CN111400862A - Method, device and equipment for predicting crystal thermal field distribution - Google Patents
Method, device and equipment for predicting crystal thermal field distribution Download PDFInfo
- Publication number
- CN111400862A CN111400862A CN202010086848.6A CN202010086848A CN111400862A CN 111400862 A CN111400862 A CN 111400862A CN 202010086848 A CN202010086848 A CN 202010086848A CN 111400862 A CN111400862 A CN 111400862A
- Authority
- CN
- China
- Prior art keywords
- temperature
- temperature distribution
- field structure
- prediction model
- temperature field
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000009826 distribution Methods 0.000 title claims abstract description 218
- 239000013078 crystal Substances 0.000 title claims abstract description 75
- 238000000034 method Methods 0.000 title claims abstract description 68
- 230000008859 change Effects 0.000 claims abstract description 24
- 238000013528 artificial neural network Methods 0.000 claims abstract description 15
- 238000012549 training Methods 0.000 claims description 18
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 13
- 238000012795 verification Methods 0.000 claims description 11
- 238000009413 insulation Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 6
- 239000011664 nicotinic acid Substances 0.000 claims description 5
- 238000004088 simulation Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 15
- 238000012544 monitoring process Methods 0.000 abstract description 7
- 230000002159 abnormal effect Effects 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 16
- HBMJWWWQQXIZIP-UHFFFAOYSA-N silicon carbide Chemical compound [Si+]#[C-] HBMJWWWQQXIZIP-UHFFFAOYSA-N 0.000 description 12
- 229910010271 silicon carbide Inorganic materials 0.000 description 11
- 238000003860 storage Methods 0.000 description 9
- 230000006872 improvement Effects 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 239000000463 material Substances 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 239000002994 raw material Substances 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000009422 external insulation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000001737 promoting effect Effects 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003592 biomimetic effect Effects 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009421 internal insulation Methods 0.000 description 1
- 230000005055 memory storage Effects 0.000 description 1
- 230000005693 optoelectronics Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 229920001296 polysiloxane Polymers 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000758 substrate Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
Abstract
The application discloses a method for predicting crystal thermal field distribution, which comprises the following steps: determining the parameter value of the crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure; inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model, and predicting the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in a temperature field structure cavity. When the temperature field structure changes, the temperature distribution corresponding to the temperature field structure can be predicted through the temperature distribution prediction model, the temperature distribution corresponding to the temperature field structure is in a monitoring state, and once the temperature distribution corresponding to the temperature field structure is predicted by the temperature distribution prediction model to be in an abnormal range, warning information can be sent out to remind a manager to process in time.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a method, an apparatus, and a device for predicting crystal thermal field distribution.
Background
At present, the method capable of preparing silicon carbide single crystals in batches is a physical vapor transport method, which is a method for promoting raw materials to sublimate upwards and crystallize on the surface of seed crystals under the action of temperature gradient after gasifying the raw materials in a low-pressure environment at high temperature. Because the stacking fault energy of the silicon carbide is very low, and a plurality of crystal forms can be stabilized only by dozens of layers of structures, the number of the silicon carbide isomerous isomers exceeds 240, the crystal lattices are kept uniform and unchanged in the crystal growth process, and stable and reasonable temperature field distribution is needed, which also becomes one of the main technical bottlenecks restricting the mass production of the silicon carbide. The growth of silicon carbide single crystal needs to be at 2200 ℃ and 10 DEG C-4The method is carried out in a Pa vacuum environment, and no reliable technical means is available at present for measuring and monitoring the temperature distribution in the crystal growth process.
Disclosure of Invention
In view of this, embodiments of the present application provide a method, an apparatus, and a device for predicting crystal thermal field distribution, so as to solve the problem in the prior art that a reliable technical means is not available to measure and monitor temperature distribution in a crystal growth process.
The embodiment of the application adopts the following technical scheme:
the embodiment of the application provides a method for predicting crystal thermal field distribution, which comprises the following steps:
determining the parameter value of the crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure;
inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model, and predicting the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
Further, before inputting the pre-trained temperature distribution prediction model, the method further includes:
inputting a plurality of groups of parameter values of the temperature field structure into the temperature field bionic simulation system to obtain a plurality of groups of temperature distributions corresponding to the parameter values of the temperature field structure;
establishing an initial temperature distribution prediction model;
and training an initial temperature distribution prediction model according to the parameter values of the multiple groups of temperature field structures and the temperature distribution corresponding to the parameter value of each temperature field structure, and obtaining the temperature distribution prediction model meeting the preset conditions.
Further, after obtaining a plurality of sets of temperature distributions corresponding to the parameter values of the temperature field structure, the method further includes:
and processing the temperature distribution result according to a preset method so as to obtain the temperature distribution with uniform numerical value.
Further, the preset method is a difference method.
Further, the training of the initial temperature distribution prediction model according to the parameter values of the plurality of groups of temperature field structures and the temperature distribution corresponding to the parameter value of each temperature field structure to obtain a temperature distribution prediction model meeting preset conditions specifically includes:
constructing a data set by a plurality of groups of parameter values of the temperature field structure and temperature distribution corresponding to the parameter value of each temperature field structure, dividing the data set into a training set and a verification set according to a preset proportion, inputting the parameter value of the temperature field structure in the training set and the temperature distribution corresponding to the parameter value of each temperature field structure into the initial temperature distribution prediction model, calculating a weight relation between the parameter value of the temperature field structure and the temperature distribution corresponding to the parameter value of the temperature field structure through neurons of a hidden layer of a full-connection layer neural network, continuously carrying out iterative adjustment, taking the parameter value of the temperature field structure in the verification set as an input after obtaining the weight relation of a preset condition, predicting the temperature distribution corresponding to the parameter value of the temperature field structure according to the weight relation, comparing the temperature distribution with the temperature distribution in the verification set, and carrying out reverse adjustment on the weight relation if an error value is larger than the preset value, and obtaining the temperature score prediction model meeting the conditions until the error value is less than or equal to the preset value.
Further, after obtaining the eligible temperature score prediction model, the method further includes:
judging whether a correlation coefficient between the temperature distribution obtained by the temperature distribution prediction model and the temperature distribution in the data set is in a preset threshold value or not;
and if the correlation coefficient between the temperature distribution obtained by the temperature distribution prediction model and the temperature distribution in the data set is not in the preset threshold value, optimizing the temperature distribution prediction model by adopting a preset algorithm so as to improve the prediction precision of the temperature distribution prediction model.
Further, the optimizing the temperature distribution prediction model by using a preset algorithm specifically includes:
and adjusting one or more of the number of training steps, the training step length, the number of neurons and the number of layers of hidden layers by adopting the preset algorithm to further optimize the temperature distribution prediction model, wherein the preset algorithm is an optimized gradient descent algorithm.
Further, the temperature field structural parameters include a coil radius, a coil external insulation distance, a coil length, a crucible diameter, a crucible wall thickness, an internal insulation thickness, an external insulation thickness, a coil height and a lower insulation thickness.
The embodiment of the present application further provides a device for predicting crystal thermal field distribution, where the device includes:
the crystal temperature field structure change determining unit is used for determining a parameter value of crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure;
the prediction unit is used for inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model and predicting the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
Embodiments of the present application also provide an apparatus for predicting crystal thermal field distribution, the apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform the following:
the crystal temperature field structure change determining unit is used for determining a parameter value of crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure;
the prediction unit is used for inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model and predicting the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
The embodiment of the application adopts at least one technical scheme which can achieve the following beneficial effects: when the temperature field structure changes, the temperature distribution corresponding to the temperature field structure can be predicted through the temperature distribution prediction model, the temperature distribution corresponding to the temperature field structure is in a monitoring state, and once the temperature distribution corresponding to the temperature field structure is predicted by the temperature distribution prediction model to be in an abnormal range, warning information can be sent out to remind a manager to process in time.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the application and together with the description serve to explain the application and not to limit the application. In the drawings:
FIG. 1 is a schematic flow chart of a method for predicting thermal field distribution of a crystal according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart of a method for predicting thermal field distribution of a crystal according to a second embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a temperature field structure provided in the second embodiment of the present disclosure;
fig. 4 is a schematic diagram of temperature distribution corresponding to a temperature field structure provided in the second embodiment of the present disclosure;
FIG. 5 is a schematic structural diagram of an initial temperature distribution prediction model provided in example two of the present specification;
FIG. 6 is a schematic structural diagram of an apparatus for predicting thermal field distribution of a crystal according to a third embodiment of the present disclosure;
fig. 7 is a schematic structural diagram of an apparatus for predicting crystal thermal field distribution according to an embodiment of the present application.
Detailed Description
Silicon carbide (SiC), also known as a wide bandgap semiconductor material, is a single crystal material composed of two elements, carbon and silicon. Compared with the traditional semiconductor material, the material has more excellent physical properties, particularly excellent properties in the aspects of forbidden band width, heat conductivity, saturated electron drift rate, breakdown field strength and the like, and can realize high-efficiency work of devices in extreme environments such as high temperature, high pressure, high frequency and the like, so that the material has important application prospects in the fields of microwave communication, power electronic devices, optoelectronic devices and the like. With the rapid development of industries such as 5G communication, electric vehicles, Internet of things and direct-current flexible power transmission and transformation, the demand of silicon carbide substrate materials is continuously expanded.
At present, the method capable of preparing silicon carbide single crystals in batches is a physical vapor transport method, which is a method for promoting raw materials to sublimate upwards and crystallize on the surface of seed crystals under the action of temperature gradient after gasifying the raw materials in a low-pressure environment at high temperature. Because the stacking fault energy of the silicon carbide is very low, and a plurality of crystal forms can be stabilized only by dozens of layers of structures, the number of the silicon carbide isomerous isomers exceeds 240, the crystal lattices are kept uniform and unchanged in the crystal growth process, and stable and reasonable temperature field distribution is needed, which also becomes one of the main technical bottlenecks restricting the mass production of the silicon carbide. The growth of silicon carbide single crystal needs to be at 2200 ℃ and 10 DEG C-4Carried out in a Pa vacuum environment, at presentThere is no reliable technical means to measure and monitor the temperature distribution during the crystal growth process.
In order to solve the problem, the conventional method is to use bionic simulation software to calculate the temperature distribution in the temperature field in a high-order equation iterative operation mode, but the method is very time-consuming, and the simulation of the temperature distribution of the temperature field once after the structure of the temperature field is changed usually needs more than several hours.
In order to make the objects, technical solutions and advantages of the present application more apparent, the technical solutions of the present application will be described in detail and completely with reference to the following specific embodiments of the present application and the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flowchart of a method for predicting thermal field distribution of a crystal according to an embodiment of the present disclosure.
The embodiment of the present specification may be implemented by a prediction system of a temperature distribution, and specifically includes:
and step S101, determining the parameter value of the crystal temperature field structure change.
Specifically, the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure.
And S102, inputting the parameter values of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model, and predicting the temperature distribution corresponding to the temperature field structure.
Specifically, the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
When the temperature field structure changes, the temperature distribution corresponding to the temperature field structure can be predicted through the temperature distribution prediction model, the temperature distribution corresponding to the temperature field structure is in a monitoring state, and once the temperature distribution corresponding to the temperature field structure is predicted by the temperature distribution prediction model to be in an abnormal range, warning information can be sent out to remind a manager to process in time.
In correspondence with the above embodiments, fig. 2 is a schematic flow chart of a method for predicting crystal thermal field distribution according to a second embodiment of the present disclosure.
The embodiment of the present specification may be implemented by a prediction system of a temperature distribution, and specifically includes:
step S201, a plurality of groups of temperature distribution corresponding to the parameter values of the temperature field structure are obtained by inputting the parameter values of the temperature field structure into the temperature field bionic simulation system.
In step S201 of the embodiment of the present specification, 200 temperature field structures that are randomly combined may be selected and respectively input into the biomimetic simulation software for calculation, so as to obtain the temperature distribution corresponding to each temperature field structure. Referring to fig. 3, the temperature field structure parameters include a coil radius 1, a coil outer insulation distance 2, a coil length 3, a crucible diameter 4, a crucible wall thickness 5, an inner insulation thickness 6, an outer insulation thickness 7, a coil height 8, and a lower insulation thickness 9, and the temperature distribution is the temperature distribution inside the temperature field structure 10, which is shown in fig. 4 as a schematic temperature distribution diagram corresponding to the temperature field structure.
Further, in step S201 of the embodiment of the present specification, the temperature distribution needs to be processed according to a preset method, so as to obtain the temperature distribution with uniform values, for example, the temperature distribution corresponding to 200 temperature field structures and each group of temperature field structures may be subjected to data interpolation processing, and since the value number of each temperature range may be different when the temperature distribution corresponding to the temperature field structures is obtained, the temperature value number of each range may be the same through a difference method.
Step S202, an initial temperature distribution prediction model is established.
In step S202 of the embodiment of the present specification, referring to fig. 5, an initial temperature distribution prediction model including a 3 × 64 fully-connected layer neural network may be established. The input part is temperature field structure parameters, the processing part is a fully-connected neural network hidden layer, and the output part is a prediction result of temperature distribution.
Step S203, according to the parameter values of the plurality of groups of temperature field structures and the temperature corresponding to the parameter value of each temperature field structure, distributing and training an initial temperature distribution prediction model to obtain a temperature distribution prediction model meeting preset conditions.
In step S203 in this embodiment of the present description, specifically, a data set is constructed by using a plurality of sets of parameter values of temperature field structures and a temperature distribution corresponding to the parameter value of each temperature field structure, and the data set is divided into a training set and a verification set according to a preset ratio, for example, the ratio of the training set to the verification set may be 8: 2;
inputting the parameter values of the temperature field structures in the training set and the temperature distribution corresponding to the parameter values of each temperature field structure into the initial temperature distribution prediction model, calculating the weight relationship between the parameter values of the temperature field structures and the temperature distribution corresponding to the parameter values of the temperature field structures through neurons of a hidden layer of a neural network of a full connection layer, continuously carrying out iterative adjustment, taking the parameter values of the temperature field structures in the verification set as input after obtaining the weight relationship of preset conditions, predicting the temperature distribution (predicted value) corresponding to the parameter values of the temperature field structures according to the weight relationship, comparing the temperature distribution (actual value calculated by bionic simulation software) in the verification set with the temperature distribution in the verification set, observing the reduction condition of the temperature distribution to measure the convergence condition of each pre-selected temperature distribution prediction model, and carrying out reverse adjustment on the weight relationship if the error value is greater than the preset value, and obtaining the temperature score prediction model meeting the conditions until the error value is less than or equal to the preset value. The weight relationship of the preset conditions may be preset, and the set weight relationship may be adjusted according to different situations. The preset value can also be set according to different situations, for example, the preset value can be set to 0.95.
Judging whether the correlation coefficient between the temperature distribution obtained by the temperature distribution prediction model and the temperature distribution in the data set is in a preset threshold value or not, wherein the calculation formula of the correlation coefficient can be as follows:
the preset threshold value can be 0.998-1;
and if the correlation coefficient between the temperature distribution obtained by the temperature distribution prediction model and the temperature distribution in the data set is not in the preset threshold value, optimizing the temperature distribution prediction model by adopting a preset algorithm so as to improve the prediction precision of the temperature distribution prediction model.
Specifically, when the temperature distribution prediction model is optimized, one or more of the structure parameters such as the number of training steps, the training step, the number of neurons, the number of layers of hidden layers, and the batch _ size (number of samples) may be adjusted, and then the temperature distribution prediction model is optimized, for example, the value is adjusted from 3 × 64 to 4 × 128, and finally, a network of 4 × 128+ epoch ═ 10000+ batch _ size ═ 256 is obtained, and this structure may obtain a better convergence effect. The preset algorithm may be an optimized gradient descent algorithm.
And step S204, determining the parameter value of the crystal temperature field structure change.
Specifically, the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure.
And S205, inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model, and predicting the temperature distribution corresponding to the temperature field structure.
Specifically, the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
When the temperature field structure changes, the temperature distribution corresponding to the temperature field structure can be predicted through the temperature distribution prediction model, the temperature distribution corresponding to the temperature field structure is in a monitoring state, and once the temperature distribution corresponding to the temperature field structure is predicted by the temperature distribution prediction model to be in an abnormal range, warning information can be sent out to remind a manager to process in time.
Corresponding to the foregoing embodiment, fig. 6 is a schematic structural diagram of an apparatus for predicting crystal thermal field distribution provided in the third embodiment of the present disclosure, specifically including: determining unit 1 and predicting unit 2.
The determining unit 1 is used for determining a parameter value of crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure.
The prediction unit 2 is used for inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model to predict the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
When the temperature field structure changes, the temperature distribution corresponding to the temperature field structure can be predicted through the temperature distribution prediction model, the temperature distribution corresponding to the temperature field structure is in a monitoring state, and once the temperature distribution corresponding to the temperature field structure is predicted by the temperature distribution prediction model to be in an abnormal range, warning information can be sent out to remind a manager to process in time.
Fig. 7 is a schematic structural diagram of an apparatus for predicting crystal thermal field distribution according to an embodiment of the present application, where the apparatus includes a memory for storing computer program instructions and a processor for executing the program instructions, and when the computer program instructions are executed by the processor, the apparatus for triggering the apparatus to execute includes a determining unit 3 and a predicting unit 4.
The determination unit 3 is used for determining the parameter value of the crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle.
The prediction unit 4 is used for inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model to predict the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
When the temperature field structure changes, the temperature distribution corresponding to the temperature field structure can be predicted through the temperature distribution prediction model, the temperature distribution corresponding to the temperature field structure is in a monitoring state, and once the temperature distribution corresponding to the temperature field structure is predicted by the temperature distribution prediction model to be in an abnormal range, warning information can be sent out to remind a manager to process in time.
In the 90 th generation of 20 th century, it is obvious that improvements in Hardware (for example, improvements in Circuit structures such as diodes, transistors and switches) or software (for improvement in method flow) can be distinguished for a technical improvement, however, as technology develops, many of the improvements in method flow today can be regarded as direct improvements in Hardware Circuit structures, designers almost all obtain corresponding Hardware Circuit structures by Programming the improved method flow into Hardware circuits, and therefore, it cannot be said that an improvement in method flow cannot be realized by Hardware entity modules, for example, Programmable logic devices (Programmable logic devices L organic devices, P L D) (for example, Field Programmable Gate Arrays (FPGAs) are integrated circuits whose logic functions are determined by user Programming of devices), and a digital system is "integrated" on a P L D "by self Programming of designers without requiring many kinds of integrated circuits manufactured and manufactured by special chip manufacturers to design and manufacture, and only a Hardware software is written in Hardware programs such as Hardware programs, software programs, such as Hardware programs, software, Hardware programs, software programs, Hardware programs, software, Hardware programs, software, Hardware programs, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software, Hardware, software.
A controller may be implemented in any suitable manner, e.g., in the form of, for example, a microprocessor or processor and a computer readable medium storing computer readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, Application Specific Integrated Circuits (ASICs), programmable logic controllers (PLC's) and embedded microcontrollers, examples of which include, but are not limited to, microcontrollers 625D, Atmel AT91SAM, Microchip PIC18F26K20 and Silicone L abs C8051F320, which may also be implemented as part of the control logic of a memory.
The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.
For convenience of description, the above devices are described as being divided into various units by function, and are described separately. Of course, the functionality of the units may be implemented in one or more software and/or hardware when implementing the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium which can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The application may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The application may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above description is only an example of the present application and is not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (10)
1. A method of predicting thermal field distribution of a crystal, the method comprising:
determining the parameter value of the crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure;
inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model, and predicting the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
2. The method of predicting crystal thermal field distribution of claim 1, wherein before inputting the pre-trained temperature distribution prediction model, the method further comprises:
inputting a plurality of groups of parameter values of the temperature field structure into the temperature field bionic simulation system to obtain a plurality of groups of temperature distributions corresponding to the parameter values of the temperature field structure;
establishing an initial temperature distribution prediction model;
and training an initial temperature distribution prediction model according to the parameter values of the multiple groups of temperature field structures and the temperature distribution corresponding to the parameter value of each temperature field structure, and obtaining the temperature distribution prediction model meeting the preset conditions.
3. The method of predicting a crystal thermal field distribution of claim 2, wherein after obtaining a plurality of sets of temperature distributions corresponding to parameter values of the temperature field structure, the method further comprises:
and processing the temperature distribution result according to a preset method so as to obtain the temperature distribution with uniform numerical value.
4. The method of predicting the thermal field distribution of a crystal according to claim 3, wherein the predetermined method is a difference method.
5. The method for predicting crystal thermal field distribution according to claim 2, wherein the training of an initial temperature distribution prediction model according to the parameter values of the plurality of groups of temperature field structures and the temperature distribution corresponding to the parameter value of each temperature field structure to obtain a temperature distribution prediction model meeting preset conditions specifically comprises:
constructing a data set by a plurality of groups of parameter values of the temperature field structure and temperature distribution corresponding to the parameter value of each temperature field structure, dividing the data set into a training set and a verification set according to a preset proportion, inputting the parameter value of the temperature field structure in the training set and the temperature distribution corresponding to the parameter value of each temperature field structure into the initial temperature distribution prediction model, calculating a weight relation between the parameter value of the temperature field structure and the temperature distribution corresponding to the parameter value of the temperature field structure through neurons of a hidden layer of a full-connection layer neural network, continuously carrying out iterative adjustment, taking the parameter value of the temperature field structure in the verification set as an input after obtaining the weight relation of a preset condition, predicting the temperature distribution corresponding to the parameter value of the temperature field structure according to the weight relation, comparing the temperature distribution with the temperature distribution in the verification set, and carrying out reverse adjustment on the weight relation if an error value is larger than the preset value, and obtaining the temperature score prediction model meeting the conditions until the error value is less than or equal to the preset value.
6. The method of predicting a crystal thermal field distribution of claim 5, wherein after the deriving the eligible temperature component prediction model, the method further comprises:
judging whether a correlation coefficient between the temperature distribution obtained by the temperature distribution prediction model and the temperature distribution in the data set is in a preset threshold value or not;
and if the correlation coefficient between the temperature distribution obtained by the temperature distribution prediction model and the temperature distribution in the data set is not in the preset threshold value, optimizing the temperature distribution prediction model by adopting a preset algorithm so as to improve the prediction precision of the temperature distribution prediction model.
7. The method for predicting the crystal thermal field distribution according to claim 6, wherein the optimizing the temperature distribution prediction model by using a preset algorithm specifically comprises:
and adjusting one or more of the number of training steps, the training step length, the number of neurons and the number of layers of hidden layers by adopting the preset algorithm to further optimize the temperature distribution prediction model, wherein the preset algorithm is an optimized gradient descent algorithm.
8. A method for predicting crystal thermal field distribution according to claim 1, wherein the thermal field structure parameters comprise coil radius, coil outer insulation distance, coil length, crucible diameter, crucible wall thickness, inner insulation thickness, outer insulation thickness, coil height and lower insulation thickness.
9. An apparatus for predicting thermal field distribution of a crystal, the apparatus comprising:
the crystal temperature field structure change determining unit is used for determining a parameter value of crystal temperature field structure change, wherein the temperature field structure is a cavity structure, and a crystal is arranged in the middle of the temperature field structure;
the prediction unit is used for inputting the parameter value of the temperature field structure change data of the crystal into a pre-trained temperature distribution prediction model and predicting the temperature distribution corresponding to the temperature field structure; the temperature distribution prediction model comprises a temperature field structure parameter input layer, a full connection layer neural network hidden layer and a prediction output layer of temperature distribution, wherein the temperature distribution is the temperature distribution in the temperature field structure cavity.
10. An apparatus for predicting thermal field distribution of a crystal, the apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform the apparatus of claim 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010086848.6A CN111400862B (en) | 2020-02-11 | 2020-02-11 | Method, device and equipment for predicting thermal field distribution of crystal |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010086848.6A CN111400862B (en) | 2020-02-11 | 2020-02-11 | Method, device and equipment for predicting thermal field distribution of crystal |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111400862A true CN111400862A (en) | 2020-07-10 |
CN111400862B CN111400862B (en) | 2024-02-23 |
Family
ID=71430304
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010086848.6A Active CN111400862B (en) | 2020-02-11 | 2020-02-11 | Method, device and equipment for predicting thermal field distribution of crystal |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111400862B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116776135A (en) * | 2023-08-24 | 2023-09-19 | 之江实验室 | Physical field data prediction method and device based on neural network model |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107868979A (en) * | 2017-08-31 | 2018-04-03 | 西安理工大学 | A kind of silicon single crystal diameter control method based on permanent casting speed control structure |
CN108754599A (en) * | 2018-05-31 | 2018-11-06 | 西安理工大学 | A kind of silicon monocrystal growth temprature control method based on finite element numerical simulation |
-
2020
- 2020-02-11 CN CN202010086848.6A patent/CN111400862B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107868979A (en) * | 2017-08-31 | 2018-04-03 | 西安理工大学 | A kind of silicon single crystal diameter control method based on permanent casting speed control structure |
CN108754599A (en) * | 2018-05-31 | 2018-11-06 | 西安理工大学 | A kind of silicon monocrystal growth temprature control method based on finite element numerical simulation |
Non-Patent Citations (3)
Title |
---|
于强;俞号峰;刘威;: "空间晶体生长炉综合控制方法" * |
王可: "直拉单晶炉热场模型参数辨识及模型预测控制" * |
郭耿涛: "晶体生长炉PID神经网络温度控制研究" * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116776135A (en) * | 2023-08-24 | 2023-09-19 | 之江实验室 | Physical field data prediction method and device based on neural network model |
CN116776135B (en) * | 2023-08-24 | 2023-12-19 | 之江实验室 | Physical field data prediction method and device based on neural network model |
Also Published As
Publication number | Publication date |
---|---|
CN111400862B (en) | 2024-02-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111400862A (en) | Method, device and equipment for predicting crystal thermal field distribution | |
KR20130139783A (en) | Production management apparatus and production management system for electronic device, and storage medium | |
CN114298091A (en) | SF6Gas flow measuring value correction method, device, equipment and storage medium | |
CN112749810A (en) | Model training method, control parameter determination method and device | |
CN113410126B (en) | Method and system for automatically regulating thickness of silicon oxide film in thermal oxidation process | |
Patra et al. | Artificial neural network‐based modelling of compensated multi‐crystalline solar‐grade silicon under wide temperature variations | |
CN110660453B (en) | Parallel computing method for solving rate theoretical equation based on exponential time difference format | |
CN111539556A (en) | Wafer scheduling method and device | |
CN115128948A (en) | RST Smith predictor | |
CN112800615A (en) | Method for predicting residual life of silicon foam material | |
CN113718329B (en) | Semiconductor device and process control method thereof | |
Latyshev et al. | Relationship between the north–south asymmetry of sunspot formation and the amplitude of 11-year solar activity cycles | |
CN113868916B (en) | LFVPSO-BPNN-based multi-loop groove cabling temperature rise prediction method | |
Jen et al. | Combining on-line experiment and process control methods for changes in a dynamic model | |
CN117350517B (en) | Control method, system, equipment and storage medium for chemical production flow | |
CN116913815B (en) | Control method, device and storage medium for high-temperature CVD production process | |
US20230051330A1 (en) | Using defect models to estimate defect risk and optimize process recipes | |
US9213322B1 (en) | Methods for providing run to run process control using a dynamic tuner | |
Gong et al. | Performance analysis of single EWMA controller subject to metrology delay under dynamic models | |
CN113326584B (en) | Electrical equipment optimization design method taking robustness and reliability into consideration | |
Smith et al. | Simulated annealing of constrained statistical functions | |
Biner et al. | Phase-field crystal modeling of material behavior | |
US20150364533A1 (en) | Methods of manufacturing polyresistors with selected tcr | |
CN116316564A (en) | Wind power prediction method, processor, system and storage medium | |
CN113762526A (en) | Federal learning method, hierarchical network system, storage medium and electronic equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: No.99, Tianyue South Road, Huaiyin District, Jinan City, Shandong Province Applicant after: Shandong Tianyue advanced technology Co.,Ltd. Address before: No.99, Tianyue South Road, Huaiyin District, Jinan City, Shandong Province Applicant before: SICC Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |