abnormal value, wherein the image crack change abnormal value calculation formula is: , where m is the number of cracks on the crystal image at the end of the set time period, kj is the length of the jth crack on the crystal image at the end of the set time period, zj is the maximum width of the jth crack on the crystal image at the end of the set time period, M is the number of cracks on the crystal image at the beginning of the set time period, lc is the length of the cth crack on the crystal image at the beginning of the set time period, fc is the width of the cth crack on the crystal image at the beginning of the set time period, and exp() is the power of the natural constant e.

[0085] In this embodiment, the process of importing the image abnormality analysis change results within a set period into the cooling gradient abnormality analysis model to perform cooling gradient abnormality analysis includes the following specific steps:

[0086] S301, obtaining the calculated image change abnormal value and image crack change abnormal value, and substituting the calculated image change abnormal value and image crack change abnormal value into the image abnormality analysis change abnormality coefficient calculation formula to calculate the image abnormality analysis change abnormality coefficient, wherein the image abnormality analysis change abnormality coefficient calculation formula is: ,in, is the proportion of abnormal values ​​in image changes;

[0087] S302. Compare the image abnormality analysis change abnormality coefficient with the set image abnormality analysis change abnormality threshold. If the image abnormality analysis change abnormality coefficient is greater than or equal to the set image abnormality analysis change abnormality threshold, it is judged that the cooling gradient is abnormal and the cooling gradient needs to be adjusted. If the image abnormality analysis change abnormality coefficient is less than the set image abnormality analysis change abnormality threshold, it is judged that the cooling gradient is normal and no cooling gradient adjustment is needed.

[0088] In this embodiment, the actual cooling gradient is obtained by importing the obtained cooling gradient abnormal analysis results into the actual cooling gradient analysis model: the calculated image abnormal analysis change abnormal coefficient, image abnormal analysis change abnormal threshold and cooling gradient are obtained and imported into the actual cooling gradient calculation formula to calculate the actual cooling gradient, wherein the actual cooling gradient calculation formula is: , where Tm is the cooling gradient, ln() is the logarithm with the natural constant e as the base, and K is the abnormal threshold of image abnormality analysis change;

[0089] The calculated actual cooling gradient is obtained and compared with the set safe cooling gradient range. If the actual cooling gradient is within the set safe cooling gradient range, the cooling gradient is adjusted to the actual cooling gradient. If the actual cooling gradient is not within the set safe cooling gradient range, the safe cooling gradient with the smallest difference from the actual cooling gradient within the safe cooling gradient range is selected and set as the second actual cooling gradient. The cooling gradient is adjusted to the second actual cooling gradient. It should be noted that the safe cooling gradient range here is the safe cooling gradient range set when the crystal growth equipment leaves the factory.

[0090] It should be noted here that the first proportion coefficient, the proportion of image change abnormal values ​​and the abnormal threshold value of image abnormal analysis change are obtained according to the experience of technical personnel in this field, and the preferred value-taking method is: 500 groups of crystal images obtained during the cooling process and the temperature change gradient data of the crystal during the cooling process are obtained at the same time, and the quality of the crystals after production is manually sorted according to good or bad, and the crystal images during the cooling process and the temperature change gradient data of the crystal during the cooling process are imported into the actual cooling gradient calculation formula to calculate the actual cooling gradient, and the actual cooling gradient and the sorting results are imported into the fitting software, and the values ​​of the first proportion coefficient, the proportion of image change abnormal values ​​and the abnormal threshold value of image abnormal analysis change that meet the highest sorting judgment accuracy are output;

[0091] It should be noted in the present embodiment that the advantages of the present embodiment over the prior art are: real-time acquisition of crystal images during the cooling process, acquisition of the temperature change gradient of the crystal during the cooling process, acquisition of crystal images within a set time period, importing the acquired crystal images into an image abnormality analysis model for image abnormality analysis, importing the image abnormality analysis change results within a set period into a cooling gradient abnormality analysis model for cooling gradient abnormality analysis, importing the obtained cooling gradient abnormality analysis results into an actual cooling gradient analysis model to obtain an actual cooling gradient, adjusting the cooling gradient during the cooling process according to the obtained actual cooling gradient, performing abnormal evaluation of the crystal images during the cooling process, thereby analyzing the changes in crystal quality with temperature during the cooling process, and accurately calculating and analyzing the required cooling gradient by comprehensively considering the crystal quality during the cooling process and the cooling safety, thereby improving the crystal quality during the cooling process while ensuring the safety during the cooling process.

[0092] Example 2

[0093] like Figure 3 and Figure 4As shown, a crystal cooling process monitoring system based on multi-source data fusion is implemented based on the above-mentioned crystal cooling process monitoring method based on multi-source data fusion, which specifically includes a data acquisition module, an image abnormality analysis module, a cooling gradient abnormality analysis module, an actual cooling gradient acquisition module, a cooling gradient adjustment module and a control module, wherein the data acquisition module is used to acquire the crystal image in the cooling process in real time, and simultaneously acquire the temperature change gradient in the cooling process of the crystal;

[0094] An image anomaly analysis module is used to obtain crystal images within a set time period and import the obtained crystal images into an image anomaly analysis model for image anomaly analysis;

[0095] A cooling gradient anomaly analysis module is used to import the image anomaly analysis change results within a set period into the cooling gradient anomaly analysis model to perform cooling gradient anomaly analysis;

[0096] An actual cooling gradient acquisition module is used to import the obtained cooling gradient anomaly analysis results into an actual cooling gradient analysis model to obtain the actual cooling gradient;

[0097] A temperature-lowering gradient adjustment module is used to adjust the temperature-lowering gradient during the temperature-lowering process according to the actual temperature-lowering gradient obtained;

[0098] The control module is used for controlling the operation of the data acquisition module, the image anomaly analysis module, the cooling gradient anomaly analysis module, the actual cooling gradient acquisition module and the cooling gradient adjustment module, wherein the data acquisition module is integrated in the acquisition terminal, the image anomaly analysis module, the cooling gradient anomaly analysis module and the actual cooling gradient acquisition module are integrated in the data analysis terminal, and the cooling gradient adjustment module and the control module are integrated in the control terminal.

[0099] Example 3

[0100] This embodiment provides an electronic device, including: a processor and a memory, wherein the memory stores a computer program that can be called by the processor;

[0101] The processor executes the above-mentioned crystal cooling process monitoring method based on multi-source data fusion by calling the computer program stored in the memory.

[0102] The electronic device may have relatively large differences due to different configurations or performances, and may include one or more processors and one or more memories, wherein the memory stores at least one computer program, which is loaded and executed by the processor to implement the crystal cooling process monitoring method based on multi-source data fusion provided by the above method embodiment. The electronic device may also include other components for implementing the functions of the device, for example, the electronic device may also have components such as a wired or wireless network interface and an input / output interface to input and output data. This embodiment will not be described in detail here.

[0103] Example 4

[0104] This embodiment provides a computer-readable storage medium having a rewritable computer program stored thereon;

[0105] When the computer program runs on a computer device, the computer device executes the above-mentioned crystal cooling process monitoring method based on multi-source data fusion.

[0106] For example, the computer readable storage medium can be a read-only memory, a random access memory, a read-only CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0107] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through a wired network or / and a wireless network. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

Claims

1. A crystal cooling process monitoring method based on multi-source data fusion, characterized in that: It includes the following specific steps: Acquire the crystal image in the cooling process in real time, and simultaneously acquire the temperature gradient of the crystal in the cooling process; Acquire crystal images within a set time period, and import the acquired crystal images into an image anomaly analysis model for image anomaly analysis; Importing the image abnormality analysis change results within a set period into the cooling gradient abnormality analysis model to perform cooling gradient abnormality analysis; According to the obtained abnormal analysis results of the cooling gradient, the actual cooling gradient is imported into the actual cooling gradient analysis model to obtain the actual cooling gradient; The step of adjusting the temperature-lowering gradient during the temperature-lowering process according to the actual temperature-lowering gradient obtained; acquiring the crystal image within the set time period, and importing the acquired crystal image into the image anomaly analysis model for image anomaly analysis includes the following specific contents: Obtain all crystal images within the set time period from the current moment, obtain the crystal image of the time node at the end of the set time period, import the crystal image of the time node at the end of the set time period into the image change abnormal value calculation formula to calculate the image change abnormal value, wherein the image change abnormal value calculation formula is: , where n is the number of crystal image pixels, xiw is the pixel value of the i-th pixel of the crystal image at the end of the set time period, xiwm is the standard pixel value of the i-th pixel of the crystal image at the end of the set time period, xiq is the pixel value of the i-th pixel of the crystal image at the start of the set time period, xiqm is the standard pixel value of the i-th pixel of the crystal image at the start of the set time period, T is the length of the set time period, xit is the pixel value of the i-th pixel of the crystal image at the time t of the set time period, dt is the time integral, xitm is the standard pixel value of the i-th pixel of the crystal image at the time t of the set time period, and a is the first accounting coefficient; the obtaining of the crystal image within the set time period and importing the obtained crystal image into the image anomaly analysis model for image anomaly analysis also includes the following specific contents: The crystal image at the time node at the end of the set time period is obtained, the crack data on the surface of the crystal image is obtained, and the crack data on the surface of the crystal image is imported into the image crack change abnormal value calculation formula to calculate the image crack change abnormal value, wherein the image crack change abnormal value calculation formula is: , where m is the number of cracks on the crystal image at the end of the set time period, kj is the length of the jth crack on the crystal image at the end of the set time period, zj is the maximum width of the jth crack on the crystal image at the end of the set time period, M is the number of cracks on the crystal image at the start of the set time period, lc is the length of the cth crack on the crystal image at the start of the set time period, fc is the width of the cth crack on the crystal image at the start of the set time period, and exp() is the power of the natural constant e; the step of importing the image anomaly analysis change results within the set period into the cooling gradient anomaly analysis model for cooling gradient anomaly analysis includes the following specific steps: Obtain the calculated image change abnormal value and image crack change abnormal value, substitute the calculated image change abnormal value and image crack change abnormal value into the image abnormality analysis change abnormality coefficient calculation formula to calculate the image abnormality analysis change abnormality coefficient, wherein the image abnormality analysis change abnormality coefficient calculation formula is: ,in, is the proportion of abnormal values ​​in image changes; The image abnormality analysis change abnormality coefficient is compared with the set image abnormality analysis change abnormality threshold. If the image abnormality analysis change abnormality coefficient is greater than or equal to the set image abnormality analysis change abnormality threshold, the cooling gradient is judged to be abnormal and the cooling gradient adjustment is required. If the image abnormality analysis change abnormality coefficient is less than the set image abnormality analysis change abnormality threshold, the cooling gradient is judged to be normal and no cooling gradient adjustment is required.

2. The crystal cooling process monitoring method based on multi-source data fusion according to claim 1, characterized in that: The actual cooling gradient is obtained by importing the obtained cooling gradient anomaly analysis result into the actual cooling gradient analysis model: the calculated image anomaly analysis change anomaly coefficient, image anomaly analysis change anomaly threshold and cooling gradient are obtained and imported into the actual cooling gradient calculation formula to calculate the actual cooling gradient, wherein the actual cooling gradient calculation formula is: , where Tm is the cooling gradient, ln() is the logarithm with the natural constant e as the base, and K is the abnormal threshold of image abnormality analysis change; The calculated actual cooling gradient is obtained and compared with the set safe cooling gradient range. If the actual cooling gradient is within the set safe cooling gradient range, the cooling gradient is adjusted to the actual cooling gradient. If the actual cooling gradient is not within the set safe cooling gradient range, the safe cooling gradient with the smallest difference from the actual cooling gradient within the safe cooling gradient range is selected and set as the second actual cooling gradient, and the cooling gradient is adjusted to the second actual cooling gradient.

3. The crystal cooling process monitoring method based on multi-source data fusion according to claim 2, characterized in that: The specific steps of acquiring the crystal image in the cooling process in real time and acquiring the temperature gradient of the crystal in the cooling process are: Acquire the crystal image during the cooling process of the set temperature change gradient, and simultaneously acquire the set real-time temperature change gradient; The acquired crystal image is subjected to a clarity preprocessing to obtain a clear crystal image, and the crack data on the surface of the crystal image and the pixel data of each pixel point are obtained, wherein the surface crack data includes the surface crack length, maximum width and position data. At the same time, the crystal safety pixel range data is obtained and stored in the storage component.

4. A crystal cooling process monitoring system based on multi-source data fusion, which is implemented based on the crystal cooling process monitoring method based on multi-source data fusion according to any one of claims 1 to 3, characterized in that: It specifically includes a data acquisition module, an image anomaly analysis module, a cooling gradient anomaly analysis module, an actual cooling gradient acquisition module, a cooling gradient adjustment module and a control module, wherein the data acquisition module is used to acquire the crystal image in the cooling process in real time and simultaneously acquire the temperature change gradient of the crystal in the cooling process; The image anomaly analysis module is used to obtain crystal images within a set time period, and import the obtained crystal images into the image anomaly analysis model to perform image anomaly analysis; The cooling gradient anomaly analysis module is used to import the image anomaly analysis change results within a set period into the cooling gradient anomaly analysis model to perform cooling gradient anomaly analysis; The actual cooling gradient acquisition module is used to import the obtained cooling gradient anomaly analysis results into the actual cooling gradient analysis model to obtain the actual cooling gradient; The temperature-lowering gradient adjustment module is used to adjust the temperature-lowering gradient during the temperature-lowering process according to the actual temperature-lowering gradient obtained; The control module is used for controlling the operation of the data acquisition module, the image anomaly analysis module, the cooling gradient anomaly analysis module, the actual cooling gradient acquisition module and the cooling gradient adjustment module.

5. An electronic device comprising: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; It is characterized in that the processor executes the crystal cooling process monitoring method based on multi-source data fusion as described in any one of claims 1 to 3 by calling the computer program stored in the memory.

6. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are executed on a computer, the computer executes the crystal cooling process monitoring method based on multi-source data fusion as described in any one of claims 1 to 3.

Images