Control method, system, and apparatus for crystal growth equipment

Abstract

Embodiments of this specification provide a method, system, and apparatus for controlling a crystal growth apparatus, the method comprising the steps of: acquiring hierarchical growth data of at least one crystal growth apparatus, wherein the hierarchical growth data includes at least hierarchical growth control data and hierarchical growth result data; and determining a control mode for a target crystal growth apparatus based on the hierarchical growth data and the target growth result, wherein the control mode includes at least one of a temperature control mode and a power control mode.

Description

Technical Field

[0001] This application relates to the field of semiconductor technology, and particularly to a control method, system, and apparatus for crystal growth equipment.

Background Art

[0002] The growth of silicon carbide crystals is difficult, and in the crystal growth process, the control requirements for the crystal growth temperature in the crystal growth equipment (for example, a growth furnace) are strict. In the crystal growth process, the temperature in the growth equipment will vary according to the growth course of the material. If the growth temperature in the growth equipment cannot be accurately controlled, quality problems (such as excessive convexity, impure crystals, etc.) will occur in the grown silicon carbide crystals.

[0003] Therefore, it is desired to provide a control method, system, and apparatus for crystal growth equipment. Thereby, monitor the crystal growth process during the silicon carbide growth process, and automatically adjust and control based on the growth status of the crystal to improve the quality of crystal growth.

Summary of the Invention

Means for Solving the Problems

[0004] An embodiment of this specification provides a control method for crystal growth equipment. The method includes: obtaining historical growth data of at least one crystal growth equipment, where the historical growth data includes at least historical growth control data and historical growth result data; and determining a control mode for the target crystal growth equipment based on the historical growth data and the target growth result, where the control mode includes at least one of a temperature control mode and a power control mode.

[0005] One embodiment of this specification provides a control system for a crystal growth apparatus, the system comprising: an acquisition module used to acquire hierarchical growth data of at least one crystal growth apparatus, wherein the hierarchical growth data includes at least hierarchical growth control data and hierarchical growth result data; and a determination module used to determine a control mode for a target crystal growth apparatus based on the hierarchical growth data and the target growth result, wherein the control mode includes at least one of a temperature control mode and a power control mode.

[0006] One embodiment of this specification provides a control device for a crystal growth apparatus, the apparatus including a processor, which is used to perform the control method for the crystal growth apparatus described in the above embodiment. [Brief explanation of the drawing]

[0007] This specification will be further described in the manner of exemplary embodiments, which will be described in detail by the drawings. These embodiments are not limiting, and in these embodiments, the same reference numerals represent the same structures, as follows:

[0008] [Figure 1] This is a schematic diagram illustrating application scenarios of the control systems for exemplary crystal growth equipment shown in some of the embodiments of this specification. [Figure 2] This is a block diagram of an exemplary control system for a crystal growth apparatus as shown in some of the embodiments of this specification. [Figure 3] This is a structural block diagram of an exemplary crystal growth apparatus shown in some of the examples described herein. [Figure 4] This is a flowchart illustrating a control method for an exemplary crystal growth apparatus as shown in some of the embodiments herein. [Figure 5] This is a schematic diagram illustrating the control mode of the target crystal growth equipment shown in some of the embodiments of this specification. [Figure 6]This flowchart exemplifies the determination of the first control mode of the target crystal growth apparatus shown in some embodiments of this specification. [Figure 7] This flowchart exemplifies the determination of the second control mode of the target crystal growth apparatus shown in some embodiments of this specification. [Figure 8] This is a schematic diagram illustrating the determination of the target growth power of the target crystal growth equipment shown in some of the embodiments of this specification. [Figure 9] This flowchart exemplifies the determination of the temperature retention parameters of the target crystal growth apparatus shown in some of the embodiments of this specification. [Figure 10] This flowchart exemplifies the determination of quality parameters of growth results as shown in some examples of this specification. [Figure 11] This is a schematic diagram illustrating the determination of the target control recipe for the target crystal growth equipment shown in some of the embodiments of this specification. [Figure 12] These are schematic diagrams of different growth stages of exemplary target crystal growth equipment as shown in some of the examples of this specification. [Modes for carrying out the invention]

[0009] To more clearly illustrate the technical solutions of the embodiments described herein, the following drawings, which are necessary for describing the embodiments, are briefly introduced. It is clear that the drawings in the following description are merely some examples or embodiments of this specification, and those skilled in the art can further apply this specification to other similar scenes based on these drawings, without requiring any creative work. Unless otherwise stated or evident from the linguistic context, the same numbers in the drawings refer to the same structure or operation.

[0010] As should be understood, the terms “system,” “apparatus,” “unit,” and / or “module” as used herein are one way of distinguishing different components, elements, members, parts, or assemblies at different levels. However, if other terms can achieve the same purpose, the above terms may be replaced by other expressions.

[0011] As specified herein and in the claims, unless the context explicitly indicates an exceptional situation, words such as “one,” “a type,” and / or “the” do not specifically refer to a singular number and may include plurals. Generally, the terms “includes” and “incorporates” indicate only the inclusion of clearly marked steps and elements, but these steps and elements do not constitute a single exclusive list, and a method or apparatus may include other steps or elements.

[0012] Flowcharts are used herein to illustrate operations performed by systems based on the embodiments described herein. It should be understood that the preceding or subsequent operations do not necessarily have to be performed precisely in order. Conversely, they may be performed in reverse order, or individual steps may be performed simultaneously. Simultaneously, other operations may be added to these processes, or some or more steps may be removed from these processes.

[0013] Figure 1 is a schematic diagram of an application scene of an exemplary crystal growth equipment control system as shown in some embodiments of this specification. In some embodiments, as shown in Figure 1, the application scene 100 of the crystal growth equipment control system may include a crystal growth machine 101, a storage device 102, a processing device 103, and a monitoring device 104.

[0014] The crystal growth apparatus 101 may be any apparatus used for crystal growth. The crystal growth apparatus 101 may include, but is not limited to, physical gas-phase transport growth apparatus, liquid-phase epitaxial growth apparatus, and Czochralski method growth apparatus.

[0015] As just one example, the crystal growth apparatus 101 may be a silicon carbide growth furnace, and as shown in Figure 3, the crystal growth apparatus 101 may include a furnace body 310, a crucible 306, a heat insulating material, and a heating coil (not shown). The furnace body 310 can accommodate other structures in the crystal growth apparatus 101. The heating coil can be used to control the temperature inside the furnace body 310. The crucible 306 is installed inside the furnace body 310. In some embodiments, the crucible 306 can be separated into a raw material chamber 309 and a growth chamber 311 by a gasket 308, the raw material chamber 309 can be used to contain silicon carbide powder material, and the growth chamber 311 can be used to grow silicon carbide crystals. In some embodiments, the growth chamber 311 may include a crystal growth position 307 (for example, a position for setting a seed). The heat-insulating material can be used to maintain or adjust the temperature inside the furnace body 310, and by setting the heat-insulating parameters of the heat-insulating material, the temperature inside the furnace body 310 and the temperature field distribution can be changed, thereby affecting the growth of silicon carbide crystals in the crystal growth equipment 101. In some embodiments, the heat-insulating material may include a top heat-insulating layer 301, a crucible top heat-insulating layer 302, a heat-insulating outer cylinder layer 303, a heat-insulating inner cylinder layer 304, and a bottom heat-insulating layer 305. In some embodiments, the thermal insulation parameters of the thermal insulation material may include one or a combination of several parameters, such as the distance D1 between the top of the crucible 306 and the top thermal insulation layer 301, the outer diameter R1 of the crucible top thermal insulation layer 302, the inner diameter R2 of the crucible top thermal insulation layer 302, the thickness H1 of the crucible top thermal insulation layer 302, the thickness H3 of the thermal insulation outer cylinder layer 303, the thickness H2 of the thermal insulation inner cylinder layer 304, and the thickness H4 of the bottom thermal insulation layer 305.

[0016] The crystal growth apparatus 101 can perform data exchange with the storage device 102. For example, the crystal growth apparatus 101 can transmit and store in real time relevant growth data (such as growth control data, growth result data, device parameters, etc.) in a storage device (such as the storage device 102) during the crystal growth process. The crystal growth apparatus 101 can perform data exchange with the processing device 103. For example, the crystal growth apparatus 101 can set or adjust its control mode, growth control parameters, etc. based on the instructions of the processing device 103.

[0017] The storage device 102 can store data, instructions, and / or any other information. In some embodiments, the storage device 102 can be used to store and execute, or use, the processing device 103 to complete the data and / or instructions of the exemplary methods described herein.

[0018] In some embodiments, the storage device 102 can store the data information of the crystal growth apparatus 101, the processing device 103, and the monitoring device 104. For example, the storage device 102 can store the historical growth data in the historical crystal growth process of the crystal growth apparatus 101. Further, for example, when the historical growth data of the crystal growth apparatus 101 and the target growth result are transmitted to the processing device 103 for further processing, the processing device 103 can store the processed data (such as the control mode and growth control parameters) in the storage device 102.

[0019] In some embodiments, the storage device 102 may include a large-capacity memory, a removable memory, a volatile read-write memory, a read-only memory (ROM), etc., or any combination thereof. In some embodiments, the storage device 102 can be realized on a cloud platform. In some embodiments, the storage device 102 may be part of the processing device 103.

[0020] The processing device 103 can process data and / or information acquired from the crystal growth device 101, the storage device 102, and / or the monitoring device 104. The processing device 103 can execute program instructions based on these data, information, and / or processing results to perform one or more functions described in the present application. For example, the processing device 103 can acquire the historical growth data of at least one crystal growth device and determine the control mode of the target crystal growth device based on the historical growth data and the target growth result. More details regarding the determination of the control mode can be referred to in FIGS. 5, 6, 7, and their related descriptions. Further, for example, the processing device 103 can determine the growth control parameters of the target crystal growth device based on the control mode of the target crystal growth device and the target growth result. When responding to the first control mode of the target crystal growth device being the power control mode, the processing device 103 can determine the first target power and the first target air pressure when the target crystal growth device is in the first control stage. When responding to the second control mode of the target crystal growth device being the temperature control mode, the processing device 103 can determine the target temperature and the target growth power when the target crystal growth device is in the second control stage. More details regarding the determination of the growth control parameters can be referred to in FIGS. 6, 7, 8, and their related descriptions.

[0021] In some embodiments, the processing unit 103 may be local or remote. In some embodiments, the processing unit 103 may include one or more sub-processing units (e.g., a single-core processing unit or a multi-core processing unit). As just one example, the processing unit 103 may include a central processor (CPU), an application-specific integrated circuit (ASIC), an application-specific instruction processor (ASIP), a graphics processor (GPU), a physical processor (PPU), a digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic circuit (PLD), a controller, a microcontroller unit, a reduced instruction set computer (RISC), a microprocessor, etc., or any combination thereof.

[0022] The monitoring device 104 can be used to acquire relevant data during the crystal growth process. For example, the monitoring device 104 can be used to acquire pressure information, temperature information, and air flow rate information, etc., from the crystal growth device 101 during the crystal growth process. In some embodiments, the monitoring device 104 may include, but is not limited to, temperature monitoring devices (e.g., temperature sensors, thermometers, etc.), pressure monitoring devices (e.g., pressure sensors, etc.), and air flow rate monitoring devices (e.g., flow meters, etc.). The monitoring device 104 can exchange data with the storage device 102 and transmit the acquired monitoring data to the storage device 102 for storage. The monitoring device 104 can exchange data with the processing device 103 and transmit the acquired monitoring data to the processing device 103 for processing.

[0023] In application scenario 100 of the control system for crystal growth equipment, data exchange between components can be achieved by a network or data connection. The network may be a wired network or a wireless network (e.g., Wi-Fi and Bluetooth).

[0024] It should be noted that the above description is provided for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make multiple variations and revisions under the guidance of the contents of this specification. Further and / or alternative exemplary embodiments can be obtained by combining and describing the features, structures, methods, and other features of the exemplary embodiments described herein in various ways. For example, the crystal growth apparatus 101, processing apparatus 103, and monitoring apparatus 104 may share a single memory device 102, or each may have its own memory device. However, these variations and revisions do not deviate from the scope of this specification.

[0025] Figure 2 is a block diagram of an exemplary control system for a crystal growth apparatus as shown in some embodiments of this specification. In some embodiments, the control system 200 for the crystal growth apparatus may include an acquisition module 201 and a determination module 202. In some embodiments, the control system 200 for the crystal growth apparatus can be implemented by a processing apparatus 103.

[0026] The acquisition module 201 can be used to acquire hierarchical growth data from at least one crystal growth machine.

[0027] The determination module 202 can be used to determine the control mode of the target crystal growth equipment based on hierarchical growth data and target growth results. In some embodiments, the determination module 202 can acquire the equipment parameters of the target crystal growth equipment and determine the control mode of the target crystal growth equipment based on the equipment parameters, hierarchical growth data, and target growth results. For more information on determining the control mode, please refer to Figure 4 and its related description.

[0028] In some embodiments, the decision module 202 can process instrument parameters and target growth results based on a control recommendation model to determine the control mode of the target crystal growth equipment, where the control recommendation model is an instrument learning model. More details about the control recommendation model can be found in Figure 5 and its related description.

[0029] In some embodiments, the determination module 202 can be further used to determine the growth control parameters of the target crystal growth equipment based on the control mode of the target crystal growth equipment and the target growth results. More details regarding the determination of growth control parameters can be found in Figure 4 and its related description.

[0030] In some embodiments, the determination module 202 can determine a first control mode when the target crystal growth equipment is in a first control stage, which is the raw material preheating stage. More details regarding the determination of the first control mode can be found in Figure 6 and its related description.

[0031] In some embodiments, the determination module 202 can determine a second control mode when the target crystal growth equipment is in a second control stage, the second control stage being the crystal growth stage. More details regarding the determination of the second control mode can be found in Figure 7 and its related description.

[0032] In some embodiments, when the second control mode is a temperature control mode, the determination module 202 can be used to determine the target temperature for the second control stage, to obtain the current growth temperature of the target crystal growth equipment, and to determine the target growth power of the target crystal growth equipment based on the current growth temperature and the target temperature. More details regarding the determination of the target growth power can be found in Figure 8 and its related description.

[0033] In some embodiments, in response to the second control mode being a power control mode, the determination module 202 can be used to determine a second target power in the second control phase, and the second target power is used to control a target crystal growth device in the second control phase.

[0034] In some embodiments, the determination module 202 can determine the target temperature field distribution for the second control stage based on the second control mode, and determine the heat retention parameters of the target crystal growth equipment based on the target temperature field distribution. For more information on determining the heat retention parameters, please refer to Figure 9 and its related description.

[0035] In some embodiments, the determination module 202 can be used to determine a target control recipe for a target crystal growth machine based on the control mode. More details regarding the determination of the target control recipe can be found in Figures 10 and 11 and their related descriptions.

[0036] In some embodiments, the determination module 202 can simulate the crystal growth process based on a target control recipe and determine quality parameters that can reflect the quality of the growth result. More details on determining the quality parameters of the growth result can be found in Figure 11 and its related description.

[0037] It should be noted that the above description of the control system 200 of the crystal growth apparatus and its modules is for ease of description only and is not limited to the embodiments cited herein. To the extent that it can be understood, a person skilled in the art, after understanding the principle of the system, may arbitrarily combine individual modules or configure subsystems to connect with other modules, without departing from this principle. In some embodiments, the acquisition module 201 and the determination module 202 may be different modules in a single system, or they may be a single module that implements the functions of two or more of the above modules. For example, individual modules may share a single storage module, or each individual module may have its own storage module. Any such variations are within the scope of protection of this specification.

[0038] Figure 4 is a flowchart of an exemplary control method for a crystal growth apparatus as shown in some embodiments of this specification.

[0039] In some embodiments, the control method for the crystal growth apparatus can be performed by a processor or a control system 200 for the crystal growth apparatus. For example, process 400 can be stored in a storage device (e.g., storage device 102) in the form of a program or instructions, and process 400 can be realized when the processor or the control system 200 for the crystal growth apparatus executes the program or instructions. The schematic diagram of process 400 operation presented below is illustrative. In some embodiments, the process can be completed by using one or more additional operations not described and / or one or more operations not considered. Furthermore, the order of operations of process 400 shown in Figure 4 and described below is not limiting.

[0040] Step 401: Obtain historical growth data for at least one crystal growth machine.

[0041] The hysteretic growth data may also be relevant data from the hysteretic crystal growth process (i.e., the hysteretic growth process) of at least one crystal growth machine.

[0042] In some embodiments, the historical growth data may include historical growth control data and historical growth result data.

[0043] The hierarchical growth control data may be growth control data from the hierarchical crystal growth process of at least one crystal growth machine. In some embodiments, the growth control data may include control modes, growth control parameters, and heat retention parameters at different control stages in the crystal growth process of the crystal growth machine. In some embodiments, the hierarchical growth control data may include hierarchical control modes, hierarchical growth control parameters, and hierarchical heat retention parameters at different control stages in the hierarchical crystal growth process of one or more crystal growth machines.

[0044] The control stages may be stages divided based on the temperature control status of the crystal growth equipment during the crystal growth process. In some embodiments, as shown in Figure 12, the control stages in the crystal growth process may include a first control stage, a second control stage, a temperature adaptation stage, and a cooling stage. The first control stage is the raw material preheating stage, in which the temperature of the crystal growth equipment is continuously raised to a certain temperature value (also called the growth temperature), so that the raw material in the crystal growth equipment gradually reaches the growth temperature. After the first control stage is completed, the second control stage is entered. The second control stage is the crystal growth stage, in which the temperature of the crystal growth equipment is maintained so that the growth temperature does not change (temperature fluctuations do not exceed ±0.1°C). After the second control stage is completed, the temperature adaptation stage is entered. In the temperature adaptation stage, the temperature of the crystal growth equipment is gradually lowered from the growth temperature to a temperature close to the crystal surface. After the temperature adaptation stage is completed, the cooling stage is entered. In the cooling stage, the temperature of the crystal growth equipment is gradually lowered to room temperature (for example, 25°C ± 5°C).

[0045] The control mode may be the operating mode used to control the crystal growth equipment. More information about the control mode can be found in step 402 and its related descriptions. The growth control parameters may be the related parameters used to control crystal growth. More information about the growth control parameters can be found in the related descriptions below. The heat retention parameters may be the related parameters used to maintain the heat of the crystal growth equipment. More information about the heat retention parameters can be found in Figure 9 and its related descriptions.

[0046] The hysteretic growth result data may also be data related to crystals (e.g., silicon carbide single crystals) produced during the hysteretic crystal growth process by at least one crystal growth instrument.

[0047] In some embodiments, the growth result data may include crystal thickness, size, convexity range, concavity range, crystal quality status, slice thickness, and number of slices. Crystal quality status may include the presence or absence of defects, the type of defects (e.g., phase transitions, dislocations, microtubules, and inclusions), and the severity of the defects. The severity of defects can be artificially labeled and categorized based on prior knowledge or experience.

[0048] In some embodiments, the hysteretic growth data may further include hysteretic instrument parameters of at least one crystal growth apparatus. The hysteretic instrument parameters may be instrument parameters used by at least one crystal growth apparatus in the hysteretic crystal growth process. Instrument parameters corresponding to different crystal growth apparatuses may be different. In some embodiments, the instrument parameters of a crystal growth apparatus may be set as needed, and the instrument parameters in the crystal growth process of different batches of the same crystal growth apparatus may be set to be the same or different. In some embodiments, the instrument parameters may include, but are not limited to, the extreme vacuum of the crystal growth apparatus, pressure control range, pressure control accuracy under different pressure levels, maximum heating temperature, temperature control accuracy, power control range, and power control accuracy.

[0049] In some embodiments, the hierarchical growth data of at least one crystal growth machine can be read from a storage device. Here, the storage device may be a storage device 102 attached to the control system of the crystal growth machine, or it may be an external storage device not belonging to the control system of the crystal growth machine, such as a hard disk and an optical disk. In some embodiments, the hierarchical growth data of at least one crystal growth machine can be read by an interface, which includes, but is not limited to, a program interface, a data interface and a transmission interface. In some embodiments, the hierarchical growth data of at least one crystal growth machine can be automatically extracted from the interface when the control system of the crystal growth machine is operating. In some embodiments, the control system of the crystal growth machine can be called to other external devices or systems, and when called, the hierarchical growth data of at least one crystal growth machine is transmitted to the control system of the crystal growth machine. In some embodiments, the hierarchical growth data of at least one crystal growth machine can be obtained using any method well known to those skilled in the art, and this specification is not limited thereto.

[0050] Step 402: Based on the historical growth data and target growth results, determine the control mode of the target crystal growth equipment.

[0051] The target growth result may be growth result data from the desired crystal growth equipment. More details about the growth result data can be found in the related descriptions above. The target growth result can be obtained through user input. For example, the user can upload the target growth result from their terminal.

[0052] The target crystal growth equipment may be the crystal growth equipment currently used for production.

[0053] The control mode may also be the operating mode of the target crystal growth equipment. Under different control modes, the target crystal growth equipment can control the temperature inside the crystal growth equipment (e.g., the growth chamber) in different ways.

[0054] In some embodiments, the control modes may include a temperature control mode and a power control mode.

[0055] In temperature control mode, the processor can automatically adjust the power of the target crystal growth equipment so that the temperature of at least one temperature measurement point does not change from its corresponding preset temperature (or the temperature fluctuation does not exceed ±0.1°C). The temperature measurement point may be located inside the crucible, which is the target crystal growth equipment, and / or inside the furnace body. Different temperature measurement points may have different preset temperatures. The location, number, and preset temperatures for different temperature measurement points may be system default values, empirical values, and artificially set values, or any combination thereof, or may be set based on actual needs, and this specification is not limited thereto.

[0056] In power control mode, the processor can directly set the power of the target crystal growth machine, causing the machine to operate according to the set power. In some embodiments, the set power may be the operating power of the heating coil in the target crystal growth machine, and the heating coil can heat the machine according to the set power. In some embodiments, the set power may be a fixed value. In some embodiments, the set power may be a power value that changes over time.

[0057] In some embodiments, the processor can determine the control mode of the target crystal growth equipment based on historical growth data and target growth results using multiple methods. For example, the processor can construct a target feature vector based on historical growth data and target growth results, match the target feature vector in a vector database to determine an association vector that fits preset conditions, and then determine the final control mode of the target crystal growth equipment based on the control mode of the target crystal growth equipment corresponding to the determined association vector.

[0058] A vector database refers to a database used to store, index, and query vectors. A vector database enables rapid similarity queries and other vector management operations when dealing with large numbers of vectors. In some embodiments, the vector database may include multiple reference feature vectors and corresponding control modes for target crystal growth equipment.

[0059] In some embodiments, a reference feature vector can be constructed by screening historical growth data from multiple crystal growth machines to satisfy the corresponding historical target growth result in the historical growth result data. For example, multiple reference feature vectors can be obtained by performing feature extraction on this portion of the historical growth data and the historical target growth result. Here, the feature extraction method may include principal component analysis and machine learning. The control mode of the target crystal growth machine corresponding to the reference feature vector can be obtained based on the historical growth data. In some embodiments, a vector database can be constructed based on multiple reference feature vectors and the corresponding control modes of the target crystal growth machine.

[0060] A preset condition may refer to a decision condition used to determine the related vector. In some embodiments, the preset condition may include the condition that the vector distance satisfies a distance threshold and that the vector distance is the minimum.

[0061] In some embodiments, the processor can acquire the instrument parameters of the target crystal growth machine and determine the control mode of the target crystal growth machine based on the instrument parameters, historical growth data, and target growth results. More details about the instrument parameters can be found in step 401 and its related descriptions.

[0062] The instrument parameters of the target crystal growth machine can be obtained by several methods. For example, the instrument parameters of the target crystal growth machine can be obtained by user input. For example, a user can upload the instrument parameters of the target crystal growth machine from a terminal. Furthermore, for example, the instrument parameters of the target crystal growth machine can be read from a storage device. Here, the storage device may be a storage device 102 attached to the control system of the crystal growth machine, or it may be an external storage device not belonging to the control system of the crystal growth machine, such as a hard disk and an optical disk. In some embodiments, the instrument parameters of the target crystal growth machine can be read by an interface, which includes, but is not limited to, a program interface, a data interface, and a transmission interface. In some embodiments, the instrument parameters of the target crystal growth machine can be automatically extracted from the interface when the control system of the crystal growth machine is operating. In some embodiments, the instrument parameters of the target crystal growth machine may be obtained using any method well known to those skilled in the art, and this specification is not limited thereto.

[0063] In some embodiments, the processor can determine the control mode of the target crystal growth equipment using a first preset comparison table, based on the equipment parameters of the target crystal growth equipment, historical growth data, and the target growth result. In some embodiments, the first preset comparison table includes multiple different reference equipment parameters, multiple different reference growth results, and the correspondence between the multiple different reference growth results and the reference equipment parameters and the reference control mode. Here, the reference growth result can be determined based on historical growth result data in the historical growth data, and the reference equipment parameters can be determined based on historical equipment parameters in the historical growth data. In some embodiments, the first preset comparison table can be obtained by constructing multiple different reference equipment parameters, multiple different reference growth results, and the correspondence between the multiple different reference growth results and the reference equipment parameters and the reference control mode based on prior knowledge or historical data (for example, screening data from historical growth data of different crystal growth equipment where the historical growth result satisfies the corresponding historical target growth result). In some embodiments, the processor searches a first preset comparison table based on the instrument parameters of the target crystal growth equipment, historical growth data, and target growth results to determine reference instrument parameters that are close to the instrument parameters, further determines the reference growth result that best matches the target growth result, and determines the reference control mode corresponding to the reference growth result as the control mode of the target crystal growth equipment.

[0064] In some embodiments, the processor can process the instrument parameters of the target crystal growth machine and the target growth result based on a control recommendation model, and determine the control mode of the target crystal growth machine, where the control recommendation model is an instrument learning model. For more details on determining the control mode based on the control recommendation model, please refer to Figure 5 and its related description.

[0065] In some embodiments, the target crystal growth apparatus may have the same or different control modes at different control stages. The control mode corresponding to the target crystal growth apparatus in the first control stage is also called the first control mode, and the control mode corresponding to the target crystal growth apparatus in the second control stage is also called the second control mode. Accordingly, the processor can determine the first control mode when the target crystal growth apparatus is in the first control stage, and the second control mode when the target crystal growth apparatus is in the second control stage. More details on determining the first control mode and determining the second control mode can be found in Figures 6 and 7 and their related descriptions, respectively.

[0066] In some embodiments of this specification, by analyzing the hierarchical growth data and target growth results of multiple crystal growth machines, the disciplines in the hierarchical crystal growth process are summarized. This allows for the automatic control of the current crystal growth process parameters based on the hierarchical data, effectively reducing errors caused by manual adjustment control. This ensures that the actual growth results of the crystal growth machine better match the target growth results, further ensuring the accuracy of automatic adjustment control and improving the quality of crystal growth. Simultaneously, determining the control mode based on the instrument parameters of the crystal growth machine allows for full consideration of the impact of instrument differences between different crystal growth machines on automatic adjustment, thereby improving the accuracy of automatic adjustment control.

[0067] In some other embodiments, the control mode of the target crystal growth machine may be determined by user input, and the processor can acquire the user-inputted control mode. In some embodiments, the user can input control modes for individual control stages of the target crystal growth machine (e.g., first control stage, second control stage, etc.), and the processor can acquire the user-inputted control modes for each individual control stage. In some embodiments, the control mode of the target crystal growth machine may be a preset default mode, and the processor can acquire the control modes for each individual control stage from the target crystal growth machine.

[0068] In some embodiments, after determining or acquiring the control mode of the target crystal growth equipment, the processor can determine the growth control parameters of the target crystal growth equipment based on the control mode of the target crystal growth equipment and the target growth results.

[0069] Growth control parameters may also be related parameters used to control crystal growth. In some embodiments, growth control parameters may include the temperature, air pressure, power, and air flow rate of the target crystal growth equipment.

[0070] In some embodiments, the processor can determine the growth control parameters of the target crystal growth equipment using a second preset comparison table, based on the control mode of the target crystal growth equipment and the target growth result. In some embodiments, the second preset comparison table includes a correspondence between multiple different reference control modes and multiple different reference growth results and reference growth control parameters. In some embodiments, the second preset comparison table can be obtained by constructing a correspondence between multiple different reference control modes and multiple different reference growth results and reference growth control parameters based on prior knowledge or historical data (for example, screening historical growth data from different crystal growth equipment to find data where the historical growth result satisfies the corresponding historical target growth result). In some embodiments, the processor can search the second preset comparison table based on the control mode of the target crystal growth equipment and the target growth result to determine the reference control mode and reference growth result that best match the control mode and reference growth result, and determine the reference growth control parameters corresponding to the reference control mode and reference growth result as the growth control parameters of the target crystal growth equipment.

[0071] In some embodiments, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results based on a control recommendation model, and determine the growth control parameters of the target crystal growth equipment. For more details on determining growth control parameters based on a control recommendation model, please refer to Figure 5 and its related description.

[0072] In some embodiments, the target crystal growth apparatus may undergo multiple control stages during the crystal growth process, and the control modes of different control stages may be different, as may the growth control parameters of different control stages. In some embodiments, the processor may first determine the control mode corresponding to a particular control stage, and then determine the growth control parameters of that control stage.

[0073] In some embodiments, when the first control mode of the target crystal growth machine is in the first control phase is a power control mode, the processor can determine a first target power for the target crystal growth machine in the first control phase. More details about the first target power can be found in Figure 6 and its related description.

[0074] In some embodiments, when the second control mode of the target crystal growth machine is in the second control stage is a temperature control mode, the processor can determine the target temperature of the target crystal growth machine in the second control stage, and determine the current growth temperature of the target crystal growth machine and the target growth power of the target crystal growth machine based on the target temperature. More details regarding the determination of the target temperature of the target crystal growth machine in the second control stage can be found in Figure 7 and its related description. More details regarding the determination of the target growth power of the target crystal growth machine in the second control stage can be found in Figure 8 and its related description.

[0075] In some embodiments, the processor can determine a second target power for the target crystal growth equipment in the second control phase when the second control mode is a power control mode while the target crystal growth equipment is in the second control phase. More details regarding the determination of the second target power for the target crystal growth equipment in the second control phase can be found in Figure 7 and its related description.

[0076] In some embodiments, the processor may further determine a second target air pressure and / or target air flow rate for the second control stage based on a second control mode. More details regarding the determination of the second target air pressure and target air flow rate can be found in Figure 7 and its related description.

[0077] Figure 5 is a schematic diagram illustrating the control mode of the target crystal growth apparatus shown in some embodiments of this specification.

[0078] In some embodiments, as shown in Figure 5, the processor can process the instrument parameters 501 and target growth results 502 of the target crystal growth equipment based on the control recommendation model 503, and determine the control mode 504 of the target crystal growth equipment.

[0079] The control recommendation model 503 may also be an instrument learning model used to determine the control mode of the target crystal growth instrument. For example, the control recommendation model 503 may include one of the following: a neural network (NN) model, a back-propagation neural network (BP) model, and a convolutional neural network (CNN) model, or any combination thereof.

[0080] In some embodiments, the control recommendation model 503 includes at least 800 multiplication operations in a single run. In some embodiments, the control recommendation model 503 is executed at least partially by the GPU.

[0081] In some embodiments, the input to the control recommendation model 503 may include the instrument parameters 501 and target growth results 502 of the target crystal growth equipment, and the output may include the control mode 504 of the target crystal growth equipment. More details about the instrument parameters, target growth results, and control modes can be found in Figure 4 and its related descriptions.

[0082] In some embodiments, the output of the control recommendation model 503 may further include growth control parameters for the target crystal growth equipment. More details about the growth control parameters can be found in Figure 4 and its related description.

[0083] In some embodiments, the control recommendation model 503 can be obtained by acquiring a plurality of first training samples 505 and corresponding first labels 506, and training the control recommendation model 503 based on the plurality of first training samples 505 and corresponding first labels 506.

[0084] As an example, multiple first training samples 505 bearing a first label 506 are input to the initial control recommendation model 507. A loss function is constructed using the first label 506 and the output of the initial control recommendation model 507. The parameters of the initial control recommendation model 507 are iteratively updated based on the loss function using gradient descent or other methods. Model training is completed and a trained control recommendation model 503 is obtained when the loss function of the initial control recommendation model 507 satisfies preset conditions. Here, the preset conditions may include convergence of the loss function and reaching a threshold number of iterations.

[0085] In some embodiments, the training samples of the control recommendation model 503 include historical growth data. This historical growth data may be data where the historical growth result data is similar to or close to the corresponding historical target growth result data.

[0086] In some embodiments, the first training sample 505 includes the historical instrument parameters of the first sample crystal growth apparatus and the historical target growth results. In some embodiments, the first sample crystal growth apparatus may include multiple crystal growth apparatuses. In some embodiments, the first sample crystal growth apparatus may include a target crystal growth apparatus.

[0087] In some embodiments, the first label 506 may include a history control mode and / or history growth control parameters of the first sample crystal growth apparatus.

[0088] In some embodiments, the first label 506 may include history control modes in individual history control stages of the first sample crystal growth apparatus. For example, the first label 506 may include a history first control mode in which the first sample crystal growth apparatus is in history first control stage, and a history second control mode in which the first sample crystal growth apparatus is in history second control stage.

[0089] In some embodiments, the first training sample 505 and the first label 506 can be determined based on screening data in the history growth data of multiple crystal growth machines in which the history growth result data is similar to or close to the corresponding history target growth result. For example, the history machine parameters and the history target growth result in the history growth data corresponding to the data can be determined as the first training sample 505, and the history control mode in the history growth data corresponding to the data can be determined as the first label 506. Furthermore, for example, the history growth control parameters in the history growth data corresponding to the data can be determined as the first label 506.

[0090] In some embodiments, the control recommendation model 503 may include an input layer, M hidden layers, and an output layer, where the number of neurons in the input layer is equal to the number of input variables, and the number of neurons in the output layer is equal to the number of outputs associated with each input.

[0091] In some embodiments, M may be equal to 2. In some embodiments, M may be an integer greater than or equal to 2. In some embodiments, the number of neurons in the hidden layer can be determined based on the number of first training samples, the number of neurons in the input layer, and the number of neurons in the output layer. An exemplary determination formula is as follows:

number

number

[0092] In some embodiments, the control recommendation model 503 includes an activation function. In some embodiments, the activation function may include, but is not limited to, a rectified linear unit (ReLU). In some embodiments, the activation function may include a first activation function and a second activation function, wherein the first and second activation functions are of the same type. In some embodiments, a single dropout layer may be included between the first and second activation functions, and the initial parameter of the dropout layer may be set to 0.2, which can be gradually reduced based on the training effect of the model.

[0093] In some embodiments of this specification, the control recommendation model can determine the control mode and growth control parameters of the target crystal growth equipment based on a large number of broad features, thereby enabling the predicted control mode and growth control parameters of the crystal growth equipment to have higher accuracy, and further enabling highly accurate and automated adjustment control of the crystal growth process in the actual production process.

[0094] Figure 6 is a flowchart illustrating the determination of a first control mode for a target crystal growth machine as shown in some embodiments of this specification. In some embodiments, the method for determining the first control mode can be performed by a processor or the control system 200 of the crystal growth machine. For example, process 600 can be stored in a storage device (e.g., storage device 102) in the form of a program or instructions, and process 600 can be realized when the processor or the control system 200 of the crystal growth machine executes the program or instructions. The schematic diagrams of process 600 operations presented below are illustrative. In some embodiments, the process can be completed by using one or more additional operations not described and / or one or more operations not considered. Furthermore, the order of operations of process 600 shown in Figure 6 and described below is not limiting.

[0095] Step 601: Determine the first control mode in which the target crystal growth machine is in the first control stage.

[0096] The first control stage may also be a raw material preheating stage. More details about the first control stage can be found in Figure 12 and its related description.

[0097] The first control mode may also be a control mode in which the target crystal growth equipment is in the first control stage.

[0098] In some embodiments, the processor can determine the first control mode in which the target crystal growth equipment is in the first control stage by several methods. For example, the processor can determine the first control mode in which the target crystal growth equipment is in the first control stage by performing processing such as data fitting and statistical analysis on historical growth data. For example, the processor can determine the first control mode in which the target crystal growth equipment is in the first control stage by processing the equipment parameters of the target crystal growth equipment and the target growth results using a control recommendation model. More details about the control recommendation model can be found in Figure 5 and its related description. Furthermore, for example, the processor can determine the first control mode in which the target crystal growth equipment is in the first control stage based on a first preset rule. An exemplary first preset rule may determine the first control mode based on the degree of accuracy of temperature measurement at the temperature measurement point. If the degree of accuracy of temperature measurement in the first control stage is less than the default value, i.e., the actual temperature value of the temperature measurement point in the first control stage cannot be accurately obtained, then the first control mode when the target crystal growth equipment is in the first control stage is the power control mode. In some embodiments, the degree of accuracy of temperature measurement at a temperature measurement point when the target crystal growth equipment is in the first control stage can be determined based on the historical growth data of the target crystal growth equipment.

[0099] In some other embodiments, the processor may determine the first control mode of the target crystal growth machine by a method of obtaining user input. In some other embodiments, the target crystal growth machine has a pre-set control mode for a first control stage, and the processor can determine the first control mode of the target crystal growth machine by obtaining information from the target crystal growth machine.

[0100] Step 602: In response to the first control mode being a power control mode, the first target power for the first control stage is determined.

[0101] In some embodiments, the first target power may be a fixed power set for the target crystal growth machine under power control mode, and the target crystal growth machine may operate directly at the fixed power in the first control stage. In some embodiments, the target crystal growth machine may also gradually increase its operating power when actually operating, in which case the first target power may be the power value corresponding to when the target crystal growth machine stops adjusting its power, that is, the target crystal growth machine can maintain operation at the first target power when it reaches the first target power.

[0102] In some embodiments, the first target power can be used to control the target crystal growth equipment in the first control stage. For example, in the first control stage, the target crystal growth equipment can be controlled to adjust its power according to a certain power adjustment range, stop adjusting the power until the power of the target crystal growth equipment increases to the first target power, and maintain the power of the target crystal growth equipment at the first target power.

[0103] In some embodiments, the processor can determine the first target power of the first control stage by multiple methods.

[0104] In some embodiments, the processor can determine the first target power for the first control phase by a control recommendation model. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results by the control recommendation model to determine the first target power for the first control phase of the target crystal growth equipment. The corresponding first label used to train the control recommendation mode, i.e., the hierarchical growth control parameters of the first sample crystal growth equipment, may include the hierarchical first target power for the hierarchical first control phase of the first sample crystal growth equipment. More details about the control recommendation model can be found in Figure 5 and its related description.

[0105] After the target crystal growth equipment reaches a certain temperature value (which may be referred to as the critical temperature), the power input is stopped. At this point, the temperature of the target crystal growth equipment rises continuously to a certain extent and finally stabilizes at the required temperature value for the first control stage.

[0106] In some embodiments, the processor can determine the critical temperature of the target crystal growth apparatus in the first control stage by multiple methods. Under power control mode, when the temperature of the target crystal growth apparatus reaches the critical temperature, the processor can control the target crystal growth apparatus to stop power input.

[0107] In some embodiments, the processor can determine the critical temperature of the first control stage by a control recommendation model. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results by the control recommendation model to determine the critical temperature of the target crystal growth equipment in the first control stage. The corresponding first label used to train the control recommendation mode, i.e., the hysteretic growth control parameters of the first sample crystal growth equipment, may include the hysteretic critical temperature of the first control stage of the hysteretic first control stage of the first sample crystal growth equipment. More details about the control recommendation model can be found in Figure 5 and its related description.

[0108] In some embodiments, the processor can determine the critical temperature of the first control stage using a critical temperature determination model. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results using a critical temperature determination model to determine the critical temperature of the target crystal growth equipment in the first control stage.

[0109] The critical temperature determination model may also be an instrument learning model, such as a CNN model or a DNN model.

[0110] In some embodiments, the critical temperature determination model includes at least 800 multiplication operations in a single run. In some embodiments, the critical temperature determination model is executed at least partially on a GPU.

[0111] In some embodiments, the critical temperature determination model can be obtained by training on a large number of fourth training samples bearing a fourth label. In some embodiments, the fourth training samples may include the hysteretic instrument parameters and hysteretic growth results of the fourth sample crystal growth instrument. The fourth label may include the hysteretic critical temperature at the hysteretic first control stage of the fourth sample crystal growth instrument. The training method of the critical temperature determination model can be based on the training method of the control recommendation model, and more details can be found in Figure 5 and its related description.

[0112] In some embodiments, during the first control phase, after stabilizing within a certain temperature range under power control mode, the temperature in the target crystal growth apparatus may not directly reach a constant temperature state. In this case, the temperature of the target crystal growth apparatus can be adjusted to a constant temperature state by adjusting the air pressure of the target crystal growth apparatus. Under a constant temperature state, the temperature at each individual location inside the target crystal growth apparatus is the same or approximately the same (for example, temperature fluctuations do not exceed ±1°C). For more details on how to determine whether a constant temperature state is reached, please refer to Figure 7 and its related description.

[0113] The following describes how to determine how to adjust the air pressure of the target crystal growth equipment by step 603. It should be noted that step 603 is not a mandatory step, and it is not necessary to perform step 603 immediately after steps 601 and 602.

[0114] Step 603: Determine the first target air pressure for the first control stage based on the first target power.

[0115] The first target air pressure may be the air pressure value that adjusts the temperature of the target crystal growth equipment to a constant temperature state.

[0116] In some embodiments, the processor can determine the first target air pressure in the first control stage by multiple methods.

[0117] In some embodiments, the processor can determine the first target air pressure for the first control stage by a control recommendation model. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results by the control recommendation model to determine the first target air pressure for the first control stage of the target crystal growth equipment. The corresponding first label used to train the control recommendation mode, i.e., the hierarchical growth control parameters of the first sample crystal growth equipment, may include the hierarchical first target air pressure for the hierarchical first control stage of the first sample crystal growth equipment. More details about the control recommendation model can be found in Figure 5 and its related description.

[0118] In some embodiments, after determining a first target air pressure, the processor can adjust the air pressure of the target crystal growth apparatus based on the first target air pressure using a preset air pressure adjustment scheme. An exemplary air pressure adjustment scheme may include changing the pressure difference between the furnace and the crucible by adjusting the furnace air pressure. For example, the furnace air pressure can be adjusted by controlling the intake volume (positive pressure) or exhaust volume (negative pressure) of the furnace.

[0119] In some other embodiments, the processor may determine, based on hierarchical growth data, that the control mode in which the target crystal growth apparatus is in a first control stage is a temperature control mode. When the first control stage is a temperature control mode, the processor may determine, based on hierarchical growth data, the relationship (e.g., a temperature-time curve) in which the temperature of the first control stage changes over time, and then further control the input power of the target crystal growth apparatus so that the temperature inside the target crystal growth apparatus changes over time precisely according to this relationship.

[0120] Figure 7 is a flowchart illustrating the determination of a second control mode for a target crystal growth machine as shown in some embodiments of this specification. In some embodiments, the method for determining the second control mode can be performed by a processor or the control system 200 of the crystal growth machine. For example, process 700 can be stored in a storage device (e.g., storage device 102) in the form of a program or instructions, and process 700 can be realized when the processor or the control system 200 of the crystal growth machine executes the program or instructions. The schematic diagrams of process 700 operations presented below are illustrative. In some embodiments, the process can be completed by using one or more additional operations not described and / or one or more operations not considered. Furthermore, the order of operations of process 700 shown in Figure 7 and described below is not limiting.

[0121] Step 701: Determine the second control mode in which the target crystal growth equipment is in the second control stage.

[0122] The second control step may be the crystal growth step. More details about the second control step can be found in Figure 12 and its related description.

[0123] The second control mode may also be a control mode in which the target crystal growth equipment is in the second control stage.

[0124] In some embodiments, the processor can determine the second control mode in which the target crystal growth equipment is in the second control stage by several methods. For example, the processor can process the equipment parameters of the target crystal growth equipment and the target growth results using a control recommendation model to determine the second control mode in which the target crystal growth equipment is in the second control stage. More details about the control recommendation model can be found in Figure 5 and its related description.

[0125] In some other embodiments, the processor may also determine the second control mode of the target crystal growth machine by obtaining user input. In some other embodiments, the target crystal growth machine has a pre-set control mode for the second control stage, and the processor can determine the second control mode of the target crystal growth machine by obtaining information from the target crystal growth machine.

[0126] One point that needs to be explained is that different growth control parameters can be determined when the second control mode is a different control mode. Below, steps 702-704 describe the situation when the second control mode is a temperature control mode, and steps 705-707 describe the situation when the second control mode is a power control mode. The order between steps 702-704 and steps 705-707 is not restrictive.

[0127] Step 702: In response to the second control mode being a temperature control mode, the target temperature for the second control stage is determined.

[0128] The target temperature may refer to a preset temperature that the target crystal growth apparatus must reach under temperature control mode. In some embodiments, the target temperature may be the temperature that the target crystal growth apparatus must maintain during the crystal growth stage. In some embodiments, the target temperature may be the same as, or approximately the same as, the constant temperature reached in the first control stage.

[0129] In some embodiments, the processor can determine the target temperature for the second control stage using multiple methods.

[0130] In some embodiments, the processor can determine the target temperature for the second control stage by a control recommendation model. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results by the control recommendation model to determine the target temperature for the second control stage of the target crystal growth equipment. The corresponding first label used to train the control recommendation mode, i.e., the hysteretic growth control parameters of the first sample crystal growth equipment, may include the hysteretic target temperature for the hysteretic second control stage of the first sample crystal growth equipment. More details about the control recommendation model can be found in Figure 5 and its related description.

[0131] Step 703: Obtain the current growth temperature of the target crystal growth equipment.

[0132] The current growth temperature may refer to the temperature of the target crystal growth equipment at the current time. In some embodiments, the current time may correspond to the moment when the target crystal growth equipment enters the second control phase. In some embodiments, the current time may be a point in time during the second control phase of the target crystal growth equipment.

[0133] In some embodiments, the processor can obtain the current growth temperature of the target crystal growth machine in real time using monitoring equipment (e.g., temperature monitoring equipment). For example, the temperature monitoring equipment can detect the temperature at the crystal growth location in the target crystal growth machine in real time and obtain the current growth temperature.

[0134] In some embodiments, the processor can obtain the current growth temperature based on at least three temperature measurement points. For example, the current growth temperature of the target crystal growth equipment can be obtained by performing processing such as weighting on the temperature values ​​detected at at least three temperature measurement points.

[0135] In some embodiments, the axial positions of the at least three temperature measurement points are different. The axial direction may be perpendicular to the bottom plane of the crystal growth apparatus. Different axial positions of the at least three temperature measurement points may mean that the at least three temperature measurement points are not in the same horizontal plane. For example, one temperature measurement point may be located at the bottom of the crucible (e.g., the center of the bottom), one at the top of the crucible (e.g., the center of the top), and the remaining temperature measurement points may be located between these two points.

[0136] Step 704: Determine the target growth power of the target crystal growth equipment based on the current growth temperature and the target temperature.

[0137] The target growth power of the target crystal growth apparatus may be the power status of the target crystal growth apparatus in the second control stage. When the second control mode in the second control stage is a temperature control mode, the target growth power may fluctuate in order to stabilize the temperature of the target crystal growth apparatus. In some embodiments, the target growth power can be used to control the target crystal growth apparatus in the second control stage. In some embodiments, the target growth power can be determined in real time based on the acquired current growth temperature and target temperature. In some embodiments, the target growth power can be represented by a power curve, where the horizontal axis is time and the vertical axis is power, and the power curve can be generated in real time based on the current growth temperature and target temperature.

[0138] In some embodiments, the processor can determine the target growth power of the target crystal growth equipment based on the current growth temperature and the target temperature using a variety of methods.

[0139] In some embodiments, the processor can process the current growth temperature and target temperature based on a power determination model to determine the target growth power of the target crystal growth equipment. For more information on determining the target growth power based on the power determination model, please refer to Figure 8 and its related description.

[0140] Step 705: In response to the second control mode being a power control mode, the second target power for the second control stage is determined.

[0141] The second target power may refer to the preset power in the second control stage of the target crystal growth apparatus under power control mode. In some embodiments, the second target power may be a constant power value. In some embodiments, the second target power may be a set power value that changes with time, where the correspondence between power and time is maintained so as not to change after being set.

[0142] In some embodiments, the second target power can be used to control the target crystal growth equipment in the second control stage. For example, the power of the target crystal growth equipment can be controlled to be maintained at the second target power in the second control stage.

[0143] In some embodiments, the processor can determine the second target power of the second control stage by multiple methods.

[0144] In some embodiments, the processor can determine the second target power for the second control phase by a control recommendation model. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results by the control recommendation model to determine the second target power for the target crystal growth equipment in the second control phase. The corresponding first label used to train the control recommendation mode, i.e., the hysteretic growth control parameters of the first sample crystal growth equipment, may include the hysteretic second target power for the hysteretic second control phase of the first sample crystal growth equipment. More details about the control recommendation model can be found in Figure 5 and its related description.

[0145] Under power control mode, the target crystal growth equipment operates according to the set power, which may prevent it from responding to emergencies that occur during the crystal growth process, resulting in poor crystal growth results. In this case, monitoring the real-time temperature conditions during the crystal growth process allows for appropriate measures to be taken based on the real-time temperature conditions. This is further explained in steps 706 and 707. It should be noted that steps 706 and 707 are not mandatory steps, and it is not necessary to perform steps 706 and 707 immediately after step 705.

[0146] Step 706: Obtain the current growth temperature of the target crystal growth equipment.

[0147] For more information on the current growth temperature and how to obtain it, please refer to step 703 and its related descriptions.

[0148] Step 707: In response to the current growth temperature fluctuation being greater than the preset fluctuation threshold, the second control mode is switched from power control mode to temperature control mode.

[0149] The current growth temperature fluctuation may refer to the temperature difference between the current growth temperature and the growth temperature at a previous point in time. Here, the previous point in time may be any point in time prior to the current point in time. In some embodiments, the time interval for acquiring the current temperature may be predetermined, for example, set to 1 second, 2 seconds, and 5 seconds.

[0150] In some embodiments, when the current growth temperature fluctuation is greater than a preset fluctuation threshold, the processor can issue an alert command and / or an alarm command, and control associated alert devices and / or alarm devices to issue an alert and / or alarm.

[0151] In some embodiments, when the current growth temperature fluctuation is greater than a preset fluctuation threshold, the processor can switch the second control mode from power control mode to temperature control mode. After the second control mode is switched to temperature control mode, the processor can automatically adjust the power of the target crystal growth equipment to maintain the temperature at at least one temperature measurement point, or at least three temperature measurement points, at the same temperature as the previous growth temperature (or the temperature fluctuation does not exceed ±0.1°C).

[0152] In some embodiments, when the second control mode is switched from power control mode to temperature control mode, the processor can be controlled to trigger an alert from the associated alert device.

[0153] In some embodiments, an emergency may occur during the crystal growth process, causing a rapid change in its growth temperature, which can significantly impact the crystal growth results. Accordingly, the processor can further determine whether the current growth temperature fluctuation is greater than a preset alarm threshold, and in response to the current growth temperature fluctuation being greater than the preset alarm threshold, the processor can control the target crystal growth equipment to stop working.

[0154] It should be noted that the preset fluctuation threshold and the preset alarm threshold may be threshold conditions relating to temperature fluctuations. Here, the preset fluctuation threshold and the preset alarm threshold increase sequentially. In some embodiments, the preset fluctuation threshold and the preset alarm threshold may be system default values, empirical values, artificially set values, or any combination thereof, or may be set based on actual needs, and this specification is not limited thereto. For example, the preset fluctuation threshold may be ±0.1°C, and the preset alarm threshold may be ±2°C.

[0155] In some embodiments of this specification, the control process of a crystal growth apparatus is divided into multiple control stages based on the temperature control status during the crystal growth process. By determining the control mode and growth control parameters for each control stage, the crystal growth process can be segmented and controlled, improving the accuracy of automatic adjustment control and thereby enhancing the quality of crystal growth. Simultaneously, in actual application processes, the crystal growth process can be monitored in real time, and by determining whether individual parameters change according to normal regulations, it is possible to accurately control in real time whether or not abnormalities occur during the crystal growth process, and to provide real-time warnings, alarms, and / or adjustments based on the actual situation.

[0156] In some embodiments, to ensure crystal growth quality, the processor can achieve automatic adjustment control of the crystal growth process in the second control stage by controlling several other parameters. For example, the processor can achieve automatic adjustment control of the crystal growth process by controlling the temperature field distribution in the second control stage of the target crystal growth equipment. More details about the controlled temperature field distribution can be found in Figure 9 and its related description. Furthermore, for example, the processor can control the air pressure and air flow rate in the second control stage of the target crystal growth equipment; specifically, see the related description below.

[0157] In some embodiments, the processor can determine a second target air pressure and / or target air flow rate for a second control stage based on a second control mode.

[0158] In some embodiments, the processor can determine the second target air pressure and / or target air flow rate of the second control stage based on the second control mode using a variety of methods.

[0159] In some embodiments, the processor can determine a second target air pressure and / or target air flow rate under different control modes in the second control stage by a control recommendation model, and further, the processor can determine a corresponding second target air pressure and / or target air flow rate based on the second control mode. For example, the processor can process the instrument parameters of a target crystal growth machine and the target growth results by a control recommendation model, and determine a second target air pressure and / or target air flow rate under different control modes in the second control stage of the target crystal growth machine.

[0160] In some embodiments, the output of the control recommendation model can be represented as a vector. For example, the output of the control recommendation model may be a vector [(m,a1,b1)(n,a2,b2)], where m represents the second control mode being power control mode, a1 represents the second target air pressure when the second control mode is power control mode, b1 represents the target air flow rate when the second control mode is power control mode, n represents the second control mode being temperature control mode, a2 represents the second target air pressure when the second control mode is temperature control mode, and b2 represents the target air flow rate when the second control mode is temperature control mode.

[0161] In some embodiments, the first label used to train the control recommendation mode, i.e., the hierarchical growth control parameters of the first sample crystal growth apparatus, may include the hierarchical second target air pressure and hierarchical target air flow rate under different control modes in the hierarchical second control stage of the first sample crystal growth apparatus. More details about the control recommendation model can be found in Figure 5 and its related description.

[0162] In some embodiments of this specification, the recommended control model allows for the precise determination of the second target air pressure and target air flow rate in the second control stage of the target crystal growth equipment. Furthermore, it enables highly accurate and automatic adjustment control of the crystal growth process in the actual production process, thereby improving crystal growth quality.

[0163] Figure 8 is a schematic diagram illustrating the determination of the target growth power for target crystal growth equipment as shown in some embodiments of this specification.

[0164] In some embodiments, when the second control mode is a temperature control mode, the processor can process the current growth temperature 801 and target temperature 802 of the target crystal growth equipment based on the power determination model 803 to determine the target growth power 804 of the target crystal growth equipment.

[0165] The power determination model 803 may also be an instrument learning model used to determine the target growth power of a target crystal growth machine. For example, the power determination model 803 may include one of the following: a DNN model, a CNN model, etc., or any combination thereof.

[0166] In some embodiments, the power determination model 803 includes at least 800 multiplication operations in a single run. In some embodiments, the power determination model 803 is executed at least partially by the GPU.

[0167] In some embodiments, the inputs to the power determination model 803 may include the current growth temperature 801 and the target temperature 802 of the target crystal growth equipment, and the output may include the target growth power 804 of the target crystal growth equipment. More details about the current growth temperature, target temperature, and target growth power can be found in Figure 7 and its related description.

[0168] In some embodiments, the power determination model 803 can be obtained by acquiring a plurality of second training samples 805 and training the power determination model 803 based on the plurality of second training samples 805.

[0169] For example, multiple second training samples 805 bearing a second label 806 are input to the initial power determination model 807, a loss function is constructed using the second label 806 and the output of the initial power determination model 807, and the parameters of the initial power determination model 807 can be iteratively updated based on the loss function by gradient descent or other methods. Model training is completed and a trained power determination model 803 is obtained when the loss function of the initial power determination model 807 satisfies preset conditions. Here, the preset conditions may be that the loss function converges, or that the number of iterations reaches a threshold, etc.

[0170] In some embodiments, the second training sample 805 includes the historical current growth temperature and the historical target temperature when the second control mode of the second sample crystal growth apparatus is the temperature control mode. In some embodiments, the second sample crystal growth apparatus may include multiple crystal growth apparatuses. In some embodiments, the second sample crystal growth apparatus may include a target crystal growth apparatus. In some embodiments, the second sample crystal growth apparatus may be the same as the first sample crystal growth apparatus.

[0171] In some embodiments, the second label 806 may include the hysteretic target growth power of the second sample crystal growth apparatus.

[0172] In some embodiments, the second training sample 805 and the second label 806 can be determined by screening hierarchical growth data from multiple crystal growth machines for data where the hierarchical growth result data is similar to or close to the corresponding hierarchical target growth result. For example, in the hierarchical growth data corresponding to the data, the hierarchical current growth temperature and the hierarchical target temperature when the second control mode is temperature control mode can be determined as the second training sample 805, and in the hierarchical growth data corresponding to the data, the hierarchical target growth power when the second control mode is temperature control mode can be determined as the second label 806.

[0173] In some embodiments of this specification, by determining the target growth power of the target crystal growth equipment using a power determination model, the control system or apparatus for the crystal growth equipment can automatically and accurately determine the growth control parameters of the crystal growth equipment and further realize adjustment control of the crystal growth process.

[0174] Figure 9 is a flowchart illustrating the determination of the temperature retention parameters of a target crystal growth apparatus as shown in some embodiments of this specification. In some embodiments, the method for determining the temperature retention parameters can be performed by a processor or the control system 200 of the crystal growth apparatus. For example, process 900 can be stored in a memory device (e.g., memory device 102) in the form of a program or instructions, and process 900 can be realized when the processor or the control system 200 of the crystal growth apparatus executes the program or instructions. The schematic diagrams of process 900 operations presented below are illustrative. In some embodiments, the process can be completed by using one or more additional operations not described and / or one or more operations not considered. Furthermore, the order of operations of process 900 shown in Figure 9 and described below is not limiting.

[0175] Step 901: Determine the target temperature field distribution for the second control stage based on the second control mode.

[0176] The temperature field distribution can influence the airflow direction within the growth equipment, which in turn affects the flow of the gaseous components within the crystal growth equipment, thereby influencing the convexity, concavity, or overall quality of the grown crystals.

[0177] The target temperature field distribution may also be the temperature field distribution necessary to ensure that the crystal achieves the target growth result.

[0178] In some embodiments, the processor can determine the target temperature field distribution of the second control stage based on the second control mode using multiple methods.

[0179] In some embodiments, the processor can determine the target temperature field distribution under different control modes in the second control stage by a control recommendation model, and further, the processor can determine the corresponding target temperature field distribution based on the second control mode. For example, the processor can process the instrument parameters of the target crystal growth equipment and the target growth results by a control recommendation model, and determine the target temperature field distribution under different control modes in the second control stage of the target crystal growth equipment.

[0180] In some embodiments, the output of the control recommendation model can be represented as a vector. For example, the output of the control recommendation model may be a vector [(m, c1)(n, c2)], where m represents the second control mode as a power control mode, c1 represents a kind of target temperature field distribution, n represents the second control mode as a temperature control mode, and c2 represents another kind of target temperature field distribution.

[0181] In some embodiments, the first label used to train the recommended control mode, i.e., the hysteretic growth control parameters of the first sample crystal growth apparatus, may include the hysteretic target temperature field distribution under different control modes in the hysteretic second control stage of the first sample crystal growth apparatus. More details about the recommended control model can be found in Figure 5 and its related description.

[0182] Step 902: Determine the heat retention parameters of the target crystal growth equipment based on the target temperature field distribution.

[0183] The heat retention parameters can be used to adjust the temperature field distribution of the target crystal growth equipment.

[0184] In some embodiments, the heat retention parameters may include at least one of the following: the outer diameter parameter of the first heat retention layer, the inner diameter parameter of the first heat retention layer, the first thickness parameter of the first heat retention layer, and the second thickness parameter of the second heat retention layer.

[0185] Referring to Figure 3, the first insulation layer may refer to the crucible top insulation layer 301 of the crystal growth equipment. The second insulation layer may refer to the outer insulation layer 303, the inner insulation layer 304, and the bottom insulation layer 305 of the crystal growth equipment. The first thickness parameter is the thickness of the first insulation layer, and the second thickness parameter is the thickness of each insulation layer in the second insulation layer.

[0186] In some embodiments, the processor can determine the heat retention parameters of the target crystal growth equipment based on the target temperature field distribution and a third preset comparison table. In some embodiments, the third preset comparison table includes the correspondence between multiple different reference temperature field distributions and reference heat retention parameters. In some embodiments, the third preset comparison table can be obtained by constructing the correspondence between multiple different reference temperature field distributions and reference heat retention parameters based on prior knowledge or historical data (for example, screening historical growth data from different crystal growth equipment to data where the historical growth results satisfy the corresponding historical target growth results). In some embodiments, the processor can search the third preset comparison table based on the target temperature field distribution to determine the reference temperature field distribution that best matches the target temperature field distribution, and determine the reference heat retention parameters corresponding to the reference temperature field distribution as the heat retention parameters of the target crystal growth equipment.

[0187] In some embodiments, the processor can determine one or more recommended insulation layer solutions based on insulation parameters.

[0188] The recommended insulation layer solution may refer to a layout solution related to the thickness and model number of the first or second insulation layer.

[0189] In some embodiments, the processor can determine one or more recommended insulation layer solutions based on insulation parameters using a variety of methods. For example, the processor can determine the model number of a replacement insulation layer to be used to replace the current first insulation layer based on the outer diameter parameter, inner diameter parameter, and first thickness parameter of the first insulation layer in the insulation parameters. Furthermore, for example, the processor can determine the model number of a candidate insulation layer for the second insulation layer based on the second thickness parameter of the second insulation layer in the insulation parameters.

[0190] In some embodiments of this specification, the temperature field distribution of a target crystal growth device can be adjusted by determining the temperature field distribution during the crystal growth stage and the target temperature field distribution in real time, thereby determining the insulation parameters and the recommended insulation layer layout solution. This reduces the influence of the airflow direction on the crystal quality and improves the crystal growth quality.

[0191] Figure 10 is a flowchart illustrating the determination of quality parameters of growth results as shown in some embodiments of this specification. In some embodiments, the method for determining quality parameters can be performed by a processor or the control system 200 of the crystal growth equipment. For example, process 1000 can be stored in a storage device (e.g., storage device 102) in the form of a program or instructions, and process 1000 can be realized when the processor or the control system 200 of the crystal growth equipment executes the program or instructions. The schematic diagrams of process 1000 operations presented below are illustrative. In some embodiments, the process can be completed by using one or more additional operations not described and / or one or more operations not considered. Furthermore, the order of operations of process 1000 shown in Figure 10 and described below is not limiting.

[0192] Step 1001: Determine the target control recipe for the target crystal growth equipment based on the control mode.

[0193] The control recipe may be a solution used to control the crystal growth equipment so that it adjusts control parameters in a timely manner during the crystal growth process. The target control recipe may be a control recipe used to control the finally determined target crystal growth equipment. In some embodiments, the target control recipe may consist of one or more control modes and one or more growth control parameters. For example, the target control recipe may be a first control mode which is a power control mode, with a first target power of j and a first target air pressure of q; a second control mode which is a temperature control mode, with a target growth power of p, a second target power of g, a target temperature field distribution of k, a heat retention parameter of h, a second target air pressure of f, and a target air flow rate of r.

[0194] In some embodiments, the processor can determine a target control recipe for a target crystal growth machine based on the control mode using multiple methods.

[0195] In some embodiments, the processor can determine the growth control parameters of a target crystal growth machine and then create a corresponding control recipe from these growth control parameters. For example, the processor can first determine the growth control parameters of a first control stage and then generate a control recipe applied to the first control stage based on the growth control parameters of the first control stage. Furthermore, for example, the processor can first determine the growth control parameters of a second control stage and then generate a control recipe applied to the second control stage based on the growth control parameters of the second control stage. Furthermore, for example, the processor can first determine the growth control parameters of both the first and second control stages and then generate a control recipe based on the growth control parameters of both the first and second control stages.

[0196] In some embodiments, the processor can determine a target control recipe from a plurality of preset control recipes based on the control mode. In some embodiments, the preset control recipe may be a preset control recipe used to control a target crystal growth machine. In some embodiments, the preset control recipe may be a pre-stored, artificially set control recipe. In some embodiments, the preset control recipe may be determined by the processor.

[0197] In some embodiments, the processor can determine preset control recipes using multiple methods. For example, the processor can use historical control recipes in historical data as preset control recipes. For example, the processor can select one preset parameter value from multiple preset parameter values ​​as one recipe option for a preset control recipe. Furthermore, for example, the processor can generate preset control recipes directly.

[0198] In some embodiments, the processor can process a control mode and at least one preset control recipe based on a recipe determination model, determine the estimated growth result for each of the at least one preset control recipes, and determine a target control recipe based on multiple estimated growth results. For more details on determining a target control recipe based on a recipe determination model, please refer to Figure 11 and its related description.

[0199] Step 1002: Simulate the crystal growth process based on the target control recipe and determine quality parameters that can reflect the quality of the growth result.

[0200] Quality parameters may be evaluation values ​​used to assess the quality of the growth results. In some embodiments, quality parameters may include evaluation values ​​for multiple evaluation items. For example, quality parameters may include evaluation values ​​corresponding to crystal thickness, crystal size, crystal convexity range, crystal concavity range, crystal quality status, crystal slice thickness, and the number of crystal slices.

[0201] In some embodiments, the quality parameters may further include one or more of the crystallization curve and the carbide curve.

[0202] A crystallization curve can reflect the crystallization status during the crystal growth process, with time on the horizontal axis and the crystallization status (e.g., crystallization radius and crystallization thickness) on the vertical axis. This crystallization curve can be generated in real time based on the current crystallization status and time during the growth process.

[0203] The carbonization curve can reflect the carbonization process during crystal growth, with time on the horizontal axis and the carbonization status (e.g., the distance from the bottom of the raw material to the center of the hourglass, and the depth of the hourglass) on the vertical axis. This carbonization curve can be generated in real time based on the current carbonization status during the growth process and time. For more information about hourglasses, please refer to the related descriptions below.

[0204] Each evaluation value in the quality parameters can be represented by a number between 0 and 10, with higher numbers indicating better growth result quality corresponding to the evaluation item.

[0205] In some embodiments, the processor simulates the crystal growth process based on a target control recipe and obtains simulated growth results. These simulated growth results include growth result data generated by the system. More details about the growth result data can be found in Figure 4 and its related descriptions.

[0206] In some embodiments, the processor can process the simulation growth results and target growth results based on a third preset rule and determine quality parameters.

[0207] In some embodiments, the exemplary third preset rule may compare the thickness difference between the crystal thickness in the simulation growth result and the crystal thickness in the target growth result with a plurality of preset thickness difference ranges, and determine the evaluation value corresponding to the preset thickness difference range to which the thickness difference belongs as the evaluation value corresponding to the crystal thickness. Here, the correspondence between the preset thickness difference ranges and the evaluation values ​​may be preset by the system or manually. The rules for determining other evaluation values ​​in quality parameters are similar to the rules for determining the evaluation value corresponding to crystal thickness, and therefore will not be described in detail again here.

[0208] In some embodiments, the processor can simulate the crystal growth process based on a target control recipe and obtain simulated reaction conditions for different growth courses. Furthermore, the processor can determine quality parameters by analyzing the simulated reaction conditions for different growth courses.

[0209] The simulated reaction state may refer to the reaction state of the raw materials in the simulated crystal growth process. For example, the raw material may be silicon carbide powder. In some examples, the simulated reaction state may include the simulated crystallization state and the simulated carbonization state.

[0210] The simulated crystallization status may refer to the situation in which crystallization appears in the raw material during the simulated crystal growth process. In some embodiments, the simulated crystallization status may include the crystallization radius and crystallization thickness at the top of the raw material at the material supply position, the crystallization radius and crystallization thickness at the bottom of the crucible, and the volatility of the raw material.

[0211] The simulated carbonization state may refer to the situation in which carbonization appears in the raw material during the simulated crystal growth process. Carbonization is a process in which carbon is gradually enclosed towards the center, resulting in the formation of hourglass-shaped carbon powder in the center of the raw material. In some examples, the simulated carbonization state may include the distance from the bottom of the raw material to the center of the hourglass, and the depth of the hourglass (the distance from the edge of the raw material to the center of the hourglass), etc.

[0212] In some embodiments, the processor can analyze and process simulated reaction conditions in different growth courses using multiple methods and determine quality parameters. For example, the processor can determine quality parameters by performing processes such as data fitting on the simulated reaction conditions in different growth courses.

[0213] In some embodiments, the processor performs a simulation process to simulate the crystal growth process based on a target control recipe, and quality parameters reflecting the growth result quality can be presented to the user via a user terminal. In some embodiments, the simulation process and simulation results can be transmitted to the user in the form of messages. For example, they can be sent to the client's mobile phone number via short message. Alternatively, they can be sent to the user's client terminal via prompt information. In some embodiments, the simulation process and simulation results can be transmitted to and stored in a storage device. Here, the storage device may be a storage device attached to the control system of the crystal growth machine, or it may be an external storage device. For example, optical discs and hard disks. In some embodiments, the simulation process and simulation results can be transmitted to a specific interface, which includes, but is not limited to, a program interface, a data interface, and a transmission interface. In some embodiments, the simulation process and simulation results may also be output using any method well known to those skilled in the art, for example, by displaying them on a local control terminal, and the present application is not limited thereto.

[0214] In some embodiments of this specification, a means is provided for monitoring the quality of crystals or growth results by performing systematic simulations against a target control recipe and predicting the reaction status of the raw materials.

[0215] It should be noted that the above descriptions relating to processes 400, 600, 700, 900, and 100 are for illustrative and explanatory purposes only and do not limit the scope of this specification. Those skilled in the art can make various modifications and changes to processes 400, 600, 700, 900, and 100 under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0216] Figure 11 is a schematic diagram illustrating the determination of a target control recipe for a target crystal growth apparatus as shown in some embodiments of this specification.

[0217] In some embodiments, the processor can process a target crystal growth equipment control mode 1101 and at least one preset control recipe 1102 based on a recipe determination model 1103, determine the estimated growth result 1104 for each of the at least one preset control recipe, and determine a target control recipe 1106 based on the multiple estimated growth results and the target growth result 1105. More details about the preset control recipes can be found in Figure 10 and its related description.

[0218] In some embodiments, the processor can determine a target control recipe 1106 based on multiple estimated growth results and the target growth result 1105 of the target crystal growth equipment using multiple methods.

[0219] In some embodiments, the processor can determine an estimated growth result that is closest to the target growth result 1105 among multiple estimated growth results, and determine a preset control recipe corresponding to that estimated growth result as the target control recipe 1106. In some embodiments, the method for determining whether the two are close can be determined by calculating the vector distance between the vectors characterizing the estimated growth result and the target growth result in the recipe determination model, and the two can be determined to be close when the vector distance between them is less than a distance threshold.

[0220] The estimated growth results may be estimated growth result data obtained by controlling the target crystal growth equipment according to a preset control recipe. For example, the estimated growth results may include the estimated crystal thickness, estimated size, estimated convexity range, estimated concavity range, estimated crystal quality status, estimated slice thickness, and estimated number of slices. More details about the growth result data can be found in Figure 4 and its related descriptions.

[0221] The recipe determination model 1103 may also be an instrument learning model used to determine the target control recipe for the target crystal growth equipment. For example, the recipe determination model 1103 may include one of the following: a DNN model, a CNN model, etc., or any combination thereof.

[0222] In some embodiments, the recipe determination model 1103 includes at least 800 multiplication operations in a single run, and in some embodiments, the recipe determination model 1103 is executed at least partially by the GPU.

[0223] In some embodiments, the input to the recipe determination model 1103 may include the control mode 1101 of the target crystal growth equipment and at least one preset control recipe 1102, and the output may include the estimated growth results 1104 of each of the at least one preset control recipe. More details about the control modes can be found in Figure 4 and its related description.

[0224] In some embodiments, the processor can process a target crystal growth machine control mode 1101, at least one preset control recipe 1102, and a target growth result 1105 based on a recipe determination model 1103 to determine a target control recipe 1106. Accordingly, the inputs to the recipe determination model 1103 may include the target crystal growth machine control mode 1101, at least one preset control recipe 1102, and the target growth result 1105 of the target crystal growth machine, and the outputs may include the target control recipe 1106 of the target crystal growth machine.

[0225] In some embodiments, the recipe determination model 1103 can be obtained by acquiring a plurality of third training samples 1107 and corresponding third labels 1108, and training the recipe determination model 1103 based on the plurality of third training samples 1107 and corresponding third labels 1108.

[0226] For example, multiple third training samples 1107 bearing a third label 1108 are input to the initial recipe determination model 1109. A loss function is constructed using the third label 1108 and the output of the initial recipe determination model 1109. The parameters of the initial recipe determination model 1108 can then be iteratively updated based on the loss function using gradient descent or other methods. Model training is completed when the loss function of the initial recipe determination model 1109 satisfies preset conditions, and a trained recipe determination model 1103 is obtained. Here, the preset conditions may include convergence of the loss function and reaching a threshold number of iterations.

[0227] In some embodiments, the third training sample 1107 includes a history control mode and a history control recipe for the third sample crystal growth apparatus. In some embodiments, the third sample crystal growth apparatus may include multiple crystal growth apparatuses. In some embodiments, the third sample crystal growth apparatus may include a target crystal growth apparatus. In some embodiments, the third sample crystal growth apparatus may be the same as the first sample crystal growth apparatus and / or the second sample crystal growth apparatus.

[0228] In some embodiments, the third label 1108 may include the historical growth result corresponding to the historical control recipe.

[0229] In some embodiments, the third label 1108 may further include a history control recipe that corresponds to the best history growth result under the circumstances of multiple history control recipes.

[0230] In some embodiments, the third training sample 1107 can be determined based on hierarchical growth data from multiple crystal growth machines. The third label 1108 can be determined by artificial labeling.

[0231] In some embodiments of this specification, the recipe determination model can determine the target control recipe for a target crystal growth machine based on a large number of broad features, making the predicted target control recipe more accurate, and further enabling highly accurate and automated adjustment control of the crystal growth process in the actual production process.

[0232] Having already described the basic concepts, it is clear that, for those skilled in the art, the above detailed disclosure is merely illustrative and does not constitute a limitation of this specification. Although not explicitly described here, those skilled in the art may make various revisions, improvements, and modifications to this specification. Such revisions, improvements, and modifications are proposed herein and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0233] At the same time, this specification uses specific terms to describe its embodiments. For example, “one embodiment,” “one embodiment,” and / or “several embodiments” means a certain feature, structure, or characteristic related to at least one embodiment of this specification. Therefore, it should be emphasized and noted that “one embodiment,” “one embodiment,” or “one alternative embodiment” that are not mentioned two or more times in different places in this specification do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments of this specification may be appropriately combined.

[0234] Furthermore, unless explicitly stated in the claims, the order of processing elements and sequences, the use of numerals and letters, or the use of other names described herein are not intended to limit the order of the processes and methods herein. While the above disclosure examines several embodiments of the invention that are considered useful today by various examples, it should be understood that such details are for illustrative purposes only, and the additional claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that fit the substance and scope of the embodiments herein. For example, the system components described above may be implemented by hardware devices, but may also be implemented solely by software solutions, for example, by installing the described system on a conventional server or mobile device.

[0235] Similarly, it should be noted that, in order to simplify the expressions disclosed herein and thereby aid in understanding one or more embodiments of the invention, multiple features may be combined into a single embodiment, drawing, or description thereof in the preceding descriptions of embodiments herein. However, such a method of disclosure does not mean that the features required by the subject matter herein are greater than the features referred to in the claims. In fact, the features of an embodiment are fewer than all the features of a single embodiment disclosed above.

[0236] In some embodiments, numbers are used to describe component and attribute quantities, and it should be understood that, in some examples, these types of numbers used in describing embodiments are modified with modifiers such as “approximately,” “approximately,” or “roughly.” Unless otherwise stated, “approximately,” “approximately,” or “roughly” clearly indicate that the above numbers may have a variation of ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are all approximations, and these approximations may be modified based on the characteristics required by the individual embodiment. In some embodiments, the numerical parameters should take into account the specified number of significant figures and use a general digit-reservation method. While the numerical ranges and parameters used to confirm their range in some embodiments herein are approximations, in specific embodiments, the setting of these types of numerical values ​​should be as precise as possible within the feasible range.

[0237] This specification incorporates, in particular, all of the patents, patent applications, patent application publications, and other materials, such as articles, books, specifications, publications, documents, etc., referenced herein, with respect to their entirety, except for any application history materials that are inconsistent with or contradictory to the content of this specification, and also excludes materials that limit the broadest scope of the claims of this specification (currently or subsequently added to this specification). It should be noted that if there is any inconsistency or contradiction between the descriptions, definitions, and / or use of terms in the supplementary materials of this specification, the descriptions, definitions, and / or use of terms in this specification shall prevail.

[0238] Finally, it should be understood that the examples described herein are merely for illustrating the principles of the examples herein. Other variations may also fall within the scope of this specification. Therefore, without limitation, alternative layouts of the examples herein may be considered consistent with the teachings herein, for example. Accordingly, the examples herein are not limited to those explicitly introduced and described herein.

Claims

1. A method for controlling a crystal growth apparatus, wherein the method is A step of acquiring historical growth data of at least one crystal growth device, wherein the historical growth data includes at least historical growth control data and historical growth result data. A method for controlling a crystal growth apparatus, comprising the step of determining a control mode for a target crystal growth apparatus based on the aforementioned historical growth data and target growth results, wherein the control mode includes at least one of a temperature control mode and a power control mode.

2. The hierarchical growth data further includes hierarchical instrument parameters of the at least one crystal growth instrument, and the step of determining the control mode of the target crystal growth instrument based on the hierarchical growth data and the target growth result is: The steps include obtaining the instrument parameters of the target crystal growth apparatus, The method according to claim 1, comprising the step of determining a control mode for the target crystal growth equipment based on the equipment parameters, the historical growth data, and the target growth result.

3. The step of determining the control mode of the target crystal growth equipment based on the equipment parameters, the historical growth data, and the target growth results is: The method according to claim 2, comprising the step of processing the instrument parameters and the target growth results based on a control recommendation model to determine the control mode of the target crystal growth instrument, wherein the control recommendation model is an instrument learning model and the training sample of the control recommendation model includes the historical growth data.

4. The aforementioned recommended control model is, A method for obtaining a plurality of first training samples and corresponding first labels, wherein the first training samples include the history instrument parameters and history growth results of a first sample crystal growth instrument, and the first labels include the history control mode and / or history growth control parameters of the first sample crystal growth instrument. The method according to claim 3, obtained by training a control recommendation model based on a plurality of first training samples and corresponding first labels.

5. The aforementioned method, The method according to claim 1, further comprising the step of determining growth control parameters for the target crystal growth apparatus based on the control mode and the target growth result.

6. The method according to claim 1, wherein the step of determining a control mode for a target crystal growth apparatus is a step of determining a first control mode when the target crystal growth apparatus is in a first control stage, the first control stage being a raw material preheating stage.

7. The aforementioned method, The method according to claim 6, further comprising the step of determining a first target power for the first control stage in response to the first control mode being the power control mode, wherein the first target power is used to control the target crystal growth apparatus in the first control stage.

8. The method according to claim 7, further comprising the step of determining a first target air pressure for the first control stage based on the first target power.

9. The method according to claim 1, wherein the step of determining a control mode for a target crystal growth apparatus is a step of determining a second control mode when the target crystal growth apparatus is in a second control stage, the second control stage being a crystal growth stage.

10. The aforementioned method, A step of determining the target temperature of the second control stage in response to the second control mode being the temperature control mode, The steps include obtaining the current growth temperature of the target crystal growth apparatus, The method according to claim 9, further comprising the step of determining a target growth power of the target crystal growth apparatus based on the current growth temperature and the target temperature, wherein the target growth power is used to control the target crystal growth apparatus in the second control step.

11. The step of obtaining the current growth temperature of the target crystal growth equipment is: The method according to claim 10, comprising the step of obtaining the current growth temperature based on at least three temperature measurement points, wherein the axial positions of the at least three temperature measurement points are different.

12. The step of determining the target growth power of the target crystal growth equipment based on the current growth temperature and the target temperature is: The method according to claim 10, comprising the step of processing the current growth temperature and the target temperature based on a power determination model to determine the target growth power of the target crystal growth equipment.

13. The aforementioned power determination model is, A method for obtaining a plurality of second training samples and corresponding second labels, wherein the second training sample includes the historical current growth temperature and historical target temperature when the second control mode of the second sample crystal growth apparatus is the temperature control mode, and the second label includes the historical target growth power. The method according to claim 12, obtained by training the power determination model based on the plurality of second training samples and the corresponding second labels.

14. The method according to claim 9, further comprising the step of determining a second target power for the second control stage in response to the second control mode being the power control mode, wherein the second target power is used to control the target crystal growth apparatus in the second control stage.

15. The aforementioned method, The steps include obtaining the current growth temperature of the target crystal growth apparatus, The method according to claim 14, further comprising the step of switching the second control mode from the power control mode to the temperature control mode in response to the current growth temperature fluctuation being greater than a preset fluctuation threshold.

16. The aforementioned method, A step of determining the target temperature field distribution of the second control stage based on the second control mode, The method according to claim 9, further comprising the step of determining the heat retention parameters of the target crystal growth apparatus based on the target temperature field distribution.

17. The method according to claim 16, wherein the heat retention parameter includes at least one of the outer diameter parameter of the first heat retention layer, the inner diameter parameter of the first heat retention layer, the first thickness parameter of the first heat retention layer, and the second thickness parameter of the second heat retention layer.

18. The method according to claim 16, further comprising the step of determining one or more recommended insulation layer solutions based on the insulation parameters.

19. The method according to claim 9, further comprising the step of determining a second target air pressure and / or target air flow rate for the second control stage based on the second control mode.

20. The aforementioned method, The method according to claim 1, further comprising the step of determining a target control recipe for the target crystal growth apparatus based on the control mode.

21. The step of determining the target control recipe for the target crystal growth equipment based on the control mode is: A step of processing the control mode and at least one preset control recipe based on a recipe determination model, and determining the estimated growth result for each of the at least one preset control recipes, The method according to claim 20, comprising the step of determining a target control recipe based on a plurality of estimated growth results and the target growth results.

22. The aforementioned recipe determination model is, A method comprising the steps of obtaining multiple third training samples and corresponding third labels, wherein the third training samples include a history control mode and a history control recipe for a third sample crystal growth apparatus, and the third labels include history growth results corresponding to the history control recipe. The method according to claim 21, obtained by training a recipe determination model based on the plurality of third training samples and corresponding third labels.

23. The step of determining the target control recipe for the target crystal growth equipment based on the control mode is: The method according to claim 20, comprising the steps of determining a target control recipe by processing the control mode, at least one preset control recipe, and the target growth result based on a recipe determination model.

24. The aforementioned method, The method according to claim 20, further comprising the step of simulating the crystal growth process based on the aforementioned target control recipe and determining quality parameters that can reflect the growth result quality.

25. The method according to claim 24, wherein the quality parameters include at least one or more of crystallization curves and carbide curves.

26. A control system for a crystal growth machine, wherein the system is An acquisition module used to acquire hierarchical growth data of at least one crystal growth machine, wherein the hierarchical growth data includes at least hierarchical growth control data and hierarchical growth result data, A control system for a crystal growth apparatus, comprising: a determination module used to determine a control mode for a target crystal growth apparatus based on the aforementioned hierarchical growth data and target growth results, wherein the control mode includes at least one of a temperature control mode and a power control mode.

27. A control device for a crystal growth machine, wherein the device includes a processor, and the processor is used to perform the control method for the crystal growth machine described in claims 1 to 25.

Images