CN117187942A - Crucible position control method and device in crystal pulling process - Google Patents
Crucible position control method and device in crystal pulling process Download PDFInfo
- Publication number
- CN117187942A CN117187942A CN202311167985.2A CN202311167985A CN117187942A CN 117187942 A CN117187942 A CN 117187942A CN 202311167985 A CN202311167985 A CN 202311167985A CN 117187942 A CN117187942 A CN 117187942A
- Authority
- CN
- China
- Prior art keywords
- image
- information
- guide cylinder
- crucible
- fitting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 94
- 239000013078 crystal Substances 0.000 title claims abstract description 55
- 239000002210 silicon-based material Substances 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 29
- 238000003709 image segmentation Methods 0.000 claims abstract description 28
- 238000002844 melting Methods 0.000 claims abstract description 20
- 230000008018 melting Effects 0.000 claims abstract description 20
- 238000012549 training Methods 0.000 claims description 43
- 238000002372 labelling Methods 0.000 claims description 32
- 239000007788 liquid Substances 0.000 claims description 25
- XUIMIQQOPSSXEZ-UHFFFAOYSA-N Silicon Chemical compound [Si] XUIMIQQOPSSXEZ-UHFFFAOYSA-N 0.000 claims description 12
- 229910052710 silicon Inorganic materials 0.000 claims description 12
- 239000010703 silicon Substances 0.000 claims description 12
- 238000003780 insertion Methods 0.000 claims description 5
- 230000037431 insertion Effects 0.000 claims description 5
- 238000003466 welding Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 9
- 239000012768 molten material Substances 0.000 description 7
- 230000011218 segmentation Effects 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 239000000155 melt Substances 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000004590 computer program Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 239000003779 heat-resistant material Substances 0.000 description 1
- 239000011261 inert gas Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000004801 process automation Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000010076 replication Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Landscapes
- Crystals, And After-Treatments Of Crystals (AREA)
Abstract
The application discloses a crucible position control method and device in the crystal pulling process, comprising the following steps: obtaining a target image of the total melting of the silicon material in the crystal pulling process; dividing an image area of a target image based on an image processing model to obtain image segmentation information, wherein the image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area; fitting according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of two fitting circles of mirror images and real images; determining the center pixel distance of the two fitting circles based on the center coordinates and the radius information of the two fitting circles; and generating control information for adjusting the position of the crucible based on the circle center pixel spacing. According to the application, the center distance is obtained through image recognition and calculation and is used as a reference value of the movable crucible, manual adjustment is not needed, and the adjustment precision and efficiency are improved.
Description
Technical Field
The application relates to the technical field of photovoltaic and semiconductor single crystal pulling, in particular to a crucible position control method and device in the crystal pulling process.
Background
In the field of single crystal growth, the process from the completion of the full melting of the silicon material to the determination of the position of the crucible is mainly performed manually, and due to the subjective factors of operators, different experience of crucible position adjustment, shaking of the melt in the crucible and other problems, the adjustment of the position of the crucible is easy to be inaccurate, and a large amount of operators are required to pay attention to the current position of the crucible at any time, so that the production cost, the yield and the production efficiency are easily affected to a certain extent.
Disclosure of Invention
In view of this, the present application provides the following technical solutions:
a crucible position control method in a crystal pulling process comprises the following steps:
obtaining a target image of the total melting of the silicon material in the crystal pulling process;
dividing an image area of the target image based on an image processing model to obtain image segmentation information, wherein the image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area, and the image processing model is a model which is obtained based on training of an overall process image from after full melting of a silicon material to before seed crystal insertion and can detect the guide cylinder real image area and the guide cylinder mirror image area;
fitting according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of two fitting circles of mirror images and real images;
determining the center pixel distance of the two fitting circles based on the center coordinates and the radius information of the two fitting circles;
and generating control information for adjusting the position of the crucible based on the circle center pixel spacing.
Optionally, the method further comprises:
obtaining an image data set, wherein the image data set comprises a plurality of images obtained by acquiring an overall process image from the end of an industrial camera after the silicon material is fully melted to the time before seed crystal is inserted;
performing feature labeling on each image in the image data set to obtain labeling information, wherein the labeling information comprises first labeling information representing the mirror image edge of the guide cylinder in the reflection of the surface of the silicon molten liquid and second labeling information representing the real image edge of the guide cylinder;
generating an image training sample based on the labeling information of each image in the image dataset and the image dataset;
and training the image training sample by taking the image characteristics of each image in the image dataset as training characteristics and the labeling information of each image as a training target to obtain an image processing model.
Optionally, the fitting to obtain the center coordinates and radius information of two fitting circles of the mirror image and the real image according to the pixel information of the real image area of the guide cylinder and the pixel information of the mirror image area of the guide cylinder respectively includes:
performing least square fitting on edge pixel points of the guide cylinder real image area according to the pixel information of the guide cylinder real image area to obtain circle center coordinates and radius information of a real image fitting circle;
and carrying out least square fitting on edge pixel points of the guide cylinder mirror image area according to the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of a mirror image fitting circle.
Optionally, the generating control information for adjusting the crucible position based on the center pixel spacing includes:
obtaining a reference circle center pixel space;
comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different;
based on the control pattern, control information is generated that adjusts the crucible position.
Optionally, the comparing the center pixel pitch obtained in real time with the reference center pixel pitch, and determining a control mode for adjusting the position of the crucible based on the comparison result includes:
calculating a difference value between the circle center pixel spacing obtained in real time and the reference circle center pixel spacing;
if the difference value is larger than a first threshold value, determining a control mode for adjusting the position of the crucible as a first mode;
if the difference is not greater than the first threshold and is greater than a second threshold, determining a control mode for adjusting the position of the crucible to be a second mode;
if the difference value is not greater than the second threshold value, determining a control mode for adjusting the position of the crucible as a third mode;
the first mode adjusts the crucible lifting speed at a first speed, the second mode adjusts the crucible lifting speed at a second speed, the third mode adjusts the crucible lifting speed at a third speed, the first speed is greater than the second speed, and the second speed is greater than the third speed.
Optionally, the method further comprises:
based on the image segmentation information, a comparison result of the inner circle space and the outer circle space is obtained in real time, and the crucible position is adjusted based on the comparison result;
when the welding process is performed, the crucible position is adjusted by using the comparison result of the absolute liquid port distance and the reference liquid port distance.
A crucible position control apparatus during a crystal pulling process, comprising:
the image acquisition unit is used for acquiring a target image of the total melting of the silicon material in the crystal pulling process;
the image processing unit is used for dividing the image area of the target image based on an image processing model to obtain image segmentation information, wherein the image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area, and the image processing model is a model which is obtained based on training of an overall process image from the time of fully melting a silicon material to the time of inserting a seed crystal and can detect the guide cylinder real image area and the guide cylinder mirror image area;
the fitting unit is used for respectively obtaining circle center coordinates and radius information of two fitting circles of the mirror image and the real image according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area in a fitting mode;
the determining unit is used for determining the circle center pixel distance of the two fitting circles based on the circle center coordinates and the radius information of the two fitting circles;
and the generating unit is used for generating control information for adjusting the position of the crucible based on the circle center pixel spacing.
Optionally, the method further comprises:
the image data set comprises a plurality of images acquired by acquiring the whole process image from the time when the silicon material is fully melted to the time when the seed crystal is inserted by the industrial camera end;
the marking unit is used for marking the characteristics of each image in the image data set to obtain marking information, and the marking information comprises first marking information representing the mirror image edge of the guide cylinder in the reflection of the surface of the silicon molten liquid and second marking information representing the real image edge of the guide cylinder;
the sample generation unit is used for generating an image training sample based on the labeling information of each image in the image data set and the image data set;
the training unit is used for taking the image characteristics of each image in the image dataset as training characteristics, taking the labeling information of each image as a training target, and training the image training sample to obtain an image processing model.
Optionally, the fitting unit is specifically configured to:
performing least square fitting on edge pixel points of the guide cylinder real image area according to the pixel information of the guide cylinder real image area to obtain circle center coordinates and radius information of a real image fitting circle;
and carrying out least square fitting on edge pixel points of the guide cylinder mirror image area according to the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of a mirror image fitting circle.
Optionally, the generating unit includes:
an acquisition subunit, configured to obtain a reference circle center pixel pitch;
the comparison subunit is used for comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different;
and the generation subunit is used for generating control information for adjusting the position of the crucible based on the control mode.
Optionally, the comparing subunit is specifically configured to:
calculating a difference value between the circle center pixel spacing obtained in real time and the reference circle center pixel spacing;
if the difference value is larger than a first threshold value, determining a control mode for adjusting the position of the crucible as a first mode;
if the difference is not greater than the first threshold and is greater than a second threshold, determining a control mode for adjusting the position of the crucible to be a second mode;
if the difference value is not greater than the second threshold value, determining a control mode for adjusting the position of the crucible as a third mode;
the first mode adjusts the crucible lifting speed at a first speed, the second mode adjusts the crucible lifting speed at a second speed, the third mode adjusts the crucible lifting speed at a third speed, the first speed is greater than the second speed, and the second speed is greater than the third speed.
According to the technical scheme, the application discloses a crucible position control method and device in the crystal pulling process, comprising the following steps: obtaining a target image of the total melting of the silicon material in the crystal pulling process; dividing an image area of a target image based on an image processing model to obtain image segmentation information, wherein the image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area; fitting according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of two fitting circles of mirror images and real images; determining the center pixel distance of the two fitting circles based on the center coordinates and the radius information of the two fitting circles; and generating control information for adjusting the position of the crucible based on the circle center pixel spacing. According to the application, the center distance is obtained through image recognition and calculation and is used as a reference value of the movable crucible, manual adjustment is not needed, and the adjustment precision and efficiency are improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only embodiments of the present application, and other drawings may be obtained according to the provided drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of a method for controlling crucible position in a crystal pulling process according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a real-time image acquired according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a marked area of a guide cylinder mirror image according to an embodiment of the present application;
fig. 4 is a schematic diagram of a real image labeling area of a guide cylinder according to an embodiment of the present application;
FIG. 5 is a schematic diagram of a segmented image effect diagram according to an embodiment of the present application;
FIG. 6 is a schematic diagram of a crucible position control apparatus in a crystal pulling process according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
The terms first and second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to the listed steps or elements but may include steps or elements not expressly listed.
In order to facilitate the explanation of the control method for fusion of the seed crystal in the crystal pulling process provided by the embodiment of the application, the related terms will be explained.
Pulling: refers to the previous working procedure in the photovoltaic field, and the silicon material is produced into a silicon rod in a single crystal furnace by a Czochralski method.
A guide shell: and the cylinder body with gradually reduced diameter from top to bottom in the single crystal furnace is used for guiding and adjusting inert gas in the cylinder body.
Crucible: vessels made of heat resistant material for holding silicon materials in the photovoltaic manufacturing field.
And (3) fully melting the silicon material: refers to a time node when the silicon material is heated and melted in the crucible until the molten material in the crucible is liquid.
Seed crystal: the single crystal with a certain crystal orientation is used as a replication sample, so that the pulled silicon ingot and the seed crystal have the same crystal orientation.
CCD camera: an industrial camera carrying charge coupled device.
Inserting seed crystals: refers to the process in which the seed crystal is ready to be inserted into the melt to start falling.
And (3) protecting equipment: protection of the relevant equipment during automatic control determination of the crucible position is meant, in this patent, in particular, to prevent the crucible from excessively rising to contact the guide cylinder and the heat shield, and avoid the danger of melting of the guide cylinder and the heat shield.
Absolute liquid port distance: with conventional image processing, the pixel spacing from the edge of the real image of the guide cylinder to the mirrored edge of the guide cylinder within the ROI (Region OfInterest ).
The embodiment of the application provides a crucible position control method in a crystal pulling process, which can solve the problem of insufficient automation for controlling a crucible to an ideal position after a silicon material is fully melted, and realize the whole process automation from the full melting of the silicon material to the insertion of a seed crystal to realize the adjustment of the crucible to the ideal position.
Referring to fig. 1, a schematic flow chart of a crucible position control method in a crystal pulling process according to an embodiment of the present application may include the following steps:
s101, obtaining a target image of the total melting of the silicon material in the crystal pulling process.
The image of the full melting of the silicon material in the crystal pulling process is acquired in real time through the industrial camera end, the acquired image is determined to be a target image, or an image acquisition period is set, and the image is acquired at intervals.
S102, dividing the image area of the target image based on the image processing model to obtain image segmentation information.
The image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area, and the image processing model is based on training of an overall process image from after full melting of a silicon material to before seed crystal insertion, and can detect the guide cylinder real image area and the guide cylinder mirror image area.
Correspondingly, the embodiment of the application also provides a model generation method, which comprises the following steps:
obtaining an image data set, wherein the image data set comprises a plurality of images obtained by acquiring an overall process image from the end of an industrial camera after the silicon material is fully melted to the time before seed crystal is inserted; performing feature labeling on each image in the image data set to obtain labeling information, wherein the labeling information comprises first labeling information representing the mirror image edge of the guide cylinder in the reflection of the surface of the silicon molten liquid and second labeling information representing the real image edge of the guide cylinder; generating an image training sample based on the labeling information of each image in the image dataset and the image dataset; and training the image training sample by taking the image characteristics of each image in the image dataset as training characteristics and the labeling information of each image as a training target to obtain an image processing model.
Specifically, in the process of obtaining an image training sample, the whole process image from the full melting of the silicon material to the time before the seed crystal is inserted down (for example, the whole process image is collected at intervals of s seconds) can be collected through an industrial camera end, and n whole process images of the real image of the guide cylinder, the mirror image of the liquid level guide cylinder in the crucible and the crucible lifting and lowering are collected together. In order to accurately train the image processing model, the training target is required to be labeled with information, namely, the image data is divided into areas, different types of labeling extraction work are carried out on the guide cylinder real images and the mirror image boundary features in each state, and then the extracted features are subjected to model training to obtain the image processing model. Further, in the process of collecting the images, the whole process image refers to all image states of the process from the process that the silicon material is fully melted to be in a liquid state in the crucible to the process that the silicon material is inserted under the seed crystal, wherein the image comprises an image when the crucible is not adjusted, and an image of the crucible position adjustment moment in the crucible position positioning process. The feature labeling is to label the image data, and labeling information is as follows: the guide cylinder is at the mirror image edge (for example, marked as first marking information and can be marked as 0) of the reflection of the silicon melt surface, and the real image edge (for example, marked as second marking information and can be marked as 1) of the guide cylinder, and the marking frame is required to be attached to the mirror image and the real image edge of the guide cylinder.
Referring to fig. 2, fig. 2 shows real-time image information captured by a CCD camera during a crucible, since the heated silicon material has a high temperature, the molten material presents a bright region of the molten material due to the high temperature, so that the guide cylinder suspended above the molten material can reflect the mirror image of the guide cylinder at the molten material level, so that there are bright and dark regions at the molten material level, the bright region being the molten material level of the portion of the guide cylinder where the guide cylinder is not covered by the reflection, and the dim region being the guide cylinder reflection covering region. Secondly, when the training image segmentation model selects the segmentation areas, the pixel value distinction between the segmentation areas is ensured to be obvious, and the labeling range between the segmentation areas is not overlapped, so that a good image segmentation model is obtained.
S103, fitting according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area respectively to obtain circle center coordinates and radius information of two fitting circles of mirror images and real images.
S104, determining the center pixel distance of the two fitting circles based on the center coordinates and the radius information of the two fitting circles.
S105, generating control information for adjusting the position of the crucible based on the circle center pixel spacing.
After the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area are obtained based on the image processing model, circle fitting can be carried out, the circle center distance is calculated, the crucible lifting is automatically controlled according to the circle center distance, and various information such as real-time parameters and the like are utilized as the limit for protecting the crucible position.
In one implementation manner of the embodiment of the present application, the fitting to obtain the center coordinates and radius information of two fitting circles of the mirror image and the real image according to the pixel information of the real image area of the guide cylinder and the pixel information of the mirror image area of the guide cylinder includes: performing least square fitting on edge pixel points of the guide cylinder real image area according to the pixel information of the guide cylinder real image area to obtain circle center coordinates and radius information of a real image fitting circle; and carrying out least square fitting on edge pixel points of the guide cylinder mirror image area according to the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of a mirror image fitting circle.
In an embodiment of the present application, the generating control information for adjusting the position of the crucible based on the center pixel pitch includes: obtaining a reference circle center pixel space; comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different; based on the control pattern, control information is generated that adjusts the crucible position.
Further, the comparing the center pixel pitch obtained in real time with the reference center pixel pitch, and determining a control mode for adjusting the position of the crucible based on the comparison result, includes: calculating a difference value between the circle center pixel spacing obtained in real time and the reference circle center pixel spacing; if the difference value is larger than a first threshold value, determining a control mode for adjusting the position of the crucible as a first mode; if the difference is not greater than the first threshold and is greater than a second threshold, determining a control mode for adjusting the position of the crucible to be a second mode; if the difference value is not greater than the second threshold value, determining a control mode for adjusting the position of the crucible as a third mode; the first mode adjusts the crucible lifting speed at a first speed, the second mode adjusts the crucible lifting speed at a second speed, the third mode adjusts the crucible lifting speed at a third speed, the first speed is greater than the second speed, and the second speed is greater than the third speed.
In the embodiment of the application, the crucible position can be positioned step by step, coarse positioning after melting material is completely melted can be performed in the mode of comparing the circle center distances, then fine positioning is performed after seed crystal is inserted, namely, based on the image segmentation information, the comparison result of the inner circle distance and the outer circle distance is obtained in real time, and the crucible position is adjusted based on the comparison result; when the welding process is performed, the crucible position may be adjusted by using the comparison result between the absolute liquid gap and the reference liquid gap.
Specifically, the circle fitting is that the divided guide cylinder real image and the mirror image edge are fitted into a circle, see fig. 5, which is a divided image effect diagram, so as to find the circle center, thereby calculating the distance between the two circle centers, and because the guide cylinder position is unchanged all the time in the crucible adjustment process, the distance can be used as a basis according to the circle center, for example, the distance between the circle center pixels of the crucible position which is manually set by the first furnace opening can be used as the distance between the target circle centers, so that the two circle centers are continuously compared to achieve the purpose of determining the crucible position.
In the embodiment of the application, a segmentation diagram is output after image segmentation model reasoning, referring to fig. 5, the segmentation diagram is divided into a green region and a red region, the green region and the red region are determined through pixel points, an edge point set is collected, and a circle is fitted on the edge by using a least square method, so that a circle center X, Y coordinate and an r radius are obtained, two fitting circle center coordinates and radius information of a mirror image and a real image can be obtained in the fitting circle mode, and a pixel distance between two circle centers is obtained by using a difference between two circle center Y coordinates.
Then, when the crucible position is controlled based on the circle center pixel spacing, the first threshold value can be set to 15, the second threshold value can be set to 5, the corresponding first speed is 100mm/min, the second speed is 50mm/min, and the third speed is 5mm/min, and specifically, the crucible lifting speed gradient control takes the proximity degree of the real-time circle center pixel spacing and the reference standard circle center pixel spacing as an adjustment basis. For example, when the real-time center pixel pitch-reference standard center pixel pitch is greater than 15, the crucible elevation speed is fast (100 mm/min), when the real-time center pixel pitch-reference standard center pixel pitch is 15 or less and greater than 5, the crucible elevation speed is medium speed (50 mm/min), and when the real-time center pixel pitch-reference standard center pixel pitch is 5 or less, the crucible elevation speed is adjusted to be slow (5 mm/min). The setting of the range between the center pixel interval and the reference center pixel interval and the lifting speed of each crucible can be flexibly set, and the specific numerical values are exemplified and do not represent final setting values. In the crucible position positioning process, the accuracy of the crucible position is improved by adopting a multiple-time crucible position positioning mode, the circle center pixel spacing is adopted for the first coarse positioning according to the set crucible position, the second fine positioning is used for crucible position adjustment by comparing the real-time inner and outer circle spacing with the reference standard inner and outer circle spacing, and the third fine positioning is started when entering the welding process, and the crucible position adjustment is performed by comparing the absolute liquid port distance with the reference standard liquid port distance.
Further, after the image segmentation information is obtained based on the image processing model, a corresponding protection mechanism can be executed based on the image segmentation information, for example, if the process parameters and other modules, such as the set values of the silicon liquid height, the liquid opening distance and the like, are utilized, and if the set values exceed the upper limit, an alarm is given; if the alarm is not triggered and the final adjustment procedure is completed, the crucible position is confirmed to be successful.
In the embodiment of the application, the marked artificial crucible position whole process image is trained through a depth image segmentation network, so that an image segmentation model is obtained and deployed on a site server. In the process of on-site production, when the process step reaches the crucible position, a CCD camera transmits a melt image in a crucible captured in real time to an image segmentation model, an image guide cylinder real image is segmented, the guide cylinder is arranged in a silicon melt mirror image area and is sent back to an upper computer for traditional image algorithm processing and calculation, the upper computer fits the segmented guide cylinder real image and mirror image edge into respective circular areas, the central positions of the circular areas are determined, the pixel difference values of the real image and the mirror image center position are calculated and compared with the pixel difference value of the reference center position, crucible lifting and stopping are judged, and the crucible lifting and stopping are used as limiting protection of the crucible moving distance according to various parameters.
In addition, when the crucible position is adjusted, the crucible lifting speed is adjusted in stages according to the real-time circle center pixel spacing value to the reference circle center pixel spacing value of the target, namely, the closer to the reference circle center pixel spacing value, the slower the crucible lifting speed.
Furthermore, in the embodiment of the application, the crucible position can be positioned for a plurality of times in a short time to achieve the accuracy of adjusting the position of the target crucible, coarse positioning is performed once after the crucible is completely melted, the crucible is adjusted and stopped after the pixel difference value of the target reference circle center is reached, fine positioning is performed once when the seed crystal is inserted, and the position of the crucible can be accurately adjusted to an ideal position after the crucible is adjusted for a short time and a plurality of times. Correspondingly, various parameters such as the silicon liquid level position, the liquid opening distance and the like can be used as protection, and the crucible is stopped to be adjusted as long as a certain parameter reaches an upper limit value.
In another implementation manner of the embodiment of the present application, there is also provided a crucible position control device in a crystal pulling process, referring to fig. 6, including:
an image acquisition unit 201, configured to acquire a target image of the silicon material melted completely during the crystal pulling process;
the image processing unit 202 is configured to perform image region division on the target image based on an image processing model to obtain image segmentation information, where the image segmentation information includes a guide cylinder real image region and a guide cylinder mirror image region, and the image processing model is a model that is obtained based on training an overall process image from after full melting of a silicon material to before seed crystal insertion and is capable of detecting the guide cylinder real image region and the guide cylinder mirror image region;
a fitting unit 203, configured to obtain center coordinates and radius information of two fitting circles of the mirror image and the real image by fitting according to the pixel information of the real image area of the guide cylinder and the pixel information of the mirror image area of the guide cylinder respectively;
a determining unit 204, configured to determine a center pixel distance of the two fitting circles based on the center coordinates and the radius information of the two fitting circles;
and a generating unit 205, configured to generate control information for adjusting the crucible position based on the center pixel pitch.
Optionally, the method further comprises:
the image data set comprises a plurality of images acquired by acquiring the whole process image from the time when the silicon material is fully melted to the time when the seed crystal is inserted by the industrial camera end;
the marking unit is used for marking the characteristics of each image in the image data set to obtain marking information, and the marking information comprises first marking information representing the mirror image edge of the guide cylinder in the reflection of the surface of the silicon molten liquid and second marking information representing the real image edge of the guide cylinder;
the sample generation unit is used for generating an image training sample based on the labeling information of each image in the image data set and the image data set;
the training unit is used for taking the image characteristics of each image in the image dataset as training characteristics, taking the labeling information of each image as a training target, and training the image training sample to obtain an image processing model.
Optionally, the fitting unit is specifically configured to:
performing least square fitting on edge pixel points of the guide cylinder real image area according to the pixel information of the guide cylinder real image area to obtain circle center coordinates and radius information of a real image fitting circle;
and carrying out least square fitting on edge pixel points of the guide cylinder mirror image area according to the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of a mirror image fitting circle.
Optionally, the generating unit includes:
an acquisition subunit, configured to obtain a reference circle center pixel pitch;
the comparison subunit is used for comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different;
and the generation subunit is used for generating control information for adjusting the position of the crucible based on the control mode.
Optionally, the generating unit includes:
an acquisition subunit, configured to obtain a reference circle center pixel pitch;
the comparison subunit is used for comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different;
and the generation subunit is used for generating control information for adjusting the position of the crucible based on the control mode.
Optionally, the comparing subunit is specifically configured to:
calculating a difference value between the circle center pixel spacing obtained in real time and the reference circle center pixel spacing;
if the difference value is larger than a first threshold value, determining a control mode for adjusting the position of the crucible as a first mode;
if the difference is not greater than the first threshold and is greater than a second threshold, determining a control mode for adjusting the position of the crucible to be a second mode;
if the difference value is not greater than the second threshold value, determining a control mode for adjusting the position of the crucible as a third mode;
the first mode adjusts the crucible lifting speed at a first speed, the second mode adjusts the crucible lifting speed at a second speed, the third mode adjusts the crucible lifting speed at a third speed, the first speed is greater than the second speed, and the second speed is greater than the third speed.
Optionally, the apparatus further comprises:
the inner circle and outer circle comparison unit is used for obtaining a comparison result of the inner circle and outer circle distance and the reference inner circle and outer circle distance in real time based on the image segmentation information, and adjusting the position of the crucible based on the comparison result;
and the liquid port distance comparison unit is used for adjusting the position of the crucible by using the comparison result of the absolute liquid port distance and the reference liquid port distance when the welding process is carried out.
It should be noted that, the specific implementation of each unit and sub-unit in this embodiment may refer to the corresponding content in the foregoing, which is not described in detail herein.
In another embodiment of the application, there is also provided a readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of crucible position control in a crystal pulling process as set forth in any one of the preceding claims.
In another embodiment of the present application, there is also provided an electronic device, which may include:
a memory for storing an application program and data generated by the operation of the application program;
a processor for executing the application program to implement the crucible position control method in the crystal pulling process as set forth in any one of the preceding claims.
It should be noted that, the specific implementation of the processor in this embodiment may refer to the corresponding content in the foregoing, which is not described in detail herein.
In the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described in a different point from other embodiments, and identical and similar parts between the embodiments are all enough to refer to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant points refer to the description of the method section.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative elements and steps are described above generally in terms of functionality in order to clearly illustrate the interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. The software modules may be disposed in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for controlling the position of a crucible during a crystal pulling process, comprising:
obtaining a target image of the total melting of the silicon material in the crystal pulling process;
dividing an image area of the target image based on an image processing model to obtain image segmentation information, wherein the image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area, and the image processing model is a model which is obtained based on training of an overall process image from after full melting of a silicon material to before seed crystal insertion and can detect the guide cylinder real image area and the guide cylinder mirror image area;
fitting according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of two fitting circles of mirror images and real images;
determining the center pixel distance of the two fitting circles based on the center coordinates and the radius information of the two fitting circles;
and generating control information for adjusting the position of the crucible based on the circle center pixel spacing.
2. The method as recited in claim 1, further comprising:
obtaining an image data set, wherein the image data set comprises a plurality of images obtained by acquiring an overall process image from the end of an industrial camera after the silicon material is fully melted to the time before seed crystal is inserted;
performing feature labeling on each image in the image data set to obtain labeling information, wherein the labeling information comprises first labeling information representing the mirror image edge of the guide cylinder in the reflection of the surface of the silicon molten liquid and second labeling information representing the real image edge of the guide cylinder;
generating an image training sample based on the labeling information of each image in the image dataset and the image dataset;
and training the image training sample by taking the image characteristics of each image in the image dataset as training characteristics and the labeling information of each image as a training target to obtain an image processing model.
3. The method according to claim 1, wherein the fitting to obtain the center coordinates and radius information of two fitting circles of the mirror image and the real image according to the pixel information of the real image area of the guide cylinder and the pixel information of the mirror image area of the guide cylinder respectively includes:
performing least square fitting on edge pixel points of the guide cylinder real image area according to the pixel information of the guide cylinder real image area to obtain circle center coordinates and radius information of a real image fitting circle;
and carrying out least square fitting on edge pixel points of the guide cylinder mirror image area according to the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of a mirror image fitting circle.
4. The method of claim 1, wherein generating control information for adjusting the crucible position based on the center pixel spacing comprises:
obtaining a reference circle center pixel space;
comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different;
based on the control pattern, control information is generated that adjusts the crucible position.
5. The method of claim 4, wherein comparing the center-pixel spacing obtained in real-time with the reference center-pixel spacing and determining a control pattern for adjusting the crucible position based on the comparison comprises:
calculating a difference value between the circle center pixel spacing obtained in real time and the reference circle center pixel spacing;
if the difference value is larger than a first threshold value, determining a control mode for adjusting the position of the crucible as a first mode;
if the difference is not greater than the first threshold and is greater than a second threshold, determining a control mode for adjusting the position of the crucible to be a second mode;
if the difference value is not greater than the second threshold value, determining a control mode for adjusting the position of the crucible as a third mode;
the first mode adjusts the crucible lifting speed at a first speed, the second mode adjusts the crucible lifting speed at a second speed, the third mode adjusts the crucible lifting speed at a third speed, the first speed is greater than the second speed, and the second speed is greater than the third speed.
6. The method according to claim 4, wherein the method further comprises:
based on the image segmentation information, a comparison result of the inner circle space and the outer circle space is obtained in real time, and the crucible position is adjusted based on the comparison result;
when the welding process is performed, the crucible position is adjusted by using the comparison result of the absolute liquid port distance and the reference liquid port distance.
7. A crucible position control apparatus in a crystal pulling process, comprising:
the image acquisition unit is used for acquiring a target image of the total melting of the silicon material in the crystal pulling process;
the image processing unit is used for dividing the image area of the target image based on an image processing model to obtain image segmentation information, wherein the image segmentation information comprises a guide cylinder real image area and a guide cylinder mirror image area, and the image processing model is a model which is obtained based on training of an overall process image from the time of fully melting a silicon material to the time of inserting a seed crystal and can detect the guide cylinder real image area and the guide cylinder mirror image area;
the fitting unit is used for respectively obtaining circle center coordinates and radius information of two fitting circles of the mirror image and the real image according to the pixel information of the guide cylinder real image area and the pixel information of the guide cylinder mirror image area in a fitting mode;
the determining unit is used for determining the circle center pixel distance of the two fitting circles based on the circle center coordinates and the radius information of the two fitting circles;
and the generating unit is used for generating control information for adjusting the position of the crucible based on the circle center pixel spacing.
8. The apparatus as recited in claim 7, further comprising:
the image data set comprises a plurality of images acquired by acquiring the whole process image from the time when the silicon material is fully melted to the time when the seed crystal is inserted by the industrial camera end;
the marking unit is used for marking the characteristics of each image in the image data set to obtain marking information, and the marking information comprises first marking information representing the mirror image edge of the guide cylinder in the reflection of the surface of the silicon molten liquid and second marking information representing the real image edge of the guide cylinder;
the sample generation unit is used for generating an image training sample based on the labeling information of each image in the image data set and the image data set;
the training unit is used for taking the image characteristics of each image in the image dataset as training characteristics, taking the labeling information of each image as a training target, and training the image training sample to obtain an image processing model.
9. The apparatus according to claim 7, wherein the fitting unit is specifically configured to:
performing least square fitting on edge pixel points of the guide cylinder real image area according to the pixel information of the guide cylinder real image area to obtain circle center coordinates and radius information of a real image fitting circle;
and carrying out least square fitting on edge pixel points of the guide cylinder mirror image area according to the pixel information of the guide cylinder mirror image area to obtain circle center coordinates and radius information of a mirror image fitting circle.
10. The apparatus of claim 7, wherein the generating unit comprises:
an acquisition subunit, configured to obtain a reference circle center pixel pitch;
the comparison subunit is used for comparing the circle center pixel spacing obtained in real time with the reference circle center pixel spacing, and determining a control mode for adjusting the position of the crucible based on a comparison result, wherein crucible adjustment speeds represented by different control modes are different;
and the generation subunit is used for generating control information for adjusting the position of the crucible based on the control mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311167985.2A CN117187942B (en) | 2023-09-11 | 2023-09-11 | Crucible position control method and device in crystal pulling process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311167985.2A CN117187942B (en) | 2023-09-11 | 2023-09-11 | Crucible position control method and device in crystal pulling process |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117187942A true CN117187942A (en) | 2023-12-08 |
| CN117187942B CN117187942B (en) | 2024-03-26 |
Family
ID=88988330
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311167985.2A Active CN117187942B (en) | 2023-09-11 | 2023-09-11 | Crucible position control method and device in crystal pulling process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117187942B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104005083A (en) * | 2014-05-20 | 2014-08-27 | 北京工业大学 | Device and method for measuring liquid level height of molten silicon in single crystal furnace |
| CN109750352A (en) * | 2017-11-07 | 2019-05-14 | 胜高股份有限公司 | The manufacturing method and device of monocrystalline |
| CN112226812A (en) * | 2020-10-20 | 2021-01-15 | 北京图知天下科技有限责任公司 | Czochralski monocrystalline silicon production method, device and system |
| CN112639175A (en) * | 2018-08-23 | 2021-04-09 | 胜高股份有限公司 | Method for growing single crystal silicon |
| CN113607241A (en) * | 2021-08-05 | 2021-11-05 | 包头美科硅能源有限公司 | Method for improving liquid level detection precision and sensitivity |
| CN114926440A (en) * | 2022-05-26 | 2022-08-19 | 之江实验室 | Method and device for detecting diameter of monocrystalline silicon based on concentric ellipse chord length ratio |
| CN114998415A (en) * | 2022-05-23 | 2022-09-02 | 无锡海纳智能科技有限公司 | Method and device for determining diameter of crystal bar pixel |
| CN115265391A (en) * | 2022-09-30 | 2022-11-01 | 杭州利珀科技有限公司 | Silicon material spacing detection method in preparation process of single crystal silicon rod |
-
2023
- 2023-09-11 CN CN202311167985.2A patent/CN117187942B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104005083A (en) * | 2014-05-20 | 2014-08-27 | 北京工业大学 | Device and method for measuring liquid level height of molten silicon in single crystal furnace |
| CN109750352A (en) * | 2017-11-07 | 2019-05-14 | 胜高股份有限公司 | The manufacturing method and device of monocrystalline |
| CN112639175A (en) * | 2018-08-23 | 2021-04-09 | 胜高股份有限公司 | Method for growing single crystal silicon |
| CN112226812A (en) * | 2020-10-20 | 2021-01-15 | 北京图知天下科技有限责任公司 | Czochralski monocrystalline silicon production method, device and system |
| CN113607241A (en) * | 2021-08-05 | 2021-11-05 | 包头美科硅能源有限公司 | Method for improving liquid level detection precision and sensitivity |
| CN114998415A (en) * | 2022-05-23 | 2022-09-02 | 无锡海纳智能科技有限公司 | Method and device for determining diameter of crystal bar pixel |
| CN114926440A (en) * | 2022-05-26 | 2022-08-19 | 之江实验室 | Method and device for detecting diameter of monocrystalline silicon based on concentric ellipse chord length ratio |
| CN115265391A (en) * | 2022-09-30 | 2022-11-01 | 杭州利珀科技有限公司 | Silicon material spacing detection method in preparation process of single crystal silicon rod |
Non-Patent Citations (1)
| Title |
|---|
| 徐亚顺;李轩;王晓杰;莫绪涛;马四海;黄仙山;: "单晶硅熔液面位置检测的图像处理方法研究", 光学技术, no. 05, 15 September 2020 (2020-09-15), pages 52 - 57 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117187942B (en) | 2024-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113215651B (en) | Crystal pulling control method and equipment, single crystal furnace and computer storage medium | |
| CN1202290C (en) | Method and system for controlling growth of a silicon crystal | |
| CN101377008B (en) | Silicon single crystal pulling method | |
| US6171391B1 (en) | Method and system for controlling growth of a silicon crystal | |
| TWI447271B (en) | Method and apparatus for controlling the growth process of a monocrystalline silicon ingot | |
| CN115125609B (en) | Sapphire seeding control process and system based on image processing algorithm | |
| CN102787353B (en) | Method for measuring crystal wire of silicon rod of single crystal furnace in non-contact mode | |
| CN117187942B (en) | Crucible position control method and device in crystal pulling process | |
| CN109829638A (en) | A kind of crucible level control device and method based on image | |
| JP2538748B2 (en) | Crystal size measuring device | |
| JP7221484B1 (en) | Single crystal pulling method and single crystal pulling apparatus | |
| KR200396732Y1 (en) | Silicon crystal growth controllar using image binarization | |
| AU775147B2 (en) | Method of and apparatus for growing ribbon of crystal | |
| CN107743531B (en) | Island position detection device and method for melting furnace | |
| JP4109843B2 (en) | Single crystal pulling apparatus and pulling method | |
| CN117187943B (en) | Melt detection method and device in crystal pulling process and electronic equipment | |
| JP4078156B2 (en) | Single crystal pulling apparatus and single crystal pulling method | |
| CN114252018B (en) | Crystal diameter detection method, system and computer program product | |
| CN118064983A (en) | Czochralski monocrystalline silicon welding method, device, equipment and storage medium | |
| CN117385459A (en) | Crystal detection method and device in crystal growth process | |
| JPH0632693A (en) | Method for controlling seeding of single crystal | |
| JP2022132995A (en) | Method for detecting state of surface of raw material melt, method for producing monocrystal, and cz monocrystal production device | |
| KR20200126082A (en) | Apparatus and method for automating dipping process of ingot growing apparatus | |
| CN105220230A (en) | For molten soup surface velocity measuring method and the long brilliant equipment of people section chief's crystalline substance | |
| CN113029014A (en) | Method and device for detecting diameter of single crystal silicon rod |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |