Drawing speed control method, device, electronic device and storage medium

Abstract

An embodiment of the present invention provides a pulling speed control method, device, equipment and medium. The method includes: during the Czochralski single crystal pulling process, obtaining characteristic data before the head stage of the isodiametric growth, inputting the characteristic data into a pulling speed prediction model, which is trained by characteristic data samples before the head stage of the isodiametric growth and the corresponding marked sample pulling speeds at the head stage of the isodiametric growth, generating a predicted pulling speed at the head stage of the isodiametric growth according to the characteristic data by the pulling speed prediction model, and before the head stage of the isodiametric growth, adjusting process parameters according to the predicted pulling speed and a preset pulling speed to control the pulling speed at the head stage of the isodiametric growth to change towards the preset pulling speed, so that the pulling speed at the head stage of the isodiametric growth can be predicted in advance, and the pulling speed is controlled accordingly, so that the pulling speed can probably fall within a suitable range, thereby reducing the occurrence rate of wire breakage problems.

Description

Technical Field

[0001] The present invention relates to the technical field of crystal preparation, and in particular, to a pulling speed control method, a pulling speed control device, an electronic device, and a storage medium. Background Art

[0002] The preparation process of monocrystalline silicon materials mainly uses the Czochralski process (CZ) to refine polysilicon raw materials into monocrystalline silicon. In the process of pulling a single crystal directly, the process of generating a rod-shaped monocrystalline silicon crystal is divided into steps such as loading, heating and melting the material, cooling and adjusting the seeding temperature, seeding, shoulder releasing, shoulder turning, constant diameter, and ending.

[0003] Among them, seeding is to bring the seed crystal (that is, a single crystal processed into a certain shape) installed at the end of the wire rope into contact with the liquid surface. At the seeding temperature, silicon molecules will grow along the lattice direction of the seed crystal to form a single crystal. Shoulder releasing is to gradually grow the crystal diameter to the required diameter. During the shoulder releasing process, a section of crystal with a gradually increasing length and a gradually increasing diameter to about the required diameter will be pulled out to eliminate crystal dislocations. When the crystal grows to the production-required diameter during the shoulder releasing process, it enters the shoulder turning process. Shoulder turning is to increase the pulling speed of the crystal and control the crystal diameter within the production-required diameter. When the shoulder turning is completed, it enters the constant diameter control step, which is one of the key links determining the growth quality of the monocrystalline silicon crystal. In this step, through the automatic control of the pulling speed and temperature, the crystal will grow with a constant diameter according to the set diameter.

[0004] In the current direct pulling process, in the head stage of constant diameter (for example, the stage from the start of constant diameter to the stage where the crystal length increases by 50 mm), it is difficult to accurately control the pulling speed. Through the setting of process parameters in the seeding, shoulder releasing, and shoulder turning processes, the pulling speed after the start of constant diameter is very randomly distributed. If the pulling speed cannot be accurately controlled after the start of constant diameter, or even the pulling speed is not within the appropriate range, the probability of wire breakage problems will increase. Summary of the Invention

[0005] In view of the above problems, embodiments of the present invention are proposed to provide a pulling speed control method that overcomes the above problems or at least partially solves the above problems, so as to solve the problems that the pulling speed cannot be accurately controlled after the start of constant diameter, or even the pulling speed is not within the appropriate range, and the probability of wire breakage problems is high.

[0006] Correspondingly, embodiments of the present invention further provide a pulling speed control device, an electronic device, and a storage medium to ensure the implementation and application of the above method.

[0007] To solve the above problems, embodiments of the present invention disclose a pulling speed control method, including:

[0008] During the Czochralski single crystal growth process, characteristic data before the head stage of isodiametric growth is obtained, where the characteristic data includes operation data related to the pulling speed in the head stage of isodiametric growth;

[0009] The characteristic data is input into a pulling speed prediction model, where the pulling speed prediction model is trained by characteristic data samples before the head stage of isodiametric growth and corresponding marked sample pulling speeds in the head stage of isodiametric growth;

[0010] Based on the characteristic data, the pulling speed prediction model generates a predicted pulling speed for the head stage of isodiametric growth;

[0011] Before the head stage of isodiametric growth, according to the predicted pulling speed and a preset pulling speed, process parameters are adjusted to control the pulling speed in the head stage of isodiametric growth to change towards the preset pulling speed.

[0012] Optionally, the obtaining of the characteristic data before the head stage of isodiametric growth includes:

[0013] Monitoring the current processing state of the process before the head stage of isodiametric growth;

[0014] When the current processing state reaches a preset processing state each time, according to the preset operation data types corresponding to the preset processing state, the operation data is collected once, and the operation data collected each time is determined as the characteristic data corresponding to the current processing state.

[0015] Optionally, the preset operation data types corresponding to multiple preset processing states are different, the pulling speed prediction model includes multiple sub-pulling speed prediction models, and the inputting of the characteristic data into the pulling speed prediction model includes:

[0016] When the current processing state reaches a preset processing state each time, the sub-pulling speed prediction model corresponding to the current processing state is called;

[0017] The characteristic data corresponding to the current processing state is input into the sub-pulling speed prediction model corresponding to the current processing state;

[0018] The generating of the predicted pulling speed for the head stage of isodiametric growth based on the characteristic data by the pulling speed prediction model includes:

[0019] Based on the characteristic data, the sub-pulling speed prediction model generates a predicted pulling speed for the head stage of isodiametric growth in the current processing state.

[0020] Optionally, the preset processing state includes at least one of the following: the crystal length reaches a preset length in the shoulder release stage of the Czochralski single crystal growth process, the end of the shoulder transition stage of the Czochralski single crystal growth process.

[0021] Optionally, before inputting the feature data into the pulling speed prediction model, it further includes:

[0022] After training the pulling speed prediction model with the feature data samples before the head stage of equal-diameter growth and the corresponding marked sample pulling speeds at the head stage of equal-diameter growth, obtain the feature weights corresponding to various feature data in the trained pulling speed prediction model;

[0023] Before the head stage of equal-diameter growth, adjusting the process parameters according to the predicted pulling speed and the preset pulling speed to control the pulling speed at the head stage of equal-diameter growth to change towards the preset pulling speed includes:

[0024] Determine that the process parameters need to be adjusted according to the difference between the predicted pulling speed and the preset pulling speed;

[0025] Calculate the adjustment amount of each process parameter according to the difference and the feature weight corresponding to each process parameter;

[0026] Adjust each process parameter according to the adjustment amount of each process parameter.

[0027] Optionally, the operating data includes at least one of the following: power, average crucible rotation, shoulder diameter, crystal diameter; the process parameters include at least one of the following: power, average crucible rotation.

[0028] Optionally, the pulling speed prediction model includes: a random forest model, or an extreme gradient boosting model, or a categorical gradient boosting model.

[0029] An embodiment of the present invention also discloses a pulling speed control device, including:

[0030] A data acquisition module, configured to acquire feature data before the head stage of equal-diameter growth during the Czochralski single crystal growth process, where the feature data includes operating data related to the pulling speed at the head stage of equal-diameter growth;

[0031] A data input module, configured to input the feature data into the pulling speed prediction model, where the pulling speed prediction model is trained with feature data samples before the head stage of equal-diameter growth and the corresponding marked sample pulling speeds at the head stage of equal-diameter growth;

[0032] A pulling speed generation module, configured to generate the predicted pulling speed at the head stage of equal-diameter growth according to the feature data by the pulling speed prediction model;

[0033] A parameter adjustment module, configured to adjust process parameters according to the predicted pulling speed and a preset pulling speed before the head stage of equal-diameter growth, so as to control the pulling speed in the head stage of equal-diameter growth to change towards the preset pulling speed.

[0034] Optionally, the data acquisition module includes:

[0035] A status monitoring sub-module, configured to monitor the current processing status of the process before the head stage of equal-diameter growth;

[0036] A parameter acquisition sub-module, configured to collect the operation data once according to the preset operation data types corresponding to the preset processing status every time the current processing status reaches a preset processing status, and determine the operation data collected each time as the characteristic data corresponding to the current processing status.

[0037] Optionally, the preset operation data types corresponding to multiple preset processing statuses are different, the pulling speed prediction model includes multiple sub-pulling speed prediction models, and the data input module includes:

[0038] A model calling sub-module, configured to call the sub-pulling speed prediction model corresponding to the current processing status every time the current processing status reaches a preset processing status;

[0039] A data input sub-module, configured to input the characteristic data corresponding to the current processing status into the sub-pulling speed prediction model corresponding to the current processing status;

[0040] The pulling speed generation module is specifically configured to:

[0041] Generate a predicted pulling speed in the head stage of equal-diameter growth in the current processing status according to the characteristic data by the sub-pulling speed prediction model.

[0042] Optionally, the preset processing status includes at least one of the following: the crystal length reaches a preset length in the shoulder releasing stage of the Czochralski single crystal growth process, and the shoulder turning stage of the Czochralski single crystal growth process ends.

[0043] Optionally, it further includes:

[0044] A weight acquisition module, configured to obtain the characteristic weights corresponding to various characteristic data in the trained pulling speed prediction model after training the pulling speed prediction model with the characteristic data samples before the head stage of equal-diameter growth and the sample pulling speeds corresponding to the marked head stages of equal-diameter growth before inputting the characteristic data into the pulling speed prediction model;

[0045] The parameter adjustment module includes:

[0046] An adjustment determination sub-module, configured to determine that the process parameters need to be adjusted according to the difference between the predicted drawing speed and the preset drawing speed;

[0047] A calculation sub-module, configured to calculate the adjustment amount of each of the process parameters according to the difference and the characteristic weights corresponding to each of the process parameters;

[0048] A parameter adjustment sub-module, configured to adjust each of the process parameters according to the adjustment amount of each of the process parameters.

[0049] Optionally, the operating data includes at least one of the following: power, average crucible rotation speed, shoulder diameter, crystal diameter; the process parameters include at least one of the following: power, average crucible rotation speed.

[0050] Optionally, the drawing speed prediction model includes: a random forest model, or an extreme gradient boosting model, or a category gradient boosting model.

[0051] An embodiment of the present invention further discloses an electronic device, which is characterized in that it includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory complete communication with each other through the communication bus;

[0052] The memory is used to store a computer program;

[0053] The processor is configured to implement the method steps as described above when executing the program stored on the memory.

[0054] An embodiment of the present invention further discloses a readable storage medium, when the instructions in the storage medium are executed by the processor of the electronic device, enabling the electronic device to execute one or more of the drawing speed control methods in the embodiments of the present invention.

[0055] The embodiments of the present invention have the following advantages:

[0056] According to an embodiment of the present invention, during the Czochralski single crystal growth process, characteristic data before the head stage of the isodiametric growth is obtained, wherein the characteristic data includes operation data related to the pulling speed at the head stage of the isodiametric growth. The characteristic data is input into a pulling speed prediction model, wherein the pulling speed prediction model is trained by characteristic data samples before the head stage of the isodiametric growth and corresponding marked sample pulling speeds at the head stage of the isodiametric growth. According to the characteristic data, the pulling speed prediction model generates the predicted pulling speed at the head stage of the isodiametric growth. Before the head stage of the isodiametric growth, according to the predicted pulling speed and a preset pulling speed, process parameters are adjusted to control the change of the pulling speed at the head stage of the isodiametric growth towards the preset pulling speed, so that the pulling speed at the head stage of the isodiametric growth can be predicted in advance, and the pulling speed is controlled accordingly, so that the pulling speed can probably fall within a suitable range, thereby reducing the occurrence rate of wire breakage problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flowchart of the steps of an embodiment of a pulling speed control method of the present invention;

[0058] Figure 2 is a flowchart of the steps of an embodiment of a pulling speed control method of the present invention;

[0059] Figure 3 is a structural block diagram of an embodiment of a pulling speed control device of the present invention;

[0060] Figure 4 is a structural block diagram of a computing device for pulling speed control shown according to an exemplary embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0061] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0062] Referring to Figure 1 , a flowchart of the steps of an embodiment of a pulling speed control method of the present invention is shown, which may specifically include the following steps:

[0063] Step 101, during the Czochralski single crystal growth process, obtain characteristic data before the head stage of the isodiametric growth, wherein the characteristic data includes operation data related to the pulling speed at the head stage of the isodiametric growth.

[0064] In the embodiments of the present invention, the Czochralski single crystal process is a process of refining raw materials into single crystals by the Czochralski method. For example, the process of Czochralski single crystal silicon. The Czochralski single crystal process can be divided into the crystal seeding stage, the shoulder releasing stage, the shoulder turning stage, the isodiameter growth stage, etc. Among them, the beginning part of the isodiameter growth stage is called the head stage of isodiameter growth (for example, the stage from the start of isodiameter to the stage where the crystal length increases by 50 mm).

[0065] In the embodiments of the present invention, the operation data during the Czochralski single crystal process includes power, average crucible rotation, shoulder diameter, crystal diameter, etc. The operation data is the data monitored during actual operation, and specifically can include any applicable data, and the embodiments of the present invention do not limit this. After analyzing a large amount of data, it is found that the pulling speed in the head stage of isodiameter growth has a key impact on isodiameter wire breakage, and the pulling speed in the head stage of isodiameter growth is related to the operation data of the part before the head stage of isodiameter growth. Through data analysis, among the operation data before the head stage of isodiameter growth, the operation data related to the pulling speed in the head stage of isodiameter growth is screened out as characteristic data.

[0066] For example, according to historical big data statistics, the operation data related to the pulling speed in the head stage of isodiameter growth is first screened. Then data analysis and data processing are carried out. Then secondary screening is carried out to perform correlation analysis between operation data and correlation analysis between operation data and pulling speed, and finally the characteristic data that can be used for data modeling is determined. The analysis and processing methods include but are not limited to big data statistics related data processing and analysis methods such as histograms, scatter plots, violin plots, heat maps, box plots, density plots, etc.

[0067] Step 102, input the characteristic data into the pulling speed prediction model, where the pulling speed prediction model is trained by the characteristic data samples before the head stage of isodiameter growth and the corresponding marked sample pulling speeds in the head stage of isodiameter growth.

[0068] In the embodiments of the present invention, to predict the pulling speed in the head stage of isodiameter growth, the correlation between the characteristic data before the head stage of isodiameter growth and the pulling speed in the head stage of isodiameter growth can be used by machine learning to obtain a pulling speed prediction model that can predict the pulling speed. To train the pulling speed prediction model, accurate sample data and corresponding label data are required, that is, the characteristic data samples before the head stage of isodiameter growth and the corresponding marked sample pulling speeds in the head stage of isodiameter growth. Specifically, the characteristic data samples and sample pulling speeds can be obtained through multiple experiments, or the characteristic data samples and sample pulling speeds can be selected from historical data.

[0069] In an alternative embodiment of the present invention, the drawing speed prediction model includes: a random forest model, or an extreme gradient boosting model, or a categorical gradient boosting model. The random forest model is a classifier that uses multiple trees to train and predict samples. The extreme gradient boosting (XGboost) model is a model that uses boosting trees for prediction in a large-scale parallel manner. The categorical gradient boosting (CATboost) model is a model of a gradient boosting algorithm that can handle categorical features well.

[0070] For example, considering that the label variable of the drawing speed prediction model is the drawing speed, which belongs to a continuous variable. Select a suitable model according to the data type of the label data. The initially screened models include, but are not limited to, models such as random forest, XGboost, and CATboost. The data used to train the model is divided into a training set and a test set. Among them, the test set generally accounts for 25% to 15%. The loss function of the model selects indicators such as MAE (Mean Absolute Error), MSE (Mean Squared Error), and R2 (R-Square). By adjusting the parameters of the model, the optimal parameter combination is found. All three models are trained according to the above method. Finally, a model with an R2 index greater than 0.4, and the larger the better, and the smallest MAE and MSE, and the smaller the better, is selected as the best comprehensive model among the three models. Finally, the best comprehensive model is used as the adopted drawing speed prediction model.

[0071] In an embodiment of the present invention, in the prediction stage, the input of the drawing speed prediction model is the feature data before the head stage of the equal-diameter growth, and the output is the predicted value of the drawing speed, denoted as the predicted drawing speed.

[0072] Step 103, according to the feature data, generate the predicted drawing speed of the head stage of the equal-diameter growth by the drawing speed prediction model.

[0073] In an embodiment of the present invention, for the current Czochralski crystal growth process, after obtaining the current feature data, the obtained feature data is input into the drawing speed prediction model to obtain the output of the drawing speed prediction model, and then the output of the drawing speed prediction model is used as the predicted drawing speed of the head stage of the equal-diameter growth.

[0074] In an embodiment of the present invention, before the head stage of the equal-diameter growth, the current feature data can be input into the drawing speed prediction model every time a preset time period elapses, or every time the process reaches a preset processing state, so as to continuously obtain updated predicted drawing speeds.

[0075] Step 104, before the head stage of equal-diameter growth, adjust the process parameters according to the predicted pulling speed and the preset pulling speed, so as to control the pulling speed in the head stage of equal-diameter growth to change towards the preset pulling speed.

[0076] In the embodiment of the present invention, when controlling the pulling speed by changing relevant process parameters, there is a delay in the change of the pulling speed. Therefore, it is necessary to control the pulling speed before the head stage of equal-diameter growth. By analyzing the relationship between the pulling speed and the wire breakage rate in the head stage of equal-diameter growth through data analysis, the preset pulling speed that makes the wire breakage rate lower or the lowest is determined in advance. Among them, the process parameters are the parameters related to the process set for the equipment, including power, average crucible rotation, etc., or any other applicable process parameters, and the embodiment of the present invention does not limit this.

[0077] In an alternative embodiment of the present invention, the operating data includes at least one of the following: power, average crucible rotation, shoulder diameter, crystal diameter; the process parameters include at least one of the following: power, average crucible rotation. The power is the power of the single crystal furnace. The average crucible rotation is the average rotation speed of stirring the melt in the crucible of the single crystal furnace. The shoulder diameter refers to the crystal diameter in the shoulder formation stage.

[0078] In the embodiment of the present invention, the process parameters are adjusted according to the predicted pulling speed and the preset pulling speed, so that the pulling speed changes towards the preset pulling speed. If the predicted pulling speed is higher than the preset pulling speed, the pulling speed is reduced by adjusting the process parameters to approach the preset pulling speed. If the predicted pulling speed is lower than the preset pulling speed, the pulling speed is increased by adjusting the process parameters to approach the preset pulling speed. When adjusting the process parameters, the adjustment amount is related to the difference between the predicted pulling speed and the preset pulling speed. Generally, the larger the absolute value of the difference between the predicted pulling speed and the preset pulling speed, the larger the adjustment amount of the process parameters. Since different process parameters have different effects on the pulling speed, one or more process parameters can be adjusted according to the preset adjustment rules to control the pulling speed in the head stage of equal-diameter growth to change towards the preset pulling speed.

[0079] According to an embodiment of the present invention, during the Czochralski single crystal growth process, characteristic data before the head stage of the constant diameter growth is obtained, wherein the characteristic data includes operation data related to the pulling speed in the head stage of the constant diameter growth. The characteristic data is input into a pulling speed prediction model, wherein the pulling speed prediction model is trained by characteristic data samples before the head stage of the constant diameter growth and the corresponding marked sample pulling speeds in the head stage of the constant diameter growth. According to the characteristic data, the predicted pulling speed in the head stage of the constant diameter growth is generated by the pulling speed prediction model. Before the head stage of the constant diameter growth, according to the predicted pulling speed and a preset pulling speed, the process parameters are adjusted to control the pulling speed in the head stage of the constant diameter growth to change towards the preset pulling speed, so that the pulling speed in the head stage of the constant diameter growth can be predicted in advance, and the pulling speed is controlled accordingly, so that the pulling speed can probably fall within a suitable range, thereby reducing the occurrence rate of wire breakage problems.

[0080] In an alternative embodiment of the present invention, before step 102, it may further include: after training the pulling speed prediction model with characteristic data samples before the head stage of the constant diameter growth and the corresponding marked sample pulling speeds in the head stage of the constant diameter growth, obtaining the characteristic weights corresponding to various characteristic data in the trained pulling speed prediction model.

[0081] Correspondingly, in a specific implementation manner, step 104 may include: determining that the process parameters need to be adjusted according to the difference between the predicted pulling speed and the preset pulling speed; calculating the adjustment amounts of the respective process parameters according to the difference and the characteristic weights corresponding to the respective process parameters; and adjusting the respective process parameters according to the adjustment amounts of the respective process parameters.

[0082] The training process of the pulling speed prediction model is a process of determining the characteristic weights corresponding to various characteristic data according to the training data. After the pulling speed prediction model is trained, the characteristic weights are determined. The characteristic weights corresponding to various characteristic data are obtained from the trained pulling speed prediction model. Among them, the greater the characteristic weight, the greater the influence of this type of characteristic data on the pulling speed, and the smaller the characteristic weight, the smaller the influence of this type of characteristic data on the pulling speed.

[0083] When adjusting process parameters, first calculate the difference between the predicted drawing speed and the preset drawing speed, and determine whether the process parameters need to be adjusted according to the difference. For example, if the absolute value of the difference is greater than the preset threshold, it is determined that the process parameters need to be adjusted; if the absolute value of the difference is not greater than the preset threshold, it is determined that the process parameters do not need to be adjusted. Then, each process parameter has a corresponding relationship with the operating data. The process parameter is a set value, while the operating data is the value during actual operation. For example, the process parameter is the set power, and the operating data is the power during actual operation. According to the difference and the characteristic weights corresponding to each process parameter, calculate the adjustment amount of each process parameter. For example, for each process parameter, multiply the difference by the characteristic weight corresponding to the process parameter, and determine the product as the adjustment amount of the process parameter. Specifically, any applicable method for calculating the adjustment amount can be adopted, and the embodiments of the present invention do not limit this. Finally, increase or decrease each process parameter according to the adjustment amount of each process parameter.

[0084] Referring to Figure 2 , a step flowchart of an embodiment of a drawing speed control method according to the present invention is shown, which may specifically include the following steps:

[0085] Step 201, monitor the current processing state of the process before the head stage of equal-diameter growth.

[0086] In the embodiments of the present invention, before the head stage of equal-diameter growth, the processing states of the process include the state of the crystal length, the end of a certain processing stage (for example, the shoulder transition stage), or other any applicable processing states, and the embodiments of the present invention do not limit this.

[0087] In the embodiments of the present invention, before the head stage of equal-diameter growth, monitor the processing state of the process to obtain the current processing state. For example, monitor the crystal length to obtain the current crystal length, or monitor whether the shoulder transition stage has ended to obtain whether the current is the end of the shoulder transition stage.

[0088] Step 202, when the current processing state reaches a preset processing state each time, collect the operating data once according to the preset type of operating data corresponding to the preset processing state, and determine the collected operating data each time as the characteristic data corresponding to the current processing state.

[0089] In the embodiments of the present invention, in order to predict the drawing speed multiple times based on the characteristic data for a period of time before the head stage of equal-diameter growth, the operating data needs to be collected at regular intervals. For this purpose, multiple preset prediction points are set in the process before the head stage of equal-diameter growth, and each preset prediction point corresponds to a preset processing state of the process. For example, the crystal length reaches the preset length, the shoulder transition stage ends, etc.

[0090] In the embodiments of the present invention, during the process before the head stage of equal-diameter growth, at different stages, the operating data related to the pulling speed in the head stage of equal-diameter growth will be different. Therefore, there are preset types of operating data corresponding to the preset processing states. Of course, the preset types of operating data are all operating data related to the pulling speed in the head stage of equal-diameter growth.

[0091] In the embodiments of the present invention, every time the current processing state reaches a preset processing state, according to the preset type of operating data corresponding to this preset processing state, the operating data of the preset type of operating data is collected once. And the collected operating data is determined as the characteristic data corresponding to the current processing state, that is, the characteristic data when the process is in the preset processing state.

[0092] In an alternative embodiment of the present invention, the preset processing state includes at least one of the following: the crystal length reaches a preset length during the shoulder release stage of the Czochralski single crystal growth process, and the shoulder turning stage of the Czochralski single crystal growth process ends. For example, during the Czochralski single crystal growth process, it is divided into 5 preset prediction points: the crystal length reaches 50mm, 60mm, 70mm, 80mm during the shoulder release stage, and the shoulder turning stage ends.

[0093] Step 203, every time the current processing state reaches a preset processing state, call the sub-pulling speed prediction model corresponding to the current processing state.

[0094] In the embodiments of the present invention, since at different stages, the operating data related to the pulling speed in the head stage of equal-diameter growth will be different. Therefore, the preset types of operating data corresponding to multiple preset processing states are different. That is to say, in one case, the preset types of operating data corresponding to each preset processing state are all different. In another case, the preset types of operating data corresponding to some two preset processing states are the same, and the preset types of operating data corresponding to some two preset processing states are different.

[0095] In the embodiments of the present invention, for different preset types of operating data, different models need to be trained for prediction. The pulling speed prediction model includes multiple sub-pulling speed prediction models. After determining the preset processing state, a sub-pulling speed prediction model that can process the characteristic data of the corresponding preset type of operating data can be obtained. When the current processing state reaches a certain preset processing state, call the sub-pulling speed prediction model corresponding to this preset processing state.

[0096] For example, the trained sub-drawing speed prediction model generates pkl format files (pkl1 file, pkl2 file, pkl3 file, pkl4 file, pkl5 file). The main program calls the corresponding pkl file for prediction according to the current processing status and outputs the prediction result. During the shoulder-forming stage, when the crystal length reaches 50mm, 60mm, 70mm, 80mm, and at the end of the shoulder-transition stage, for 5 preset prediction points, the pkl1 file, pkl2 file, pkl3 file, pkl4 file, and pkl5 file are called in sequence for result prediction, so as to achieve the purpose of predicting the drawing speed at the head stage of equal-diameter growth.

[0097] Step 204: Input the feature data corresponding to the current processing status into the sub-drawing speed prediction model corresponding to the current processing status.

[0098] In the embodiment of the present invention, after the feature data is collected in the current processing status, it is input into the sub-drawing speed prediction model corresponding to the current processing status, so as to obtain the predicted drawing speed in the current processing status through the sub-drawing speed prediction model.

[0099] Step 205: According to the feature data, the sub-drawing speed prediction model generates the predicted drawing speed at the head stage of equal-diameter growth in the current processing status.

[0100] In the embodiment of the present invention, according to the feature data collected in the current processing status, the sub-drawing speed prediction model can generate the predicted drawing speed, that is, the predicted drawing speed at the head stage of equal-diameter growth in the current processing status.

[0101] Step 206: Before the head stage of equal-diameter growth, according to the predicted drawing speed and the preset drawing speed, adjust the process parameters to control the drawing speed at the head stage of equal-diameter growth to change towards the preset drawing speed.

[0102] In the embodiment of the present invention, the specific implementation manner can refer to the description in the foregoing embodiments and will not be elaborated herein.

[0103] According to an embodiment of the present invention, by monitoring the current processing state of the process before the head stage of equal-diameter growth, when the current processing state reaches a preset processing state each time, according to the preset type of operation data corresponding to the preset processing state, the operation data is collected once, and the operation data collected each time is determined as the characteristic data corresponding to the current processing state. When the current processing state reaches a preset processing state each time, the sub-drawing speed prediction model corresponding to the current processing state is called, and the characteristic data corresponding to the current processing state is input into the sub-drawing speed prediction model corresponding to the current processing state. According to the characteristic data, the sub-drawing speed prediction model generates a predicted drawing speed of the head stage of equal-diameter growth in the current processing state. Before the head stage of equal-diameter growth, according to the predicted drawing speed and the preset drawing speed, the process parameters are adjusted to control the drawing speed in the head stage of equal-diameter growth to change towards the preset drawing speed, so that the drawing speed in the head stage of equal-diameter growth can be predicted in advance, and accordingly the drawing speed is controlled, so that the drawing speed can probably fall within a suitable range, thereby reducing the incidence of wire breakage problems.

[0104] It should be noted that for the method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the embodiments of the present invention are not limited by the described action sequence, because according to the embodiments of the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0105] Refer to Figure 3 , which shows a structural block diagram of an embodiment of a drawing speed control device of the present invention, and specifically may include the following modules:

[0106] The data acquisition module 301 is used to acquire characteristic data before the head stage of equal-diameter growth during the Czochralski single crystal growth process, wherein the characteristic data includes operation data related to the drawing speed of the head stage of equal-diameter growth;

[0107] The data input module 302 is used to input the characteristic data into the drawing speed prediction model, wherein the drawing speed prediction model is trained by characteristic data samples before the head stage of equal-diameter growth and corresponding marked sample drawing speeds of the head stage of equal-diameter growth;

[0108] The drawing speed generation module 303 is used to generate a predicted drawing speed of the head stage of equal-diameter growth by the drawing speed prediction model according to the characteristic data;

[0109] A parameter adjustment module 304, configured to adjust process parameters according to the predicted pulling speed and a preset pulling speed before the head stage of equal-diameter growth, so as to control the pulling speed in the head stage of equal-diameter growth to change towards the preset pulling speed.

[0110] In an embodiment of the present invention, optionally, the data acquisition module includes:

[0111] A status monitoring sub-module, configured to monitor the current processing status of the process before the head stage of equal-diameter growth;

[0112] A parameter acquisition sub-module, configured to collect the operation data once according to the preset operation data types corresponding to the preset processing status every time the current processing status reaches a preset processing status, and determine the operation data collected each time as the characteristic data corresponding to the current processing status.

[0113] In an embodiment of the present invention, optionally, the preset operation data types corresponding to multiple preset processing statuses are different, the pulling speed prediction model includes multiple sub-pulling speed prediction models, and the data input module includes:

[0114] A model calling sub-module, configured to call the sub-pulling speed prediction model corresponding to the current processing status every time the current processing status reaches a preset processing status;

[0115] A data input sub-module, configured to input the characteristic data corresponding to the current processing status into the sub-pulling speed prediction model corresponding to the current processing status;

[0116] The pulling speed generation module is specifically configured to:

[0117] Generate a predicted pulling speed of the head stage of equal-diameter growth in the current processing status according to the characteristic data by the sub-pulling speed prediction model.

[0118] In an embodiment of the present invention, optionally, the preset processing status includes at least one of the following: the crystal length reaches a preset length in the shoulder releasing stage of the Czochralski single crystal growth process, and the shoulder turning stage of the Czochralski single crystal growth process ends.

[0119] In an embodiment of the present invention, optionally, it further includes:

[0120] A weight acquisition module, configured to, before inputting the characteristic data into the pulling speed prediction model, after training the pulling speed prediction model with the characteristic data samples before the head stage of equal-diameter growth and the corresponding sample pulling speeds of the head stage of equal-diameter growth with corresponding labels, acquire the characteristic weights corresponding to various characteristic data in the trained pulling speed prediction model;

[0121] The parameter adjustment module includes:

[0122] An adjustment determination sub-module, configured to determine that the process parameters need to be adjusted according to the difference between the predicted drawing speed and the preset drawing speed;

[0123] A calculation sub-module, configured to calculate the adjustment amount of each of the process parameters according to the difference and the characteristic weight corresponding to each of the process parameters;

[0124] A parameter adjustment sub-module, configured to adjust each of the process parameters according to the adjustment amount of each of the process parameters.

[0125] In an embodiment of the present invention, optionally, the operation data includes at least one of the following: power, average crucible rotation, shoulder diameter, crystal diameter; the process parameters include at least one of the following: power, average crucible rotation.

[0126] In an embodiment of the present invention, optionally, the drawing speed prediction model includes: a random forest model, or an extreme gradient boosting model, or a categorical gradient boosting model.

[0127] According to an embodiment of the present invention, during the Czochralski single crystal growth process, characteristic data before the head stage of the equal-diameter growth is obtained, wherein the characteristic data includes operation data related to the drawing speed in the head stage of the equal-diameter growth, and the characteristic data is input into a drawing speed prediction model, wherein the drawing speed prediction model is trained by characteristic data samples before the head stage of the equal-diameter growth and the corresponding marked sample drawing speed in the head stage of the equal-diameter growth. According to the characteristic data, the prediction drawing speed in the head stage of the equal-diameter growth is generated by the drawing speed prediction model. Before the head stage of the equal-diameter growth, according to the prediction drawing speed and the preset drawing speed, the process parameters are adjusted to control the drawing speed in the head stage of the equal-diameter growth to change towards the preset drawing speed, so that the drawing speed in the head stage of the equal-diameter growth can be predicted in advance, and accordingly the drawing speed is controlled, so that the drawing speed can probably fall within a suitable range, thereby reducing the occurrence rate of wire breakage problems.

[0128] For the apparatus embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for the related parts, please refer to the partial description of the method embodiment.

[0129] Figure 4 It is a structural block diagram of an electronic device 400 for drawing speed control shown according to an exemplary embodiment. For example, the electronic device 400 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0130] Refer to Figure 4, the electronic device 400 may include one or more of the following components: a processing component 402, a memory 404, a power component 406, a multimedia component 408, an audio component 410, an input / output (I / O) interface 412, a sensor component 414, and a communication component 416.

[0131] The processing component 402 generally controls the overall operation of the electronic device 400, such as operations associated with display, telephone calls, data communications, camera operations, and recording operations. The processing component 402 may include one or more processors 420 to execute instructions to complete all or part of the steps of the above-mentioned drawing speed control method. In addition, the processing component 402 may include one or more modules to facilitate the interaction between the processing component 402 and other components. For example, the processing unit 402 may include a multimedia module to facilitate the interaction between the multimedia component 408 and the processing component 402.

[0132] The memory 404 is configured to store various types of data to support the operation of the device 400. Examples of these data include instructions for any application or method operating on the electronic device 400, contact data, phone book data, messages, pictures, videos, etc. The memory 404 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disks, or optical disks.

[0133] The power component 404 provides power to various components of the electronic device 400. The power component 404 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the electronic device 400.

[0134] The multimedia component 408 includes a screen that provides an output interface between the electronic device 400 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of the touch or swipe actions but also detect the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 408 includes a front camera and / or a rear camera. When the electronic device 400 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capabilities.

[0135] The audio component 410 is configured to output and / or input audio signals. For example, the audio component 410 includes a microphone (MIC) that is configured to receive external audio signals when the electronic device 400 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals can be further stored in the memory 404 or transmitted via the communication component 416. In some embodiments, the audio component 410 further includes a speaker for outputting audio signals.

[0136] The I / O interface 412 provides an interface between the processing component 402 and a peripheral interface module, and the peripheral interface module can be a keyboard, a click wheel, buttons, etc. These buttons can include but are not limited to: a home button, a volume button, a power button, and a lock button.

[0137] The sensor component 414 includes one or more sensors for providing status assessments of various aspects of the electronic device 400. For example, the sensor component 414 can detect the on / off state of the device 400, the relative positioning of components, such as the display and the keypad of the electronic device 400. The sensor component 414 can also detect a change in the position of the electronic device 400 or a component of the electronic device 400, the presence or absence of user contact with the electronic device 400, the orientation or acceleration / deceleration of the electronic device 400, and the temperature change of the electronic device 400. The sensor component 414 can include a proximity sensor that is configured to detect the presence of nearby objects without any physical contact. The sensor component 414 can also include a light sensor, such as a CMOS or a CCD image sensor, for use in imaging applications. In some embodiments, the sensor component 414 can further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0138] The communication component 416 is configured to facilitate communication between the electronic device 400 and other devices in a wired or wireless manner. The electronic device 400 can access a communication standard-based wireless network, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 414 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 414 further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0139] In an exemplary embodiment, the electronic device 400 can be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above-mentioned pulling speed control method.

[0140] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 404 including instructions, and the above instructions can be executed by a processor 420 of the electronic device 400 to complete the above-mentioned pulling speed control method. For example, the non-transitory computer-readable storage medium can be a ROM, Random Access Memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0141] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a terminal, enables the terminal to execute a pulling speed control method, and the method includes:

[0142] During the Czochralski single crystal growth process, characteristic data before the head stage of isothermal growth is acquired, where the characteristic data includes operation data related to the pulling speed in the head stage of isothermal growth;

[0143] The characteristic data is input into a pulling speed prediction model, where the pulling speed prediction model is trained by characteristic data samples before the head stage of isothermal growth and corresponding marked sample pulling speeds in the head stage of isothermal growth;

[0144] According to the characteristic data, a predicted pulling speed in the head stage of isothermal growth is generated by the pulling speed prediction model;

[0145] Before the head stage of isothermal growth, according to the predicted pulling speed and a preset pulling speed, process parameters are adjusted to control the pulling speed in the head stage of isothermal growth to change towards the preset pulling speed.

[0146] Optionally, obtaining the characteristic data before the head stage of equal-diameter growth includes:

[0147] Monitoring the current processing state of the process before the head stage of equal-diameter growth;

[0148] When the current processing state reaches each preset processing state, collect the operation data once according to the preset operation data types corresponding to the preset processing state, and determine the operation data collected each time as the characteristic data corresponding to the current processing state.

[0149] Optionally, the preset operation data types corresponding to multiple preset processing states are different, the pulling speed prediction model includes multiple sub-pulling speed prediction models, and inputting the characteristic data into the pulling speed prediction model includes:

[0150] When the current processing state reaches each preset processing state, call the sub-pulling speed prediction model corresponding to the current processing state;

[0151] Input the characteristic data corresponding to the current processing state into the sub-pulling speed prediction model corresponding to the current processing state;

[0152] Generating the predicted pulling speed of the head stage of equal-diameter growth from the pulling speed prediction model according to the characteristic data includes:

[0153] Generating the predicted pulling speed of the head stage of equal-diameter growth in the current processing state from the characteristic data by the sub-pulling speed prediction model.

[0154] Optionally, the preset processing state includes at least one of the following: the crystal length reaches a preset length in the shoulder releasing stage of the Czochralski single crystal process, and the shoulder turning stage of the Czochralski single crystal process ends.

[0155] Optionally, before inputting the characteristic data into the pulling speed prediction model, it further includes:

[0156] After training the pulling speed prediction model with the characteristic data samples before the head stage of equal-diameter growth and the sample pulling speeds of the corresponding marked head stages of equal-diameter growth, obtain the characteristic weights corresponding to various characteristic data in the trained pulling speed prediction model;

[0157] Before the head stage of equal-diameter growth, adjusting the process parameters according to the predicted pulling speed and the preset pulling speed to control the pulling speed in the head stage of equal-diameter growth to change towards the preset pulling speed includes:

[0158] Determine that the process parameters need to be adjusted according to the difference between the predicted pulling speed and the preset pulling speed;

[0159] Calculate the adjustment amount of each of the process parameters according to the difference value and the characteristic weights corresponding to the respective process parameters;

[0160] Adjust each of the process parameters according to the adjustment amount of each of the process parameters.

[0161] Optionally, the operating data includes at least one of the following: power, average crucible rotation speed, shoulder diameter, crystal diameter; the process parameters include at least one of the following: power, average crucible rotation speed.

[0162] Optionally, the pulling speed prediction model includes: a random forest model, or an extreme gradient boosting model, or a category gradient boosting model.

[0163] Each embodiment in this specification is described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the embodiments, reference may be made to each other.

[0164] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a device, or a computer program product. Therefore, the embodiments of the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the embodiments of the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal devices generate a device for implementing the specified functions in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0166] These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal devices to work in a predictive manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the specified functions in Figure 1 one process or multiple processes and / or blocksFigure 1 The functions specified in one or more boxes.

[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable terminal device provide for implementing the steps of the functions specified in one or more processes and / or boxes. Figure 1 One process or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes.

[0168] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

[0169] Finally, it should also be noted that in this text, 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. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article or terminal device comprising the element.

[0170] The above has introduced in detail a drawing speed control method and device, an electronic device, and a storage medium provided by the present invention. Specific examples are used in this text to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A drawing speed control method, characterized in that, Including: During the Czochralski single crystal growth process, characteristic data before the head stage of the isodiameter growth is obtained, where the characteristic data includes operation data related to the pulling speed in the head stage of the isodiameter growth, and the operation data includes at least one of the following: power, average crucible rotation, shoulder diameter, crystal diameter; Inputting the characteristic data into a pulling speed prediction model, where the pulling speed prediction model is trained by characteristic data samples before the head stage of the isodiameter growth and corresponding marked sample pulling speeds in the head stage of the isodiameter growth; Generating a predicted pulling speed in the head stage of the isodiameter growth based on the characteristic data by the pulling speed prediction model; Before the head stage of the isodiameter growth, adjusting process parameters according to the predicted pulling speed and a preset pulling speed to control the pulling speed in the head stage of the isodiameter growth to change towards the preset pulling speed; where the process parameters include at least one of the following: power, average crucible rotation; Before inputting the characteristic data into the pulling speed prediction model, it further includes: After training the pulling speed prediction model by characteristic data samples before the head stage of the isodiameter growth and corresponding marked sample pulling speeds in the head stage of the isodiameter growth, obtaining the characteristic weights corresponding to various characteristic data in the trained pulling speed prediction model; The adjusting process parameters according to the predicted pulling speed and the preset pulling speed before the head stage of the isodiameter growth to control the pulling speed in the head stage of the isodiameter growth to change towards the preset pulling speed includes: Determining that the process parameters need to be adjusted according to the difference between the predicted pulling speed and the preset pulling speed; Calculating the adjustment amount of each process parameter according to the difference and the characteristic weights corresponding to each process parameter; Adjusting each process parameter according to the adjustment amount of each process parameter.

2. The method according to claim 1, wherein The obtaining the characteristic data before the head stage of the isodiameter growth includes: Monitoring the current processing state of the process before the head stage of the isodiameter growth; When the current processing state reaches a preset processing state each time, collecting the operation data once according to the preset operation data type corresponding to the preset processing state, and determining the operation data collected each time as the characteristic data corresponding to the current processing state.

3. The method according to claim 2, wherein The preset operation data types corresponding to multiple preset processing states are different, the pulling speed prediction model includes multiple sub - pulling speed prediction models, and the inputting the characteristic data into the pulling speed prediction model includes: When the current processing state reaches a preset processing state each time, calling the sub - pulling speed prediction model corresponding to the current processing state; Inputting the characteristic data corresponding to the current processing state into the sub - pulling speed prediction model corresponding to the current processing state; The generating a predicted pulling speed in the head stage of the isodiameter growth based on the characteristic data by the pulling speed prediction model includes: Generating a predicted pulling speed in the head stage of the isodiameter growth in the current processing state by the sub - pulling speed prediction model based on the characteristic data.

4. The method according to claim 2 or 3, characterized in that, The preset processing state includes at least one of the following: the crystal length reaches a preset length in the shoulder releasing stage of the Czochralski single crystal growth process, and the shoulder turning stage of the Czochralski single crystal growth process ends.

5. The method according to claim 1, wherein The pulling speed prediction model includes: a random forest model, or an extreme gradient boosting model, or a categorical gradient boosting model.

6. A casting speed control device, characterized in that, Including: A data acquisition module, configured to acquire feature data before the head stage of the isodiametric growth during the Czochralski single crystal growth process, where the feature data includes operation data related to the pulling speed of the head stage of the isodiametric growth, and the operation data includes at least one of the following: power, average crucible rotation, shoulder diameter, crystal diameter; A data input module, configured to input the feature data into the pulling speed prediction model, where the pulling speed prediction model is trained by feature data samples before the head stage of the isodiametric growth and corresponding labeled sample pulling speeds of the head stage of the isodiametric growth; A pulling speed generation module, configured to generate a predicted pulling speed of the head stage of the isodiametric growth according to the feature data by the pulling speed prediction model; A parameter adjustment module, configured to adjust process parameters according to the predicted pulling speed and a preset pulling speed before the head stage of the isodiametric growth, so as to control the pulling speed in the head stage of the isodiametric growth to change towards the preset pulling speed; where the process parameters include at least one of the following: power, average crucible rotation; The device further includes: A weight acquisition module, configured to acquire feature weights corresponding to various feature data in the trained pulling speed prediction model after training the pulling speed prediction model by feature data samples before the head stage of the isodiametric growth and corresponding labeled sample pulling speeds of the head stage of the isodiametric growth, before inputting the feature data into the pulling speed prediction model; The parameter adjustment module includes: An adjustment determination sub-module, configured to determine that the process parameters need to be adjusted according to the difference between the predicted pulling speed and the preset pulling speed; A calculation sub-module, configured to calculate the adjustment amount of each of the process parameters according to the difference and the feature weights corresponding to each of the process parameters; A parameter adjustment sub-module, configured to adjust each of the process parameters according to the adjustment amount of each of the process parameters.

7. An electronic device, characterized in that, Including a processor, a communication interface, a memory, and a communication bus, where the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store a computer program; The processor is configured to implement the method steps described in any one of claims 1-5 when executing the program stored on the memory.

8. A readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the electronic device, the electronic device is enabled to execute the pulling speed control method described in any one of claims 1-5 of the method claims.

Images