A method for calibrating CCD temperature measurement in a crystal pulling process

Abstract

The application belongs to the technical field of crystal pulling. The application discloses a CCD temperature measurement correction method in a crystal pulling process, and the steps are as follows: a database of a seed crystal feature point standard image at a seed crystal pulling moment and a standard liquid temperature is established; the seed crystal feature point is an edge of the seed crystal appearing in the image; a deep learning model is established based on the database; a seed crystal edge image and a liquid surface temperature are collected through a visual system of a CCD; feature point judgment is performed on the seed crystal edge image based on the deep learning model; if the seed crystal pulling condition is met, it is determined whether the measured liquid surface temperature is equal to the standard liquid surface temperature; if not, the CCD is corrected according to the measured liquid surface temperature, so that the measured liquid surface temperature is close to or equal to the standard liquid surface temperature. The method automatically judges whether the CCD temperature measurement correction is needed through the seed crystal edge image, which can avoid the subjective factors, experience and other problems of manual judgment, and the situation that the correction opportunity is missed due to the judgment error.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of crystal pulling, and particularly relates to a temperature correction method in a crystal pulling process. BACKGROUND

[0002] In the process of pulling single crystal by a single crystal furnace, the power value of a heater is often controlled by controlling the temperature of a thermal field to ensure that the crystal growth temperature in the single crystal furnace is in dynamic balance. At present, most of the silicon crystal manufacturers use a CCD to measure the liquid surface temperature, and determine the crystal pulling parameters according to the liquid surface temperature detected by the CCD to ensure the crystal growth rate of the crystal pulling. However, due to the measurement error and other factors, it is difficult to ensure the consistency of the liquid surface temperature detected by the CCD and the actual liquid surface temperature, resulting in the inconsistency between the actual liquid surface temperature and the temperature measured by the CCD, and therefore the temperature measurement correction of the CCD is needed.

[0003] At present, whether the correction is needed is determined by the experience of the staff according to the crystal pulling situation.

[0004] However, the manual determination has subjective factors and insufficient experience, which may lead to incorrect judgment, inaccurate correction timing, correction errors and other problems, and finally results in waste of working hours and manpower, and affects the crystal growth rate of the single crystal pulling. SUMMARY

[0005] In view of the above problems that the existing manual judgment of whether the temperature measurement correction of the CCD is needed has the problems of inaccurate correction timing and correction errors, and finally results in waste of working hours and manpower and affects the crystal growth rate of the single crystal pulling, one of the purposes of the present application is to provide a CCD temperature measurement correction method in a crystal pulling process, which can automatically determine whether the correction is needed, thereby improving the efficiency and accuracy of the correction.

[0006] To achieve the above purpose, the present application adopts the following technical scheme:

[0007] A CCD liquid temperature correction method in a crystal pulling process, comprising the following steps:

[0008] S1, establishing a database of standard images of seed crystal feature points at the moment of crystal pulling and standard liquid temperatures; the seed crystal feature points are the corners of the seed crystal appearing in the images;

[0009] S2, establishing a deep learning model based on the database of step S1;

[0010] S3, collecting the corner images of the seed crystal after temperature adjustment and the liquid surface temperature by the visual system of the CCD;

[0011] S4, judging the feature points of the corner images of the seed crystal of step S3 based on the deep learning model of step S2;

[0012] If the seeding condition is met, it is determined whether the measured liquid surface temperature is equal to the standard liquid surface temperature;

[0013] If not, the CCD is temperature-corrected according to the measured liquid surface temperature so that the measured liquid surface temperature of the CCD is close to or equal to the standard liquid surface temperature;

[0014] The step S3 comprises:

[0015] When the seed crystal off-angle has four points, the seed crystal off-angle image of the seed crystal rotating one round is collected by the visual system of the CCD; the collection of the seed crystal off-angle image of the seed crystal rotating one round comprises collection of 50 seed crystal off-angle images; the seeding condition is that the qualified rate of the seed crystal off-angle image of the seed crystal rotating one round is greater than or equal to 90%.

[0016] Preferably, the step S4 further comprises:

[0017] After the temperature correction of the CCD according to the measured liquid surface temperature so that the measured liquid surface temperature of the CCD is close to or equal to the standard liquid surface temperature, a temperature correction alarm of the CCD is given.

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

[0019] S5, obtaining the liquid surface temperature at the beginning and the end of the seeding, and determining whether the temperature difference between the liquid surface temperature at the beginning of the seeding and the liquid surface temperature at the end of the seeding meets a preset condition;

[0020] If not, an alarm is given.

[0021] Preferably, the preset condition is that the temperature difference is less than or equal to 1℃.

[0022] Preferably, the temperature correction of the CCD is performed before the seeding starts or after the seeding is performed for 100mm.

[0023] From the above description, it can be seen that the application has the following beneficial effects:

[0024] According to the seeding feature point image, it is determined whether the temperature correction of the CCD is needed, so as to reduce the participation of manual work, avoid the subjective factors and experience of manual determination, and avoid the situation that the correction opportunity is missed due to the determination error; at the same time, the temperature correction of the CCD is performed according to the measured liquid surface temperature and the actual liquid surface temperature, which is beneficial to improve the accuracy of the correction; finally, a self-checking processing step is provided, so as to improve the accuracy of the correction, which is beneficial to improve the work efficiency and the production efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the related art description. Obviously, the drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative effort based on these drawings.

[0026] Figure 1 The logical flowchart of the present application. DETAILED DESCRIPTION

[0027] In the following, only some exemplary embodiments are simply described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present application. Therefore, the drawings and the description are considered to be exemplary in nature rather than limiting.

[0028] The terms "comprising" and "having" and any variations thereof in the present application are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but can optionally further include steps or units not listed, or can optionally further include other steps or units inherent to these processes, methods, products or devices.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terminology used in the description of the present application only for the purpose of describing specific embodiments and is not intended to limit the present application. The term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0030] In the present application, referring to "embodiments" means that the specific features, structures or properties described in connection with the embodiments can be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily mean that it refers to the same embodiment, nor is it independent or alternative to other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0031] The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0032] The embodiments of the present application disclose a CCD temperature measurement correction method in a crystal pulling process, specifically comprising the following steps:

[0033] S1, a database of a seed crystal feature point standard image at a seed crystal pulling moment and a standard liquid surface temperature is established; the seed crystal feature point is an edge of the seed crystal appearing in the image at the seed crystal pulling moment.

[0034] Specifically, a large number of seed crystal off-angle images at the seeding moment are collected by the visual system of the CCD to construct a database of seed crystal off-angle images and liquid surface temperature.

[0035] S2, a deep learning model is established based on the database of step S1.

[0036] More specifically, the deep learning model of step S2 includes an image data processing process and a temperature data processing process. It can be understood that the deep learning model also needs to be simulated and trained and run to continuously optimize.

[0037] S3, the seed crystal off-angle image and the liquid surface temperature after temperature adjustment are collected by the CCD.

[0038] Specifically, when there are four points of seed crystal off-angle, the seed crystal off-angle image of one rotation of the seed crystal is collected by the visual system of the CCD. In this embodiment, a total of 50 adjacent seed crystal off-angle images are collected, that is, the CCD visual system collects a seed crystal off-angle image every 7.2 degrees of rotation of the seed crystal.

[0039] S4, the seed crystal off-angle image of step S3 is logically judged based on the deep learning model of step S2.

[0040] If the seeding condition is met, it is judged whether the liquid surface temperature measured by the CCD is equal to the standard liquid surface temperature;

[0041] If not, the CCD is temperature-corrected according to the measured liquid surface temperature to make the measured liquid surface temperature close to or equal to the standard liquid surface temperature.

[0042] Specifically, the seeding condition is that the CCD visual system collects the seed crystal off-angle image, and the qualified rate of the 50 seed crystal off-angle images is ≥90%, and the liquid surface temperature decreases by <0.3 in 5 minutes; when the above conditions are met, the seeding operation is performed.

[0043] More specifically, after the CCD temperature correction of step S4, the CCD temperature correction alarm is triggered.

[0044] In some embodiments, the following steps are further included:

[0045] S5, the liquid surface temperature at the beginning of seeding and the end of seeding is obtained, and it is judged whether the temperature difference between the liquid surface temperature at the beginning of seeding and the liquid surface temperature at the end of seeding meets a preset condition, so as to perform self-checking processing on the correction result; if not, an alarm is prompted to remind the staff to adjust the seeding operation, so as to improve the work efficiency and production efficiency.

[0046] Specifically, the preset condition is that the temperature difference is ≤1℃.

[0047] In some embodiments, the CCD temperature correction is performed before the crystal pulling starts or 100 mm after the crystal pulling starts.

[0048] The working process of the application is described below with a specific embodiment.

[0049] First, the liquid surface temperature value at the crystal pulling time in the crystal pulling system is set to 1450±0.5℃, so that the corresponding crystal pulling feature point image can be determined through the database;

[0050] When the image collected by the visual system of the CCD has an out-corner point, the crystal pulling feature point judgment is entered; 50 seed crystal out-corner images and liquid surface temperatures are collected by the visual system of the CCD in one rotation of the seed crystal;

[0051] The 50 seed crystal out-corner images are identified and logically judged by the deep learning model, and when at least 45 seed crystal out-corner images are qualified compared with the crystal pulling feature point image, that is, the actual liquid surface temperature has approached or equal to the set liquid surface temperature;

[0052] It is judged whether the liquid surface temperature measured by the CCD is equal to or close to the set liquid surface temperature value 1450±0.5℃; if not, the temperature correction is performed according to the liquid surface temperature measured by the visual system of the CCD at this time, so that the liquid surface temperature measured by the CCD is close to or equal to the set liquid surface temperature 1450±0.5℃, and the CCD temperature correction alarm is triggered;

[0053] Finally, the self-checking process is performed through the temperature difference of the liquid surface temperature at the start of the crystal pulling and the end of the crystal pulling, and when the temperature difference exceeds the preset condition, the alarm is triggered, the crystal pulling operation is adjusted, and the working efficiency and production efficiency are improved.

[0054] As can be seen from the above description, the beneficial effects of the embodiments of the application are:

[0055] By using the above method, it is judged whether the CCD temperature correction is needed according to the crystal pulling feature point image, so as to reduce the participation of manual work, avoid the situation that the judgment error is caused by subjective factors, experience and other problems in manual judgment, and miss the correction opportunity; at the same time, the CCD is temperature corrected according to the measured liquid surface temperature and the actual liquid surface temperature, which is beneficial to improve the accuracy of correction; finally, the self-checking step is also provided, so as to improve the accuracy of correction, which is beneficial to improve the working efficiency and production efficiency.

[0056] The above specific embodiments further illustrate the purpose, technical solutions and beneficial effects of the application, and it should be understood that the above description is only a specific embodiment of the application and is not used to limit the protection scope of the application, and any modification, equivalent replacement, improvement, etc. within the spirit and principle of the application should be included in the protection scope of the application.

Claims

1. A method of CCD temperature measurement correction in a crystal pulling process, characterized by, The method comprises the following steps: S1, establishing a database of standard images of seeding feature points at seeding time and standard liquid temperature; the seeding feature points are the corners of the seed crystal appearing in the image; S2, establishing a deep learning model based on the database of step S1; S3, collecting the corner image of the seed crystal and the liquid surface temperature after temperature adjustment through the visual system of the CCD; S4, judging the feature points of the corner image of the seed crystal of step S3 based on the deep learning model of step S2; If the seeding condition is met, it is judged whether the measured liquid surface temperature is equal to the standard liquid surface temperature; If not, the CCD is corrected according to the measured liquid surface temperature, so that the liquid surface temperature measured by the CCD is close to or equal to the standard liquid surface temperature; The step S3 comprises: When there are four corners of the seed crystal, the corner image of the seed crystal rotating one round is collected through the visual system of the CCD; the collection of the corner image of the seed crystal rotating one round comprises collecting 50 corner images of the seed crystal; the seeding condition is that the qualified rate of the corner image of the seed crystal rotating one round is greater than or equal to 90%.

2. The method of claim 1, wherein, The step S4 further comprises: After the CCD is corrected according to the measured liquid surface temperature, so that the liquid surface temperature measured by the CCD is close to or equal to the standard liquid surface temperature, the CCD temperature correction alarm is prompted.

3. The method of CCD pyrometry calibration in a crystal pulling process according to claim 1 or 2, characterized in that, The method further comprises the following steps: S5, obtaining the liquid surface temperature at the beginning and the end of seeding, and judging whether the temperature difference between the liquid surface temperature at the beginning of seeding and the liquid surface temperature at the end of seeding meets the preset condition; If not, an alarm is prompted.

4. The method of claim 3, wherein, The preset condition is that the temperature difference is less than or equal to 1℃.

5. The method of claim 1, wherein, The CCD temperature correction is performed before the seeding starts or 100mm after the seeding starts. The method further comprises the following steps: S6, obtaining the liquid surface temperature at the beginning and the end of the seeding, and judging whether the temperature difference between the liquid surface temperature at the beginning of the seeding and the liquid surface temperature at the end of the seeding meets the preset condition; If not, an alarm is prompted. The preset condition is that the temperature difference is less than or equal to 1℃. The CCD temperature correction is performed before the seeding starts or 100mm after the seeding starts.

Images