A method, apparatus and device for temperature control in a closed high temperature cavity

Abstract

The application discloses a temperature control method, device and equipment in a closed high-temperature cavity, and the method comprises the following steps: acquiring a first parameter set of a first device in a current processing process of a first object in the cavity, wherein a first parameter contained in the first parameter set is a parameter affecting the temperature in the cavity; inputting the first parameter set into a trained prediction model to obtain a first temperature sequence output by the prediction model, wherein the first temperature sequence is a sequence of the temperature in the cavity changing with time; and based on the first temperature sequence, the value of the first parameter is regulated to control the temperature in the cavity.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of terminals, and relates to, but is not limited to, a temperature control method, device and equipment in a closed high-temperature cavity. BACKGROUND

[0002] In the related art, when an object is processed through a closed high-temperature cavity of a device, due to the requirement of the object on the closed environment, the temperature in the closed high-temperature cavity is difficult to be measured in real time, and the temperature in the cavity cannot be accurately predicted.

[0003] The temperature in most cavities is adjusted depending on the long-term personal operation experience of a process engineer, resulting in low adjustment accuracy and large deviation. SUMMARY

[0004] In view of this, the present application provides a temperature control method, device and equipment in a closed high-temperature cavity.

[0005] In a first aspect, the present application provides a temperature control method in a closed high-temperature cavity, the method comprising: obtaining a first parameter set in a current processing process of a first object in the cavity by a first device, wherein a first parameter contained in the first parameter set is a parameter affecting the temperature in the cavity; inputting the first parameter set into a trained prediction model to obtain a first temperature sequence output by the prediction model, wherein the first temperature sequence is a sequence of the temperature in the cavity changing with time; and based on the first temperature sequence, adjusting the value of the first parameter to control the temperature in the cavity.

[0006] In a second aspect, the present application provides a temperature control device in a closed high-temperature cavity, comprising: a first obtaining module configured to obtain a first parameter set in a current processing process of a first object in the cavity by a first device, wherein a first parameter contained in the first parameter set is a parameter affecting the temperature in the cavity; an input module configured to input the first parameter set into a trained prediction model to obtain a first temperature sequence output by the prediction model, wherein the first temperature sequence is a sequence of the temperature in the cavity changing with time; and an adjusting module configured to adjust the value of the first parameter based on the first temperature sequence to control the temperature in the cavity.

[0007] In a third aspect, the present application provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, and the processor implements the steps in the temperature control method in a closed high-temperature cavity according to the present application when executing the program. BRIEF DESCRIPTION OF DRAWINGS

[0008] Figure 1A flowchart of a temperature control method in a closed high-temperature cavity according to an embodiment of the present application is shown in

[0009] Figure 2 A schematic diagram of a prediction model according to an embodiment of the present application is shown in

[0010] Figure 3 A flowchart of a VD furnace vacuum temperature dynamic prediction method according to an embodiment of the present application is shown in

[0011] Figure 4 A schematic diagram of a temperature control device in a closed high-temperature cavity according to an embodiment of the present application is shown in

[0012] Figure 5 A hardware entity diagram of an electronic device according to an embodiment of the present application is shown in DETAILED DESCRIPTION

[0013] The technical solutions of the present application will be further described in detail below in combination with the drawings and embodiments.

[0014] Figure 1 A flowchart of a temperature control method in a closed high-temperature cavity according to an embodiment of the present application is shown in Figure 1 As shown, the method comprises the following steps:

[0015] Step 102: Obtain a first parameter set of a first device in a current processing process of a first object in the cavity, wherein the first parameter in the first parameter set is a parameter affecting the temperature in the cavity;

[0016] Since impurities in the air enter the first object, it will affect the quality of the first object, therefore, the first object needs to be processed in the closed high-temperature cavity of the first device, the temperature in the cavity is an important indicator affecting the quality of the first object, and the requirement of the first object for the closed environment makes it difficult to measure the temperature in the closed high-temperature cavity in real time, so the temperature in the closed high-temperature cavity needs to be predicted.

[0017] Step 104: Input the first parameter set into the trained prediction model to obtain a first temperature sequence output by the prediction model, wherein the first temperature sequence is a sequence of the temperature in the cavity changing with time;

[0018] The first temperature sequence can be a sequence composed of temperatures at multiple future time points in a future period of time. Since the first parameter in the first parameter set is a parameter that affects the temperature in the cavity, i.e., the first parameter and the temperature in the cavity have a correlation, a plurality of historical parameter sets and a historical temperature sequence actually measured under each historical parameter set can be obtained. The initial prediction model is trained by using the plurality of historical parameter sets and the historical temperature sequence corresponding to each historical parameter set, and a trained prediction model is obtained. The trained prediction model can analyze the input first parameter set and predict the first temperature sequence corresponding to the first parameter set.

[0019] Step 106: based on the first temperature sequence, regulating the value of the first parameter to control the temperature in the cavity.

[0020] The parameter regulation range can be recommended according to the difference between the predicted temperature value at each time point in the predicted first temperature sequence and the expected temperature value of the user at the corresponding time point. The first parameter is regulated according to the parameter regulation range, so that the temperature in the cavity approaches the expected temperature value.

[0021] In the embodiments of the present application, the prediction model is used to predict the temperature in the closed high-temperature cavity in a future period of time (for example, 20 minutes or 30 minutes), so that the temperature in the closed high-temperature cavity during the processing of the first object can be more accurately predicted, the frequency of manual sampling can be reduced, and the product yield can be improved. The parameters are regulated according to the predicted temperature to control the temperature in the cavity. The temperature in the cavity is controlled by adjusting the parameters, and the best parameter combination is given, so that ideal temperature control can be achieved.

[0022] In some embodiments, the first device can be a metallurgical device, the first object can be a smelting object, the closed high-temperature cavity can be vacuum, and the processing process can be a smelting process, i.e., the metallurgical device smelts the smelting object in a vacuum. The first temperature sequence is a sequence composed of temperatures at multiple future time points in a future period of time under the vacuum state of the metallurgical device.

[0023] In the embodiments of the present application, the prediction model is used to predict the temperature in the closed high-temperature vacuum cavity in a future period of time, so that the temperature in the closed high-temperature vacuum cavity during the smelting process of the smelting object can be more accurately predicted, the frequency of manual sampling can be reduced, and the product yield can be improved. The parameters are regulated according to the predicted temperature to control the temperature in the cavity. The temperature in the cavity is controlled by adjusting the parameters, and the best parameter combination is given, so that ideal temperature control can be achieved.

[0024] In some embodiments, the method further comprises:

[0025] Step 1012: obtaining, for each of the at least one historical processing process of the first object by the first device, a second parameter set and a corresponding second temperature sequence;

[0026] The second parameter set includes a second parameter that affects the temperature in the cavity, and the second temperature sequence is a sequence of actually measured temperature changes in the cavity over time.

[0027] The second parameter includes at least one of the operation record parameter of the first device, the basic parameter of the first object, and the real-time operation parameter in each of the historical processing processes; the first device includes a vacuum degassing furnace (VD), and the first object includes steel.

[0028] In some embodiments, for each historical processing process, a second parameter set in the historical processing process can be obtained, and the temperature in the cavity actually measured under the second parameter set.

[0029] In some embodiments, VD furnace historical data in the past two years can be obtained, including the operation record parameter of the VD furnace, the basic parameter of the steel, and the real-time operation parameter. The operation record data of the VD furnace can include the start-up time, the maintenance record, and the device configuration (the size and voltage of the VD furnace, etc.), the basic parameter of the steel can include the ladle condition, the ladle age, the steel grade, the smelting time, and the power-on time, etc., and the real-time operation data can include the temperature drop rate, the pressure, the vacuum nitrogen flow, and the furnace aluminum amount, etc. VD furnace historical data in each historical smelting process can be obtained, and a temperature sequence actually measured under the corresponding VD furnace historical data can be obtained.

[0030] Step 1014: training an initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain the trained prediction model.

[0031] In the embodiments of the present application, by analyzing the process conditions of the metallurgical process and the historical data of the smelting process, a prediction model is constructed to predict the future VD furnace vacuum temperature, which is important for mastering and controlling product quality. By predicting the temperature of the cavity in a closed vacuum condition in the future, the furnace temperature can be more accurately controlled, the precise temperature control process in the ladle refining process can be realized, the frequency of manual temperature sampling can be reduced, and the product yield can be improved.

[0032] In some embodiments, the method further comprises:

[0033] Step 10131: constructing at least one fitting equation based on a first correlation relationship between the second parameters in the second parameter set of each of the historical processing processes.

[0034] In some embodiments, the correlation between the second parameters can be determined according to the process mechanism of the metallurgical process, and the fitting equations such as the steel grade temperature fitting equation, the vacuum temperature drop rate equation, the nitrogen and argon time cumulative effect curve, and the temperature drop and steel loss correlation curve can be simulated and fitted.

[0035] Step 10132: generating a corresponding mechanism model based on each of the fitting equations;

[0036] Step 1014: training an initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain the trained prediction model, including:

[0037] Step 10141: training an initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain a target prediction model;

[0038] Step 10142: model fusion of at least one mechanism model and the target prediction model to obtain the trained prediction model.

[0039] In some embodiments, a linear regression model can be constructed for the fitted mechanism model and the target prediction model for model fusion.

[0040] In the embodiments of the present application, a machine learning deep learning model is constructed using deep learning combined with steel metallurgical mechanism technology, which gets rid of the mechanism model fitted from the mechanism perspective, avoids the failure of the fitted mechanism model caused by the sudden change of one or more factors affecting the vacuum temperature of the VD furnace, and has higher prediction accuracy of the predicted temperature and more applicable scenarios of the prediction model. When some factors change suddenly, the prediction model still has good performance, and the prediction result is closer to the true result.

[0041] In some embodiments, the step 104 of "regulating the value of the first parameter based on the first temperature sequence" includes:

[0042] Step 1041: determining a target temperature sequence that meets the expected condition from the first temperature sequence based on the processing quality of the first object;

[0043] In which, the target temperature sequence that meets the user's expected condition can be determined according to whether the first temperature sequence predicted by the prediction model is close to the expected temperature; the better the processing quality, the closer the first temperature sequence to the target temperature sequence.

[0044] Step 1042: determining a target parameter set corresponding to the target temperature sequence;

[0045] In some embodiments, step 1042 "determining a target parameter set corresponding to the target temperature sequence" comprises:

[0046] Step 10421: determining a second correlation between the value of the first parameter and the temperature in the cavity based on a process control principle.

[0047] In some embodiments, the process control principle can be a second correlation between the value of the first parameter and the temperature in the cavity, for example, to control the temperature rise, the power-on time needs to be adjusted up, instead of adjusting other parameters, or the power-on time needs to be shortened.

[0048] Step 10422: determining a target parameter set corresponding to the target temperature sequence based on the target temperature sequence and the second correlation.

[0049] Step 1043: sending the target parameter set to the control personnel, so that the control personnel controls the value of the first parameter based on the target parameter set to control the temperature in the cavity.

[0050] Wherein, the target parameter set can be a combination of parameters such as pre-vacuum temperature drop, vacuum temperature drop, ladle slag overflow vacuum degree, vacuum nitrogen flow, furnace aluminum addition amount, and pressure; the target parameter set corresponding to the target temperature sequence meeting the expected condition can be determined according to the process control principle, the target parameter set can be more accurately determined according to the target temperature sequence, the target parameter set can be recommended to the field operator in real time based on the parameter adjustment control range, and the field operator can adjust the value of the first parameter according to the operation suggestion to achieve ideal temperature control.

[0051] In some embodiments, the method further comprises:

[0052] Step 108: inputting the target parameter set and the target temperature sequence into the trained prediction model to update the trained prediction model.

[0053] In the embodiments of the present application, the prediction model can be positively fed back according to the target parameter set and the target temperature sequence meeting the expected condition; the prediction model based on online learning technology is established for periodic training optimization.

[0054] In some embodiments, the prediction model is a recurrent neural network model.

[0055] In the embodiment of the present application, a recurrent neural network model (RNN) is used to model the data, which has better learning ability for time series data than a BP (Back Propagation) neural network and other machine learning algorithms, and can not only estimate the current VD furnace vacuum temperature, but also dynamically predict the temperature in the next 20 minutes, which has a stronger guiding significance for display.

[0056] Steel production is a very complex continuous process, and the VD furnace vacuum temperature is an important indicator of steel quality. Due to the characteristics of VD furnace smelting under vacuum, it is difficult to realize real-time measurement of the smelting temperature parameter, and it is impossible to accurately predict the vacuum temperature. Usually, the vacuum temperature of most VD furnaces depends on the long-term personal operation experience of engineers for adjustment, which leads to low adjustment accuracy, large deviation and other defects affecting the operation of the smelting process. Therefore, by combining the metallurgical equipment operation record data and real-time operation data, the future temperature of the DV vacuum is predicted, and a parameter optimization algorithm (such as a tabu search algorithm) based on operational optimization is constructed to recommend the parameter control range in real time, so as to achieve ideal temperature control.

[0057] The embodiment of the present application can provide a VD furnace vacuum temperature dynamic prediction device based on deep learning and online learning based on historical data and real-time operation data. By analyzing the core parameters affecting the VD furnace vacuum temperature, a time series prediction model based on RNN is constructed to dynamically predict the VD furnace vacuum temperature in the next 20 minutes, and a parameter optimization algorithm based on operational optimization is constructed to dynamically adjust the optimal parameter combination, accurately control the furnace temperature, realize accurate temperature control process in the ladle refining process, reduce the frequency of manual temperature sampling, and improve the product yield.

[0058] The embodiment of the present application provides a VD furnace vacuum temperature dynamic prediction method, which comprises the following steps:

[0059] Step S201: Obtain VD furnace historical data in the past two years, including operation record data such as start-up time and maintenance record, basic parameter data such as equipment configuration, ladle condition, ladle age, steel grade, smelting time, power-on time, and real-time operation data such as temperature drop, pressure, vacuum nitrogen flow, and furnace aluminum addition.

[0060] Step S202: Construct a fitting mechanism characteristic equation to realize a mechanism model.

[0061] The mechanism model can include a steel grade temperature drop fitting equation, a vacuum temperature drop rate equation, a nitrogen argon time cumulative effect curve, and a temperature drop and steel loss correlation curve.

[0062] Step S203: Establish a deep learning RNN model for predicting the vacuum temperature of a VD furnace;

[0063] like Figure 2 As shown, the prediction model includes an input layer 21, a hidden layer 22, and an output layer 23. The input layer 21 includes nodes 1, 2, and 3; the hidden layer 22 includes nodes 4, 5, 6, and 7; and the output layer includes nodes 8 and 9. Historical VD furnace data X can be input into the input layer 21, and the VD furnace vacuum temperature Y can be output by the output layer 23. W can be used... 41 W represents the weight between node 1 and node 4. 42 W represents the weight between node 2 and node 4. 43 W represents the weight between node 3 and node 4. 84 W represents the weight between node 8 and node 4. 85 W represents the weight between node 8 and node 5. 86 W represents the weight between node 8 and node 6. 87 This represents the weight between node 8 and node 7.

[0064] Step S204: Construct a linear regression model based on the mechanism model and the time series prediction model of furnace vacuum temperature using RNN, and perform model fusion;

[0065] Step S205: Establish a prediction model based on online learning technology and perform timed training and optimization;

[0066] Step S206: Based on the process control principles and model results, find the optimal combination of parameter control;

[0067] Specifically, the target temperature sequence with the smallest difference can be determined based on the difference between the first temperature sequence output by the prediction model and the expected temperature sequence. Based on the process control principle, the optimal set of target parameters corresponding to the target temperature sequence can be determined. The target parameter combination can be a combination of parameters such as the temperature drop before deep vacuum, the temperature drop upon reaching vacuum, the vacuum degree of slag overflow in the ladle, the vacuum nitrogen flow rate, the amount of aluminum added to the furnace, and the pressure at their optimal values.

[0068] Step S207: Based on the optimal vacuum temperature of the VD furnace stabilized by the combined parameters, recommend the range of parameter combinations to the on-site operators in real time;

[0069] Step S208: On-site operators adjust the equipment operating parameters according to the operating suggestions to achieve ideal temperature control.

[0070] In the embodiments of the present application, according to the metallurgical process mechanism, multi-process mechanism quantification is carried out, and mechanism formula fitting such as steel grade temperature fitting equation, vacuum temperature drop rate equation, nitrogen and argon time cumulative effect curve, and temperature drop and steel loss correlation curve is simulated and fitted; the VD furnace vacuum temperature time series prediction model is constructed based on the time series prediction model of RNN, and the VD furnace vacuum temperature in the future 5, 10 and 20 minutes is dynamically predicted.

[0071] Based on the four equations of the fitted metallurgical process mechanism and the VD furnace vacuum temperature time series results, the linear regression model is constructed for model fusion in the embodiments of the present application.

[0072] Based on the completed model fusion, the parameter optimization is realized by constructing the tabu search based on operational optimization in the embodiments of the present application. In the VD furnace vacuum temperature parameter optimization, the optimization is carried out on the basis of the parameters of the last round of iteration, and the influence of the last round of trend is fully considered, so as to avoid random initial parameters and achieve rapid convergence of the model. The best parameter combination is dynamically adjusted, and the furnace temperature is accurately controlled. The accurate temperature control process in the ladle refining process is realized, the frequency of manual temperature sampling is reduced, and the product yield is improved.

[0073] In the steel smelting direction, the four mechanism equations in the steel smelting industry are fitted and deep learning is re-fused in the embodiments of the present application, which breaks away from the complete mechanism from the perspective of fitting the mechanism model, and avoids the failure of the fitted mechanism model caused by the sudden change of one or more factors affecting the VD vacuum temperature.

[0074] The embodiments of the present application realize multi-process mechanism quantification in the application mode combination, and the mechanism formula fitting such as steel grade temperature drop fitting equation, vacuum temperature drop rate equation, nitrogen and argon time cumulative effect curve, and temperature drop and steel loss correlation curve, which greatly improves the feature combination validity and guarantees the temperature prediction accuracy; through the analysis of the process conditions of the metallurgical process and the historical data of the smelting process, the time series prediction model based on RNN is constructed, the future VD furnace vacuum temperature is predicted, and it has important guidance for mastering and controlling product quality.

[0075] The embodiment of the application models data by using RNN, which has better learning ability for time series data than BP neural network and other machine learning algorithms. Not only can the current VD furnace vacuum temperature be predicted, but also the temperature in the next twenty minutes can be dynamically predicted, which has a stronger guiding significance for display. According to the real-time prediction data, a parameter optimization optimization algorithm based on operational optimization is constructed to give the best parameter combination to achieve ideal temperature control. The model avoids the influence of equipment operation and process improvement on the model through self-learning algorithm, and the active learning cooperates with optimization to solve the problem of model mutation for small batch steel products and process improvement. In the parameter optimization, the last used parameters are introduced, and iterative optimization is performed on this basis, fully considering the trend influence of the last round, so as to avoid random initial parameters and achieve fast convergence of the model.

[0076] The embodiment of the application has higher accuracy than the fitting mechanism model prediction in the related art, and the model is applicable to more scenarios. When some factors change suddenly, the model still has good performance; the prediction result is closer to the true result.

[0077] Figure 3 A flowchart of a VD furnace vacuum temperature dynamic prediction method according to an embodiment of the application is shown in FIG. 1, which includes the following steps: Figure 3

[0078] Step 301: Obtain VD furnace historical data;

[0079] Step 302: Construct a fitting mechanism characteristic equation to realize a mechanism model;

[0080] Among them, a temperature fitting equation for different steel grades can be fitted; a vacuum temperature drop rate equation can be fitted; a nitrogen and argon time cumulative effect curve can be fitted; a temperature drop and steel loss correlation curve can be fitted;

[0081] Step 303: Construct a deep learning RNN VD furnace vacuum temperature prediction model;

[0082] Step 304: Construct a linear regression model based on the mechanism model and the RNN furnace vacuum temperature time series prediction model;

[0083] Step 305: Construct an online learning prediction model;

[0084] Step 306: Find the optimal control parameter combination;

[0085] Among them, different parameter combinations can be generated first, and then the best parameter combination can be found from the different parameter combinations according to the control parameter combination corresponding to the optimal temperature sequence;

[0086] Step 307: Real-time recommend parameter control range to the field operator;

[0087] ​Step 308: adjust the device operating parameter.

[0088] It should be noted that, in the embodiments of the present application, if the above-mentioned temperature control method in the closed high-temperature cavity is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing an electronic device (which can be a mobile phone, a tablet computer, a desktop computer, a personal digital assistant, a navigator, a digital telephone, a video telephone, a television, a sensing device, etc.) to execute all or part of the methods described in the embodiments of the present application. The aforementioned storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various storage medium capable of storing program codes. Thus, the embodiments of the present application are not limited to any specific hardware and software combination.

[0089] Figure 4 A schematic diagram of a structure of a temperature control device in a closed high-temperature cavity according to an embodiment of the present application is shown in FIG. 4. As shown in FIG. 4, the device 400 includes a first obtaining module 401, an input module 402, and a control module 403, wherein: Figure 4

[0090] The first obtaining module 401 is configured to obtain a first parameter set in a current processing process of a first object in the cavity by a first device, wherein a first parameter included in the first parameter set is a parameter affecting the temperature in the cavity.

[0091] The input module 402 is configured to input the first parameter set into a trained prediction model to obtain a first temperature sequence output by the prediction model, wherein the first temperature sequence is a sequence of the temperature in the cavity changing with time.

[0092] The control module 403 is configured to control the value of the first parameter based on the first temperature sequence to control the temperature in the cavity.

[0093] In some embodiments, the device further includes a second obtaining module configured to obtain a second parameter set and a corresponding second temperature sequence of each historical processing process of the first object in at least one historical processing process of the first object by the first device, wherein a second parameter included in the second parameter set is a parameter affecting the temperature in the cavity, and the second temperature sequence is a sequence of the actually measured temperature in the cavity changing with time; and a training module configured to train an initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain the trained prediction model. ​

[0094] In some embodiments, the second parameters include at least one of a running record parameter of the first device, a base parameter of the first object, and a real-time running parameter in each of the historical processing processes; and the first device includes a vacuum degassing furnace VD, and the first object includes steel.

[0095] In some embodiments, the apparatus further includes a construction module configured to construct at least one fitting equation based on a first correlation between second parameters in the second parameter set of each of the historical processing processes; and a generation module configured to generate a corresponding mechanism model based on each of the fitting equations; and the training module includes a training submodule configured to train an initial prediction model based on the second parameter set and a corresponding second temperature sequence to obtain a target prediction model; and a fusion submodule configured to perform model fusion between the at least one mechanism model and the target prediction model to obtain the trained prediction model.

[0096] In some embodiments, the regulation module 403 includes a first determination submodule configured to determine a target temperature sequence that satisfies an expected condition from the first temperature sequence based on a processing quality of the first object; a second determination submodule configured to determine a target parameter set corresponding to the target temperature sequence; and a regulation submodule configured to send the target parameter set to a regulation personnel for regulating a value of the first parameter based on the target parameter set to control the temperature in the cavity.

[0097] In some embodiments, the apparatus further includes an update module configured to input the target parameter set and the target temperature sequence into the trained prediction model to update the trained prediction model.

[0098] In some embodiments, the second determination submodule includes a first determination unit configured to determine a second correlation between the value of the first parameter and the temperature in the cavity based on a process control principle; and a second determination unit configured to determine the target parameter set corresponding to the target temperature sequence based on the target temperature sequence and the second correlation.

[0099] In some embodiments, the prediction model is a recurrent neural network model.

[0100] The above apparatus embodiments are similar to the above method embodiments in description and have similar beneficial effects. For technical details not disclosed in the apparatus embodiments of the present application, please refer to the description of the method embodiments of the present application.

[0101] Correspondingly, the embodiments of the present application provide an electronic device, Figure 5Fig. 1 shows a schematic diagram of a hardware entity of an electronic device according to an embodiment of the present application. Figure 5 As shown in Fig. 1, the hardware entity of the device 500 includes a memory 501 and a processor 502, the memory 501 stores a computer program executable on the processor 502, and the processor 502 implements the steps of the temperature control method in the closed high-temperature cavity in the above embodiments when executing the program.

[0102] The memory 501 is configured to store instructions and applications executable by the processor 502, and can also cache data (for example, image data, audio data, voice communication data and video communication data) to be processed by the processor 502 and modules in the device 500, which can be implemented by FLASH or Random Access Memory (RAM).

[0103] Correspondingly, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the temperature control method in the closed high-temperature cavity provided in the above embodiments.

[0104] It should be noted that the description of the above storage medium and device embodiments is similar to the description of the above method embodiments, and has similar beneficial effects to the device embodiments. For technical details of the storage medium and method embodiments of the present application that are not disclosed, please refer to the description of the device embodiments of the present application for understanding.

[0105] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily mean the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of each process does not mean the execution order, and the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The sequence number of the above embodiments of the present application is only for description, not representing the advantages and disadvantages of the embodiments.

[0106] It should be noted that, in the present document, the terms "comprises", "comprising", or any other variations 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 can also include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0107] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other manners. The described device embodiments are merely illustrative. For example, the division of the units is only a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed coupling between the components can be indirect coupling or direct coupling through some interface, or communication connection, which can be electrical, mechanical or in other forms.

[0108] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units; they can be located in one place, or distributed on multiple network units; and some or all of the units can be selected according to actual needs to achieve the purpose of the present embodiment. In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in the form of hardware, or in the form of hardware plus software functional units.

[0109] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware, and the foregoing program can be stored in a computer readable storage medium, and the program performs the steps of the above-mentioned method embodiments when executed; and the foregoing storage medium includes a mobile storage device, a read only memory (ROM), a magnetic disc or an optical disc and various storage medium capable of storing program codes. Alternatively, when the integrated units of the present application are realized in the form of software function modules and sold or used as independent products, they can also be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, includes a plurality of instructions for causing a computer device (which can be a mobile phone, a tablet computer, a desktop computer, a personal digital assistant, a navigator, a digital telephone, a video telephone, a television, a sensor device, etc.) to execute all or part of the methods described in the embodiments of the present application. The foregoing storage medium includes a mobile storage device, a ROM, a magnetic disc or an optical disc and various storage medium capable of storing program codes.

[0110] The methods disclosed in the several method embodiments provided by the present application can be combined arbitrarily without conflict to obtain new method embodiments. The features disclosed in the several product embodiments provided by the present application can be combined arbitrarily without conflict to obtain new product embodiments. The features disclosed in the several method or device embodiments provided by the present application can be combined arbitrarily without conflict to obtain new method embodiments or device embodiments.

[0111] The above is only the implementation of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for controlling temperature in a high-temperature cavity, comprising: obtaining a first parameter set of a first device in a current processing process of a first object in the cavity, wherein a first parameter in the first parameter set is a parameter affecting temperature in the cavity; inputting the first parameter set into a trained prediction model to obtain a first temperature sequence output by the prediction model, wherein the first temperature sequence is a sequence of temperature in the cavity changing over time; based on the first temperature sequence, adjusting a value of the first parameter to control temperature in the cavity; the step of adjusting the value of the first parameter based on the first temperature sequence comprises: determining a target temperature sequence satisfying an expected condition from the first temperature sequence based on processing quality of the first object, wherein the target temperature sequence is used to represent a temperature sequence corresponding to an optimal processing quality of the first object; determining a target parameter set corresponding to the target temperature sequence; and sending the target parameter set to an adjustment personnel for adjusting the value of the first parameter based on the target parameter set.

2. The method of claim 1, wherein, The method further comprises: obtaining a second parameter set and a corresponding second temperature sequence of each historical processing process of the first object in at least one historical processing process of the first object by the first device; wherein a second parameter in the second parameter set is a parameter affecting temperature in the cavity, and the second temperature sequence is a sequence of actually measured temperature in the cavity changing over time; training an initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain the trained prediction model.

3. The method of claim 2, wherein, The second parameter comprises at least one of a running record parameter of the first device, a basic parameter of the first object, and a real-time running parameter in each historical processing process; the first device comprises a vacuum degassing furnace (VD), and the first object comprises steel.

4. The method of claim 2, wherein, The method further comprises: constructing at least one fitting equation based on a first correlation relationship between second parameters in the second parameter set of each historical processing process; generating a corresponding mechanism model based on each fitting equation; the step of training the initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain the trained prediction model comprises: training an initial prediction model based on the second parameter set and the corresponding second temperature sequence to obtain a target prediction model; and performing model fusion on at least one mechanism model and the target prediction model to obtain the trained prediction model.

5. The method of claim 1, wherein, The method further comprises: inputting the target parameter set and the target temperature sequence into the trained prediction model to update the trained prediction model.

6. The method of claim 1, wherein, The step of determining the target parameter set corresponding to the target temperature sequence comprises: determining a second correlation relationship between the value of the first parameter and temperature in the cavity based on a process control principle; and determining the target parameter set corresponding to the target temperature sequence based on the target temperature sequence and the second correlation relationship.

7. The method of any one of claims 1 to 6, wherein, The prediction model is a recurrent neural network model.

8. An apparatus for controlling temperature in a high-temperature cavity, the apparatus comprising: a first obtaining module configured to obtain a first parameter set of a first device in a current processing of a first object in the cavity, the first parameter set comprising a first parameter affecting temperature in the cavity; an inputting module configured to input the first parameter set into a trained prediction model to obtain a first temperature sequence output by the prediction model, the first temperature sequence being a sequence of temperature in the cavity over time; a regulating module configured to regulate a value of the first parameter based on the first temperature sequence to control temperature in the cavity; the regulating module comprising: a first determining submodule configured to determine a target temperature sequence satisfying a desired condition from the first temperature sequence based on a processing quality of the first object, the target temperature sequence being used to represent a temperature sequence corresponding to an optimal processing quality of the first object; a second determining submodule configured to determine a target parameter set corresponding to the target temperature sequence; and a regulating submodule configured to send the target parameter set to a regulator for regulating the value of the first parameter based on the target parameter set.

9. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, The processor implements the steps in the method for controlling temperature in a high-temperature cavity according to any one of claims 1 to 7 when executing the program.

Images