A transmission rotating movable crystal growth high pressure furnace

Abstract

The present invention relates to the technical field of crystal growth furnaces, and in particular to a transmission rotating movable crystal growth high-pressure furnace, comprising a crystal growth furnace, a base is rotatably connected to the bottom plate inside the crystal growth furnace, a graphite crucible is fixed on the base, a quartz crucible is fixed inside the graphite crucible, an electric heating plate is arranged on the inner wall of the crystal growth furnace on the outside of the graphite crucible, a driving mechanism is arranged at the bottom of the crystal growth furnace, a linkage mechanism is arranged on the base, and the linkage mechanism is transmission-connected to the electric heating plate. The transmission rotating movable crystal growth high-pressure furnace provided by the present invention improves the uniformity of heating of the quartz crucible by controlling the rotation of the quartz crucible, and on the basis of controlling the rotation of the quartz crucible, synchronously adjusts the position of the electric heating plate on the quartz crucible, so that the heat radiated from the electric heating plate to the crucible is more uniform, and the temperature in the furnace is evenly distributed, thereby improving the quality of crystal growth and shortening the growth cycle.

Description

Technical Field

[0001] The invention relates to the technical field of crystal growth furnaces, in particular to a transmission rotating movable crystal growth high pressure furnace. Background Art

[0002] A crystal growth furnace usually consists of a high-temperature furnace body and a container for melting furnace materials. First, the raw material or furnace material is heated to a high temperature to melt it into a molten liquid. Then, a crystal seed or substrate is placed in the molten liquid in an appropriate manner and gradually extracted from it. As the crystal seed is extracted, the molten liquid crystallizes on the seed to form a single crystal. The crystal growth furnace controls the growth rate and quality of the crystal by controlling parameters such as temperature, pressure and extraction speed.

[0003] A Chinese patent with authorization announcement number CN114318510B discloses an indium antimonide crystal growth method and a crystal growth furnace, which can reduce the stress of indium antimonide crystals. The method is implemented based on a crystal growth furnace, and the crystal growth furnace includes a furnace body, a crucible arranged in the furnace body, and a heat shield located above the crucible. The crucible contains indium antimonide polycrystalline raw materials, and an electromagnetic induction heating coil is arranged inside the heat shield. The electromagnetic induction heating coil is sleeved on the periphery of the indium antimonide crystal and corresponds to the local interval of the indium antimonide crystal; the crystal growth furnace is used to grow indium antimonide crystals based on the Czochralski method. When growing, the indium antimonide crystal passes through the screen opening of the heat shield and the electromagnetic induction heating coil in sequence, and the indium antimonide crystal is periodically heated by the electromagnetic induction heating coil.

[0004] As mentioned in the above patent, among the existing crystal growth furnaces using the melting method, the more common one is the static high-pressure furnace. However, these static high-pressure furnaces have obvious deficiencies in the crystal growth process, which are mainly manifested in the following aspects: uneven crystal growth. Due to the uneven temperature distribution in the furnace, the crystal has inconsistent morphology during the growth process, which affects the quality and performance of the crystal. Secondly, the static high-pressure furnace usually adopts fixed growth parameters and cannot be adjusted in real time according to the actual needs of crystal growth, resulting in a long growth cycle. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a transmission rotating movable crystal growth high-pressure furnace, including a crystal growth furnace, a base is rotatably connected to the bottom plate inside the crystal growth furnace, a graphite crucible is fixed on the base, a quartz crucible is fixed in the graphite crucible, an electric heating plate is arranged on the inner wall of the crystal growth furnace on the outside of the graphite crucible, a driving mechanism is arranged at the bottom of the crystal growth furnace, a linkage mechanism is arranged on the base, and the linkage mechanism is transmission-connected to the electric heating plate, when the driving mechanism drives the base to rotate, the linkage mechanism synchronously drives the electric heating plate to reciprocate and rise and fall in the crystal growth furnace.

[0006] As a further description of the above technical solution: an upper furnace body is arranged above the crystal growth furnace, and the upper furnace body is sealed and connected to the crystal growth furnace, thereby forming an upper chamber above the crystal growth furnace, a hydraulic telescopic rod is installed on the top of the upper chamber, and a seed crystal rod is fixed to the bottom end of the movable end of the hydraulic telescopic rod, a seed crystal chuck is installed at the bottom end of the seed crystal rod, and a seed crystal is installed on the seed crystal chuck.

[0007] As a further description of the above technical solution: a base is welded to the bottom of the crystal growth furnace, the driving mechanism is a driving motor, the driving motor is installed on the inner top plate of the base, and the output shaft of the driving motor extends into the crystal growth furnace and is fixedly connected to the base.

[0008] As a further description of the above technical solution: the electric heating plate is a ring-shaped structure, and a resistance heating body is embedded and installed in the electric heating plate along the vertical direction. There are a number of resistance heating bodies, and the resistance heating bodies are distributed in a ring shape with equal intervals. A slider is welded on the outer wall of the electric heating plate, and the slider is slidably connected to a strip slide groove opened on the inner wall of the crystal growth furnace.

[0009] As a further description of the above technical solution: the linkage mechanism includes a guide groove and a fixed block, the guide groove is opened on the inner wall of the electric heating plate, the guide groove is a wave-shaped annular structure connected end to end, the fixed block is welded on the base, the fixed block is located on the outside of the graphite crucible, there are a plurality of fixed blocks, and the plurality of fixed blocks are distributed on the base in a ring shape with equal intervals, and sliding seats are welded on the plurality of fixed blocks, and the sliding seats are slidably connected to the guide groove.

[0010] As a further description of the above technical solution: it also includes a furnace control unit, and the furnace control unit includes:

[0011] The first data acquisition module collects historical control data training parameters of the crystal growth furnace, the historical control data training parameters are collected when the crystal preparation meets the standards, and the historical control data training parameters include comprehensive parameters of crystal raw materials and growth parameters of prepared crystals;

[0012] The comprehensive parameters of the crystal raw materials include the quality and composition ratio of the crystal raw materials;

[0013] The growth parameters of the prepared crystal are the heating temperature data of the electric heating plate and the pressure data in the crystal growth furnace;

[0014] The model training module trains the machine learning model for predicting growth parameters based on the historical control data training parameters, collects the real-time comprehensive parameters of the crystal raw materials, inputs them into the trained machine learning model for predicting growth parameters, obtains the growth parameter set label, and then obtains the growth parameter set corresponding to the growth parameter set label, that is, obtains the corresponding growth parameters in the generation parameter set;

[0015] A second data acquisition module collects comprehensive parameters of the melt in real time, wherein the comprehensive parameters of the melt include viscosity data of the melt and diffusion coefficient data of solutes in the melt;

[0016] The comprehensive analysis module generates a boosting speed coefficient based on the growth parameters and the comprehensive parameters of the melt, and generates a boosting speed grade according to the boosting speed coefficient.

[0017] As a further description of the above technical solution: the growth parameter set is {(a1, b1,), (a2, b2), ..., (a n , b n ,)},a n is the heating temperature data of the electric heating plate corresponding to the set label n, b n is the pressure data in the crystal growth furnace corresponding to the set label n;

[0018] Methods for training machine learning models to predict growth parameters include:

[0019] Set the growth parameter sets with corresponding numbers in advance;

[0020] The comprehensive parameters of the crystal raw materials are converted into a corresponding set of feature vectors, and each set of feature vectors is used as the input of the machine learning model. The machine learning model uses the number of a set of growth parameters corresponding to each set of comprehensive parameters of the crystal raw materials as output, and the number of a set of growth parameters actually corresponding to each set of comprehensive parameters of the crystal raw materials as the prediction target, and minimizing the loss function value of the machine learning model as the training target; when the loss function value of the machine learning model is less than or equal to the preset target loss value, the training is stopped.

[0021] As a further description of the above technical solution: the method for generating the lifting speed coefficient based on the comprehensive parameters of the growth parameters and the melt includes:

[0022]

[0023] In the formula, TSxs is the lifting speed coefficient, WDsj is the heating temperature data of the electric heating plate, YLsj is the pressure data in the crystal growth furnace, NDsj is the viscosity data of the object, and KSxs is the diffusion coefficient data of the solute in the melt. and is the weight factor, and and Both are greater than 0.

[0024] As a further description of the above technical solution: the lifting speed levels include a first-level lifting speed, a second-level lifting speed and a third-level lifting speed, and the lifting speeds of the first-level lifting speed, the second-level lifting speed and the third-level lifting speed increase in sequence.

[0025] As a further description of the above technical solution: the method for generating a boost speed level according to a boost speed coefficient includes:

[0026] Preset lifting speed coefficient threshold TS l and TS2, where TS1 < TS2;

[0027] When TSxs<TS1, the first level of lifting speed is generated;

[0028] When TS1≤TSxs<TS2, the secondary lifting speed is generated;

[0029] When TSxs>TS2, a three-level boost speed is generated.

[0030] Beneficial effects:

[0031] The present invention provides a transmission rotary movable crystal growth high pressure furnace. When the base drives the quartz crucible to rotate, the linkage assembly is used to synchronously drive the electric heating plate to reciprocate and rise and fall in the crystal growth furnace. That is, the transmission rotary movable crystal growth high pressure furnace controls the rotation of the quartz crucible to improve the uniformity of its heating. On the basis of controlling the rotation of the quartz crucible, the position of the electric heating plate on the quartz crucible is synchronously adjusted to make the heat radiated from the electric heating plate to the crucible more uniform, thereby achieving uniform distribution of temperature in the furnace, improving the quality of crystal growth, shortening the growth cycle, and reducing the energy consumption of the crystal growth furnace.

[0032] Furthermore, by collecting historical control data training parameters, a machine learning model for predicting growth parameters is trained, and then real-time comprehensive parameters of crystal raw materials are collected and input into the trained machine learning model for predicting growth parameters to obtain a growth parameter set label, and then the growth parameter set corresponding to the growth parameter set label is obtained to achieve intelligent control of growth parameters and ensure the growth efficiency of the crystal. Further, a lifting speed coefficient is generated based on the growth parameters of the crystal and the comprehensive parameters of the melt, and a lifting speed level is generated according to the lifting speed coefficient, so that the lifting speed of the seed rod 62 is controlled based on the generated lifting speed level, thereby ensuring the stability of the crystal growth process and the product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments:

[0034] Figure 1 A cross-sectional schematic diagram of a transmission rotating movable crystal growth high pressure furnace provided by the present invention;

[0035] Figure 2 A schematic cross-sectional view of a base provided by the present invention;

[0036] Figure 3 A schematic diagram of the structure of the electric heating plate provided by the present invention;

[0037] Figure 4 A schematic diagram of the connection structure between the base and the graphite crucible provided by the present invention;

[0038] Figure 5 A schematic diagram of the structure of the linkage assembly provided by the present invention;

[0039] Figure 6 This is a module connection diagram of the furnace control unit provided by the present invention.

[0040] In the figure: 1. crystal growth furnace; 11. base; 2. pedestal; 3. graphite crucible; 4. quartz crucible; 5. electric heating plate; 51. resistance heating body; 52. slider; 6. upper furnace body; 61. hydraulic telescopic rod; 62. seed crystal rod; 63. seed crystal chuck; 64. seed crystal; 7. driving motor; 81. fixed block; 82. sliding seat; 83. guide groove. DETAILED DESCRIPTION

[0041] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention is further described below with reference to specific diagrams. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0042] Example 1

[0043] See also Figure 1-Figure 5 The embodiment of the present invention provides a technical solution: a transmission rotating movable crystal growth high pressure furnace, comprising a crystal growth furnace 1, a base 2 is rotatably connected to the bottom plate inside the crystal growth furnace 1, a graphite crucible 3 is fixed on the base 2, a quartz crucible 4 is fixed inside the graphite crucible 3, an electric heating plate 5 is arranged on the inner wall of the crystal growth furnace 1 on the outside of the graphite crucible 3, a driving mechanism is arranged at the bottom of the crystal growth furnace 1, a linkage mechanism is arranged on the base 2, and the linkage mechanism is transmission-connected to the electric heating plate 5, when the driving mechanism drives the base 2 to rotate, the linkage mechanism synchronously drives the electric heating plate 5 to reciprocate and rise and fall in the crystal growth furnace 1.

[0044] A base 11 is welded to the bottom of the crystal growth furnace 1 , and the driving mechanism is a driving motor 7 , which is mounted on the inner top plate of the base 11 , and the output shaft of the driving motor 7 extends into the crystal growth furnace 1 and is fixedly connected to the base 2 .

[0045] The electric heating plate 5 is a ring-shaped structure, and a resistance heating body 51 is embedded and installed in the electric heating plate 5 along the vertical direction. There are a number of resistance heating bodies 51, and the resistance heating bodies 51 are distributed in a ring shape with equal intervals. A slider 52 is welded on the outer wall of the electric heating plate 5, and the slider 52 is slidably connected to a strip groove opened on the inner wall of the crystal growth furnace 1.

[0046] The linkage mechanism includes a guide groove 83 and a fixed block 81. The guide groove 83 is provided on the inner wall of the electric heating plate 5. The guide groove 83 is a wave-shaped annular structure connected end to end. The fixed block 81 is welded on the base 2. The fixed block 81 is located on the outside of the graphite crucible 3. There are a plurality of fixed blocks 81, and the plurality of fixed blocks 81 are distributed on the base 2 in a ring-shaped manner with equal intervals. A sliding seat 82 is welded on the plurality of fixed blocks 81, and the sliding seat 82 is slidably connected to the guide groove 83.

[0047] Specifically, when the transmission rotary movable crystal growth high pressure furnace is used, the raw materials for preparing the crystal are added into the quartz crucible 4, and then the electric heating plate 5 is controlled to heat the quartz crucible 4 to form a melt, and the seed crystal 64 is controlled to be placed in the melt and gradually extracted therefrom. As the seed crystal 64 is extracted, the melt will crystallize on the seed crystal 64 to form a crystal;

[0048] Furthermore, during use, the transmission rotary movable crystal growth high-pressure furnace controls the rotation of the base 2 through a driving mechanism, and the rotation of the base 2 drives the quartz crucible 4 to rotate, and through the setting of the linkage mechanism, when the base 2 drives the quartz crucible to rotate, the linkage assembly synchronously drives the electric heating plate 5 to move back and forth in the crystal growth furnace 1, that is, the transmission rotary movable crystal growth high-pressure furnace improves the uniformity of heating by controlling the rotation of the quartz crucible 4, and on the basis of controlling the rotation of the quartz crucible 4, synchronously adjusts the position of the electric heating plate 5 on the quartz crucible 4, so that the heat radiated from the electric heating plate 5 to the crucible is more uniform, thereby achieving uniform distribution of temperature in the furnace, improving the quality of crystal growth, shortening the growth cycle, and reducing the energy consumption of the crystal growth furnace.

[0049] Furthermore, the specific manner in which the driving mechanism drives the quartz crucible 4 to rotate is to control the driving motor 7 to drive the base 2 to rotate, and the rotation of the base 2 drives the quartz crucible 4 to rotate through the graphite crucible 3, thereby improving the uniformity of heating of the quartz crucible 4, and when the base 2 rotates, the fixed block 81 provided on the base 2 is synchronously driven to move, and the fixed block 81 drives the sliding seat 82 to move in the guide groove 83 provided on the electric heating plate 5, thereby driving the electric heating plate 5 to move back and forth.

[0050] Example 2

[0051] See also Figure 1The present embodiment is based on the above-mentioned embodiment and is implemented by adding an upper furnace body 6. The upper furnace body 6 is arranged above the crystal growth furnace 1, and the upper furnace body 6 is sealed and connected to the crystal growth furnace 1, so as to form an upper chamber above the crystal growth furnace 1. A hydraulic telescopic rod 61 is installed on the top of the upper chamber, and a seed crystal rod 62 is fixed to the bottom end of the movable end of the hydraulic telescopic rod 61, a seed crystal chuck 63 is installed at the bottom end of the seed crystal rod 62, and a seed crystal 64 is installed on the seed crystal chuck 63.

[0052] Specifically, on the basis of Example 1, raw materials for preparing crystals are added into the quartz crucible 4, and then the electric heating plate 5 is controlled to heat the quartz crucible 4 to form a melt, and the seed crystal 64 is controlled to be placed in the melt and gradually extracted therefrom. As the seed crystal 64 is extracted, the melt crystallizes on the seed crystal 64 to form a crystal;

[0053] The method of controlling the extraction of the seed crystal 64 is to control the contraction of the hydraulic telescopic rod 61 to drive the seed crystal rod 62 to rise, and the seed crystal 64 is moved upward by the lifting of the seed crystal rod 62, so that the molten liquid will crystallize on the seed crystal 64 to form a crystal.

[0054] Example 3

[0055] See also Figure 1 and Figure 6 , and also includes a furnace control unit, the furnace control unit includes:

[0056] The first data acquisition module collects historical control data training parameters of the crystal growth furnace, the historical control data training parameters are collected when the crystal preparation meets the standards, and the historical control data training parameters include comprehensive parameters of crystal raw materials and growth parameters of prepared crystals;

[0057] The comprehensive parameters of crystal raw materials include the quality and composition ratio of crystal raw materials;

[0058] The growth parameters for preparing the crystal are the heating temperature data of the electric heating plate 5 and the pressure data in the crystal growth furnace 1;

[0059] It should be noted that the crystal growth furnace 1 is provided with a vacuum pump (the figure shows the vacuum pump) for adjusting the pressure in the crystal growth furnace 1 .

[0060] The model training module trains the machine learning model for predicting growth parameters based on the historical control data training parameters, collects the real-time comprehensive parameters of the crystal raw materials, inputs them into the trained machine learning model for predicting growth parameters, obtains the growth parameter set label, and then obtains the growth parameter set corresponding to the growth parameter set label, that is, obtains the corresponding growth parameters in the generation parameter set;

[0061] The second data acquisition module collects comprehensive parameters of the melt in real time, and the comprehensive parameters of the melt include viscosity data of the melt and diffusion coefficient data of solutes in the melt;

[0062] It should be noted that the molten state formed by adding the raw materials for preparing the crystal into the quartz crucible 4 and then heating is the melt.

[0063] It should be noted that the viscosity data of the melt can be obtained by installing an ultrasonic sensor, emitting ultrasonic waves and receiving reflected signals, calculating the viscosity based on the propagation speed and attenuation, and realizing non-contact measurement to reduce interference with the melt.

[0064] The method for obtaining the diffusion coefficient data of the solute in the melt can be obtained by the neutron scattering method, that is, by setting a neutron source to irradiate the melt, monitoring the changes in the scattering signal, and calculating the diffusion coefficient based on the scattering spectrum. It should be noted that obtaining the diffusion coefficient data of the solute in the melt by the neutron scattering method is an existing technology.

[0065] It should be noted that the larger the diffusion coefficient of the solute in the melt, the more conducive it is to the transport of the solute from the main solution to the growth interface, thereby increasing the growth rate. That is, the larger the diffusion coefficient of the solute in the melt, the faster the crystal grows.

[0066] The larger the viscosity of the melt, the more difficult it is for the solute molecules to move quickly to the crystal growth interface, resulting in a lower growth rate. That is, the larger the viscosity of the melt, the slower the crystal growth rate.

[0067] The comprehensive analysis module generates a boosting speed coefficient based on the growth parameters and the comprehensive parameters of the melt, and generates a boosting speed grade according to the boosting speed coefficient.

[0068] The growth parameter set is {(a1, b1,), (a2, b2), ..., (a n , b n ,)},a n is the heating temperature data of the electric heating plate 5 corresponding to the set label n, b n is the pressure data in the crystal growth furnace 1 corresponding to the set label n;

[0069] Methods for training machine learning models to predict growth parameters include:

[0070] Set the growth parameter sets with corresponding numbers in advance;

[0071] The comprehensive parameters of the crystal raw materials are converted into a corresponding set of feature vectors, and each set of feature vectors is used as the input of the machine learning model. The machine learning model uses the number of a set of growth parameters corresponding to each set of comprehensive parameters of the crystal raw materials as output, and the number of a set of growth parameters actually corresponding to each set of comprehensive parameters of the crystal raw materials as the prediction target, and minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0072] The machine learning model may be one of support vector machine regression, random forest regression or neural network regression models.

[0073] The loss function value of the machine learning model is the mean square error.

[0074] Mean square error is one of the commonly used loss functions. The model is trained with minimization as the goal, so that the machine learning model can better fit the data, thereby improving the performance and accuracy of the model.

[0075] In the loss function, MSE is the feature vector group number; x is the feature vector group number; m is the number of feature vector groups; y is the number of feature vector groups; x is the number of the growth parameter set corresponding to the xth group of eigenvectors, is the number of the growth parameter set corresponding to the x-th group of eigenvectors in real time.

[0076] Other model parameters of the machine learning model, target loss value, optimization algorithm, training set test set validation set ratio, and loss function optimization are all achieved through actual engineering implementation and continuous experimental tuning.

[0077] Methods for generating the acceleration coefficient based on the comprehensive parameters of growth parameters and melt include:

[0078]

[0079] Wherein, TSxs represents the lifting speed coefficient, WDsj is the heating temperature data of the electric heating plate 5, YLsj is the pressure data in the crystal growth furnace 1, NDsj is the viscosity data of the object, KSxs is the diffusion coefficient data of the solute in the melt, and is the weight factor, and and Both are greater than 0.

[0080] It should be noted that the size of the weight coefficient is a specific value obtained by quantifying each data, which is convenient for subsequent comparison. The size of the weight coefficient depends on the number of comprehensive parameters and the preliminary setting of the corresponding weight coefficient for each group of comprehensive parameters by those skilled in the art;

[0081] It should be noted that within the reasonable heating temperature data range of the electric heating plate 5, when the heating temperature data of the electric heating plate 5 is higher, the growth rate of the crystal is faster, that is, the faster the lifting speed that needs to be controlled is, because the temperature increase will increase the solubility of the solute in the melt, produce supersaturation, thereby promoting crystal growth, and increasing the temperature helps to increase the diffusion rate and solubility of the solute, thereby promoting crystal growth;

[0082] Within a reasonable range of pressure data in the crystal growth furnace 1 , when the pressure data in the crystal growth furnace 1 is higher, the growth rate of the crystal is faster, because under high pressure, the solution can maintain a higher solute concentration, promote the formation of a supersaturated state, and thus accelerate crystal growth.

[0083] The lifting speed levels include a first-level lifting speed, a second-level lifting speed and a third-level lifting speed, and the lifting speeds of the first-level lifting speed, the second-level lifting speed and the third-level lifting speed increase in sequence.

[0084] Methods for generating a boost speed grade based on a boost speed factor include:

[0085] Preset the lifting speed coefficient thresholds TS1 and TS2, where TS1 < TS2;

[0086] When TSxs<TS1, the first level of lifting speed is generated;

[0087] When TS1≤TSxs<TS2, the secondary lifting speed is generated;

[0088] When TSxs>TS2, a three-level boost speed is generated.

[0089] It should be noted that, based on the generated lifting speed level, the retraction speed of the hydraulic telescopic rod 61 is controlled, that is, the lifting speed of the crystal is controlled;

[0090] Specifically, a furnace control unit is provided, and historical control data training parameters are collected to train a machine learning model for predicting growth parameters. Then, real-time comprehensive parameters of crystal raw materials are collected and input into the trained machine learning model for predicting growth parameters to obtain a growth parameter set label, and then the growth parameter set corresponding to the growth parameter set label is obtained, that is, the heating temperature data of the electric heating plate 5 for controlling the crystal growth furnace 1 and the pressure data in the crystal growth furnace 1 are obtained, so as to realize intelligent control of growth parameters and ensure the growth efficiency of the crystal, and further generate a lifting speed coefficient based on the growth parameters of the crystal and the comprehensive parameters of the melt, and generate a lifting speed level according to the lifting speed coefficient, so as to control the lifting speed of the seed rod 62 based on the generated lifting speed level, thereby ensuring the stability of the crystal growth process and the product quality.

[0091] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and the description in the specification are only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A transmission rotary movable crystal growth high pressure furnace, comprising a crystal growth furnace (1), characterized in that: A base (2) is rotatably connected to the bottom plate inside the crystal growth furnace (1), a graphite crucible (3) is fixed on the base (2), a quartz crucible (4) is fixed inside the graphite crucible (3), an electric heating plate (5) is arranged on the inner wall of the crystal growth furnace (1) outside the graphite crucible (3), a driving mechanism is arranged at the bottom of the crystal growth furnace (1), a linkage mechanism is arranged on the base (2), and the linkage mechanism is transmission-connected to the electric heating plate (5), when the driving mechanism drives the base (2) to rotate, the linkage mechanism synchronously drives the electric heating plate (5) to reciprocate and rise and fall inside the crystal growth furnace (1); The linkage mechanism comprises a guide groove (83) and a fixed block (81); the guide groove (83) is provided on the inner wall of the electric heating plate (5); the guide groove (83) is a wave-shaped annular structure connected end to end; the fixed block (81) is welded on the base (2); the fixed block (81) is located outside the graphite crucible (3); a plurality of the fixed blocks (81) are provided, and the plurality of fixed blocks (81) are distributed on the base (2) in an annular shape with equal spacing; a sliding seat (82) is welded on the plurality of fixed blocks (81); the sliding seat (82) is slidably connected to the guide groove (83).

2. The transmission rotating movable crystal growth high pressure furnace according to claim 1, characterized in that: An upper furnace body (6) is arranged above the crystal growth furnace (1), and the upper furnace body (6) is sealedly connected to the crystal growth furnace (1), thereby forming an upper chamber above the crystal growth furnace (1), a hydraulic telescopic rod (61) is installed at the top of the upper chamber, and a seed crystal rod (62) is fixed to the bottom end of the movable end of the hydraulic telescopic rod (61), a seed crystal chuck 63 is installed at the bottom end of the seed crystal rod (62), and a seed crystal (64) is installed on the seed crystal chuck (63).

3. The transmission rotating movable crystal growth high pressure furnace according to claim 1, characterized in that: A base (11) is welded to the bottom of the crystal growth furnace (1); the driving mechanism is a driving motor (7); the driving motor (7) is mounted on an internal top plate of the base (11); and an output shaft of the driving motor (7) extends into the crystal growth furnace (1) and is fixedly connected to the base (2).

4. The transmission rotating movable crystal growth high pressure furnace according to claim 3, characterized in that: The electric heating plate (5) is an annular structure, and a resistance heating body (51) is embedded and installed in the electric heating plate (5) along the vertical direction. A plurality of resistance heating bodies (51) are provided, and the plurality of resistance heating bodies (51) are distributed in an annular shape with equal spacing. A slider (52) is welded on the outer wall of the electric heating plate (5), and the slider (52) is slidably connected to a strip-shaped slide groove provided on the inner wall of the crystal growth furnace (1).

5. The transmission rotating movable crystal growth high pressure furnace according to claim 1, characterized in that: It also includes a furnace control unit, the furnace control unit comprising: The first data acquisition module collects historical control data training parameters of the crystal growth furnace, the historical control data training parameters are collected when the crystal preparation meets the standards, and the historical control data training parameters include comprehensive parameters of crystal raw materials and growth parameters of prepared crystals; The comprehensive parameters of the crystal raw materials include the quality and composition ratio of the crystal raw materials; The growth parameters for preparing the crystal are heating temperature data of the electric heating plate (5) and pressure data in the crystal growth furnace (1); The model training module trains the machine learning model for predicting growth parameters based on the historical control data training parameters, collects the real-time comprehensive parameters of the crystal raw materials, inputs them into the trained machine learning model for predicting growth parameters, obtains the growth parameter set label, and then obtains the growth parameter set corresponding to the growth parameter set label, that is, obtains the corresponding growth parameters in the generation parameter set; A second data acquisition module collects comprehensive parameters of the melt in real time, wherein the comprehensive parameters of the melt include viscosity data of the melt and diffusion coefficient data of solutes in the melt; The comprehensive analysis module generates a boosting speed coefficient based on the growth parameters and the comprehensive parameters of the melt, and generates a boosting speed grade according to the boosting speed coefficient.

6. The transmission rotating movable crystal growth high pressure furnace according to claim 5, characterized in that: The growth parameter set is {(a1,b1,),(a2,b2),...,(a n ,b n ,)},a n is the heating temperature data of the electric heating plate (5) corresponding to the set label n, b n is the pressure data in the crystal growth furnace (1) corresponding to the set label n; Methods for training machine learning models to predict growth parameters include: Set the growth parameter sets with corresponding numbers in advance; The comprehensive parameters of the crystal raw materials are converted into a corresponding set of feature vectors, and each set of feature vectors is used as the input of the machine learning model. The machine learning model uses the number of a set of growth parameters corresponding to each set of comprehensive parameters of the crystal raw materials as output, and the number of a set of growth parameters actually corresponding to each set of comprehensive parameters of the crystal raw materials as the prediction target, and minimizing the loss function value of the machine learning model as the training target; when the loss function value of the machine learning model is less than or equal to the preset target loss value, the training is stopped.

7. The transmission rotating movable crystal growth high pressure furnace according to claim 6, characterized in that: The method for generating a lifting speed coefficient based on the comprehensive parameters of the growth parameters and the melt comprises: Wherein, TSxs represents the lifting speed coefficient, WDsj represents the heating temperature data of the electric heating plate (5), YLsj represents the pressure data in the crystal growth furnace (1), NDsj represents the viscosity data of the object, KSxs represents the diffusion coefficient data of the solute in the melt, and is the weight factor, and and Both are greater than 0.

8. The transmission rotating movable crystal growth high pressure furnace according to claim 7, characterized in that: The lifting speed levels include a primary lifting speed, a secondary lifting speed and a tertiary lifting speed, and the lifting speeds of the primary lifting speed, the secondary lifting speed and the tertiary lifting speed increase in sequence.

9. The transmission rotating movable crystal growth high pressure furnace according to claim 8, characterized in that: The method for generating a boost speed level according to a boost speed coefficient comprises: Preset the lifting speed coefficient thresholds TS1 and TS2, where TS1 < TS2; When TSxs<TS 1， Then a first-level boost speed is generated; When TS1≤TSxs<TS2, the secondary lifting speed is generated; When TSxs>TS2, a three-level boost speed is generated.

Images