High-flux heat treatment system and method

Abstract

The invention provides a high-flux heat treatment system and method, and relates to the technical field of heat treatment, and the system comprises a heat treatment device, a control panel and an analysis processing device; the heat treatment device comprises a controller, a honeycomb structure storage area and a plurality of gradient temperature areas, the honeycomb structure storage area comprises a plurality of sample storage honeycomb holes and a plurality of air channels, each sample storage honeycomb hole is provided with the multiple air channels in a circle, and the multiple sample storage honeycomb holes are distributed in the different gradient temperature areas; a honeycomb structure of sample placing honeycomb holes and an air duct is constructed, the air duct has a chimney effect and is provided with a plurality of gradient temperature zones with different temperatures, and position information and heating information of the sample placing honeycomb holes are input into an analysis processing device to generate a heating control instruction for heating a sample; the control panel controls the temperature of the gradient temperature area through the heating control instruction to heat the heating samples, multiple heating samples can be heated at the same time, high-flux heating is formed, the product heating efficiency can be optimized, and the production cycle can be shortened.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heat treatment, in particular to a high-throughput heat treatment system and method. BACKGROUND

[0002] Heat treatment is an important link in the process of material research and development, through which the internal structure of materials can be changed to improve their process performance and use performance. Traditional material research and development often adopts the "trial and error method", which requires multiple cycles to obtain a heat treatment process and materials that meet the requirements, resulting in a huge amount of experiments and low research and development efficiency.

[0003] High-throughput experiments can complete the preparation and characterization of a large number of samples at the same time in a short time, thereby effectively improving the efficiency of material research and development, and gradually achieving the goal of "rational design" of materials. In the prior art, microwave energy field heating can be used to rapidly prepare small-size bulk combination materials of multiple components at the same temperature field at one time; however, due to the limitation of heating methods, the sample size is limited. Alternatively, multiple independent heaters can be used to achieve high-throughput processing of samples in three or more heating zones. The above methods have very high performance requirements for the interval heat materials between different temperature zones, and the cost of multiple heating systems is high, which cannot effectively utilize thermal energy. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a high-throughput heat treatment system and method to overcome the shortcomings of the prior art.

[0005] The technical solution for solving the above technical problem is as follows: a high-throughput heat treatment system, comprising a heat treatment device, a control panel and an analysis processing device; the heat treatment device comprises a controller, a honeycomb structure placement area and multiple gradient temperature zones, the honeycomb structure placement area comprises multiple sample placement honeycomb holes and multiple air ducts, each sample placement honeycomb hole is surrounded by multiple air ducts, and multiple sample placement honeycomb holes are distributed in different gradient temperature zones.

[0006] The control panel is used to input at least one sample placement honeycomb hole position information and heating information of a heated sample;

[0007] The analysis processing device is used to input the sample placement honeycomb hole position information and the heating information into a pre-trained multilayer perception model, generate a heating control instruction corresponding to the heated sample, and send the heating control instruction to the controller of the heat treatment device through the control panel;

[0008] The controller of the heat treatment device is used to control the temperature of the gradient temperature zone corresponding to the heated sample and the flow rate of the fluid in the air duct according to the heating control instruction.

[0009] The beneficial effects of the present application are: the honeycomb structure of the sample placement honeycomb hole and the air duct is constructed, the air duct has a chimney effect, and a plurality of temperature gradient temperature zones with different temperatures are arranged, the position information and the heating information of the sample placement honeycomb hole are input into the analysis processing device to generate a heating control instruction for heating the sample, and the control panel controls the temperature of the gradient temperature zone to heat the sample according to the heating control instruction, so that a plurality of samples can be heated at the same time, high-throughput heating is formed, the product heating efficiency can be optimized, and the production cycle can be shortened.

[0010] Based on the above technical solutions, the present application can be further improved as follows.

[0011] Further, the heat treatment device further comprises a plurality of temperature measuring thermocouples; the plurality of temperature measuring thermocouples are respectively installed in each sample placement honeycomb hole;

[0012] The temperature measuring thermocouples are used to collect the temperature of the heated sample in the corresponding sample placement honeycomb hole, and send the temperature of the heated sample to the analysis processing device through the control panel;

[0013] The analysis processing device is further used to input the temperature of the heated sample into the multi-layer perception model, generate a new heating control instruction corresponding to the heated sample, and send the new heating control instruction to the controller of the heat treatment device through the control panel;

[0014] The controller of the heat treatment device is further used to adjust the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the new heating control instruction.

[0015] Further, the analysis processing device is further used to judge whether the temperature of the heated sample meets a preset furnace discharge temperature value when the temperature of the heated sample is received, and if so, generate a furnace discharge instruction corresponding to the sample placement honeycomb hole of the heated sample, and send the furnace discharge instruction to the controller of the heat treatment device through the control panel;

[0016] The controller of the heat treatment device is further used to control the gradient temperature zone corresponding to the heated sample to stop heating according to the furnace discharge instruction.

[0017] Further, the high-throughput heat treatment system further comprises a model construction module, and the model construction module comprises a data set construction unit, a training unit and an optimization unit;

[0018] The data set construction unit is configured to import the gradient temperature zone temperature, the sample placement honeycomb hole position, the fluid type in the air duct, the inlet gas temperature, the outlet temperature and flow rate, the air duct height and size, the heating temperature, the air duct cross-sectional area size, and the inlet gas flow as basic data, construct an initial data set, set multiple temperature sections at a set interval temperature, and set multiple flow sizes at a set air duct opening degree, perform data augmentation on the initial data set through the multiple temperature sections and the multiple flow sizes, and obtain a training data set.

[0019] The training unit is configured to construct an initial multi-layer perception model, use the gradient temperature zone temperature, the sample placement honeycomb hole position, the fluid type in the air duct, the inlet gas temperature, the outlet temperature and flow rate, the air duct height and size in the training data set as input, and use the heating temperature, the air duct cross-sectional area size, and the inlet gas flow as output to preliminarily train the initial multi-layer perception model.

[0020] The optimization unit is configured to randomly extract a set proportion of data from the training data set, and retrain the preliminarily trained multi-layer perception model through the extracted data.

[0021] The data set construction unit is further configured to reconstruct a new training data set.

[0022] The optimization unit is further configured to randomly extract a set proportion of data from the new training data set, train the retrained multi-layer perception model through the reextracted data, and iteratively iterate until the heating temperature control error is within a set error range, to obtain a pre-trained multi-layer perception model.

[0023] Further, the sample placement honeycomb hole and the air duct are both hexagonal cylindrical structures, and one sample placement honeycomb hole is surrounded by six air ducts to form a honeycomb structure.

[0024] Further, the inner wall of the sample placement honeycomb hole is surrounded by a cylindrical high-temperature-resistant insulation material, and the middle of the sample placement honeycomb hole forms a cavity for placing a heated sample.

[0025] Further, the heights of the multiple air ducts change in a gradient, and each air duct is provided with a one-way valve connected to a controller of the heat treatment device to receive a control signal of the controller and control the flow rate of the fluid in the air duct according to the control signal.

[0026] Another technical solution of the present application to solve the above technical problems is as follows: a high-throughput heat treatment method applied to a high-throughput heat treatment device, wherein the high-throughput heat treatment device comprises a heat treatment device, a control panel and an analysis processing device; the heat treatment device comprises a controller, a honeycomb structure sample placement area and a plurality of gradient temperature zones, the honeycomb structure sample placement area comprises a plurality of sample placement honeycomb holes and a plurality of air ducts, each of the sample placement honeycomb holes is surrounded by a plurality of air ducts, and the plurality of sample placement honeycomb holes are distributed in different gradient temperature zones;

[0027] The control panel inputs at least one sample placement honeycomb hole position information and heating information of a heated sample;

[0028] The analysis processing device inputs the sample placement honeycomb hole position information and the heating information into a pre-trained multi-layer perception model to generate a heating control instruction corresponding to the heated sample, and sends the heating control instruction to the controller of the heat treatment device through the control panel;

[0029] The controller of the heat treatment device controls the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the heating control instruction.

[0030] Further, the method further comprises the steps of:

[0031] The temperature measuring thermocouple collects the temperature of the heated sample in the corresponding sample placement honeycomb hole and sends the temperature of the heated sample to the analysis processing device through the control panel;

[0032] The analysis processing device inputs the temperature of the heated sample into the multi-layer perception model to generate a new heating control instruction corresponding to the heated sample, and sends the new heating control instruction to the controller of the heat treatment device through the control panel;

[0033] The controller of the heat treatment device adjusts the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the new heating control instruction.

[0034] Further, the method further comprises the step of constructing a pre-trained multi-layer perception model:

[0035] The gradient temperature zone temperature, sample placement honeycomb hole position, fluid type in the air duct, inlet gas temperature, outlet temperature and flow rate, air duct height and size, heating temperature, air duct cross-sectional area size and gas flow are imported as basic data to construct an initial data set, a plurality of temperature sections are set at a set interval temperature, a plurality of flow rates are set at a set air duct opening degree, the initial data set is data augmented through the plurality of temperature sections and the plurality of flow rates, and a training data set is obtained;

[0036] An initial multilayer sensing model is constructed, using the temperature of the gradient temperature zone in the training dataset, the location of the sampled honeycomb holes, the fluid type in the duct, the gas temperature at the inlet, the temperature and velocity at the outlet, and the height and size of the duct as inputs, and the heating temperature, the cross-sectional area of ​​the duct, and the flow rate of the introduced gas as outputs to initially train the initial multilayer sensing model.

[0037] A set proportion of data is randomly extracted from the training dataset, and the pre-trained multilayer perception model is retrained using the extracted data.

[0038] Rebuild the new training dataset;

[0039] A set proportion of data is randomly extracted from the new training dataset, and the retrained multilayer sensing model is trained using the extracted data. This process is repeated until the temperature control error of the heating temperature is within a set error range, thus obtaining a pre-trained multilayer sensing model. Attached Figure Description

[0040] Figure 1 This is a functional block diagram of a high-throughput heat treatment system provided in an embodiment of the present invention;

[0041] Figure 2 This is a schematic diagram of the structure of the heat treatment apparatus provided in an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of the honeycomb structure for sample placement provided in an embodiment of the present invention;

[0043] Figure 4 This is a schematic diagram showing the distribution of all air ducts provided in the embodiments of the present invention;

[0044] Figure 5 This is a schematic diagram of the air duct structure provided in an embodiment of the present invention;

[0045] Figure 6 This is a schematic flowchart illustrating a high-throughput heat treatment method provided in an embodiment of the present invention.

[0046] In the attached diagram, the component names represented by each label are as follows:

[0047] 1. Heat treatment device; 2. Control panel; 3. Analysis and processing device; 11. Honeycomb structure storage area; 12. Gradient temperature zone; 111. Sample placement honeycomb holes; 112. Air duct; 1112. High temperature resistant insulation material; 1113. Cavity; 1121. One-way valve. Detailed Implementation

[0048] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0049] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application. In the description of the present application, unless otherwise specified, " / " represents a "or" relationship between the objects before and after the " / ", for example, A / B can represent A or B; "and / or" in the present application is only a description of the relationship between the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural.

[0050] In addition, in the description of the present application, unless otherwise specified, "multiple" means two or more than two. "At least one of the following" or the like means any combination of the items, including any combination of single item or multiple items. For example, at least one of a, b, or c can represent: a, b, c, a-b, a-c, b-c, or a-b-c, where a, b, and c can be single or multiple.

[0051] Meanwhile, in the embodiments of the present application, the words "exemplary" or "for example" are used to represent as an example, illustration, or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. Rather, the words "exemplary" or "for example" are used to present the relevant concept in a specific manner, for ease of understanding.

[0052] As shown in Figure 1 , Figure 2 The embodiment of the present application provides a high-throughput heat treatment system, which comprises a heat treatment device 1, a control panel 2 and an analysis processing device 3; the heat treatment device 1 comprises a controller, a honeycomb structure sample placement area 11 and a plurality of gradient temperature areas 12, the honeycomb structure sample placement area 11 comprises a plurality of sample placement honeycomb holes 111 and a plurality of air ducts 112, each of the sample placement honeycomb holes 111 is surrounded by a plurality of air ducts 112, and a plurality of sample placement honeycomb holes 111 are distributed in different gradient temperature areas 12.

[0053] The control panel 2 is used to input at least one sample placement honeycomb hole 111 position information and heating information of a heated sample;

[0054] The analysis processing device 3 is used to input the sample placement honeycomb hole 111 position information and the heating information into a pre-trained multi-layer perception model, generate a heating control instruction corresponding to the heated sample, and send the heating control instruction to the controller of the heat treatment device 1 through the control panel 2;

[0055] The controller of the heat treatment device 1 is configured to control the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct 112 according to the heating control instruction.

[0056] Specifically, nine gradient temperature zones 12 are arranged from top to bottom in the honeycomb structure sample placement area 11.

[0057] In the above embodiment, the honeycomb structure is constructed by placing sample honeycomb holes and air ducts, the air duct has a chimney effect, and is provided with a plurality of gradient temperature zones with different temperatures, and the sample placement honeycomb hole position information and heating information are input into the analysis processing device to generate a heating control instruction for heating the sample, and the control panel controls the temperature of the gradient temperature zone to heat the heated sample through the heating control instruction, which can simultaneously heat multiple heated samples, form high-throughput heating, optimize product heating efficiency, and shorten the production cycle.

[0058] Preferably, the heat treatment device 1 further comprises a plurality of temperature measuring thermocouples; and the plurality of temperature measuring thermocouples are respectively installed in each sample placement honeycomb hole 111.

[0059] The temperature measuring thermocouples are configured to collect the temperature of the heated sample in the sample placement honeycomb hole 111 and send the temperature of the heated sample to the analysis processing device 3 through the control panel 2.

[0060] The analysis processing device 3 is further configured to input the temperature of the heated sample into the multilayer perception model, generate a new heating control instruction corresponding to the heated sample, and send the new heating control instruction to the controller of the heat treatment device 1 through the control panel 2.

[0061] The controller of the heat treatment device 1 is further configured to adjust the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct 112 according to the new heating control instruction.

[0062] Specifically, each temperature measuring thermocouple is connected to the control panel through a thermocouple compensation lead.

[0063] In the above embodiment, the temperature of the heated sample in the sample placement honeycomb hole 111 is collected by the temperature measuring thermocouple, and the temperature is input as feedback information into the multilayer perception model, and the multilayer perception model generates a new heating control instruction to adjust the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct 112, which can realize intelligent gradient temperature regulation.

[0064] Preferably, the analysis processing device 3 is further configured to receive the temperature of the heated sample, determine whether the temperature of the heated sample meets a preset furnace-out temperature value, generate a furnace-out instruction corresponding to the sample-holding cell hole 111 of the heated sample if the temperature meets the preset furnace-out temperature value, and send the furnace-out instruction to a controller of the heat treatment device 1 through the control panel 2.

[0065] The controller of the heat treatment device 1 is further configured to control the gradient temperature zone corresponding to the heated sample to stop heating according to the furnace-out instruction.

[0066] In the above embodiment, the temperature of the heated sample collected by the temperature measuring thermocouple is used to determine whether the heating is completed, and the entire automatic heating process is realized.

[0067] Preferably, the high-throughput heat treatment system further comprises a model construction module, and the model construction module comprises a data set construction unit, a training unit and an optimization unit.

[0068] The data set construction unit is configured to import the gradient temperature zone temperature, the sample-holding cell hole position, the fluid type in the air duct, the gas temperature at the air inlet, the temperature and flow rate at the air outlet, the air duct height and size, the heating temperature, the air duct cross-sectional area size and the gas flow rate as basic data, construct an initial data set, set a plurality of temperature segments at a set interval temperature, set a plurality of flow rates at a set air duct opening degree, perform data augmentation on the initial data set through the plurality of temperature segments and the plurality of flow rates, and obtain a training data set; specifically, in a controlled temperature range, 50℃ is taken as an interval (for example, 400-1200℃ interval, 12 temperature segments), in the case of air duct opening degrees of 100% and 50%, and in the case of gas (for example, nitrogen) flow rates of 100%, 50% and 0 of the maximum flow rate, 72 groups of series data about the heating temperature, the test sample zone temperature, the gas temperature at the air inlet, the temperature and flow rate at the air outlet, the air duct height and size, etc. are obtained through orthogonal test.

[0069] Then, the training unit is configured to build an initial multilayer perception model, and the initial multilayer perception model is preliminarily trained with the gradient temperature zone temperature, the sample placement honeycomb hole position, the fluid type in the air duct, the gas temperature at the air inlet, the outlet temperature and flow rate, the air duct height and size in the training data set as input and the heating temperature, the air duct cross-sectional area size and the gas flow as output. Specifically, based on the obtained 72 groups of data, based on the classic multilayer perception model (Multilayer Perceptron, MLP model), the basic data set is trained, the heating temperature, the gradient temperature zone temperature, the sample placement honeycomb hole position, the fluid type in the air duct, the gas temperature at the air inlet, the outlet temperature and flow rate, the air duct height and size are input, and the air duct cross-sectional area size and the gas flow are output. The X-MLP model for pre-training is established by adjusting the parameters. Based on the established X-MLP model, 320 groups of serial data about the heating temperature, the test sample zone temperature, the gas temperature at the air inlet, the outlet temperature and flow rate, the air duct height and size, etc. are automatically generated at intervals of 10℃ (such as 400-1200℃ interval, 80 temperature segments), respectively in the case of 75%, 25% air duct opening degree, and the case of 75%, 25% gas (such as nitrogen) flow.

[0070] Then, the optimization unit is configured to randomly extract a set proportion of data from the training data set, and re-train the preliminarily trained multilayer perception model by the extracted data.

[0071] The data set construction unit is further configured to reconstruct a new training data set.

[0072] The optimization unit is further configured to randomly extract a set proportion of data from the new training data set, and train the re-trained multilayer perception model by the re-extracted data, and iteratively iterate until the heating temperature control error is within a set error range, to obtain a pre-trained multilayer perception model.

[0073] Specifically, 5% of the data is randomly extracted from the 320 groups of data, and the X-MLP model is corrected through experimental verification. Then, the 320 groups of data are regenerated according to the above method, 5% of the data is randomly extracted, and the SX-MLP model (i.e. the pre-trained multilayer perception model) is established through experimental verification and repeated iteration until the test sample zone temperature control error is controlled within 0.5℃.

[0074] Based on the established SX-MLP model, the mapping relationship between the model input parameters (gradient temperature zone temperature, sample placement honeycomb hole position, fluid type in air duct, gas temperature at air inlet, outlet temperature and flow rate, air duct height and size) and the model output parameters (heating temperature, air duct cross-sectional area size, gas flow) is established, so as to realize intelligent gradient temperature control.

[0075] like Figure 2 As shown, preferably, both the sample placement honeycomb hole 111 and the air duct 112 are hexagonal cylindrical structures, and six air ducts 112 are arranged around one of the sample placement honeycomb holes 111 to form a honeycomb structure.

[0076] In the above embodiment, the air duct 112, which surrounds the sample, can uniformly heat the sample placed in the honeycomb hole 111.

[0077] like Figure 3 As shown, preferably, the inner wall of the sample placement honeycomb hole 111 is surrounded by a cylindrical high-temperature resistant insulating material 1112, and a cavity 1113 for placing heated samples is formed in the middle of the sample placement honeycomb hole 111.

[0078] In the above embodiments, it is possible to ensure that the sample is heated uniformly.

[0079] like Figure 4 , Figure 5 As shown, preferably, the heights of the plurality of air ducts 112 vary in a gradient, and each air duct 112 is equipped with a one-way valve 1121. The one-way valve is used to connect to the controller of the heat treatment device 1 and receive the control signal of the controller to control the fluid flow rate in the air duct 112.

[0080] In the above embodiments, the fluid velocity within the air duct is controlled by automatically adjusting the cross-sectional size of the air duct.

[0081] like Figure 6 As shown, this embodiment of the invention provides a high-throughput heat treatment method applied to a high-throughput heat treatment apparatus. The high-throughput heat treatment apparatus includes a heat treatment device, a control panel, and an analysis and processing device. The heat treatment device includes a controller, a honeycomb structure placement area, and multiple gradient temperature zones. The honeycomb structure placement area includes multiple sample placement honeycomb holes and multiple air ducts. Each sample placement honeycomb hole has multiple air ducts arranged around its perimeter, and the multiple sample placement honeycomb holes are distributed within different gradient temperature zones. The method includes:

[0082] The control panel should input the location information of at least one sample placement honeycomb hole containing a heated sample and the heating information.

[0083] The analysis and processing device inputs the sample honeycomb hole position information and the heating information into the pre-trained multilayer sensing model, generates a heating control command corresponding to the heated sample, and sends the heating control command to the controller of the heat treatment device through the control panel;

[0084] The controller of the heat treatment device controls the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the heating control command.

[0085] Preferably, the heat treatment apparatus further includes a plurality of temperature-measuring thermocouples; the plurality of temperature-measuring thermocouples are respectively installed in each sample placement honeycomb hole; and the apparatus further includes the step of:

[0086] The temperature measuring thermocouple collects the temperature of the heated sample inside the corresponding sample honeycomb hole, and sends the temperature of the heated sample to the analysis and processing device through the control panel;

[0087] The analysis and processing device inputs the temperature of the heated sample into the multilayer sensing model, generates a new heating control command corresponding to the heated sample, and sends the new heating control command to the controller of the heat treatment device through the control panel;

[0088] The controller of the heat treatment device adjusts the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the new heating control command.

[0089] Preferably, the method further includes the step of constructing a pre-trained multilayer perceptron model:

[0090] The initial dataset is constructed by importing the gradient temperature zone temperature, the location of the sample honeycomb holes, the fluid type in the duct, the inlet gas temperature, the outlet temperature and velocity, the duct height and size, the heating temperature, the duct cross-sectional area, and the incoming gas flow rate as basic data. Then, multiple temperature ranges are set at set intervals, and multiple flow rates are set according to a set duct opening degree. The initial dataset is augmented by the multiple temperature ranges and the multiple flow rates to obtain the training dataset.

[0091] An initial multilayer sensing model is constructed, using the temperature of the gradient temperature zone in the training dataset, the location of the sampled honeycomb holes, the fluid type in the duct, the gas temperature at the inlet, the temperature and velocity at the outlet, and the height and size of the duct as inputs, and the heating temperature, the cross-sectional area of ​​the duct, and the flow rate of the introduced gas as outputs to initially train the initial multilayer sensing model.

[0092] A set proportion of data is randomly extracted from the training dataset, and the pre-trained multilayer perception model is retrained using the extracted data.

[0093] Rebuild the new training dataset;

[0094] A set proportion of data is randomly extracted from the new training dataset, and the retrained multilayer sensing model is trained using the extracted data. This process is repeated until the temperature control error of the heating temperature is within a set error range, thus obtaining a pre-trained multilayer sensing model.

[0095] The advantages of this invention are mainly reflected in:

[0096] By cleverly utilizing the honeycomb structure and chimney effect, a heating zone with a single heating system and independent temperature measurement structure was set up, enabling the acquisition of heat-treated samples with multiple processes across nine gradient temperature zones in a single heat treatment experiment. Furthermore, an intelligent gradient temperature control model was established based on deep learning methods. Therefore, this invention can significantly improve the efficiency of heat treatment process optimization experiments in materials research and development, shorten the experimental cycle, and thus accelerate the pace of new material development.

[0097] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0098] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and unit can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0099] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.

[0100] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0101] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0102] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0103] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A high-throughput thermal processing system, characterized by, The high-throughput thermal treatment system comprises a thermal treatment device, a control panel and an analysis processing device; the thermal treatment device comprises a controller, a honeycomb structure sample placement area and a plurality of gradient temperature zones; the honeycomb structure sample placement area comprises a plurality of sample placement honeycomb holes and a plurality of air ducts; each sample placement honeycomb hole is surrounded by a plurality of air ducts; and the plurality of sample placement honeycomb holes are distributed in different gradient temperature zones. The control panel is used for inputting at least one sample placement honeycomb hole position information and heating information of a heated sample. The analysis processing device is used for inputting the sample placement honeycomb hole position information and the heating information into a pre-trained multi-layer perception model, generating a heating control instruction corresponding to the heated sample, and sending the heating control instruction to the controller of the thermal treatment device through the control panel. The controller of the thermal treatment device is used for controlling the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the heating control instruction.

2. The high-throughput thermal processing system of claim 1, wherein, The thermal treatment device further comprises a plurality of temperature measuring thermocouples; and the plurality of temperature measuring thermocouples are respectively installed in each sample placement honeycomb hole. The temperature measuring thermocouples are used for collecting the temperature of the heated sample in the corresponding sample placement honeycomb hole and sending the temperature of the heated sample to the analysis processing device through the control panel. The analysis processing device is further used for inputting the temperature of the heated sample into the multi-layer perception model, generating a new heating control instruction corresponding to the heated sample, and sending the new heating control instruction to the controller of the thermal treatment device through the control panel. The controller of the thermal treatment device is further used for adjusting the temperature of the gradient temperature zone corresponding to the heated sample and the fluid flow rate in the air duct according to the new heating control instruction.

3. The high-throughput thermal processing system of claim 2, wherein, The analysis processing device is further used for judging whether the temperature of the heated sample meets a preset furnace discharge temperature value when the temperature of the heated sample is received; if yes, generating a furnace discharge instruction corresponding to the sample placement honeycomb hole of the heated sample, and sending the furnace discharge instruction to the controller of the thermal treatment device through the control panel. The controller of the thermal treatment device is further used for controlling the gradient temperature zone corresponding to the heated sample to stop heating according to the furnace discharge instruction.

4. The high-throughput thermal processing system of claim 2, wherein, The high-throughput thermal treatment system further comprises a model construction module; the model construction module comprises a data set construction unit, a training unit and an optimization unit. The data set construction unit is used for importing the temperature of the gradient temperature zone, the sample placement honeycomb hole position, the fluid type in the air duct, the gas temperature at the air inlet, the temperature and flow rate at the air outlet, the height and size of the air duct, the heating temperature, the cross-sectional area size of the air duct and the gas flow rate as basic data, constructing an initial data set, setting a plurality of temperature sections at a set interval temperature and a plurality of flow rates at a set air duct opening degree, performing data augmentation on the initial data set through the plurality of temperature sections and the plurality of flow rates, and obtaining a training data set. The training unit is configured to build an initial multi-layer perception model, and the initial multi-layer perception model is preliminarily trained by taking the temperature of the gradient temperature zone, the sample placement cell position, the fluid type in the air duct, the gas temperature of the air inlet, the temperature and flow rate of the air outlet, the height and size of the air duct in the training data set as input and taking the heating temperature, the air duct cross-sectional area size and the gas flow as output. The optimization unit is configured to randomly extract a set proportion of data from the training data set, and re-train the preliminarily trained multi-layer perception model by using the extracted data. The data set construction unit is further configured to reconstruct a new training data set. The optimization unit is further configured to randomly extract a set proportion of data from the new training data set, and train the re-trained multi-layer perception model by using the re-extracted data, and iteratively iterate until the temperature control error of the heating temperature is within a set error range, to obtain a pre-trained multi-layer perception model.

5. The high-throughput thermal processing system of claim 1, wherein, The sample placement cell and the air duct are both hexagonal cylindrical structures, and six air ducts are arranged around one sample placement cell to form a honeycomb structure.

6. The high-throughput thermal processing system of claim 5, wherein, The inner wall of the sample placement cell is surrounded by a cylindrical high-temperature-resistant insulation material, and a cavity for placing a heated sample is formed in the middle of the sample placement cell.

7. The high-throughput thermal processing system of claim 5, wherein, The heights of the plurality of air ducts are gradiently changed, and a one-way valve is installed on each air duct, and the one-way valve is connected with the controller of the heat treatment device to receive a control signal of the controller and control the flow rate of the fluid in the air duct according to the control signal.

8. A high-throughput thermal processing method applied to a high-throughput thermal processing device, the high-throughput thermal processing device comprising a thermal processing device, a control panel and an analysis processing device; the thermal processing device comprising a controller, a honeycomb structure sample placement area and a plurality of gradient temperature zones, the honeycomb structure sample placement area comprising a plurality of sample placement honeycomb holes and a plurality of air ducts, each of the plurality of sample placement honeycomb holes being surrounded by a plurality of air ducts, the plurality of sample placement honeycomb holes being distributed in different gradient temperature zones; characterized in that, The method comprises: The control panel inputs at least one sample placement cell position information and heating information of a heated sample; The analysis processing device inputs the sample placement cell position information and the heating information into the pre-trained multi-layer perception model to generate a heating control instruction corresponding to the heated sample, and sends the heating control instruction to the controller of the heat treatment device through the control panel; The controller of the heat treatment device controls the temperature of the gradient temperature zone corresponding to the heated sample and the flow rate of the fluid in the air duct according to the heating control instruction.

9. The high-throughput thermal processing method of claim 8, wherein, The heat treatment device further comprises a plurality of temperature measuring thermocouples, and each temperature measuring thermocouple is installed in each sample placement cell; and the method further comprises the following steps: The temperature measuring thermocouple collects the temperature of the heated sample in the corresponding sample placement cell and sends the temperature of the heated sample to the analysis processing device through the control panel; The analysis processing device inputs the temperature of the heated sample into the multi-layer perception model to generate a new heating control instruction corresponding to the heated sample, and sends the new heating control instruction to the controller of the heat treatment device through the control panel; The controller of the heat treatment device adjusts the temperature of the gradient temperature zone corresponding to the heated sample and the flow rate of the fluid in the air duct according to the new heating control instruction.

10. The high-throughput thermal processing method of claim 9, wherein, The method further comprises the following steps of building a pre-trained multi-layer perception model: The initial data set is constructed by taking the gradient temperature zone temperature, the sample placement honeycomb hole position, the fluid type in the air duct, the gas temperature at the air inlet, the outlet temperature and flow rate, the air duct height and size, the heating temperature, the air duct cross-sectional area size, and the gas flow rate as basic data. Multiple temperature sections are set at a set interval temperature, and multiple flow rates are set at a set air duct opening degree. The initial data set is data augmented by the multiple temperature sections and the multiple flow rates to obtain a training data set; An initial multi-layer perception model is constructed. The gradient temperature zone temperature, the sample placement honeycomb hole position, the fluid type in the air duct, the gas temperature at the air inlet, the outlet temperature and flow rate, the air duct height and size in the training data set are taken as inputs, and the heating temperature, the air duct cross-sectional area size, and the gas flow rate are taken as outputs to preliminarily train the initial multi-layer perception model; A set proportion of data is randomly extracted from the training data set, and the preliminarily trained multi-layer perception model is retrained by the extracted data; A new training data set is reconstructed; A set proportion of data is randomly extracted from the new training data set, and the retrained multi-layer perception model is trained by the reextracted data. Iterative training is performed until the heating temperature control error is within a set error range to obtain a pre-trained multi-layer perception model.

Images