Casting method crystal growth control method, device, equipment and medium

Abstract

The invention discloses a casting method crystal growth control method, device and equipment and a medium, and relates to the field of crystal preparation. The casting method crystal growth control method comprises the following steps: acquiring an area curve of a crystal; according to the area curve, adopting a growth neuron network model to obtain a heating power curve in the crystal growth process, controlling a casting crystal growth module to perform crystal growth, and obtaining an image of the crystal in the growth process in real time; analyzing the image to obtain parameter data of the crystal in the growth process, and obtaining the quality of the crystal in the growth process by using a quality neural network model; if the mixed crystal exists in the crystal, controlling the casting crystal growth module to melt the crystal until the mixed crystal region is melted; and correcting the heating power curve according to the area of the crystal after the mixed crystal region is melted, and continuing the crystal growth process. According to the invention, the heating power curve required by crystal growth control can be rapidly generated, and the crystal quality can be monitored in real time.

Description

Technical Field

[0001] The present application relates to the field of crystal preparation, and in particular to a method, device, equipment and medium for controlling crystal growth by a casting method. Background Art

[0002] Gallium oxide (β-Ga2O3) is a very promising ultra-wide bandgap semiconductor material. Due to its many advantages such as a bandgap width of 4.8eV, fast saturated electron drift velocity, high breakdown field strength, and stable chemical properties, β-Ga2O3 has broad application prospects in power electronic devices, radio frequency devices, and other fields, such as communications, radar, aerospace, high-speed rail, and new energy vehicles. Compared with third-generation semiconductors such as silicon carbide and gallium nitride, gallium oxide can be grown by melt method at normal pressure, and has the advantages of high quality and low cost. The power loss of gallium oxide is only one-tenth of that of silicon carbide, and the cost is also lower than that of silicon carbide, so it has great prospects for industrialization.

[0003] The casting method for growing gallium oxide single crystals can achieve rapid enlargement of the crystal surface and complete the growth of 6-inch and larger ingots with greater thickness. However, the existing technology requires multiple modeling and parameter optimization during the growth optimization process, which is time-consuming and costly. Changes in conditions such as the thermal field structure, the material coefficient of the insulation material at high temperature, and the electromagnetic field of induction heating during the growth process may cause edge impurities, and the operator's relatively lagging processing leads to an increase in growth costs. Summary of the invention

[0004] The purpose of this application is to provide a casting method, device, equipment and medium for controlling crystal growth, which can quickly generate the heating power curve required for crystal growth control and monitor the crystal quality in real time.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for controlling crystal growth by casting, the method comprising:

[0007] Get the area curve of the crystal;

[0008] According to the area curve, a growing neural network model is used to obtain a heating power curve of a crystal growth process;

[0009] Controlling the casting crystal growth module to grow the crystal according to the heating power curve, and acquiring the image of the crystal during the growth process in real time;

[0010] According to the image, using a quality neural network model, quality prediction data of the crystal during the growth process is obtained; the quality prediction data indicates whether the crystal has a target region, and the target region is a miscellaneous crystal region or a region where the half-height width of the crystal face rocking curve is greater than or equal to 500";

[0011] If the target region exists in the crystal, the casting crystal growth module is controlled to melt the crystal until the target region is melted;

[0012] The heating power curve is corrected according to the area of ​​the crystal after the target area is melted, and the step of "controlling the casting crystal growth module to grow the crystal according to the heating power curve, and acquiring the image of the crystal during the growth process in real time" is returned to be executed until the growth of the crystal is completed.

[0013] In a second aspect, the present application provides a casting method crystal growth control device, which includes: a casting crystal growth module, a visualization system, a computer processing system and a growth control system.

[0014] The visualization system is connected to the computer processing system, and the visualization system is used to obtain images of crystals in the growth process in real time.

[0015] The computer processing system is connected to the growth control system, and the computer processing system is used to generate a heating power curve using the above-mentioned casting method crystal growth control method.

[0016] The growth control system is connected to the casting crystal growth module, and the growth control system is used to control the casting crystal growth module according to the heating power curve.

[0017] In a third aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned casting method for controlling crystal growth.

[0018] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned casting method for controlling crystal growth.

[0019] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0020] The present application provides a method, device, equipment and medium for controlling crystal growth by casting method. The present application can directly generate a heating power curve based on the area curve of the crystal using a growth neural network model, without the need for multiple modeling and parameter optimization, and by acquiring images of the crystal during the growth process in real time, quality monitoring is performed using image analysis and a quality neural network model. When there are impure crystals in the crystal, the casting crystal growth module is directly controlled to melt the impure crystal area, thereby ensuring the quality of the crystal and avoiding the defect of increased growth cost caused by delayed processing by the operator. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0022] Figure 1 A schematic flow chart of a method for controlling crystal growth by a casting method provided in one embodiment of the present application;

[0023] Figure 2 A schematic diagram of data collection in the process of constructing sample data provided in one embodiment of the present application;

[0024] Figure 3 A schematic diagram of a field of view of an image of a crystal during growth provided by an embodiment of the present application;

[0025] Figure 4 A schematic diagram of the structure of a crystal growth control device using a casting method provided in one embodiment of the present application;

[0026] Figure 5 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0028] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0029] In an exemplary embodiment, Figure 1As shown, a casting method for controlling crystal growth is provided, comprising the following steps 101 to 106. Among them:

[0030] Step 101, obtaining an area curve of a crystal.

[0031] Step 102: Obtain a heating power curve of a crystal growth process using a growing neural network model according to the area curve.

[0032] Step 103, controlling the casting crystal growth module to grow the crystal according to the heating power curve, and acquiring an image of the crystal during the growth process in real time.

[0033] Step 104, according to the image, using the quality neural network model, obtain quality prediction data of the crystal during the growth process; the quality prediction data indicates whether the crystal has a target area, and the target area is a miscellaneous crystal area or an area where the half-height width of the crystal plane rocking curve is greater than or equal to 500".

[0034] Step 105: If the target region exists in the crystal, the casting crystal growth module is controlled to melt the crystal until the target region is melted away.

[0035] Step 106, correcting the heating power curve according to the area of ​​the crystal after melting the target area, and returning to execute the step of "controlling the casting crystal growth module to grow the crystal according to the heating power curve, and acquiring the image of the crystal during the growth process in real time" until the growth of the crystal is completed.

[0036] By implementing the above steps 101 to 106, the heating power curve required for crystal growth control can be quickly generated, and the crystal quality can be monitored in real time, thereby avoiding the defect of increased growth cost caused by operator processing delays.

[0037] The growth neural network model and the quality neural network model in the embodiments of the present application are both obtained by training the neural network model.

[0038] In the training process of the growing neuron network model, the relationship curve between crystal growth and temperature in the melt and the relationship curve between temperature and heating power are obtained during the crystal growth process, and the relationship curve between crystal growth and heating power is further obtained. The growth neuron network model is constructed by machine learning method to achieve the purpose of controlling crystal growth. The specific steps are: obtaining images at different times during the crystal growth process as training images; using contour extraction method to determine the solid-liquid boundary and thermal field edge line in each training image; calculating the integral of the distance between the solid-liquid boundary and the thermal field edge line in each training image to obtain the melt area in each training image; calculating the thermal field area in each training image based on the thermal field edge line in each training image; calculating the difference between the thermal field area and the melt area in each training image as the crystal area in each training image; calculating the crystal area difference between two adjacent training images according to the crystal area of ​​each training image as the area increase per unit time; taking the area increase per unit time and the average value of heating power as input and output respectively, and constructing sample data. Further, the neural network model is trained according to the sample data to obtain the growing neuron network model.

[0039] In the process of constructing the above sample data, the boundary between the crystal and the melt is firstly obtained by contour extraction technology, which is used to calculate the change in the area of ​​the crystal nucleus, and the growth control data (i.e., heating power) is learned by the machine learning module to obtain the optimal growth parameters (i.e., heating power curve), and the growth parameters are transmitted to the growth control system to achieve the purpose of growth process control; in this embodiment, the change in crystal area is directly related to temperature, and temperature is directly related to heating power (such as Figure 2 As shown in the figure, the indirect relationship between the change of crystal area and heating power can be found through machine learning to directly generate the heating power curve; the crystal area calculation can be calculated by the integral method. In particular, before starting the area calculation, the thermal field edge curve and area S need to be measured, such as Figure 3 As shown, since the crystal will drift in the melt, the crystal area is obtained by integrating the distance L from the solid-liquid interface of the crystal to the edge curve of the thermal field, and subtracting the melt area from S; the area increase of the crystal per unit time δS=S-S1, where S is the currently measured crystal area, and S1 is the crystal area measured at the previous time point. The growth process is controlled by controlling the area increase per unit time; wherein, the time interval can be set according to 0.1-10s.

[0040] In an embodiment of the present application, the actual heating power and crystal area S during the crystal growth process are collected from the historical crystal growth process to establish a training set and a verification set, the parameters of the training set are input into the neural network model for training, and then the verification set is input into the trained neural network model, and it is continuously iterated and optimized to obtain the most ideal growth neural network model, set the ideal curve of the crystal area S, and output the heating power curve.

[0041] During the training process of the quality neural network model, the crystal surface, solid-liquid interface data, and crystal quality are obtained. Through machine learning, the relationship between the crystal surface, solid-liquid interface data and the crystal quality is learned, and a closed loop is formed with the growth control system to achieve the purpose of controlling the crystal quality. Specifically, the following steps are performed: collecting crystal parameter data, solid-liquid interface, and crystal surface data during crystal growth, including the solid-liquid interface aperture width, whether there are bright spots at the solid-liquid interface, and the brightness of the crystal as labels to characterize the quality of the crystal, establishing a training set and a validation set, inputting the parameters of the training set into the neural network model for training, and finally inputting the validation set into the trained neural network model. Continuously iterate and optimize, and finally obtain a quality neural network model that can control the crystal quality, and finally avoid the occurrence of impure crystals and crystals of poor quality.

[0042] In the embodiment of the present application, a quality neural network model is used by utilizing images to obtain the quality of the crystal during the growth process; the quality refers to whether there are impurities, crystal face data, and the half-width of the crystal in each area; in the embodiment of the present application, an ideal quality neural network model is trained by machine learning image data and the half-width quality data of the corresponding area of ​​the actual crystal. In actual growth, the quality can be judged by the image data.

[0043] In the embodiment of the present application, since the gallium oxide crystal is a monoclinic system, the refractive index and transmittance between different crystal planes are different. Therefore, when crystals with different crystal orientations appear in the same field of view, there will be brightness differences between the crystal planes and grain boundaries, which can be used to identify whether there are impurities in the crystal.

[0044] In order to ensure the quality of the crystal in the embodiment of the present application, when the presence of impure crystals is detected, the casting crystal growth module is controlled to melt the impure crystal area, and the heating power curve is corrected based on the area of ​​the crystal after the impure crystal area is melted, and the crystal of the current area is used as the starting heating position to continue crystal growth.

[0045] In an exemplary embodiment, Figure 4 As shown, a casting method crystal growth control device is provided, and the casting method crystal growth control device includes: a casting crystal growth module, a visualization system, a computer processing system and a growth control system.

[0046] Among them, the casting crystal growth module includes a furnace shell, an insulating hot field, an induction coil, a crucible, a gallium oxide melt or crystal, a cooling system, an atmosphere and air pressure control system; the coil is installed in the furnace shell, the insulating hot field is installed concentrically at the center of the coil, the crucible is placed in the center of the insulating hot field, and gallium oxide raw materials are added to the crucible. The cooling system provides cooling water for the furnace shell and the induction coil, and the atmosphere and air pressure control system controls the growth atmosphere and furnace pressure in the furnace shell.

[0047] The visualization system includes: an optical imaging bundle, a protective device, a lens cleaning device, a CCD camera, an image processing system, and a storage system; specifically: the objective lens of the CCD camera is at the front end of the optical imaging bundle, closer to the melt, and is connected to the CCD camera through the optical imaging bundle. The image is coupled into the optical imaging bundle through the objective lens and transmitted to the CCD camera, exposed by the CCD camera, and then obtained through the image processing system, and connected to the computer processing system through a network or a data transmission line.

[0048] Among them, the optical image transmission bundle and the objective lens are fixed as a whole on the upper part of the hot field; the objective lens is an optical combination lens, the purpose of which is to couple the light in a larger field of view in the melt into the optical image transmission bundle, and the objective lens itself needs water cooling or air cooling protection; the optical image transmission bundle can be a sapphire optical fiber or a quartz optical fiber, the sapphire optical fiber does not need a protective device, and the quartz optical fiber can be protected by water cooling or air cooling; the lens cleaning device is a specific gas and gas path, located at the front end of the objective lens, and the purpose of passing the flowing atmosphere is to prevent the decomposed gallium oxide from sticking to the objective lens; the cooling atmosphere is the same as the main atmosphere of the growth chamber. The CCD camera can realize the collection of melt interface data. In order to increase the clarity and accuracy of the field of view, the melt interface is equalized by the histogram method, and the gamma transformation is used to increase the grayscale contrast between the crystal and the melt, so that the growth interface is clearer, which is conducive to the capture of the solid-liquid interface.

[0049] Among them, the growth control system mainly controls the crystal growth module through parameterization, and stores the control and feedback data in the computer processing system; in particular, it includes the heating power of the induction coil, atmosphere flow, thermocouple temperature, cooling flow, etc.; the computer processing system includes an image processing system and a memory.

[0050] Among them, the computer processing system includes an image processing system, a memory, a display, an input device, an output device, and a machine learning module; the input device includes a mouse, a keyboard, and various data connectors connected to the computer processing system; the memory includes non-volatile storage, internal memory, and cloud storage on the computer.

[0051] In an exemplary embodiment, a method for performing crystal growth based on the casting method crystal growth control device of the above embodiment is provided, comprising the following steps:

[0052] Step 1: Construction of thermal field for growing gallium oxide crystals: Figure 4 As shown, the crucible is concentrically placed in the center of the heat preservation field, a block of compressed gallium oxide powder is added, an optical imaging beam is installed, the furnace door is closed, and a mechanical pump is used to evacuate the furnace to ≤1Pa, and then a mixture of one or more of N2, CO2 or AR is filled in to normal pressure.

[0053] After installation, the special optical image bundle needs to be adjusted in time with an external light source to confirm whether the edge of the crucible is completely visible.

[0054] Step 2, optical image transmission bundle is turned on: turn on the power of the optical image transmission bundle, open the cleaning air path of the lens cleaning device, turn on the water cooling protection of the optical image transmission bundle; set the furnace chamber pressure to be constant at 1-10Kpa.

[0055] In particular, when the temperature is lower than 800°C, the condition inside the crucible cannot be displayed because there is no light source in the optical image beam field of view. When the temperature is higher than 800°C, thermal field observation can be achieved by changing the exposure and focal length.

[0056] The target cleaning gas is continuously filled into the cleaning gas path of the lens cleaning device at a constant pressure, and the exhaust gas is controlled by the exhaust valve to achieve the requirement of air pressure balance.

[0057] Step 3, heating the chemical material: set the power of the induction coil in the computer control system, increase the temperature at a rate of 100-200°C / hour, use thermocouple temperature measurement and optical imaging beam interface to observe the chemical material status, and keep the temperature at 1500°C, 1600°C, 1700°C, and 1793°C for 1 hour respectively during the heating process, and introduce 2% volume fraction of oxygen at 1500°C; keep the temperature for 1-2 hours after the material block in the field of view of the optical imaging beam is completely chemically processed.

[0058] The special thermocouple enters the upper part of the thermal insulation field through the hole of the optical imaging bundle and is placed as low as possible without affecting the observation field, generally within the range of 0-20mm from the lower surface of the upper cover of the thermal insulation field.

[0059] The material of the thermocouple is platinum-rhodium alloy thermocouple, and the ambient temperature for long-term use is below 1800°C. Since oxygen will oxidize iridium at 1200°C-1500°C, oxygen is introduced after 1500°C. The purpose of introducing oxygen is to reduce the decomposition of gallium oxide and prevent the metal Ga decomposed from gallium oxide from corroding the crucible.

[0060] Step 4, crystal growth control: reduce the heating power to cool the melt to the melting point of gallium oxide, maintain for 30 minutes, until crystal nuclei appear in the melt, set the growth power curve in advance according to the power curve output by the growth neuron network model, start the automatic growth control program and the crystal quality control module, control the growth of the crystal nucleus, and complete the crystallization of the last melt in the field of view as the end point; during the crystal growth process, if bright spots appear at the solid-liquid interface, or impure crystals or poor quality points appear on the crystal surface, the system increases the power to melt the poor quality back and grow again, and the change in power is adjusted in real time according to the change in the melting speed.

[0061] In particular, when the temperature is lowered in the melt state in the crucible, spontaneous nucleation and crystallization will occur in the melt. The cooling curve is set according to the heating power curve obtained by machine learning during the crystal growth process, such as Figure 2 shown.

[0062] like Figure 3 As shown, during the growth of the crystal nucleus, a closed-loop control is established between the area increase per unit time δS and the heat generation power; in particular, if there are stray crystals at the edge of the crystal, the melting back is carried out in the opposite direction of the growth.

[0063] Step 5, cooling after growth is completed: After confirming the completion of crystal growth using the image transmitted by the optical imaging beam, the temperature is lowered to 1500°C at a rate of 100-200°C / h, and the gas in the furnace is replaced with argon.

[0064] Step 6: Figure 2 The curve shown sets a linear temperature drop, and finally a fully grown crystal is obtained.

[0065] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Wherein, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Wherein, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data determined by complementarity. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a casting method crystal growth control method is implemented.

[0066] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0067] In an exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the steps in the above-mentioned casting method crystal growth control method embodiment are implemented.

[0068] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, which, when executed by a processor, implements the steps in the above-mentioned embodiment of the casting method crystal growth control method.

[0069] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0070] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, etc., but is not limited thereto.

[0071] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0072] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for controlling crystal growth by casting method, It is characterized in that The control method comprises: Get the area curve of the crystal; According to the area curve, a growing neural network model is used to obtain a heating power curve of a crystal growth process; Controlling the casting crystal growth module to grow the crystal according to the heating power curve, and acquiring the image of the crystal during the growth process in real time; According to the image, using a quality neural network model, quality prediction data of the crystal during the growth process is obtained; the quality prediction data indicates whether the crystal has a target region, and the target region is a miscellaneous crystal region or a region where the half-height width of the crystal face rocking curve is greater than or equal to 500"; If the target region exists in the crystal, the casting crystal growth module is controlled to melt the crystal until the target region is melted; The heating power curve is corrected according to the area of ​​the crystal after the target area is melted, and the step of "controlling the casting crystal growth module to grow the crystal according to the heating power curve, and acquiring the image of the crystal during the growth process in real time" is returned to be executed until the growth of the crystal is completed.

2. The casting method for controlling crystal growth according to claim 1, It is characterized in that The growth neural network model and the quality neural network model are both obtained by training the neural network model.

3. The method for controlling crystal growth by casting according to claim 1, It is characterized in that The sample data used to train the growing neural network model is constructed as follows: Acquire multiple groups of images at different moments in the crystal growth process using the casting method as training images; Determine the solid-liquid boundary line and the thermal field edge line in each of the training images by using a contour extraction method; Calculating the integral of the distance between the solid-liquid boundary line and the thermal field edge line in each of the training images to obtain the melt region area in each of the training images; Calculate the thermal field area in each of the training images based on the thermal field edge line in each of the training images; Calculating the difference between the area of ​​the thermal field region and the area of ​​the melt region in each of the training images as the crystal area in each of the training images; According to the crystal area of ​​each of the training images, calculating the crystal area difference between two adjacent training images as the area increase per unit time; The area increase per unit time and the average heating power are used as input and output respectively to construct sample data.

4. The method for controlling crystal growth by casting according to claim 1, It is characterized in that The heating power curve of the crystal growth process is obtained by using a growing neural network model according to the area curve, specifically comprising: According to the area curve, calculating the area increase at different unit times; The area increase per unit time is input into the growing neuron network model to obtain the heating power per unit time; The heating power curve is generated according to the heating power per unit time.

5. The method for controlling crystal growth by casting according to claim 1, It is characterized in that The heating power curve is corrected according to the area of ​​the crystal after the target area is melted, specifically including: Determine the time coordinate point corresponding to the area of ​​the crystal after the target area is melted in the area curve; The portion of the heating power curve before the time coordinate point is deleted.

6. A casting method crystal growth control device, It is characterized in that The casting method crystal growth control device comprises: a casting crystal growth module, a visualization system, a computer processing system and a growth control system; The visualization system is connected to the computer processing system, and the visualization system is used to obtain images of crystals in the growth process in real time; The computer processing system is connected to the growth control system, and the computer processing system is used to generate a heating power curve using the casting method crystal growth control method according to any one of claims 1 to 5; The growth control system is connected to the casting crystal growth module, and the growth control system is used to control the casting crystal growth module according to the heating power curve.

7. The casting method crystal growth control device according to claim 6, It is characterized in that The casting crystal growth module comprises: a furnace shell, a heat preservation field, an induction coil, a crucible, a cooling system and an atmosphere and air pressure control system; The heat preservation thermal field is arranged in the furnace shell; The induction coil is arranged in the furnace shell, and the crucible is arranged on the induction coil; The cooling system is arranged between the induction coil and the furnace shell; The atmosphere and air pressure control system is connected to the interior of the furnace shell through a pipeline; The control end of the induction coil is connected to the growth control system.

8. The casting method crystal growth control device according to claim 6, It is characterized in that The visualization system comprises: an optical image transmission bundle, an objective lens, a lens cleaning device and a CCD camera; The objective lens is connected to the CCD camera via the optical image transmission bundle; The lens cleaning device is arranged at the front end of the objective lens; The CCD camera is connected to the computer processing system.

9. A computer device, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the casting method for controlling crystal growth according to any one of claims 1 to 5.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the casting method crystal growth control method according to any one of claims 1 to 5 is implemented.

Images