Silicon wafer cut-through judgment method and device, electronic equipment and storage medium

Abstract

The invention provides a silicon wafer cut-through judgment method and device, electronic equipment and a storage medium. The silicon wafer cut-through judgment method comprises the steps of obtaining a feed parameter of a current cutter feed position and a cutter discharge parameter of a current cutter discharge position; according to the feed parameter, obtaining a target feed classification matched with the feed parameter from pre-divided M feed classifications; judging whether the average value of the actual cutter feeding force is smaller than a cutting judgment threshold value corresponding to the target cutter feeding classification or not; wherein the cutting judgment threshold value is determined according to the minimum value of the mean values of the feeding force of the cutting times in the historical cutting data corresponding to the target cutting feed classification; or the target feed classification is determined according to the maximum average value in the feed force average values of non-cutting times in the historical cutting data corresponding to the target feed classification; and if yes, determining that the silicon rod corresponding to the current cutting time is in a cut-through state. Whether the silicon rod is cut through or not can be automatically judged, and the judgment efficiency is improved.

Description

Technical Field

[0001] The present application relates to the technical field of silicon wafer cutting. Specifically, it relates to a method, device, electronic device and storage medium for judging silicon wafer penetration. Background Art

[0002] At present, the slicing machine adopts the reverse cutting process. One disadvantage of the reverse cutting process is that it will form a wire bow, that is, the cutting depths of the steel wires on the right and left sides of the silicon rod are different, resulting in an uneven wire mesh. Therefore, after normal shutdown according to the process recipe during cutting, there will be a problem that the whole silicon rod is not cut through.

[0003] At present, the method for judging whether the silicon rod is cut through is to manually open the slicing machine, use a strong flashlight to irradiate and observe the wire mesh, and visually observe whether it is cut through. This method relies on manual operation and has low judgment efficiency. Summary of the Invention

[0004] The purpose of the embodiments of the present application is to provide a method, device, electronic device and computer-readable storage medium for judging silicon wafer penetration, so as to solve the problem of low judgment efficiency when judging whether the silicon rod is cut through in the related art.

[0005] The embodiments of the present application provide a method for judging silicon wafer penetration, including: obtaining the feed parameters at the feed position of the current cutting pass and the discharge parameters at the discharge position of the current cutting pass, where the feed parameters include the average actual feed force at the feed position of the current cutting pass, and the discharge parameters include the average actual discharge force at the discharge position of the current cutting pass; obtaining a target feed classification that matches the feed parameters from M pre-divided feed classifications; where the M feed classifications are determined according to the historical feed parameters at the historical feed positions of previous cutting passes, and M is an integer greater than 1; judging whether the average actual discharge force is less than the additional cutting judgment threshold corresponding to the target feed classification; where the additional cutting judgment threshold is determined according to the minimum value of the average feed force of the additional cutting passes in the historical cutting data corresponding to the target feed classification; or determined according to the maximum value of the average feed force of the non-additional cutting passes in the historical cutting data corresponding to the target feed classification; if it is less than, it is determined that the silicon rod corresponding to the current cutting pass is in a cut-through state.

[0006] In the above implementation process, the additional cutting determination threshold determined by using the minimum value among the average values of the feed forces during the additional cutting tool passes in the historical cutting data corresponding to the target feed classification, or the additional cutting determination threshold determined by using the maximum value among the average values of the feed forces during the non-additional cutting tool passes in the historical cutting data corresponding to the target feed classification, is used to judge the average value of the feed force when the tool exits for the current tool pass belonging to the target feed classification. Furthermore, when the average value of the feed force when the tool exits for the current tool pass is less than the additional cutting determination threshold, it is determined that the silicon rod corresponding to the current tool pass is in a fully cut state. Since the minimum value among the average values of the feed forces during the additional cutting tool passes can reflect the minimum average feed force required historically in the case of additional cutting, and the maximum value among the average values of the feed forces during the non-additional cutting tool passes can reflect the maximum average feed force required historically in the case of no additional cutting, the additional cutting determination threshold determined based on the minimum value among the average values of the feed forces during the additional cutting tool passes, or based on the maximum value among the average values of the feed forces during the non-additional cutting tool passes, can be used as a relatively accurate basis for judging whether additional cutting is required for the current tool pass, that is, it can relatively accurately determine whether the silicon rod corresponding to the current tool pass is in a fully cut state. Thus, compared with the related art, automatic judgment of whether the silicon rod is fully cut can be realized, improving the judgment efficiency.

[0007] In addition, in the above implementation method, M feed classifications are pre-divided based on the historical feed parameters of the historical tool pass feed positions. Then, the target feed classification is determined from the M feed classifications based on the feed parameters of the current tool pass. Furthermore, the additional cutting determination threshold corresponding to the target feed classification is used for judgment. That is, in the above implementation method, the cutting situation is also classified according to the feed parameters, and then the additional cutting determination threshold of the feed classification that best matches the cutting situation of the current tool pass is used for judgment, thereby improving the reliability of the additional cutting determination threshold and further improving the judgment accuracy of whether the silicon rod is in a fully cut state.

[0008] Furthermore, obtaining the target feed classification that matches the feed parameters from the pre-divided M feed classifications according to the feed parameters includes: if the feed parameters include the current machine tool and the current silicon wafer number corresponding to the current tool pass, then obtaining the target feed classification from the M feed classifications according to the current machine tool, the current silicon wafer number, and the actual average feed force when the tool enters.

[0009] It can be understood that there may be differences in the cutting conditions corresponding to different machines and wafers of different specifications. In the above implementation, obtaining the target feed classification from the M feed classifications based on the current machine, the current wafer number, and the average actual feed force can make the target feed classification the feed classification that best matches the current machine, the current wafer number, and the average actual feed force. Furthermore, the additional cutting determination threshold is more in line with the cutting conditions of the current tool pass, improving the reliability of the additional cutting determination threshold and further enhancing the accuracy of the judgment on whether the silicon rod is in a cut-through state.

[0010] Further, the steps for obtaining the M feed classifications include: obtaining the historical feed parameters of the historical tool pass feed positions; wherein, the historical feed parameters include the historical machine, the historical wafer, and the average historical feed force corresponding to the historical tool pass feed position; wherein, the number of historical machines is multiple, and the number of historical wafers is multiple; grouping according to the historical machine and the historical wafer to obtain a set of machine-wafer groupings; for each machine-wafer grouping in the set of machine-wafer groupings, classifying using the average historical feed force of the machine-wafer grouping to obtain N feed classifications for the machine-wafer grouping; wherein, N is an integer greater than 1; and obtaining the M feed classifications according to the N feed classifications of each machine-wafer grouping.

[0011] In the above implementation, grouping is performed through the historical machine and the historical wafer corresponding to the historical tool pass feed position to obtain a set of machine-wafer groupings, and then for each machine-wafer grouping, classification is carried out using the average historical feed force of the machine-wafer grouping, so that N feed classifications for each machine-wafer grouping can be obtained, and thus a total of M feed classifications can be obtained. The M feed classifications obtained in this way can distinguish different cutting conditions corresponding to different machines and wafers of different specifications, making the feed classification more comprehensive and detailed. Furthermore, after determining the target feed classification, the additional cutting determination threshold of the target feed classification is more in line with the cutting conditions of the current tool pass.

[0012] Further, for each machine-wafer grouping in the set of machine-wafer groupings, classifying using the average historical feed force of the machine-wafer grouping to obtain N feed classifications for the machine-wafer grouping includes: for each machine-wafer grouping, classifying the average historical feed force of the machine-wafer grouping using a quantile classification method to obtain N feed classifications for the machine-wafer grouping.

[0013] Further, obtaining the target feed classification from the M feed classifications according to the current machine tool, the current wafer number, and the average actual feed force includes: obtaining a target machine tool wafer group corresponding to the current tool feed from the machine tool wafer grouping set according to the current machine tool and the current wafer number; determining, according to the average actual feed force, a feed classification corresponding to the current tool feed from the N feed classifications of the target machine tool wafer group as the target feed classification.

[0014] In the above implementation, first, according to the current machine tool and the current wafer number, a target machine tool wafer group corresponding to the current tool feed is obtained from the machine tool wafer grouping set, and then, according to the average actual feed force, a feed classification corresponding to the current tool feed is determined from the N feed classifications of the target machine tool wafer group as the target feed classification. The target feed classification determined in this way is more in line with the cutting situation of the current tool feed.

[0015] Further, determining, according to the average actual feed force, a feed classification corresponding to the current tool feed from the N feed classifications of the target machine tool wafer group as the target feed classification includes: obtaining the average feed force range of each feed classification in the N feed classifications of the target machine tool wafer group; determining, from the N feed classifications of the target machine tool wafer group, a feed classification whose average feed force range includes the average actual feed force as the target feed classification.

[0016] In the above implementation, by obtaining the average feed force range of each feed classification in the N feed classifications of the target machine tool wafer group and taking the feed classification whose average feed force range includes the actual quantile value as the target feed classification, the target feed classification can be quickly determined, and the solution is simple and reliable to implement.

[0017] Further, determining whether the average actual tool-out force is less than the additional cutting determination threshold corresponding to the target feed classification includes: if no additional cutting tool feeds exist in all historical tool feeds corresponding to the target feed classification, determining the additional cutting determination threshold as the maximum value of the average value corresponding to the target feed classification; determining whether the average actual tool-out force is less than the maximum value of the average value corresponding to the target feed classification.

[0018] In the above implementation, when there is no additional cutting pass among all the historical passes corresponding to the target feed classification, the maximum value among the average feed forces of non-additional cutting passes can reflect the maximum average feed force required historically without additional cutting. Using this as the additional cutting determination threshold can more accurately determine whether the silicon rod corresponding to the current pass is in a cut-through state. Therefore, compared with the related art, it is possible to achieve an automated determination of whether the silicon rod is cut through, improving the determination efficiency.

[0019] Further, determining whether the actual feed force average of the actual tool exit is less than the additional cutting determination threshold corresponding to the target feed classification includes: if there is an additional cutting pass among all the historical passes corresponding to the target feed classification, determining the additional cutting determination threshold as the minimum value among the average feed forces of the target feed classification; and determining whether the actual feed force average of the actual tool exit is less than the minimum value among the average feed forces of the target feed classification.

[0020] In the above implementation, when there is an additional cutting pass among all the historical passes corresponding to the target feed classification, the minimum value among the average feed forces of the additional cutting passes can reflect the minimum average feed force required historically when additional cutting is needed. Using this as the additional cutting determination threshold can more accurately determine whether additional cutting is required for the current pass, that is, it can more accurately determine whether the silicon rod corresponding to the current pass is in a cut-through state. Therefore, compared with the related art, it is possible to achieve an automated determination of whether the silicon rod is cut through, improving the determination efficiency.

[0021] Further, if no target feed classification matching the feed parameters is obtained among the M feed classifications, the method further includes: determining that the silicon rod corresponding to the current pass is in a non-cut-through state.

[0022] In the above implementation, if there is no matching target feed classification for the current pass among the M feed classifications, this means that the feed situation of the current pass is not within the historical situation. Then, it can be considered that there is a feed anomaly, such as an abnormal feed speed, etc. At this time, determining that the silicon rod corresponding to the current pass is in a non-cut-through state can prevent the situation where the silicon rod is not cut through when performing additional cutting.

[0023] Further, the method further includes: determining whether the current pass has an abnormal state, where the abnormal state includes a wire break state and a state where the downtime exceeds a preset downtime threshold; if the current pass has the abnormal state, determining that the silicon rod corresponding to the current pass is in the non-cut-through state.

[0024] In the above implementation, when a wire break occurs in the current cutting pass or the downtime exceeds the preset duration threshold, it indicates that an abnormality has occurred in the cutting process. At this time, it is determined that the silicon rod corresponding to the current pass is in a non-penetrated state, and additional cutting can be performed to prevent the silicon rod from not being cut through.

[0025] An embodiment of the present application also provides a silicon wafer penetration judgment device, including: an acquisition module, configured to acquire the feed parameters of the feed position of the current pass and the feed parameters of the cut position of the current pass, where the feed parameters include the average actual feed force at the feed position of the current pass, and the cut parameters include the average actual cut force at the cut position of the current pass; the acquisition module is further configured to obtain, according to the feed parameters, a target feed classification that matches the feed parameters from M pre-divided feed classifications; where the M feed classifications are determined according to the historical feed parameters of the historical feed positions of the historical passes, and M is an integer greater than 1; a judgment module, configured to judge whether the average actual cut force is less than the additional cutting judgment threshold corresponding to the target feed classification; if less, it is determined that the silicon rod corresponding to the current pass is in a penetrated state; where the additional cutting judgment threshold is determined according to the minimum value of the average feed force of the additional cutting passes in the historical cutting data corresponding to the target feed classification; or determined according to the maximum value of the average feed force of the non-additional cutting passes in the historical cutting data corresponding to the target feed classification.

[0026] An embodiment of the present application also provides an electronic device, including a processor and a memory, where a computer program is stored in the memory, and the processor executes the computer program to implement any one of the above silicon wafer penetration judgment methods.

[0027] An embodiment of the present application also provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, and when the computer program is executed by at least one processor, any one of the above silicon wafer penetration judgment methods is implemented. Description of the Drawings

[0028] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required to be used in the embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0029] Figure 1 It is a schematic flow chart of a silicon wafer penetration judgment method provided by an embodiment of the present application;

[0030] Figure 2A schematic flowchart of the acquisition process of M feed classifications provided by an embodiment of the present application;

[0031] Figure 3 Another schematic flowchart of the acquisition process of M feed classifications provided by an embodiment of the present application;

[0032] Figure 4 A schematic structural diagram of a silicon wafer penetration judgment device provided by an embodiment of the present application;

[0033] Figure 5 A schematic structural diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0034] Next, the technical solutions in the embodiments of the present application will be described with reference to the accompanying drawings in the embodiments of the present application.

[0035] To solve the problem of low judgment efficiency in judging whether a silicon rod is penetrated in the related art, an embodiment of the present application provides a silicon wafer penetration judgment method. It can be seen in Figure 1 shown as Figure 1 A schematic flowchart of the silicon wafer penetration judgment method provided by an embodiment of the present application, including:

[0036] S101: Obtain the feed parameters at the feed position of the current tool feed and the retraction parameters at the retraction position of the current tool feed.

[0037] In the embodiment of the present application, the feed parameters include the average actual feed force at the feed position of the current tool feed, and the retraction parameters include the average actual retraction force at the retraction position of the current tool feed.

[0038] In the embodiment of the present application, the actual feed force at the feed position and the actual retraction force at the retraction position can be obtained by setting sensors at the feed position and the retraction position. In addition, the actual feed force at the feed position and the actual retraction force at the retraction position can also be calculated based on the radius of the diamond wire used for cutting, the cutting speed at the feed position, the cutting speed at the retraction position, etc. The present application does not limit the acquisition methods of the actual feed force at the feed position and the actual retraction force at the retraction position.

[0039] After obtaining the actual feed force at the feed position and the actual retraction force at the retraction position, the average actual feed force at the feed position and the average actual retraction force at the retraction position can be obtained by respectively calculating the average values of the actual feed force at the feed position and the actual retraction force at the retraction position.

[0040] In the embodiment of the present application, the actual feed force at the feed position can be the feed force within a certain range starting from the feed position. For example, it can be the feed force collected in the interval from 10 mm to 30 mm of the feed position. Similarly, the actual feed force at the tool withdrawal position can be the feed force within a certain range starting from the tool withdrawal position.

[0041] S102: Obtain the target feed classification that matches the feed parameters from the pre-divided M feed classifications according to the feed parameters.

[0042] In the embodiment of the present application, the M feed classifications are determined according to the historical feed parameters of the historical feed positions of the tool passes, and M is an integer greater than 1. The historical feed parameters include the average value of the historical feed forces at the feed positions of each tool pass in history.

[0043] In the embodiment of the present application, the value of M can be set according to actual needs. For example, it can be set to 5. It can be understood that the larger the value of M is set, the more detailed the feed classification will be, but the corresponding workload of obtaining the M feed classifications will be greater.

[0044] In the embodiment of the present application, the feed forces at the feed positions of each tool pass in history can be obtained, and then the average value of the historical feed forces at the feed positions of each tool pass can be obtained based on the feed forces at the feed positions of each tool pass.

[0045] In an embodiment of the present application, as shown in Figure 2 The M feed classifications can be obtained through the following method:

[0046] S201: Obtain the historical feed parameters of the historical feed positions of the tool passes. Among them, the historical feed parameters include the average value of the historical feed forces corresponding to the historical feed positions of the tool passes.

[0047] S202: Classify using the average value of the historical feed forces to obtain M feed classifications.

[0048] Optionally, when classifying using the average value of the historical feed forces, the average value of the historical feed forces can be classified by using the quantile classification method to obtain M feed classifications. For example, the average value of the historical feed forces can be sorted in ascending order, and then the sorted average value of the historical feed forces can be evenly divided into M parts according to the number of the average value of the historical feed forces to obtain M feed classifications. It can be understood that in the embodiment of the present application, other classification methods can also be used to classify the average value of the historical feed forces to obtain M feed classifications, and the embodiment of the present application does not limit this.

[0049] In the above embodiments, when obtaining a target feed classification that matches the feed parameters from M pre-divided feed classifications according to the feed parameters, the average feed force range of each feed classification in the M feed classifications can be obtained, and then, from the M feed classifications, the feed classification whose average feed force range includes the actual average feed force is determined as the target feed classification.

[0050] In the embodiments of the present application, the average feed force range of each feed classification can be determined according to the quantile value when performing feed classification. The so-called quantile value refers to the value for dividing the average feed force.

[0051] For example, assume that the historical average feed forces are a1, a2, a3, a4, a5, a6 in sequence, and a1 < a2 < a3 < a4 < a5 < a6. Assume that the quantile classification method is used for quantile division, and M is set to 2. Then, the quantile value corresponding to the 50% quantile can be calculated as: (a3 + a4) / 2. Then, the historical average feed force corresponding to the first feed classification is (a1, a2, a3), and the corresponding average feed force range is [a1, (a3 + a4) / 2]. The historical average feed force corresponding to the second feed classification is (a4, a5, a6), and the corresponding average feed force range is ((a3 + a4) / 2, a6]. Assume that the actual average feed force of the current tool pass is A, and a4 < A < a5. Then, the second feed classification can be determined as the target feed classification.

[0052] In another embodiment of the present application, reference can be made to Figure 3 as shown, the M feed classifications can also be obtained in the following manner:

[0053] S301: Obtain the historical feed parameters of the historical tool pass feed position.

[0054] Among them, the historical feed parameters include the historical machine tool, the historical silicon wafer, and the historical average feed force corresponding to the historical tool pass feed position. Among them, the number of historical machine tools is multiple, and the number of historical silicon wafers is multiple.

[0055] S302: Group according to the historical machine tool and the historical silicon wafer to obtain a machine tool silicon wafer grouping set.

[0056] For example, assume that there are m (m ≥ 2) machine tools and n (n ≥ 2) silicon wafer specifications. Then, the historical average feed forces of the same silicon wafer specification corresponding to each machine tool can be divided into a machine tool silicon wafer group. Assume that these m machine tools can all cut silicon wafers of these n silicon wafer specifications. Then, m * n machine tool silicon wafer groups can be obtained.

[0057] S303: For each grouped wafer set of a machine tool, classify them using the historical feed force average value of the grouped wafer set of the machine tool to obtain N feed classifications of the grouped wafer set of the machine tool. Based on the N feed classifications of each grouped wafer set of the machine tool, M feed classifications can be obtained. Here, N is an integer greater than 1.

[0058] It can be understood that assuming there are m machine tools, and these m machine tools can all cut wafers of n wafer specifications, then m * n grouped wafer sets of the machine tool can be obtained, and a total of m * n * N feed classifications can be obtained. That is, M is equal to m * n * N.

[0059] In the embodiments of the present application, the value of N can be set according to actual needs. For example, it can be set to 5. It can be understood that the larger the value of N is set, the more detailed the feed classification will be, but the corresponding workload for obtaining M feed classifications will be greater.

[0060] Optionally, for each grouped wafer set of the machine tool, a quantile classification method can be used to classify the historical feed force average value of the grouped wafer set of the machine tool to obtain N feed classifications of the grouped wafer set of the machine tool.

[0061] Correspondingly, the process of obtaining the target feed classification matching the feed parameters from the pre-divided M feed classifications can include: If the feed parameters include the current machine tool and the current wafer number corresponding to the current tool pass, then based on the current machine tool, the current wafer number, and the actual feed force average value, obtain the target feed classification from the M feed classifications.

[0062] Exemplarily, based on the current machine tool and the current wafer number, the target grouped wafer set corresponding to the current tool pass can be obtained from the grouped wafer set of the machine tool, and then based on the actual feed force average value, determine the feed classification corresponding to the current tool pass from the N feed classifications of the target grouped wafer set as the target feed classification.

[0063] In the embodiments of the present application, the wafer number is the number of the wafer specification, which reflects what specification of wafer is produced. Therefore, based on the current machine tool and the current wafer number, the target grouped wafer set corresponding to the current tool pass can be obtained from the grouped wafer set of the machine tool.

[0064] In the above embodiments, when determining the feed classification corresponding to the current tool pass from the N feed classifications of the target grouped wafer set as the target feed classification based on the actual feed force average value, the feed force average value range of each feed classification in the N feed classifications of the target grouped wafer set can be obtained, and then from the N feed classifications of the target grouped wafer set, determine the feed classification whose feed force average value range contains the actual feed force average value as the target feed classification.

[0065] Among them, the determination method of the average value range of the feed force can be referred to the description in the previous text.

[0066] It can be understood that there may be differences in the cutting conditions corresponding to different machines and silicon wafers of different specifications. For example, there may be certain differences in the operating speeds of different machines. In addition, for silicon wafers of different specifications, there may be certain differences in the process requirements (such as cutting speed, etc.) during cutting. Through the above embodiments, the differences between different machines and silicon wafers of different specifications are considered when dividing the feed classification, so as to classify the historical feed force average values of the same machine-silicon wafer group based on the machine-silicon wafer grouping, making the classification more detailed. At the same time, since a feed classification corresponds to the historical cutting data of a machine-silicon wafer group, the additional cutting determination threshold corresponding to a feed classification is obtained based on the historical cutting data saved when cutting the same specification of silicon wafers on the same machine, which makes the additional cutting determination threshold corresponding to the feed classification more accurate. Correspondingly, when determining the target feed classification, first determine the target machine-silicon wafer group according to the current machine and the current silicon wafer number of the current tool feed, and then determine the feed classification corresponding to the current tool feed among the N feed classifications corresponding to the target machine-silicon wafer group as the target feed classification, and then use the additional cutting determination threshold corresponding to the target feed classification for determination, so as to obtain a more accurate judgment result on whether the silicon rod is cut through.

[0067] S103: Determine whether the average value of the actual tool feed force is less than the additional cutting determination threshold corresponding to the target feed classification.

[0068] In the embodiments of the present application, the additional cutting determination threshold is determined according to the minimum value of the average values of the feed forces of the additional cutting tool feeds in the historical cutting data corresponding to the target feed classification; or determined according to the maximum value of the average values of the feed forces of the non-additional cutting tool feeds in the historical cutting data corresponding to the target feed classification.

[0069] In some embodiments, the process of determining whether the average value of the actual tool feed force is less than the additional cutting determination threshold corresponding to the target feed classification may include:

[0070] If there is an additional cutting tool feed among all the historical tool feeds corresponding to the target feed classification, determine the additional cutting determination threshold as the minimum value of the average values corresponding to the target feed classification; determine whether the average value of the actual tool feed force is less than the minimum value of the average values corresponding to the target feed classification.

[0071] In some embodiments, the process of determining whether the average value of the actual tool feed force is less than the additional cutting determination threshold corresponding to the target feed classification may also include:

[0072] If there is no additional cutting tool pass among all the historical tool passes corresponding to the target feed classification, determine the additional cutting determination threshold as the maximum value of the non-additional cutting tool exit mean corresponding to the target feed classification; determine whether the actual tool exit feed force mean is less than the maximum value of the mean corresponding to the target feed classification.

[0073] S104: If the actual tool exit feed force mean is less than the additional cutting determination threshold corresponding to the target feed classification, determine that the silicon rod corresponding to the current tool pass is in a fully cut state.

[0074] In the embodiment of the present application, if the actual tool exit feed force mean is greater than or equal to the additional cutting determination threshold corresponding to the target feed classification, it can be determined that the silicon rod corresponding to the current tool pass is in a non-fully cut state. In the case of determining that the silicon rod corresponding to the current tool pass is in a non-fully cut state, the slicing machine can be controlled to perform additional cutting on the silicon rod.

[0075] Optionally, in some embodiments of the present application, if no target feed classification matching the feed parameters is obtained among the M feed classifications, it can be determined that the silicon rod corresponding to the current tool pass is in a non-fully cut state. It can be understood that if there is no matching target feed classification for the current tool pass among the M feed classifications, this means that the feed situation of the current tool pass is not within the historical situation, then it can be considered that there is a feed anomaly, such as an abnormal feed speed, etc. At this time, it is determined that the silicon rod corresponding to the current tool pass is in a non-fully cut state, so that additional cutting can prevent the situation that the silicon rod is not fully cut.

[0076] Optionally, in some embodiments of the present application, it can be determined whether the current tool pass has an abnormal state. If the current tool pass has an abnormal state, it is determined that the silicon rod corresponding to the current tool pass is in the non-fully cut state. Among them, the abnormal state includes a wire break state and a state where the shutdown duration exceeds a preset duration threshold. When there is a wire break in the current cutting tool pass, or when the shutdown duration exceeds the preset duration threshold, it indicates that there is an abnormality in the cutting process that is sufficient to cause the cutting process to be unable to continue. At this time, it is determined that the silicon rod corresponding to the current tool pass is in a non-fully cut state, so that additional cutting can prevent the situation that the silicon rod is not fully cut.

[0077] Among them, the preset duration threshold can be set by the engineer according to the actual situation. For example, it can be set to 30 minutes.

[0078] The silicon wafer cut-through judgment method provided in the embodiment of the present application uses the minimum mean value of the feed force mean values ​​of the added cutting times in the historical cutting data corresponding to the target feed classification to determine the added cutting judgment threshold, or uses the maximum mean value of the feed force mean values ​​of the non-added cutting times in the historical cutting data corresponding to the target feed classification to determine the added cutting judgment threshold, to judge the mean value of the feed force of the current cut that belongs to the target feed classification, and then when the mean value of the feed force of the current cut is less than the added cutting judgment threshold, it is determined that the silicon rod corresponding to the current cut is in a cut-through state. Since the minimum mean value of the feed force average values ​​of the additional cutting times can reflect the minimum mean feed force required in history when additional cutting is required, and the maximum mean value of the feed force average values ​​of the non-additional cutting times can reflect the maximum mean feed force required in history when additional cutting is not required, the additional cutting judgment threshold determined based on the minimum mean value of the feed force average values ​​of the additional cutting times, or based on the maximum mean value of the feed force average values ​​of the non-additional cutting times, can be used more accurately as a basis for judging whether additional cutting is required for the current cut, that is, it can more accurately judge whether the silicon rod corresponding to the current cut is in a cut-through state. Compared with the related technology, it can realize automatic judgment of whether the silicon rod is cut through, thereby improving the judgment efficiency.

[0079] In addition, in the above implementation, M feed categories are pre-divided based on the historical feed parameters of the historical knife feed positions, and then the target feed category is determined from the M feed categories based on the current knife feed parameters, and then the additional cutting determination threshold corresponding to the target feed category is used for judgment. That is, in the above implementation, the cutting situation is also classified according to the feed parameters, and then the additional cutting determination threshold of the feed category that best matches the current knife cutting situation is implemented for judgment, thereby improving the reliability of the additional cutting determination threshold, and further improving the accuracy of judging whether the silicon rod is in a cut-through state.

[0080] Based on the same inventive concept, the present application also provides a silicon wafer cut-through judgment device 400. Figure 4 As shown, Figure 4 Shows the use of Figure 1 The silicon wafer through-cut judgment device of the method shown. It should be understood that the specific functions of the device 400 can be referred to the description above, and the detailed description is appropriately omitted here to avoid repetition. The device 400 includes at least one software function module that can be stored in a memory in the form of software or firmware or fixed in the operating system of the device 400. Specifically:

[0081] See also Figure 4 As shown, the device 400 includes: an acquisition module 401 and a judgment module 402. Among them:

[0082] The obtaining module 401 is configured to obtain the feed parameters at the current tool feed position and the retraction parameters at the current tool retraction position. The feed parameters include the average actual feed force at the current tool feed position, and the retraction parameters include the average actual retraction force at the current tool retraction position.

[0083] The obtaining module 401 is further configured to obtain a target feed classification that matches the feed parameters from M pre-divided feed classifications according to the feed parameters. The M feed classifications are determined according to the historical feed parameters at the historical tool feed positions, and M is an integer greater than 1.

[0084] The decision module 402 is configured to determine whether the average actual retraction force is less than the additional cutting determination threshold corresponding to the target feed classification. If it is less, it is determined that the silicon rod corresponding to the current tool is in a cut-through state. The additional cutting determination threshold is determined according to the minimum value of the average feed forces of the additional cutting tool passes in the historical cutting data corresponding to the target feed classification, or according to the maximum value of the average feed forces of the non-additional cutting tool passes in the historical cutting data corresponding to the target feed classification.

[0085] In a feasible implementation manner of the embodiment of the present application, the obtaining module 401 is specifically configured to: if the feed parameters include the current machine tool and the current silicon wafer number corresponding to the current tool pass, obtain the target feed classification from the M feed classifications according to the current machine tool, the current silicon wafer number, and the average actual feed force.

[0086] In the above feasible implementation manner, the obtaining module 401 is further configured to obtain the M feed classifications according to the following obtaining steps:

[0087] Obtain the historical feed parameters at the historical tool feed positions. The historical feed parameters include the historical machine tool, the historical silicon wafer, and the average historical feed force corresponding to the historical tool feed position. The number of historical machine tools is multiple, and the number of historical silicon wafers is multiple.

[0088] Group according to the historical machine tool and the historical silicon wafer to obtain a machine tool-silicon wafer grouping set.

[0089] For each machine tool-silicon wafer grouping in the machine tool-silicon wafer grouping set, classify using the average historical feed force of the machine tool-silicon wafer grouping to obtain N feed classifications of the machine tool-silicon wafer grouping. N is an integer greater than 1.

[0090] Obtain the M feed classifications according to the N feed classifications of each machine tool-silicon wafer grouping.

[0091] In the above feasible implementation manner, the obtaining module 401 is specifically configured to, for each grouped wafer of a machine tool, classify the historical average feed force of the grouped wafers of the machine tool by using a quantile classification method, so as to obtain N feed classifications of the grouped wafers of the machine tool.

[0092] In the above feasible implementation manner, the obtaining module 401 is specifically configured to, according to the current machine tool and the current wafer number, obtain a target grouped wafer of the machine tool corresponding to the current tool feed from the grouped wafer set of the machine tool, and determine, according to the actual average feed force, a feed classification corresponding to the current tool feed from the N feed classifications of the target grouped wafer of the machine tool as the target feed classification.

[0093] In the above feasible implementation manner, the obtaining module 401 is specifically configured to obtain the average feed force range of each feed classification in the N feed classifications of the target grouped wafer of the machine tool, and determine, from the N feed classifications of the target grouped wafer of the machine tool, a feed classification whose average feed force range includes the actual average feed force as the target feed classification.

[0094] In a feasible implementation manner of the embodiment of the present application, the decision module 402 is specifically configured to: if there is no additional cutting tool feed in all historical tool feeds corresponding to the target feed classification, determine the additional cutting determination threshold as the maximum value of the average value corresponding to the target feed classification; and determine whether the actual feed force of the tool out is less than the maximum value of the average value corresponding to the target feed classification.

[0095] In a feasible implementation manner of the embodiment of the present application, the decision module 402 is specifically configured to: if there is an additional cutting tool feed in all historical tool feeds corresponding to the target feed classification, determine the additional cutting determination threshold as the minimum value of the average value corresponding to the target feed classification; and determine whether the actual feed force of the tool out is less than the minimum value of the average value corresponding to the target feed classification.

[0096] In a feasible implementation manner of the embodiment of the present application, the decision module 402 is further configured to, if no target feed classification matching the feed parameters is obtained among the M feed classifications, determine that the silicon rod corresponding to the current tool feed is in a non-penetrating state.

[0097] In a feasible implementation manner of the embodiment of the present application, the decision module 402 is further configured to determine whether the current tool feed has an abnormal state, and if the current tool feed has the abnormal state, determine that the silicon rod corresponding to the current tool feed is in the non-penetrating state. Wherein, the abnormal state includes a wire break state and a state in which the shutdown duration exceeds a preset duration threshold.

[0098] It should be understood that, for the sake of brevity of description, the content described in some method embodiments will not be repeated in the apparatus embodiments.

[0099] Based on the same inventive concept, an embodiment of the present application also provides an electronic device. Refer to Figure 5 as shown, which includes a processor 501 and a memory 502. Among them:

[0100] A computer program is stored in the memory 502, and the processor 501 is configured to execute one or more computer programs stored in the memory 502 to implement the above-mentioned silicon wafer penetration judgment method.

[0101] It can be understood that the processor 501 can be a processor core or a processor chip, or other circuits that can be programmed and run. The memory 502 can be RAM (Random Access Memory), ROM (Read-Only Memory), flash memory, etc., but this is not a limitation.

[0102] It can also be understood that Figure 5 the structure shown is only schematic, and the electronic device may further include more or fewer components than those shown in Figure 5 or have a different configuration from that shown in Figure 5 For example, it may also have an internal communication bus for realizing communication between the processor 501 and the memory 502; for another example, it may also have an external communication interface, such as a USB (Universal Serial Bus) interface, a CAN (Controller Area Network) bus interface, etc.; for another example, it may also have an information display component such as a display screen, but this is not a limitation.

[0103] Based on the same inventive concept, this embodiment also provides a computer-readable storage medium, such as a floppy disk, an optical disc, a hard disk, a flash memory, a USB flash drive, an SD (Secure Digital Memory Card) card, an MMC (Multimedia Card) card, etc. One or more computer programs for implementing the above-mentioned steps are stored in the computer-readable storage medium, and these one or more computer programs can be executed by one or more processors to implement the above-mentioned silicon wafer penetration judgment method. Details are not described herein again.

[0104] The embodiments in the embodiments of the present application can be combined with each other without conflict to obtain new embodiments.

[0105] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For another example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some communication interfaces. The indirect coupling or communication connection of the devices or units can be in electrical, mechanical or other forms.

[0106] In addition, the units described as separate components may or may not be physically separated.

[0107] Furthermore, in each embodiment of the present application, the functional modules can be integrated together to form an independent part, or each module can exist alone, or two or more modules can be integrated to form an independent part.

[0108] In this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0109] In this article, "a plurality of" means two or more.

[0110] The above are only the embodiments of the present application and are not used to limit the protection scope of the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for judging silicon wafer penetration, characterized in that, Including: Obtain the feed parameters of the current tool feed position and the retraction parameters of the current tool retraction position. Among them, the feed parameters include the average actual feed force at the current tool feed position, and the retraction parameters include the average actual retraction force at the current tool retraction position; According to the feed parameters, obtain the target feed classification that matches the feed parameters from M pre-divided feed classifications; among them, the M feed classifications are determined according to the historical feed parameters of the historical tool feed positions, and M is an integer greater than 1; Judge whether the average actual retraction force is less than the additional cutting determination threshold corresponding to the target feed classification; among them, the additional cutting determination threshold is determined according to the minimum value of the average feed force of the additional cutting tool passes in the historical cutting data corresponding to the target feed classification; or determined according to the maximum value of the average feed force of the non-additional cutting tool passes in the historical cutting data corresponding to the target feed classification; If it is less, it is determined that the silicon rod corresponding to the current tool pass is in a cut-through state.

2. The method according to claim 1, characterized in that According to the feed parameters, obtaining the target feed classification that matches the feed parameters from M pre-divided feed classifications includes: If the feed parameters include the current machine tool and the current silicon wafer number corresponding to the current tool pass, then according to the current machine tool, the current silicon wafer number, and the average actual feed force, obtain the target feed classification from the M feed classifications.

3. The method according to claim 2, wherein The obtaining steps of the M feed classifications include: Obtain the historical feed parameters of the historical tool feed position; among them, the historical feed parameters include the historical machine tool, the historical silicon wafer, and the average historical feed force corresponding to the historical tool feed position; among them, the number of historical machine tools is multiple, and the number of historical silicon wafers is multiple; Group according to the historical machine tool and the historical silicon wafer to obtain a machine tool-silicon wafer grouping set; For each machine tool-silicon wafer grouping in the machine tool-silicon wafer grouping set, classify using the average historical feed force of the machine tool-silicon wafer grouping to obtain N feed classifications of the machine tool-silicon wafer grouping; where N is an integer greater than 1; According to the N feed classifications of each machine tool-silicon wafer grouping, obtain the M feed classifications.

4. The method according to claim 3, characterized in that, For each machine tool-silicon wafer grouping in the machine tool-silicon wafer grouping set, classifying using the average historical feed force of the machine tool-silicon wafer grouping to obtain N feed classifications of the machine tool-silicon wafer grouping includes: For each machine tool-silicon wafer grouping, classify the average historical feed force of the machine tool-silicon wafer grouping using the quantile classification method to obtain N feed classifications of the machine tool-silicon wafer grouping.

5. The method according to claim 4, wherein According to the current machine tool, the current silicon wafer number, and the average actual feed force, obtaining the target feed classification from the M feed classifications includes: According to the current machine tool and the current silicon wafer number, obtain the target machine tool-silicon wafer grouping corresponding to the current tool pass from the machine tool-silicon wafer grouping set; According to the average actual feed force, determine the feed classification corresponding to the current tool pass from the N feed classifications of the target machine tool-silicon wafer grouping as the target feed classification.

6. The method according to claim 5, characterized in that, Determining the feed classification corresponding to the current tool pass as the target feed classification from the N feed classifications of the silicon wafer grouping of the target machine tool according to the actual feed force mean value of the current tool pass includes: Obtaining the feed force mean value range of each feed classification among the N feed classifications of the silicon wafer grouping of the target machine tool; Determining, from the N feed classifications of the silicon wafer grouping of the target machine tool, the feed classification whose feed force mean value range contains the actual feed force mean value of the current tool pass as the target feed classification.

7. The method according to claim 1, characterized in that, Judging whether the actual out-feed force mean value is less than the additional cutting determination threshold corresponding to the target feed classification, including: If there is no additional cutting tool pass among all the historical tool passes corresponding to the target feed classification, determining the additional cutting determination threshold as the maximum value of the mean values corresponding to the target feed classification; Judging whether the actual out-feed force mean value is less than the maximum value of the mean values corresponding to the target feed classification.

8. The method according to claim 1, wherein Judging whether the actual out-feed force mean value is less than the additional cutting determination threshold corresponding to the target feed classification, including: If there is an additional cutting tool pass among all the historical tool passes corresponding to the target feed classification, determining the additional cutting determination threshold as the minimum value of the mean values corresponding to the target feed classification; Judging whether the actual out-feed force mean value is less than the minimum value of the mean values corresponding to the target feed classification.

9. The method according to any one of claims 1-8, characterized in that, If no target feed classification matching the feed parameters is obtained among the M feed classifications, the method further includes: Determining that the silicon rod corresponding to the current tool pass is in a non-penetrated state.

10. The method according to any one of claims 1-8, characterized in that, The method further includes: Judging whether the current tool pass has an abnormal state, where the abnormal state includes a wire break state and a state where the downtime exceeds a preset duration threshold; If the current tool pass has the abnormal state, determining that the silicon rod corresponding to the current tool pass is in a non-penetrated state.

11. A silicon wafer penetration judgment device, characterized in that, Including: An acquisition module for acquiring the feed parameters of the feed position of the current tool pass and the out-feed parameters of the out-feed position of the current tool pass, where the feed parameters include the actual feed force mean value of the feed position of the current tool pass, and the out-feed parameters include the actual out-feed force mean value of the out-feed position of the current tool pass; The acquisition module is further configured to, according to the feed parameters, acquire a target feed classification matching the feed parameters from M pre-divided feed classifications; where the M feed classifications are determined according to the historical feed parameters of the historical feed positions of the historical tool passes, and M is an integer greater than 1; A decision module for judging whether the actual out-feed force mean value is less than the additional cutting determination threshold corresponding to the target feed classification; if it is less, determining that the silicon rod corresponding to the current tool pass is in a penetrated state; where the additional cutting determination threshold is determined according to the minimum value of the mean values of the feed forces of the additional cutting tool passes in the historical cutting data corresponding to the target feed classification; or determined according to the maximum value of the mean values of the feed forces of the non-additional cutting tool passes in the historical cutting data corresponding to the target feed classification.

12. An electronic device, characterized in that, Including a processor and a memory, where a computer program is stored in the memory, and the processor executes the computer program to implement the method according to any one of claims 1-10.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by at least one processor, implements the method according to any one of claims 1-10.

Images