Control method for rotating motor of crystal growing furnace

Abstract

The invention provides a control method for a rotating motor of a crystal growing furnace, and relates to the technical field of crystal growing furnaces. According to the control method for the rotating motor of the crystal growing furnace, the interaction between a furnace body structure and crystal / melt temperature distribution and the influence of a thermal field structure on a growing interface are deeply researched, so that the thermal field design is successfully optimized, and the temperature distribution in the furnace is accurately simulated by utilizing a numerical calculation and analysis technology; the temperature uniformity in the crystal growth process is ensured, so that the crystal quality is improved, the energy consumption is reduced, in addition, an artificial neural network model is adopted to predict growth conditions and optimize related parameters, the fine control of the crystal growth process is realized, and finally, through experimental verification and adjustment, the simulation accuracy is ensured, and the simulation efficiency is improved. The applicability and accuracy of the model are improved, the scientific research and production efficiency is remarkably improved through the automatic and optimized control process, the research and development cycle is shortened, and the marketing of new products is accelerated.

Description

Technical Field

[0001] The present invention relates to the technical field of crystal growth furnaces, and in particular to a method for controlling a rotating motor of a crystal growth furnace. Background Art

[0002] A crystal growth furnace is an electric furnace used to melt and purify materials under specific conditions, and then grow them into single crystals or polycrystalline materials. It is primarily used to grow crystals of semiconductor materials (such as silicon and germanium), laser materials (such as yttrium aluminum garnet), and high-purity metal materials (such as tungsten and molybdenum). A crystal growth furnace typically consists of a vacuum chamber, a vacuum and gas systems, a heating system, a lifting and rotating mechanism, temperature and pressure measurement and control systems, and other auxiliary equipment. Extremely precise control and transmission are required. Crystal growth furnaces are categorized as single crystal furnaces or polycrystalline furnaces based on the crystal structure being grown. Single crystal furnaces are further divided into Czochralski and zone-melting methods based on the purification method, while polycrystalline silicon ingot furnaces are the most commonly used. The Czochralski method, also known as the Czochralski method, is a commonly used crystal growth method suitable for mass production of large, perfect crystals. This method controls the rotation and pulling speed of a seed crystal to cause the molten material to crystallize along the seed crystal and grow into rod-shaped single crystals. The zone melting method is a method for achieving high purity in single crystals during crystallization. By moving a material rod or heating element, the melt zone moves, leading to crystallization and ultimately forming a single crystal rod. Crystal growth furnaces are used in a wide range of applications, including semiconductor manufacturing, optical device fabrication, and inorganic material preparation. These devices are crucial to modern industry and technology because they can produce high-quality single crystals, which are essential for many high-tech products and research fields.

[0003] Under current technological conditions, the growth process of silicon carbide crystals is usually carried out in a high-temperature environment exceeding 2000°C. Due to the limitations of high temperature and confined space, traditional temperature monitoring methods are difficult to apply, which makes it difficult to directly observe and monitor the changes in the temperature field during the growth process. When growing silicon carbide crystals using the physical vapor transport (PVT) method, the gas phase components tend to crystallize and accumulate at the center of the temperature measuring hole in the lowest temperature zone of the growth chamber. This phenomenon makes it impossible to directly observe the temperature of the seed crystal cap, and as the accumulation at the center of the temperature measuring hole increases, the temperature field of the crystal growth interface will also change, which may eventually lead to the formation of defects such as polytype inclusions and stress. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides a method for controlling the rotating motor of a crystal growth furnace, which solves the problem that it becomes difficult to directly observe and monitor the changes in the temperature field during the growth process, and that the temperature field at the crystal growth interface also changes, which may lead to the formation of defects such as polymorphic inclusions and stress.

[0006] (2) Technical solution

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: A method for controlling a rotating motor of a crystal growth furnace, specifically comprising the following steps:

[0008] S1. Establish geometric and physical models

[0009] According to the actual structure of the crystal growth furnace, a geometric model is established, including the crucible, heater and insulation material. At the same time, a physical model is established, including heat conduction, fluid dynamics and mass transfer. The Boltzmann equation and lattice Boltzmann method are used to simulate crystal growth.

[0010] S2. Data calculation and result analysis

[0011] The finite volume method is used to discretize the governing equations, which is:

[0012]

[0013] Where f represents the velocity, enthalpy, and turbulence parameters, u is the velocity field, D is the diffusion coefficient, and S is the source term. The complex function method is then used to solve the electromagnetic field to ensure the correctness of the model. A global data simulation is then performed to analyze the temperature distribution in the crystal and the solution, the argon convection between the heat shield and the quartz crucible, and the thermal stress in the crystal.

[0014] S3. Optimize thermal field structure

[0015] By changing the crucible shape, heater position, and insulation cage separator position parameters, the crystal growth process was optimized using computer numerical simulation. The effects of the optimized heat shield on the heater, the effect of the carbon felt on the sidewalls on preventing heat loss, and the effect of the diversion system on reducing SiO deposition on the upper wall were analyzed.

[0016] S4. Process parameter optimization

[0017] Artificial neural networks and genetic algorithms were used to optimize the process parameters of pulling speed, crystal rotation speed, ambient gas, and crucible wall temperature to obtain a tight and flat solid-liquid interface. The response surface methodology was used to determine the crystal rotation and crucible rotation under the optimal thermal field distribution. Specifically, the gradient descent method was used to update the network weights:

[0018] Δw=-ηj'(w)

[0019] Where η is the learning rate, j ’ (w) is the gradient of the loss with respect to the parameters;

[0020] S5. Machine Learning Model Building and Training

[0021] Based on CFD simulation data, an artificial neural network is used to build a model of the crystal growth process. The neural network model is trained to enable it to predict the CFD simulation results of SiC crystal growth and optimize the growth conditions.

[0022] S6. Rotating motor control

[0023] The controller controls the rotary motor to make the pulling device rotate the seed crystal, causing the crystal to rotate. The torque is calculated using the BLDC motor sinusoidal commutation formula:

[0024] Shaft torque = K t [I R sin(θ)+I S sin(θ+120°)+I T sin(θ+240°)]

[0025] Where θ is the electrical angle of the shaft, K t is the torque constant of the motor, I R ,I S ,I T is the phase current. In the isodiameter growth stage, the controller controls the rotary motor to make the pulling device rotate the seed crystal, and at the same time controls the heating device to heat the crucible;

[0026] S7. Data feedback and closed-loop control

[0027] The encoder is used to feedback the position of the crucible rod and seed crystal rod to achieve closed-loop control and display. The parameters are set and displayed through the touch screen, the real-time status of the motor is monitored, and the control parameters are adjusted according to the feedback. Specifically, the PID control algorithm is used to perform closed-loop control of the key parameters in the crystal growth process:

[0028]

[0029] Among them, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, e(t) is the deviation;

[0030] S8. Experimental verification and adjustment

[0031] Crystal growth experiments are carried out to verify the accuracy of numerical simulations, and model parameters are adjusted based on the realized results to improve the reliability of the simulation.

[0032] Preferably, the S1. establishing geometric and physical models specifically includes the following steps: geometric modeling, basic assumptions, establishment of a control equation group, setting of boundary conditions and initial conditions, mathematical modeling, grid generation, equation discretization, selection of numerical solution methods and coupled solution of physical quantities.

[0033] Preferably, the S5. machine learning model establishment and training specifically includes the following steps: problem definition, data preparation, feature selection, model selection, training and verification, parameter tuning, model training, data set partitioning, model method selection, definition of hyperparameters, data loading, network initialization, definition of loss and optimizer, network training, model evaluation, model saving and model use.

[0034] (3) Beneficial effects

[0035] The present invention provides a method for controlling a rotating motor of a crystal growth furnace. The method has the following beneficial effects:

[0036] This invention provides a method for controlling the rotating motor of a crystal growth furnace. This method uses numerical simulation technology to study the relationship between the furnace structure and the crystal / melt temperature distribution, the evolution of the interface at different growth stages, and the mechanism by which the thermal field structure affects the growth interface, aiming to optimize the thermal field design. Specifically, through numerical calculations and result analysis, the temperature distribution within the crystal growth furnace is accurately simulated and analyzed to ensure temperature uniformity during the crystal growth process. This process is crucial for improving crystal quality. Optimizing the thermal field structure improves thermal efficiency, reduces energy consumption, and provides a more stable and uniform growth environment, which helps improve crystal uniformity and repeatability. By optimizing thermal field parameters and process parameters, the crystal growth rate and efficiency can be increased, the growth cycle can be shortened, and production costs can be reduced. Data feedback and closed-loop control enable real-time monitoring and adjustment of the growth process, ensuring process stability and reducing growth defects caused by operational errors or environmental changes. Precise rotary motor control and torque calculation help maintain the precise position and posture of the crystal during growth, reduce crystal stress and defects, and improve overall crystal quality. Automated control reduces the need for manual operation, reduces the likelihood of operational errors, and improves production reliability. Through precise numerical simulation and optimization algorithms, the consistency of each growth process can be ensured, improving the repeatability of the production process. Optimized thermal field structures and process parameters can reduce energy consumption, lower production costs, and minimize environmental impact. Through the optimization of machine learning models, the system can adapt to different growth conditions and requirements, improving the system's flexibility and adaptability. Artificial neural network models can be used to predict crystal growth conditions, optimize growth parameters, and achieve more precise control. Experimental verification and adjustment ensure the accuracy of simulation results, and model parameters can be further adjusted based on experimental results, improving the applicability and accuracy of the model. The automation and optimization of the entire control process can significantly improve the efficiency of scientific research and production, shorten the R&D cycle, and speed up the launch of new products. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] Example 1: Independent Design of CdZnTe Crystal Growth Furnace and Temperature Control Performance Experiment

[0040] Implementation steps:

[0041] The invention discloses a CdZnTe crystal growth furnace based on a moving heating method.

[0042] The furnace heating unit has four specifications and is divided into six temperature control sections. The servo motor drives the ball screw linear guide to achieve lifting and lowering by an industrial computer.

[0043] The furnace cavity is equipped with heating tubes composed of corundum ceramic tubes and high-temperature metal heat pipes, and the heating unit is controlled by high-precision platinum-rhodium-platinum thermocouples, Continental, transformers and thyristors.

[0044] Fuzzy + PID control algorithm is used to adjust and control the temperature distribution of the heating furnace.

[0045] The stability and control performance experiments during the temperature heating process were carried out.

[0046] Experimental results:

[0047] The heating temperature inside the furnace chamber is continuously controlled for 200 hours, the temperature fluctuation at the same position is ±0.005℃, and the heating temperature deviation is ≤±0.1℃.

[0048] The lengths of the upper and lower constant temperature zones of the furnace chamber are 400 mm and 240 mm respectively.

[0049] The length of the temperature gradient zone in the middle of the furnace chamber is about 136 mm.

[0050] The constant temperature zone at the bottom of the furnace cavity is 240 mm long.

[0051] At a heating temperature of approximately 1098°C, the temperature gradient is 0.92°C·mm-1.

[0052] The experimental results show that the furnace meets the independent design and temperature control performance requirements of CdZnTe crystal growth furnace.

[0053] Example 2: Growth of large-size YVO crystals by automatic diameter-controlled Czochralski method

[0054] Implementation steps:

[0055] Maintain appropriate temperature field distribution in space.

[0056] The medium frequency induction power of the crystal furnace is gradually increased to heat and melt the polycrystalline material. After adding the polycrystalline material three times and performing high-temperature melting, a power reduction program is set to slightly reduce the heating temperature.

[0057] (001) oriented YVO4 rod-shaped seed crystals are used for seeding growth. The seed crystals fixed on the seed crystal rods are moved down to the melt surface. The seed crystals are first melted and washed in a rotating pulling state, and then the appropriate temperature is adjusted for proper seed crystal welding.

[0058] After the crystal diameter is slightly relieved and the growth aperture is stable, it switches to automatic diameter control growth under program module adjustment.

[0059] The crystal rotation speed is maintained at 8-15 r / min and the pulling speed is 0.8-2.0 mm / h.

[0060] Experimental results:

[0061] Through the above method, the automatic diameter-controlled pulling method growth of large-size YVO crystals was achieved, the crystal growth process was stable, and the diameter control was precise.

[0062] Example 3: Application in SiC (Silicon Carbide) Crystal Growth Furnace Control System

[0063] Implementation steps:

[0064] During the SiC crystallization process, three stepper motors are used. Two stepper motors drive the lead screws to move the crucible rod and seed crystal rod up and down, and the other motor drives the crucible to rotate.

[0065] Two LM3106ACPU modules are selected, one is used to control the stepper motors of the crucible rod and the seed crystal rod, and the other is used to control the motor for the crucible rotation.

[0066] Two two-pole reducers are selected between the stepper motor and the lead screw of the crucible rod and the seed crystal rod, with the first stage being 80:1 and the second stage being 100:1.

[0067] The crucible rotation uses a 12:1 first-stage reducer.

[0068] Hetech's touch screen is selected for parameter setting and display, and the two PLC slaves are connected via the RS485 port using the ModBus protocol.

[0069] Experimental results:

[0070] The crucible rod and seed crystal rod stepper motor parameters are set to 2000Pulse / r, the screw lead is 5mm, the reduction ratio is 1:8000 for slow speed and 1:80 for fast speed.

[0071] The speed range is 0.002~0.50mm / hour at slow speed and 1~60mm / min at fast speed.

[0072] The encoder resolution is 100ppr, which means 100 pulses correspond to 5 mm.

[0073] Experimental results show that the system can accurately control the rotation and stretching movements of SiC crystals during growth, ensuring that the crystal texture meets the requirements.

[0074] In the embodiment of the present invention, the specific method is as follows:

[0075] S1. Establish geometric and physical models

[0076] According to the actual structure of the crystal growth furnace, a geometric model is established, including the crucible, heater, and insulation material. At the same time, a physical model is established, including heat conduction, fluid dynamics, and mass transfer. The Boltzmann equation and lattice Boltzmann method are used to simulate crystal growth. The geometric and physical models are established. The specific steps include:

[0077] Geometric modeling research:

[0078] Build a geometric model of an industrial crystal furnace used in actual crystal growth. This model can be imported from a computer-aided design (CAD) file or independently constructed in simulation software.

[0079] Explanation of the physical process:

[0080] The key physical mechanisms in the crystal growth process are elaborated in detail. Taking the PVT process as an example, a medium-frequency induction coil or resistance heater generates heat energy, creating axial and radial temperature gradients within the crucible. The raw material decomposes upon heating, and the resulting gaseous components are transported by the temperature gradient to the low-temperature seed crystal surface at the top of the crucible, where they recrystallize and form a single crystal.

[0081] Basic assumptions:

[0082] When constructing the physical model, the silicon melt is regarded as an incompressible Newtonian fluid, whose physical properties are constant and satisfy the Boussinesq approximation; argon is regarded as an ideal gas, using a low Mach number approximation, which conforms to the ideal gas state equation; the absorption of radiation energy by the gas is ignored, and all radiating surfaces are regarded as gray body surfaces; during the melting process of the silicon raw material, its density remains consistent with that of the melt.

[0083] Construction of the control equations:

[0084] A set of control equations including the thermal field of the solid area, melt thermal convection and radiation calculation in the furnace cavity is constructed. In terms of melt thermal convection, the control equations include momentum equation, energy equation and continuity equation.

[0085] Boundary conditions and initial conditions settings:

[0086] The corresponding boundary conditions and initial conditions are set for the control equations. The boundary conditions of the melt surface and solidification interface involve temperature, velocity and radiation conditions.

[0087] Mathematical modeling process:

[0088] Convert the physical model into a mathematical model, including the governing equations and boundary conditions. This process involves expressing the physical process in mathematical language and defining the domain and boundaries of the equations.

[0089] Mesh generation technology:

[0090] Generate meshes for numerical calculations on the geometric model. The quality and quantity of the meshes will directly affect the accuracy and computational efficiency of the numerical simulation.

[0091] Discretization of the equation:

[0092] The control equations are discretized and converted into a set of algebraic equations that can be solved on a computer.

[0093] Physical quantity coupling solution strategy:

[0094] The SIMPLE method algorithm is used to solve the coupled physical quantities.

[0095] S2. Data calculation and result analysis

[0096] The finite volume method is used to discretize the governing equations, which is:

[0097]

[0098] Where f represents the velocity, enthalpy, and turbulence parameters, u is the velocity field, D is the diffusion coefficient, and S is the source term. The complex function method is then used to solve the electromagnetic field to ensure the correctness of the model. A global data simulation is then performed to analyze the temperature distribution in the crystal and the solution, the argon convection between the heat shield and the quartz crucible, and the thermal stress in the crystal.

[0099] S3. Optimize thermal field structure

[0100] By changing the crucible shape, heater position, and insulation cage separator position parameters, the crystal growth process was optimized using computer numerical simulation. The effects of the optimized heat shield on the heater, the effect of the carbon felt on the sidewalls on preventing heat loss, and the effect of the diversion system on reducing SiO deposition on the upper wall were analyzed.

[0101] S4. Process parameter optimization

[0102] Artificial neural networks and genetic algorithms were used to optimize the process parameters of pulling speed, crystal rotation speed, ambient gas, and crucible wall temperature to obtain a tight and flat solid-liquid interface. The response surface methodology was used to determine the crystal rotation and crucible rotation under the optimal thermal field distribution. Specifically, the gradient descent method was used to update the network weights:

[0103] Δw=-ηj'(w)

[0104] Where η is the learning rate and j'(w) is the gradient of the loss with respect to the parameters;

[0105] S5. Machine Learning Model Building and Training

[0106] Based on the CFD simulation data, an artificial neural network is used to build a model of the crystal growth process. The neural network model is trained to enable it to predict the CFD simulation results of SiC crystal growth and optimize the growth conditions. The machine learning model is established and trained, specifically including the following steps:

[0107] Problem Definition:

[0108] Define the problem to be solved, including determining the target variable, evaluation metrics, and expected outcomes.

[0109] Data preparation:

[0110] Collect, clean, and preprocess data, including handling missing values, outliers, and noise, to ensure data quality and consistency.

[0111] Feature Selection:

[0112] Identifying and selecting the data columns that are most helpful in predicting the target, including feature engineering such as creating interaction terms, polynomial features, or applying dimensionality reduction techniques.

[0113] Model selection:

[0114] Depending on the nature of the problem, choose an appropriate machine learning algorithm, including decision trees, random forests, support vector machines, and neural networks.

[0115] Training and validation:

[0116] A training dataset is used to train a model, and a validation set is used to evaluate its performance. Cross-validation is a common technique used to ensure that a model performs well on more than just a fixed subset of data.

[0117] Parameter tuning:

[0118] Finding the optimal hyperparameter combination through techniques such as grid search, random search, or Bayesian optimization can significantly improve the performance of the model.

[0119] Model training:

[0120] The process of using established model methods to learn data experience also needs to be combined with model evaluation to adjust the algorithm's hyperparameters and ultimately select a model with better performance.

[0121] Dataset partitioning: The dataset is divided into a training set and a test set, and the training set can be further divided into a training set and a validation set.

[0122] Model method selection:

[0123] Choose an appropriate model structure based on the problem, such as a convolutional neural network (CNN).

[0124] Define hyperparameters:

[0125] Hyperparameters are configuration settings used to adjust how a model is trained, such as learning rate, batch size, and training epochs.

[0126] Loading data:

[0127] Use the torchvision.datasets module to download and load the dataset, and use DataLoader to handle batching and shuffling.

[0128] Initialize the network:

[0129] Instantiate the neural network and move it to the device (GPU or CPU).

[0130] Define the loss and optimizer:

[0131] Choose a loss function (such as cross entropy loss) and an optimizer (such as Adam optimizer).

[0132] Train the network:

[0133] The dataset is cycled through multiple times (epochs) and the model weights are updated based on the loss.

[0134] Model Evaluation:

[0135] Define a function to check the accuracy of the model on the training and test datasets.

[0136] Model saving:

[0137] Use a library like joblib or pickle to save the model to a file for future use.

[0138] Using the model:

[0139] Apply the model to new data to make predictions, including loading the model and predicting the new data;

[0140] S6. Rotating motor control

[0141] The controller controls the rotary motor to make the pulling device rotate the seed crystal, causing the crystal to rotate. The torque is calculated using the BLDC motor sinusoidal commutation formula:

[0142] Shaft torque = K t [I R sin(θ)+I S sin(θ+120°)+I T sin(θ+240°)]

[0143] Where θ is the electrical angle of the shaft, K t is the torque constant of the motor, I R ,IS ,I T is the phase current. In the isodiameter growth stage, the controller controls the rotary motor to make the pulling device rotate the seed crystal, and at the same time controls the heating device to heat the crucible;

[0144] S7. Data feedback and closed-loop control

[0145] The encoder is used to feedback the position of the crucible rod and seed crystal rod to achieve closed-loop control and display. The parameters are set and displayed through the touch screen, the real-time status of the motor is monitored, and the control parameters are adjusted according to the feedback. Specifically, the PID control algorithm is used to perform closed-loop control of the key parameters in the crystal growth process:

[0146]

[0147] Among them, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, e(t) is the deviation;

[0148] S8. Experimental verification and adjustment

[0149] Crystal growth experiments are carried out to verify the accuracy of numerical simulations, and model parameters are adjusted based on the realized results to improve the reliability of the simulation.

[0150] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling a rotating motor of a crystal growth furnace, characterized in that: The specific steps include: S1. Establish geometric and physical models According to the actual structure of the crystal growth furnace, a geometric model is established, including the crucible, heater and insulation material. At the same time, a physical model is established, including heat conduction, fluid dynamics and mass transfer. The Boltzmann equation and lattice Boltzmann method are used to simulate crystal growth. S2. Data calculation and result analysis The control equation is discretized using the finite volume method, and the formula is: Where f represents the physical quantities of velocity, enthalpy and turbulence parameters, u is the velocity field, D is the diffusion coefficient, and S is the source term. The complex function method is then used to solve the electromagnetic field to ensure the correctness of the model. Global data simulation is then performed to analyze the temperature distribution in the crystal and the solution, the argon convection between the heat shield and the quartz crucible, and the thermal stress in the crystal. S3. Optimize thermal field structure By changing the crucible shape, heater position, and insulation cage partition block position parameters, the crystal growth process was optimized using computer numerical simulation, and the influence of the optimized heat shield on the heater, the effect of the side wall carbon felt on preventing heat loss, and the effect of the flow guide system on reducing SiO deposition on the upper wall were analyzed. S4. Process parameter optimization Artificial neural network and genetic algorithm are used to optimize the process parameters of pulling speed, crystal rotation speed, ambient gas and crucible wall temperature to obtain a tight and flat solid-liquid interface. The crystal rotation and crucible rotation under the optimal thermal field distribution are determined by response surface method. Specifically, the gradient descent method is used to update the network weights: Δw=-ηj'(w) Where η is the learning rate and j'(w) is the gradient of the loss with respect to the parameter; S5. Machine learning model building and training Based on the CFD simulation data, an artificial neural network is used to build a model of the crystal growth process, and the neural network model is trained to enable it to predict the CFD simulation results of SiC crystal growth and optimize the growth conditions; S6. Rotating motor control The controller controls the rotary motor to make the pulling device rotate the seed crystal, so that the crystal rotates, and the torque is calculated using the BLDC motor sinusoidal commutation formula: Shaft torque = K t [I R sin(θ) + I S sin(θ + 120°) + I T sin(θ + 240°)] Where θ is the electrical angle of the shaft, K t is the torque constant of the motor, I R ,I S ,I T is the phase current. In the equal diameter growth stage, the controller controls the rotary motor to make the pulling device rotate the seed crystal, and controls the heating device to heat the crucible at the same time; S7. Data feedback and closed-loop control The encoder is used to feedback the position of the crucible rod and the seed crystal rod to achieve closed-loop control and display. The parameters are set and displayed through the touch screen, the real-time status of the motor is monitored, and the control parameters are adjusted according to the feedback. Specifically, the PID control algorithm is used to perform closed-loop control of the key parameters in the crystal growth process: Among them, K p is the proportionality coefficient, K i is the integration coefficient, K d is the differential coefficient, e(t) is the deviation; S8. Experimental verification and adjustment Crystal growth experiments are implemented to verify the accuracy of numerical simulations, and model parameters are adjusted based on the achieved results to improve the reliability of the simulation.

2. A method for controlling a rotating motor of a crystal growth furnace according to claim 1, characterized in that: The S1. establishing geometric and physical models specifically includes the following steps: geometric modeling, basic assumptions, establishment of control equations, setting of boundary conditions and initial conditions, mathematical modeling, grid generation, equation discretization, selection of numerical solution methods and coupled solution of physical quantities.

3. A method for controlling a rotating motor of a crystal growth furnace according to claim 1, characterized in that: The S5. machine learning model establishment and training specifically includes the following steps: problem definition, data preparation, feature selection, model selection, training and verification, parameter tuning, model training, data set partitioning, model method selection, definition of hyperparameters, data loading, network initialization, definition of loss and optimizer, network training, model evaluation, model preservation and model use.

Images