Method for estimating ease of fine powder occurrence in polysilicon, estimation device, method for producing polysilicon, trained model generation device and generation method

MYPI2025004906A0Undetermined Publication Date: 2024-08-14TOKUYAMA CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
MYPI2025004906
Authority / Receiving Office
MY · MY
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-02-14
Filing Date
2024-01-18
Publication Date
2024-08-14

AI Technical Summary

Technical Problem

Existing methods fail to effectively estimate and mitigate the generation of fine powder in polysilicon during vibration, leading to handling difficulties and contamination, as they do not specifically address the surface unevenness causing this issue.

Method used

A method using a trained model to estimate the ease of fine powder generation in polysilicon based on imaging data and vibration conditions, allowing for sorting and separation of polysilicon to minimize fine powder production.

Benefits of technology

Accurately estimates and reduces the generation of fine powder in polysilicon, improving handling and reducing contamination risks, thereby enhancing manufacturing efficiency and product quality.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure PI2025004906_DRAWING_001
    Figure PI2025004906_DRAWING_001
  • Figure PI2025004906_DRAWING_002
    Figure PI2025004906_DRAWING_002
  • Figure PI2025004906_DRAWING_003
    Figure PI2025004906_DRAWING_003
Patent Text Reader

Abstract

A likelihood of generation of fine powder of polysilicon is estimated. A method for estimating a likelihood of generation of fine powder of polysilicon includes: an estimation step of estimating, with use of a first trained model and from second captured image data including an image of second polysilicon that is a target of estimation, a likelihood of generation of fine powder of the second polysilicon when vibration is applied to the second polysilicon, the first trained model being generated with use of, as training data, (i) determination data indicating a determination of whether or not an amount in which fine powder of first polysilicon is generated due to vibration applied to the first polysilicon is not more than a predetermined first threshold, the vibration satisfying a predetermined condition and (ii) first captured image data including an image of the first polysilicon.
Need to check novelty before this filing date? Find Prior Art

Description

Method and device for estimating likelihood of fine powder generation in polysilicon, method for manufacturing polysilicon, and device and method for generating trained model

[0001] The present invention relates to a method and an apparatus for estimating the likelihood of fine powder generation in polysilicon, a method for manufacturing polysilicon, and an apparatus and method for generating a trained model.

[0002] In polysilicon, fine powder may be generated due to vibration during transportation, for example. Such fine powder may make handling of polysilicon difficult and may also cause contamination of the polysilicon.

[0003] The generation of such fine powder is thought to be due in part to unevenness on the surface caused by the state of deposition of polysilicon. Conventionally, methods for classifying polysilicon based on its surface state have been proposed (for example, Patent Documents 1 to 3).

[0004] Japan Special Table No. 2022-537014 Publication Japanese Special Table No. 2016-516666 Publication Japanese Special Table No. 2009-532319

[0005] However, the techniques disclosed in Patent Documents 1 to 3 do not focus on the fine powder generated from polysilicon, and therefore cannot be said to be suitable for separating polysilicon, which is prone to generating fine powder.

[0006] An object of one aspect of the present invention is to provide a method or the like that can estimate the likelihood of fine powder generation in polysilicon.

[0007] In order to solve the above-mentioned problem, a method for estimating the likelihood of fine powder generation in polysilicon according to one embodiment of the present invention includes an estimation step of estimating the likelihood of fine powder generation in second polysilicon when vibration is applied from second imaging data including an image of second polysilicon that is the estimation target, using a first trained model generated using, as training data, determination data on whether the amount of fine powder generation due to vibration of a first polysilicon applied to the first polysilicon is equal to or less than a predetermined first threshold value, and first imaging data including an image of the first polysilicon.

[0008] In order to solve the above-mentioned problems, a method for producing polysilicon according to one aspect of the present invention includes an estimation process for estimating the likelihood of fine powder generation in second polysilicon when vibration is applied from second imaging data including an image of second polysilicon that is an estimation target, using a first trained model generated using, as training data, determination data on whether the amount of fine powder generation due to vibration of a first polysilicon applied under predetermined conditions is equal to or less than a predetermined first threshold value and first imaging data including an image of the first polysilicon; and a first sorting process for sorting the second polysilicon based on the likelihood of fine powder generation estimated in the estimation process.

[0009] In order to solve the above-mentioned problem, an estimation device for estimating the likelihood of fine powder generation in polysilicon according to one embodiment of the present invention includes an estimation unit that uses a first trained model generated using, as training data, determination data on whether the amount of fine powder generated by vibration of a first polysilicon to which vibration under predetermined conditions is applied is equal to or less than a predetermined first threshold, and first imaging data including an image of the first polysilicon, to estimate the likelihood of fine powder generation in the second polysilicon when vibration is applied, from second imaging data including an image of the second polysilicon that is the estimation target.

[0010] In order to solve the above-mentioned problem, a trained model generation device according to one embodiment of the present invention includes an acquisition unit that acquires, as training data, determination data indicating whether the amount of fine powder generated by vibration of a first polysilicon to which vibration of a predetermined condition is applied is equal to or less than a predetermined first threshold value, and first imaging data including an image of the first polysilicon, and a generation unit that generates, from input second imaging data including an image of a second polysilicon that is the subject of estimation, a first trained model that outputs an estimation result of the likelihood of fine powder generation in the second polysilicon when vibration is applied.

[0011] In order to solve the above-mentioned problem, a method for generating a trained model according to one embodiment of the present invention includes an acquisition step of acquiring, as training data, judgment data indicating whether the amount of fine powder generated by vibration of a first polysilicon to which vibration of a predetermined condition is applied is equal to or less than a predetermined first threshold value, and first imaging data including an image of the first polysilicon, and a generation step of generating, from the input second imaging data including an image of the second polysilicon that is the object of estimation, a first trained model that outputs an estimation result of the likelihood of fine powder generation in the second polysilicon when vibration is applied.

[0012] According to one aspect of the present invention, it is possible to realize a method for estimating the likelihood of generation of fine powder in polysilicon.

[0013] 1 is a schematic diagram showing an overview of a part of a method for producing polysilicon. FIG. 2 is a block diagram showing the functional configurations of a generating device, an estimating device, and a sorting control device according to an embodiment of the present invention. FIG. 3 is a flowchart showing a method for producing polysilicon according to an embodiment of the present invention. FIG. 4 is a diagram showing an example of classifying polysilicon according to surface condition. FIG. 5 is a diagram showing the distribution of the size of crushed polysilicon pieces packed in a PE bag. FIG. 6 is a diagram showing the direction and intensity of vibration applied to crushed polysilicon pieces. FIG. 7 is a diagram showing the amount of fine powder in a PE bag packed with crushed polysilicon pieces before and after test transportation. FIG. 8 is a schematic diagram showing an overview of the first acquisition step to the estimation step shown in FIG. 3. FIG. 9 is a schematic diagram showing an overview of the third acquisition step to the calculation step of the area ratio of uneven regions shown in FIG. 3. FIG. 9 is a schematic diagram showing an overview of the third acquisition step to the calculation step of the number of depressions shown in FIG. 3. FIG. 10 is a diagram showing an example of threshold setting in sorting based on the total area of ​​polysilicon and the area ratio of uneven regions. FIG. 11 is a diagram showing an example of threshold setting in sorting based on the total area and number of depressions of polysilicon.

[0014] One embodiment of the present invention relates to a method for producing polysilicon. Polysilicon can be produced by depositing trichlorosilane as a raw material using, for example, a bell jar. The deposited rod-shaped polysilicon may have variations in surface condition depending on the deposition position, etc.

[0015] Specifically, the surface of the deposited polysilicon may be uneven, with the degree of unevenness varying depending on the position on the polysilicon surface. Such unevenness is sometimes called popcorn or cone because of its shape resembling popcorn.

[0016] Portions of polysilicon with large surface irregularities are inferior in quality. For example, polysilicon portions with large surface irregularities are often brittle and are at high risk of cracking during transportation. Furthermore, portions with large irregularities are prone to generating fine powder due to friction between polysilicon particles or between polysilicon and packaging. Such fine powder not only makes polysilicon difficult to handle, but can also cause contamination of the polysilicon.

[0017] The fine powder may be polysilicon fragments or powder smaller than a predetermined size. For example, when the size of the polysilicon to be produced is more than 5 mm and not more than 120 mm, the fine powder may be polysilicon having a size of 5 mm or less.

[0018] In this way, the size of the fine powder may be determined appropriately depending on the size of the polysilicon. The sizes of the polysilicon and the fine powder each refer to the major axis. The major axis refers to the distance between two points at which the distance is greatest in the polysilicon or the fine powder. In addition, since it may be difficult to accurately measure the size of fine powder having a major axis of 5 mm or less, for example, when the fine powder is defined as having a size of 5 mm or less, it may be considered that the fine powder passes through a 5 mm sieve.

[0019] The rod-shaped polysilicon may be divided into segments by a dividing process such as crushing or cutting for storage or transportation. The surface condition of such segments varies depending on the position in the rod-shaped polysilicon before division. Therefore, the physical properties such as the tendency to generate fine powder also differ from segment to segment. In this specification, when simply referring to "polysilicon," the shape is not particularly limited, and the polysilicon may be in the form of either a rod or segments.

[0020] FIG. 1 shows an example of a polysilicon crushing method to which a polysilicon manufacturing method according to one embodiment of the present invention can be applied. Rod-shaped polysilicon P has an uneven region P1 on a portion of its surface, where unevenness of a predetermined roughness or more exists. In order to obtain crushed polysilicon that is less likely to generate fine powder, it is preferable to remove the crushed material containing the uneven region P1 from the crushed material group. In FIG. 1, crushed material containing the uneven region P1 is shown as "with cone," crushed material not containing the uneven region P1 is shown as "without cone," and a crushed material group containing a mixture of crushed material containing the uneven region P1 and crushed material not containing the uneven region P1 is shown as "cone mix."

[0021] In order to remove the fragments including the uneven region P1, the above-mentioned crushing method includes a rough crushing step S1, a crushing step S2, and a chipping step S3.

[0022] The rough crushing step S1 is a step for separating the polysilicon P into a portion having the rough region P1 on its surface and a portion not having the rough region P1 in order to remove the rough region P1. The crushing step S2 is a step for crushing the portion having the rough region P1 on its surface and the portion not having the rough region P1 obtained in the rough crushing step S1, respectively, to obtain crushed objects. When the portion of the polysilicon P not having the rough region P1 is further crushed, a group of crushed objects not having the rough region P1 is obtained.

[0023] On the other hand, when the portion of the polysilicon P that includes the uneven region P1 is crushed, a group of crushed materials is obtained that is a mixture of crushed materials that include the uneven region P1 and crushed materials that do not include the uneven region P1. The crushed materials that include the uneven region P1 are further crushed and subjected to a chipping process S3 in which only the uneven region P1 is scraped off, so that the crushed materials can be separated into crushed materials that do not include the uneven region P1 and crushed materials that are mostly made up of the uneven region P1.

[0024] The classification of the crushed materials in the chipping step S3 can be performed by a human being by visually inspecting the surface condition of the crushed materials. However, it was not clear to what extent, among the various surface conditions shown in Figure 4, the contribution of these to the generation of fine powder by vibration. In addition, there was a problem in that it was difficult to standardize the criteria for judgment by visual inspection, etc.

[0025] According to one embodiment of the present invention, it is possible to sort polysilicon based on criteria for the correlation between the uneven area P1 and the likelihood of fine powder generation, which is learned based on the likelihood of fine powder generation under any vibration conditions.

[0026] To achieve this crushing method, the polysilicon manufacturing method exemplified in the first embodiment of the present invention described below uses a trained model generated using images of polysilicon that are prone to generating fine powder as training data to estimate the likelihood of generating fine powder in polysilicon. Then, the polysilicon can be sorted based on the results of this estimation. With this configuration, polysilicon that does not include an uneven region P1 on its surface and is less likely to generate fine powder can be selected with high accuracy.

[0027] Furthermore, the polysilicon manufacturing method exemplified in the second embodiment of the present invention described below uses a trained model generated using training data on the surface condition of polysilicon prone to fine powder generation to estimate the area ratio of the uneven region P1 on the polysilicon surface. By sorting the polysilicon based on the area ratio, etc., it is possible to select polysilicon suitable for the chipping step S3. This configuration allows the uneven region P1 to be efficiently removed, greatly improving the efficiency of polysilicon manufacturing.

[0028] Here, extremely high purity is required for polysilicon products. For this reason, polysilicon is etched for cleaning in the final stage of commercialization. To maintain the cleanliness of the surface, it is desirable to package the etched polysilicon so that it does not come into contact with anything other than polysilicon.

[0029] Inspecting the surface condition before etching and before crushing in order to remove the uneven region P1 is a labor-intensive task. Therefore, all of the conventional surface condition inspection methods target polysilicon after etching, and serve as a final product inspection, with the objective of inspecting the surface condition as well as preventing the inclusion of foreign matter.

[0030] Therefore, there are many chances that the inspection worker's gloves and the like come into contact with the surface during the inspection work after etching, and there is a risk that the cleanliness of the surface may be impaired during the inspection process after etching.

[0031] As an example, the weight percentage of polysilicon P before the rough crushing step S1 is set to 100%. After the rough crushing step S1, the crushing step S2, and the chipping step S3, the weight percentage of the fragments containing the uneven regions P1 unsuitable for use as a product is, for example, about 25%, depending on the deposition conditions of the polysilicon P. If the fragments containing the uneven regions P1 can be detected before etching, inspection of the surface condition after etching is unnecessary, and the surface cleanliness is not impaired, resulting in improved surface cleanliness of the polysilicon compared to conventional methods. Furthermore, the burden of inspecting the surface condition of the polysilicon can be reduced.

[0032] According to a method for producing polysilicon according to an embodiment of the present invention, the likelihood of generating fine powder can be estimated before etching the polysilicon, and the polysilicon can be sorted based on the results of the estimation. This is because the shape of the unevenness on the surface of the polysilicon does not change significantly before and after etching, and therefore the likelihood of generating fine powder in the polysilicon product after etching can be accurately estimated by inspection before etching.

[0033] This allows polysilicon that is prone to generating fine powder to be excluded from the etching process. To obtain more valuable polysilicon for semiconductors, it is sufficient to etch only the polysilicon that is less likely to generate fine powder. This significantly improves the efficiency of polysilicon production compared to conventional methods. These effects also contribute to the achievement of Goal 12 of the United Nations' Sustainable Development Goals (SDGs), including "Ensure sustainable consumption and production patterns."

[0034] 3 , a method for producing polysilicon according to this embodiment includes at least an estimation step S14 and a first classification step S15, and may further include a first acquisition step S11, a first generation step S12, and a second acquisition step S13. In the estimation step S14, for example, estimation is performed using the first trained model obtained in the first generation step S12. First, a trained model generation method according to this embodiment for acquiring the first trained model will be described.

[0035] <Generation of First Trained Model> The trained model generation method according to this embodiment may be executed by, for example, a trained model generation device 10 shown in Fig. 2. Hereinafter, the trained model generation method will be described using an example in which the method is executed by the generation device 10.

[0036] 2 , the generating device 10 has a first acquisition unit (acquisition unit) 11 and a first generation unit (generation unit) 12. The generating device 10 may be a control device capable of executing the processes of the first acquisition unit 11 and the first generation unit 12, and may be, for example, a CPU (Central Processing Unit). Furthermore, the generating device 10 may be a device that includes such a control device, as well as a storage device, a communication device, and the like.

[0037] (First Acquisition Step and First Teacher Data) As shown in FIG. 3 , the first acquisition unit 11 executes a first acquisition step (acquisition step) S11 to acquire first teacher data used to generate a first trained model. The first teacher data is data including judgment data and first imaging data. The judgment data is data indicating whether the amount of fine powder generated by vibration of the first polysilicon to which vibration under predetermined conditions is applied is equal to or less than a predetermined first threshold. The first imaging data is data including an image of the first polysilicon.

[0038] The first polysilicon is polysilicon whose tendency to generate fine powder due to vibration is known. For example, it is preferable that the first polysilicon has information on the amount of fine powder generated when vibration under predetermined conditions is applied.

[0039] The first polysilicon may be, for example, a final polysilicon product, and the amount of fine powder generated before and after transportation is known. Alternatively, the first polysilicon may be a polysilicon sample or the like, and the amount of fine powder generated before and after transportation is known, and vibration is applied under conditions similar to those in the transportation of the product. Note that the vibration applied to the first polysilicon does not have to be under conditions similar to those in the transportation of the product, and may be vibration under conditions different from those in the transportation of the product, such as an accelerated test.

[0040] The first threshold value in the determination data may be, for example, an allowable value for the amount of fine powder generated by vibration under predetermined conditions. In this case, if the amount of fine powder generated is equal to or less than the first threshold value, the first polysilicon can be said to be polysilicon that is less likely to generate fine powder. The method for setting the first threshold value is not particularly limited, and a value suitable for distinguishing polysilicon that is less likely to generate fine powder may be set as appropriate.

[0041] The determination of whether the amount of fine powder generated in the first polysilicon is equal to or less than the first threshold value may be made for each first polysilicon based on a measurement value of the actual amount of fine powder generated. Alternatively, the determination may be made based on the surface condition of the first polysilicon. This is based on the finding experimentally revealed by the present inventors that the rougher the polysilicon surface, the more likely fine powder is generated.

[0042] As shown in FIG. 4 , the inventors classified the polysilicon fragments into groups based on their surface condition. Specifically, the group with the smoothest surface was assigned Level 1, and the group with the roughest surface was assigned Level 8, and the polysilicon fragments were relatively classified into eight groups. FIG. 4 shows an example of images of the surfaces of the polysilicon fragments in each group taken at the same magnification. The inventors found that the polysilicon fragments classified into Levels 5 to 8 generated a larger amount of fine powder than the polysilicon fragments classified into Levels 1 to 4. In other words, the inventors found that if the polysilicon has a surface condition classified into Level 4 or lower, the amount of fine powder generated is likely to be within the acceptable range for commercialization.

[0043] Specifically, as shown in Figure 5, 5 kg of nuggets were prepared to have the weight distribution (weight %, left vertical axis) and cumulative weight distribution (weight %, right vertical axis) of the crushed polysilicon (nuggets) versus size (long diameter, horizontal axis). Each nugget was packed into a PE (polyethylene) bag. The size of the nuggets was obtained by measuring the long diameter, i.e., the maximum side length, with a vernier caliper. The nugget sizes ranged from 10 to 110 mm, with a median of approximately 70 mm. The PE bags were packed into cardboard boxes, and a round-trip transportation test of approximately 400 km (approximately 200 km one way) was conducted. These conditions are equivalent to those for transporting polysilicon products. The cardboard boxes were loaded into the rear of the loading space of a transport truck, and a vibrometer equipped with a triaxial acceleration sensor was installed on top of the cardboard boxes.

[0044] Figure 6 shows the results of measuring the acceleration (G) in each direction over time using a vibrometer, with the direction of travel of the transport truck designated as the X axis, the left-right direction relative to the direction of travel designated as the Y axis, and the up-down direction designated as the Z axis. For example, when a transport truck stops while traveling, acceleration occurs in the direction of travel, resulting in a positive value on the X axis, and when the truck starts moving again, a negative value. The Y axis becomes a positive value when the transport truck curves left while traveling in the direction of travel, resulting in acceleration to the right, and a negative value when the truck curves right. The Z axis is approximately -1G when stationary due to the influence of gravity. When a cardboard box bounces upward, the Z axis becomes a value more negative than -1G, and when the cardboard box is pushed from above, the Z axis becomes a value more positive than -1G.

[0045] As shown in Figure 6, the cardboard box containing the nuggets moved significantly up and down, and tilted more left and right than forward and backward. It is believed that the generation of fine powder from the nuggets occurs when the nuggets are subjected to such vibrations during transportation, causing them to collide or rub against each other. It is believed that the greater the irregularities on the nugget surface, the more likely it is that fine powder will be generated by the collisions or rubbing.

[0046] Furthermore, if there are depressions such as holes or cracks on the surface of the nugget, fine powder that cannot be removed by etching may remain in the depressions. The fine powder in such depressions is also expelled from the depressions by vibration, which causes the generation of fine powder.

[0047] Figure 7 shows the results of two tests conducted under conditions similar to those used in product transportation. In Test 1, a comparison was made between conditions in which nuggets with surface conditions classified as levels 1 to 4 were packed in PE bags and conditions in which nuggets with surface conditions classified as levels 1 to 8 were packed in PE bags. In Test 1, 25 PE bags (N = 25) were randomly sampled before transportation (before shipping) and 14 PE bags (N = 14) were randomly sampled after the test, and the amount of fine powder was measured.

[0048] Test 2 was conducted on a different day from Test 1 under the same transportation conditions, and a comparison was made between a condition in which nuggets whose surface conditions were classified as levels 1 to 6 were packed in PE bags and a condition in which nuggets whose surface conditions were classified as levels 5 or 6 were packed in PE bags. In Test 2, 10 PE bags (N=10) were randomly sampled both before transportation (before shipping) and after the test, and the amount of fine powder was measured.

[0049] The nuggets in the PE bag were sorted based on size. First, the nuggets in the PE bag were sorted using a 5 mm sieve. The fine powder that passed through the 5 mm sieve was further sorted using a 1 mm sieve. This resulted in three groups: fine powder of 1 mm or less, fine powder of 1 to 5 mm, and nuggets larger than 5 mm. The weight (g) of the fine powder group was measured.

[0050] As shown in Figure 7, Test 1 showed that the amount of fine powder that increased after the test was greater in groups 1 to 8 of levels compared to groups 1 to 4. This is thought to be because groups 1 to 8 of levels contained many nuggets with rough surfaces, which resulted in an increased amount of fine powder generated compared to groups 1 to 4 of levels.

[0051] Furthermore, Test 2 showed that the amount of fine powder increased more in the Level 5 and 6 groups after the test compared to the Level 1 to 6 groups. This is thought to be because the proportion of nuggets with rough surfaces was higher in the Level 5 and 6 groups than in the Level 1 to 6 groups, which increased the amount of fine powder generated.

[0052] As described above, it has been experimentally shown that the rougher the surface condition of polysilicon, the more likely it is that fine powder will be generated by vibration. Based on this finding, whether the amount of fine powder generated by vibration in the first polysilicon is equal to or less than a predetermined first threshold value may be determined in a pseudo manner based on whether the surface condition of the first polysilicon is equal to or less than a predetermined roughness.

[0053] The first imaging data may include at least one image of the first polysilicon. Preferably, the first imaging data is a plurality of images of the first polysilicon taken from different directions. More specifically, the first imaging data preferably includes an image of a first surface, which is one surface of the first polysilicon, and an image of a second surface, which is a surface including the reverse side of the first surface.

[0054] According to this configuration, the first imaging data includes images of the front and back surfaces of the first polysilicon, and therefore serves as an index showing the correlation between the surface condition of the first polysilicon and the likelihood of fine powder generation of the first polysilicon. Therefore, according to this first imaging data, it is possible to generate a first trained model that can estimate with high accuracy the likelihood of fine powder generation of the second polysilicon to be estimated.

[0055] The first imaging data preferably includes a plurality of images of the first surface of the first polysilicon taken from a plurality of directions and a plurality of images of the second surface taken from a plurality of directions. Furthermore, the images of the first polysilicon included in the first imaging data preferably have as few shadow areas as possible. For example, the imaging of the first polysilicon is preferably performed under a condition where light is irradiated onto the first polysilicon from a plurality of directions by using a method such as providing a plurality of light sources or diffusing light inside a dome.

[0056] When the first imaging data includes multiple images, the first imaging data may be a composite image obtained by combining the multiple images. If the first imaging data is such a composite image, the first generation unit 12 only needs to process a relatively small number of images as the first training data, thereby reducing the processing load. Note that the first imaging data may be archive-format data that individually includes multiple images.

[0057] (First Generation Step) Next, the first generation unit 12 executes a first generation step (generation step) S12 of generating a first trained model using the judgment data and the first imaging data acquired by the first acquisition unit 11 as teacher data (first teacher data). The first trained model functions as an algorithm that outputs an estimation result of the likelihood of fine powder generation in the second polysilicon when vibration is applied, from the input second imaging data including an image of the second polysilicon that is the estimation target.

[0058] The second polysilicon is polysilicon for which the likelihood of generating fine powder is to be estimated using the first trained model. The second polysilicon is preferably polysilicon for which the likelihood of generating fine powder is unknown, but is not limited thereto, and may be, for example, polysilicon for which the likelihood of generating fine powder has been estimated by some method.

[0059] The second imaging data is data including an image of the second polysilicon. The second imaging data may include at least one image of the second polysilicon. The second imaging data is preferably a plurality of images of the second polysilicon, each image being taken from a different direction. More specifically, the second imaging data preferably includes an image of a third surface, which is one surface of the second polysilicon, and an image of a fourth surface, which is a surface including the back side of the third surface.

[0060] According to this configuration, the second imaging data includes images of the front and back surfaces of the second polysilicon, and therefore serves as an index showing the correlation between the surface condition of the second polysilicon and the likelihood of the second polysilicon generating fine powder. Therefore, such second imaging data is preferable as input data for estimating the likelihood of the second polysilicon generating fine powder with high accuracy.

[0061] The second imaging data preferably includes a plurality of images of the third surface of the second polysilicon taken from a plurality of directions and a plurality of images of the fourth surface taken from a plurality of directions. Similarly to the image of the first polysilicon included in the first imaging data, the images of the second polysilicon included in the second imaging data preferably have as few shadow areas as possible. For example, the imaging of the second polysilicon is preferably performed under a condition where light is irradiated onto the second polysilicon from a plurality of directions, for example, by providing a plurality of light sources or by diffusing light within a dome.

[0062] The image format of the second imaging data is preferably the same as the image format of the first imaging data. For example, if the first imaging data is a composite image of multiple images, the second imaging data is also preferably a composite image of multiple images. In this case, the number of images included in the second imaging data is preferably the same as the number of images included in the first imaging data. With this configuration, estimation can be easily performed from the second imaging data using the first trained model.

[0063] The first generation step S12 executed by the first generation unit 12 can be said to be a step of optimizing the machine learning algorithm by inputting the first teacher data into the machine learning algorithm. The first trained model can be said to be a machine learning algorithm optimized for estimating the likelihood of fine powder generation of the second polysilicon by inputting the first teacher data. Such a machine learning algorithm is not particularly limited, but may be, for example, a neural network. The machine learning algorithm may be, for example, a CNN (convolutional neural network).

[0064] <Estimation Using First Trained Model> The method for estimating the likelihood of fine powder generation in polysilicon according to this embodiment is performed using a first trained model. The estimation method may be performed by an estimation device 20 for the likelihood of fine powder generation in polysilicon shown in Figure 2. Below, the method for estimating the likelihood of fine powder generation in polysilicon will be described using an example in which the method is performed by the estimation device 20.

[0065] 2 , the estimation device 20 includes a second acquisition unit 21 and an estimation unit 22. The estimation device 20 may be a control device capable of executing the processes of the second acquisition unit 21 and the estimation unit 22, and may be, for example, a central processing unit (CPU). Furthermore, the estimation device 20 may be a device that includes such a control device, as well as a storage device, a communication device, and the like.

[0066] 3, the second acquisition unit 21 executes a second acquisition step S13 of acquiring second imaging data. The second imaging data has already been described, and therefore will not be described here.

[0067] Next, the estimation unit 22 executes an estimation step S14 in which the estimation unit 22 uses the first trained model to estimate the likelihood of generation of fine powder in the second polysilicon when vibration is applied, from the second imaging data acquired by the second acquisition unit 21. The first trained model used by the estimation unit 22 may be stored in advance in the estimation device 20, or may be acquired by the second acquisition unit 21 from the generation device 10 as needed.

[0068] The estimation result of the tendency for fine powder to be generated in the second polysilicon may be, for example, an estimation result of whether or not the amount of fine powder generated when vibration under predetermined conditions is applied to the second polysilicon is equal to or less than a first threshold value corresponding to the vibration under the predetermined conditions. The amount of fine powder generated from the second polysilicon is considered to be positively correlated with the strength of the applied vibration. Therefore, such an estimation result is useful as an index of the tendency for fine powder to be generated in the second polysilicon, regardless of the actual strength of the vibration.

[0069] In other words, the first trained model is generated based on vibrations under specific conditions and a first threshold value for the amount of fine powder generated corresponding to the conditions. However, by using such a first trained model, the estimation unit 22 can accurately estimate the relative likelihood of fine powder generation in the second polysilicon regardless of the vibration conditions applied to the second polysilicon.

[0070] 8 shows an overview of the steps from the first acquisition step S11 to the estimation step S14. In the example shown in FIG. 8, the first imaging data is a composite image including multiple images of the first polysilicon. The determination data is data showing the results of determining that the amount of fine powder generated is equal to or less than a first threshold when the surface condition of the first polysilicon is at level 1 to 4, and that the amount of fine powder generated exceeds the first threshold when the surface condition is at level 5 to 8.

[0071] The first generation unit 12 generates a first trained model from the first imaging data and the judgment data having such a correspondence relationship. The estimation unit 22 inputs a composite image including multiple images of the second polysilicon as second imaging data into the first trained model and estimates the likelihood of the second polysilicon generating fine powder. Specifically, the estimation unit 22 may display a result indicating whether the surface state of the second polysilicon is level 4 or lower as an estimated result of the likelihood of the second polysilicon generating fine powder.

[0072] <Separation of Second Polysilicon> The first separation step S15 is a step of separating the second polysilicon based on the likelihood of fine powder generation estimated in the estimation step S14. The first separation step S15 may be performed by the separation control device 30 shown in FIG. 2. The first separation step S15 may be a step performed on the crushed polysilicon obtained in the crushing step S2 shown in FIG. 1.

[0073] 2, the sorting control device 30 has a first sorting unit 31. The sorting control device 30 may be a control device capable of executing the processing of the first sorting unit 31, and may be, for example, a CPU (Central Processing Unit). Furthermore, the sorting control device 30 may be a device that includes such a control device as well as a storage device, a communication device, and the like.

[0074] The first sorting unit 31 may perform control to sort third polysilicon, which is second polysilicon estimated in the estimation step S14 to have a tendency to generate fine powder that exceeds a predetermined second threshold, from other second polysilicon. The second threshold may be a value that indicates that the second polysilicon is unsuitable for commercialization when the tendency to generate fine powder exceeds the second threshold, for example.

[0075] For example, if the estimation result from the estimation process S14 indicates that the surface condition of the second polysilicon is below a predetermined roughness, the first sorting unit 31 may determine that the likelihood of fine powder generation of the second polysilicon is below a second threshold value.

[0076] Regarding the control for separating the second polysilicon, the first sorting unit 31 may send a control signal to a physical mechanism for separating the second polysilicon. Such a physical mechanism may be included in the sorting control device 30, or may be included in a second polysilicon sorting device (not shown) different from the sorting control device 30.

[0077] According to the polysilicon manufacturing method of this embodiment having the configuration exemplified above, it is possible to accurately estimate the likelihood of generating fine powder for the second polysilicon. Furthermore, by predicting in advance the second polysilicon that is less likely to generate fine powder, it is possible to perform etching only on the second polysilicon, thereby significantly improving the efficiency of polysilicon manufacturing.

[0078] 2 may be realized by a single control device. Also, the control blocks of the generation device 10 and the estimation device 20 may be realized by a single control device, or the control blocks of the estimation device 20 and the separation control device 30 may be realized by a single control device.

[0079] [Embodiment 2] A second embodiment of the present invention will be described below. For ease of explanation, the same reference numerals will be used to designate components having the same functions as those described in the first embodiment, and the description thereof will not be repeated.

[0080] The method for producing polysilicon according to this embodiment includes a method for efficiently performing the chipping step S3 on the third polysilicon, which is estimated to be prone to generating fine powder in the method for producing polysilicon according to the first embodiment.

[0081] 3 , the polysilicon manufacturing method according to this embodiment includes, in addition to the steps of the polysilicon manufacturing method according to embodiment 1, a third acquisition step S16, a second generation step S17, an area ratio calculation step S18, a depression number calculation step S19, and a second sorting step S20. The area ratio calculation step S18 and the depression number calculation step S19 each perform a calculation process using, for example, the second trained model generated in the second generation step S17. First, a method for generating the second trained model will be described.

[0082] <Generation of Second Trained Model> (Second Acquisition Process and Second Teacher Data) The second trained model functions as an algorithm for estimating whether the third polysilicon is suitable as a target for the chipping process S3. The generation of the second trained model may be executed by the generation device 10. As shown in FIG. 2 , the generation device 10 may have a second generation unit 13 for generating the second trained model.

[0083] 3, in this embodiment, the first acquisition unit 11 may execute a third acquisition step S16 of acquiring second training data to be used for generating the second trained model. The second training data includes surface condition data indicating a surface condition of the first polysilicon in which the amount of fine powder generated exceeds a first threshold. Such a surface condition of the first polysilicon may be, for example, a surface condition having irregularities or depressions, as shown as levels 5 to 8 in FIG. 4.

[0084] 9, the surface condition data may be data including information identifying (annotating) as an uneven region an area where the surface unevenness corresponds to levels 5 to 8 in the first imaging data of the first polysilicon in which the amount of fine powder generated exceeds a first threshold. The annotation may be performed by a user inputting the results of visual identification, or may be performed mechanically by image processing.

[0085] Furthermore, as shown in FIG. 10 , the surface condition data may include information identifying depressions, such as holes or cracks, formed on the surface of the first polysilicon, in addition to information identifying the uneven regions. It is preferable that the image included in the first imaging data does not include shadow areas. However, depending on the lighting conditions or the surface condition during imaging, it may be difficult to completely eliminate shadow areas from the image included in the first imaging data. Therefore, it may be difficult to determine whether a dark area in the image included in the first imaging data is a depression or a shadow.

[0086] Therefore, the surface condition data may be data including information identifying (annotating) recessed portions in the first imaging data of the first polysilicon in which the amount of generated fine powder exceeds the first threshold value. The annotation may be performed by a user inputting the results of visual identification, or may be performed mechanically by image processing.

[0087] Such surface condition data is suitable as second training data for creating a second trained model that can identify areas of the third polysilicon that have a surface condition that may cause the generation of fine powder, as shown in Figures 9 and 10.

[0088] The surface condition data may be acquired by the first acquisition unit 11 at the same time as acquiring the determination data and the first imaging data in the first acquisition step S11. In this case, the third acquisition step S16 may be omitted.

[0089] (Second Generation Step) Next, the second generation unit 13 executes a second generation step S17 in which the second generation unit 13 uses the surface condition data acquired by the first acquisition unit 11 as training data (second training data) to generate a second trained model. Specifically, the second trained model functions as an algorithm for calculating the area ratio of at least a portion of the surface of the third polysilicon that is estimated to be similar to the surface condition of the first polysilicon in which the amount of fine powder generated exceeds the first threshold, relative to the total area of ​​the surface of the third polysilicon. The calculation of the area ratio will be described later in the explanation of the area ratio calculation step S18.

[0090] The second generation step S17 executed by the second generation unit 13 can be said to be a step of optimizing the machine learning algorithm by inputting the second teacher data into the machine learning algorithm. The second trained model can be said to be a machine learning algorithm optimized for estimating the portion of the third polysilicon that causes the generation of fine powder by inputting the second teacher data.

[0091] Such a machine learning algorithm is not particularly limited, and may be, for example, a neural network or a convolutional neural network (CNN). The machine learning algorithm used to generate the second trained model may be the same as or different from the machine learning algorithm used to generate the first trained model.

[0092] <Estimation Using Second Trained Model> The second trained model is used in estimation processing using the second imaging data of the third polysilicon as input in the area ratio calculation step S18 and the depression number calculation step S19. The area ratio calculation step S18 and the depression number calculation step S19 may be executed by the estimation device 20. As shown in FIG. 2 , the estimation device 20 may include an area ratio calculation unit 23 that executes the area ratio calculation step S18 and a depression number calculation unit 24 that executes the depression number calculation step S19.

[0093] (Area ratio calculation process) The area ratio calculation unit 23 calculates, from the second imaging data of the third polysilicon, the area ratio of at least a portion of the surface of the third polysilicon that is estimated to be similar to the surface state of the first polysilicon in which the amount of fine powder generated exceeds the first threshold value, relative to the total area of ​​the surface of the third polysilicon.

[0094] The second imaging data of the third polysilicon may be data acquired by the second acquisition unit 21 in the second acquisition step S13. In this case, the area ratio calculation unit 23 may select only the second imaging data of the third polysilicon for which the likelihood of fine powder generation is estimated to exceed a predetermined second threshold from the second imaging data of the second polysilicon for which the estimation result estimated in the estimation step S14 exists. Furthermore, the second acquisition unit 21 may acquire the second imaging data of the third polysilicon again before the area ratio calculation step S18.

[0095] The area ratio calculation unit 23 calculates the total area of ​​the projected portions of the third polysilicon based on the second imaging data of the third polysilicon captured at a magnification of 1, preferably 1 to 3. The total area of ​​the projected portions of the third polysilicon can be calculated from the image included in the second imaging data. In this case, the area ratio calculation unit 23 may calculate an approximate total area two-dimensionally from the image included in the second imaging data, or may calculate a more accurate total area by estimating the three-dimensional shape of the third polysilicon.

[0096] For example, the area ratio calculation unit 23 may calculate the sum of the areas of the projected portions of the third polysilicon imaged from multiple directions by algebraically calculating the number of pixels of the projected portions of the third polysilicon. Then, the value obtained by dividing the sum by the number of imaging directions may be used as the index of the total area of ​​the surface of the third polysilicon. However, the method of calculating the total area of ​​the surface of the third polysilicon by the area ratio calculation unit 23 is not limited to this.

[0097] In addition, if the image included in the second imaging data does not include a portion of the surface of the third polysilicon, the area ratio calculation unit 23 may consider the total area of ​​the surface appearing in the image included in the second imaging data to be the total area of ​​the surface of the third polysilicon.

[0098] Next, the area ratio calculation unit 23 inputs the second imaging data of the third polysilicon to the second trained model, and estimates the position of an uneven region, which is a surface region similar to the surface state of the first polysilicon in which the amount of fine powder generated exceeds the first threshold. The uneven region in the third polysilicon can be said to be a surface region of at least a part of the third polysilicon, which is a portion estimated to have a large amount of fine powder generated.

[0099] The area ratio calculation unit 23 calculates the area of ​​the estimated uneven region. The area ratio calculation unit 23 calculates the area of ​​the uneven region using the same method as that used to calculate the total area. When the area ratio calculation unit 23 estimates that a plurality of independent uneven regions exist in the third polysilicon, the area ratio calculation unit 23 may individually calculate the area of ​​each uneven region and its total area, or may calculate only the total area of ​​all the uneven regions.

[0100] For example, similar to the surface area, the area ratio calculation unit 23 may use the number of pixels in the portion regarded as the uneven region as an algebra for the sum of the areas of the portions regarded as the uneven region, and divide the sum by the number of imaging directions to obtain the exponent of the area of ​​the uneven region. However, the method of calculating the area of ​​the uneven region on the surface of the third polysilicon by the area ratio calculation unit 23 is not limited to this.

[0101] The second trained model is trained using the surface condition data of the first polysilicon in which the amount of fine powder generated exceeds the first threshold as training data, and therefore, by using the second trained model, the area ratio calculation unit 23 can accurately estimate which parts of the surface of the third polysilicon are uneven regions.

[0102] The area ratio calculation unit 23 then calculates the area ratio of the uneven region to the total area of ​​the third polysilicon. The area ratio of the uneven region is used to separate the third polysilicon in the second separation step S20, which will be described later.

[0103] (Count of dents calculation step) The count of dents calculation unit 24 executes a count of dents calculation step S19 in which the number of dents is calculated, which is the number of portions estimated to be dents formed on the surface of the third polysilicon, from the second imaging data of the third polysilicon using the second trained model. Note that the count of dents calculation step S19 may be omitted in the method for manufacturing polysilicon according to this embodiment.

[0104] The second trained model is configured to detect depressions such as holes or cracks formed on the surface of the first polysilicon (for example, a depression of 1 mm in an image captured at 1x magnification). 2 Surface condition data including information identifying pits (holes or cracks having a total area equal to or greater than 100 mm) may be generated as training data. In this case, the pit number calculation unit 24 can distinguish between pits and shadows formed on the surface of the third polysilicon by using the second trained model, and can estimate the pits with high accuracy. As described above, pits formed on the surface of the third polysilicon cause the generation of fine powder after etching, and therefore, the greater the number of pits, the more likely the third polysilicon is to generate fine powder.

[0105] The number of recesses calculated by the recess number calculation unit 24 may be used to separate the third polysilicon in the second separation step S20 described later.

[0106] <Separation of Third Polysilicon> The second separation step S20 is a step of separating the third polysilicon based on the total area and area ratio calculated in the area ratio calculation step S18. The second separation step S20 may also be a step of separating the third polysilicon based on the number of recesses calculated in the recess number calculation step S19, in addition to the total area and area ratio. The separation in the second separation step S20 may be aimed at separating the third polysilicon to be subjected to the chipping step S3 shown in FIG. 1 .

[0107] The second sorting step S20 may be performed by a sorting control device 30 shown in Fig. 2. As shown in Fig. 2, the sorting control device 30 may have a second sorting unit 32 that performs the second sorting step S20.

[0108] The third polysilicon to be subjected to the chipping step S3 preferably has a total area equal to or greater than a predetermined third threshold value. This is because, when the portion including the uneven region is to be scraped off from the third polysilicon by crushing or the like, if the third polysilicon is too small, the work becomes difficult. Therefore, it is preferable that the second sorting unit 32 use the total area of ​​the third polysilicon as the sorting criterion for the third polysilicon.

[0109] The third threshold value may be set appropriately depending on the method of the chipping step S3. For example, the third threshold value may be set to a different value when the chipping step S3 is performed manually using a hammer or the like and when the chipping step S3 is performed mechanically using a crushing device.

[0110] Furthermore, the third polysilicon to be subjected to the chipping step S3 preferably has an area ratio of the uneven region equal to or less than a predetermined fourth threshold. If the area ratio of the uneven region is too large, most of the surface of the third polysilicon will be uneven, and even if the chipping step S3 is performed, almost no third polysilicon containing uneven regions will be obtained. From the viewpoint of manufacturing efficiency, it is preferable that the second sorting unit 32 use the area ratio of the uneven region in the third polysilicon as a sorting criterion for the third polysilicon.

[0111] The fourth threshold value may be set appropriately depending on the target product size of the third polysilicon. The fourth threshold value may also be adjusted appropriately depending on the total area of ​​the third polysilicon. If the total area of ​​the third polysilicon is sufficiently large, a predetermined amount or more of third polysilicon without uneven regions can be obtained after the chipping step S3 even if the fourth threshold value is set to a large value.

[0112] Furthermore, the third polysilicon to be subjected to the chipping step S3 preferably has a number of recesses equal to or less than a predetermined fifth threshold. Furthermore, the third polysilicon to be subjected to the chipping step S3 preferably has a number of recesses equal to or greater than a predetermined sixth threshold. In other words, the third polysilicon to be subjected to the chipping step S3 preferably has a number of recesses equal to or greater than the sixth threshold and equal to or less than the fifth threshold.

[0113] Third polysilicon with a large number of recesses is likely to generate a large amount of fine powder regardless of the area ratio of the uneven region because fine powder that cannot be removed by etching may exist in the recesses. Furthermore, when the number of recesses is extremely small, the estimation result of the estimation step S14 for the third polysilicon, that the likelihood of fine powder generation exceeds the second threshold, is considered to be due to an erroneous detection of a shadow or the like included in the second imaging data. For these reasons, it is preferable that the second sorting unit 32 use the number of recesses in the third polysilicon as the sorting criterion for the third polysilicon.

[0114] The fifth and sixth thresholds may be set empirically as appropriate. The fifth and sixth thresholds may also be adjusted as appropriate based on the total area of ​​the third polysilicon. For a given number of depressions, the larger the total area of ​​the third polysilicon, the fewer the number of depressions per unit area. It is preferable to consider the likelihood of powder generation in the third polysilicon per unit area.

[0115] 11 and 12 show the relationship between the results of visually separating polysilicon and the total area, area ratio, and number of depressions calculated by the area ratio calculation unit 23 and the depression number calculation unit 24. Here, the results of separating the "cone mix" obtained in the crushing step S2 shown in FIG. 1 , which is a polysilicon crushed material containing third polysilicon, are shown. Each graph in FIG. 11 shows the correlation between the total area and the area ratio of the uneven region for each separated group, and each graph in FIG. 12 shows the correlation between the total area and the number of depressions for each separated group. The total area of ​​polysilicon is expressed as an index of the total area, using the number of pixels of the polysilicon projection portion calculated from the image included in the second imaging data.

[0116] 11, graph 111 shows the distribution of all crushed objects, graph 112 shows the distribution of groups determined to not require chipping step S3, graph 113 shows the distribution of groups determined to be suitable for chipping step S3, and graph 114 shows the distribution of groups determined to be unsuitable for chipping step S3 despite including portions prone to generating fine powder.

[0117] 12, graph 121 shows the distribution of groups obtained by excluding, from all crushed objects, third polysilicon having a total area index of 10 or less. Graphs 122 to 124 also show the same in that third polysilicon having a total area index of 10 or less has been excluded. Graph 122 shows the distribution of groups determined to not require the chipping step S3, graph 123 shows the distribution of groups determined to be suitable for the chipping step S3, and graph 124 shows the distribution of groups determined to be unsuitable for the chipping step S3 despite including portions that are likely to generate fine powder.

[0118] The region shown as the "threshold range" in Figure 11 is a region where the index of the total area is greater than 10 and the area ratio of the uneven regions is 65% or less. Polysilicon classified within this region has neither too small a total area nor an excessively large area ratio of the uneven regions, and is therefore highly likely to be suitable for the chipping step S3. This threshold range may be an example of setting the third and fourth thresholds.

[0119] 12, the region where the index of the total area is greater than 10 and is sandwiched between the "threshold range (upper limit)" and the "threshold range (lower limit)" is an empirically determined region where fragments that are less likely to generate fine powder are likely to be plotted. Polysilicon classified within this region has neither too few nor too many recesses relative to the total area, indicating that it is highly likely to be suitable for the chipping step S3. Such regions may be examples of setting the third, fifth, and sixth thresholds.

[0120] Each of the third to sixth thresholds may be set based on the results of actual measurements as described above, may be set empirically, or may be set by other methods.

[0121] The second sorting unit 32 may sort the third polysilicon based on the total area and the area ratio, for example, as shown in Fig. 11. Furthermore, the second sorting unit 32 may sort the third polysilicon sorted based on the total area and the area ratio by further sorting the third polysilicon based on the total area and the number of depressions, as shown in Fig. 12.

[0122] The second sorting unit 32 may send a control signal to a physical mechanism for sorting the third polysilicon. Such a physical mechanism may be included in the sorting control device 30, or may be included in a third polysilicon sorting device (not shown) that is different from the sorting control device 30.

[0123] According to the polysilicon manufacturing method of this embodiment having the configuration exemplified above, the third polysilicon suitable for the chipping step S3 can be efficiently separated, thereby reducing both the labor cost and production loss required to obtain polysilicon that is less likely to generate fine powder.

[0124] As an example, the weight percentage of polysilicon before the rough crushing step S1 is assumed to be 100%. At this time, the polysilicon to be subjected to the chipping step S3 accounts for a relatively large proportion, for example, about 15%, although this depends on the polysilicon deposition conditions. Therefore, reducing the polysilicon loss rate through the chipping step S3 is important for improving manufacturing efficiency. Therefore, the polysilicon manufacturing method according to this embodiment, which can improve the efficiency of the chipping step S3, significantly contributes to improving manufacturing efficiency.

[0125] [Example of implementation by software] The functions of the generation device 10, the estimation device 20, or the separation control device 30 (hereinafter referred to as "device") can be realized by a program that causes a computer to function as the device, and a program that causes a computer to function as each control block of the device.

[0126] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The functions described in each of the above embodiments are realized by executing the program using the control device and storage device.

[0127] The program may be non-transitory and may be recorded on one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be supplied to the device via any wired or wireless transmission medium.

[0128] Furthermore, some or all of the functions of the control blocks can be realized by logic circuits. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present invention. In addition, the functions of the control blocks can also be realized by, for example, a quantum computer.

[0129] [Additional Notes] The present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0130] REFERENCE SIGNS LIST 10 Generation device 11 First acquisition unit (acquisition unit) 12 First generation unit (generation unit) 20 Estimation device 22 Estimation unit 23 Area ratio calculation unit 24 Depression number calculation unit

Claims

1. A method for estimating the likelihood of fine powder generation in polysilicon, comprising an estimation step of estimating the likelihood of fine powder generation in second polysilicon when vibration is applied from second imaging data including an image of second polysilicon that is the estimation target, using a first trained model generated using judgment data on whether or not an amount of fine powder generation due to vibration under specified conditions is equal to or less than a specified first threshold value in a first polysilicon to which vibration under specified conditions is applied, and first imaging data including an image of the first polysilicon as training data.

2. The estimation method according to claim 1, wherein the first imaging data includes a plurality of images of the first polysilicon taken from different directions, and the second imaging data includes a plurality of images of the second polysilicon taken from different directions.

3. The estimation method described in claim 1 or 2, wherein the first imaging data includes an image of a first surface of the first polysilicon and an image of a second surface including a back side of the first surface, and the second imaging data includes an image of a third surface of the second polysilicon and an image of a fourth surface including a back side of the third surface.

4. A method for producing polysilicon, comprising: an estimation step of estimating the likelihood of fine powder generation in second polysilicon when vibration is applied from second imaging data including an image of second polysilicon to be estimated, using a first trained model generated using as training data determination data on whether or not an amount of fine powder generation due to vibration under predetermined conditions is equal to or less than a predetermined first threshold value in first polysilicon to which vibration under predetermined conditions is applied, and first imaging data including an image of the first polysilicon; and a first sorting step of sorting the second polysilicon based on the likelihood of fine powder generation estimated in the estimation step.

5. The method for producing polysilicon according to claim 4, further comprising: an area ratio calculation step of calculating, from the second imaging data of third polysilicon, which is the second polysilicon estimated by the estimation step to have a surface state of the first polysilicon in which the amount of fine powder generation exceeds the first threshold, an area ratio of at least a portion of the surface of the third polysilicon, which is estimated to be similar to the surface state of the first polysilicon in which the amount of fine powder generation exceeds the first threshold, to a total area of ​​the surface of the third polysilicon, using a second learned model generated using as teacher data the surface state of the first polysilicon in which the amount of fine powder generation exceeds the first threshold; and a second sorting step of sorting the third polysilicon based on the total area and the area ratio.

6. The method for producing polysilicon described in claim 5, further comprising a dent number calculation step of calculating a dent number, which is the number of portions estimated to be dents formed on a surface of the third polysilicon, from the second imaging data of the third polysilicon using the second trained model, and the second classification step further classifies the third polysilicon based on the dent number in addition to the total area and the area ratio.

7. An estimation device for estimating the likelihood of fine powder generation in polysilicon, comprising an estimation unit that uses a first trained model generated using, as training data, determination data on whether or not an amount of fine powder generation due to vibration of a first polysilicon to which vibration of a predetermined condition is applied is equal to or less than a predetermined first threshold value, and first imaging data including an image of the first polysilicon, to estimate the likelihood of fine powder generation in the second polysilicon when vibration is applied from second imaging data including an image of the second polysilicon that is the estimation target.

8. A device for generating a trained model, comprising: an acquisition unit that acquires, as teacher data, determination data on whether an amount of fine powder generated due to vibration of a first polysilicon applied with predetermined conditions is equal to or less than a predetermined first threshold value, and first imaging data including an image of the first polysilicon; and a generation unit that generates a first trained model from input second imaging data including an image of a second polysilicon that is the subject of estimation, and outputs an estimation result of the likelihood of fine powder generation in the second polysilicon when vibration is applied.

9. A method for generating a trained model, comprising: an acquisition step of acquiring, as teacher data, determination data as to whether or not an amount of fine powder generated due to vibration of a first polysilicon applied with predetermined conditions is equal to or less than a predetermined first threshold value, and first imaging data including an image of the first polysilicon; and a generation step of generating, from input second imaging data including an image of a second polysilicon that is the subject of estimation, a first trained model that outputs an estimation result of the likelihood of fine powder generation in the second polysilicon when vibration is applied.