VD furnace liquid level control method and related equipment

Abstract

The invention is suitable for the technical field of metal smelting, and provides a VD furnace liquid level control method and related equipment, and the method comprises the steps that gas flow data in a VD furnace within a preset time period and liquid level data corresponding to the gas flow data are obtained; training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model; obtaining a predicted liquid level under a preset gas flow in a blowing process to be predicted based on the target prediction model, and determining prediction line data of the preset gas flow and the predicted liquid level; and control information for the liquid level of the VD furnace is obtained, the target gas flow corresponding to the control information is determined according to the prediction line data and the control information, and control over the liquid level of the VD furnace is completed based on the target gas flow. The liquid level rising height is predicted, the uniformity degree of molten steel stirring is improved, the liquid level height in a steel ladle is conveniently and accurately controlled, and the uncertainty of operation of a refining worker is avoided.

Description

Technical Field

[0001] The present application belongs to the technical field of metal smelting, and in particular, relates to a VD furnace liquid level control method and related equipment. Background Art

[0002] The vacuum environment effectively avoids oxidation by air, so that the VD refining process can use a large stirring intensity, and the slag reaction has good dynamic conditions, which can greatly promote the removal of inclusions. Vacuum degassing devices include VD, vacuum arc degassing furnace, RH vacuum degassing device, VOD refining furnace, etc. Among them, VD furnace has been widely used because of its low cost and good degassing, desulfurization and inclusion removal effects.

[0003] The main chemical reactions in the VD furnace involve gases such as O2, H2, and N2. The VD refining process is accompanied by high temperature and high pressure. By blowing gas into the bottom permeable bricks to stir the molten steel, the molten steel in the furnace can fully contact and react with the slag, promoting the floating of inclusions. In the early stage of vacuum refining, the refining slag above the molten steel surface has low temperature and high viscosity, and has strong bubble wrapping ability. With the introduction of bottom blowing gas, the pressure in the furnace increases, and the floating area of ​​the slag level on the molten steel surface increases. However, due to the low vacuum degree above, the bubbles overflow slowly, causing the emulsified area to rise as a whole. If the gas flow introduced at this time is too large, the rising speed of the emulsified area will be rapidly accelerated; in the middle stage of vacuum refining, with the increase of gas blowing volume, the emulsified area will quickly approach the upper edge of the ladle. If no air-breaking means are taken, the liquid level will quickly overflow the ladle, causing the molten steel to be oxidized by the outside world, increasing the difficulty of furnace maintenance in the later stage; in the later stage of vacuum refining, the viscosity of the refining slag covering the molten steel decreases and the air permeability is good. At this time, the bottom blowing bubbles can quickly pass through the slag layer. At this time, the size of the gas flow blown in has little effect on the liquid level height, and the liquid surface is not enough to overflow. Therefore, the size of the gas blowing volume has a significant impact on the early and middle stages of vacuum refining. Reasonable control of the bottom blowing gas flow rate to stabilize the liquid level at a reasonable height in the ladle is the key to controlling the purity of the refined molten steel in the VD furnace. It plays an important role in fully stirring the molten steel and preventing slag from overflowing.

[0004] In order to achieve the goal of eliminating abnormal splashing of molten steel, reducing manpower input, and improving the efficiency of vacuum refining, the "A control system for bottom blowing gas in a ladle" proposed in Chinese patent CN202022129012.8 proposes a method of using the PLC controller in the gas control cabinet as the core to control the gas flow rate. Although this method reduces the working hours and difficulty of workers and realizes bottom blowing automation, the PLC controller used in this patent cannot realize intelligent regulation of the liquid level. If the liquid level is not properly controlled, it may cause the impurity content in the molten steel to be too high, affecting the life of the steel. Summary of the invention

[0005] The embodiments of the present application provide a VD furnace liquid level control method and related equipment, which address the problem of lack of intelligent control of liquid level in the prior art.

[0006] The present application embodiment provides a VD furnace liquid level control method, comprising the following steps:

[0007] Acquire gas flow data and liquid level data corresponding to the gas flow data in the VD furnace within a preset time period;

[0008] Training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model;

[0009] Fitting the target prediction model to obtain prediction line data of the gas flow data and the liquid level data;

[0010] The control information for the liquid level of the VD furnace is obtained, and the target gas flow corresponding to the control information is determined according to the prediction line data and the control information, and the control of the liquid level of the VD furnace is completed based on the target gas flow.

[0011] Optionally, the step of training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model includes:

[0012] Based on the gas flow data and the liquid level data, a gradient descent optimization algorithm is used to train the preset deep learning model until the training of the preset number of iterations is completed to obtain a target prediction model, wherein the mathematical representation of the gradient descent optimization algorithm is:

[0013]

[0014] Among them, x t represents the gradient at time step t, a represents the learning rate, Represents the loss function J(x) for parameter x t The gradient of Bt represents the small batch consisting of training data sample indexes. Represents x t-1 The gradient at Represents the mini-batch stochastic gradient obtained by repeated sampling.

[0015] Optionally, the step of acquiring control information for the VD furnace liquid level and determining a target gas flow rate corresponding to the control information according to the prediction line data and the control information includes:

[0016] Sending the predicted line data to a preset receiving end and obtaining user request information for the VD furnace liquid level;

[0017] The user request information is used as the control information, and a target gas flow rate corresponding to the control information is determined according to the prediction line data and the control information.

[0018] Optionally, the step of acquiring control information for the VD furnace liquid level and determining a target gas flow rate corresponding to the control information according to the prediction line data and the control information further includes:

[0019] Acquiring threshold data of the VD furnace liquid level and actual liquid level data of the VD furnace liquid level;

[0020] Matching the threshold data with the actual liquid level data, if the threshold data is greater than the actual liquid level data, generating control information for the VD furnace liquid level based on the threshold data, and determining a target gas flow corresponding to the threshold data according to the prediction line data and the threshold data;

[0021] If the threshold data is less than or equal to the actual liquid level data, an alarm message is generated, and the alarm message is sent according to a preset rule to obtain feedback information for the alarm message;

[0022] The feedback information is used as control information for the VD furnace liquid level, and a target gas flow rate corresponding to the feedback information is determined according to the prediction line data and the control information.

[0023] Optionally, the step of acquiring gas flow data in the VD furnace within a preset time period and liquid level data corresponding to the gas flow data includes:

[0024] Acquire the operating parameters of the gas control device corresponding to the VD furnace within a preset time period, and determine the gas flow data in the VD furnace within the preset time period based on the operating parameters;

[0025] Acquire image data in the VD furnace within a preset time period, and enhance the image data using a DDE image algorithm to obtain a target image;

[0026] Liquid level data corresponding to the gas flow data within a preset time period is determined based on the target image.

[0027] Optionally, the step of acquiring gas flow data in the VD furnace within a preset time period and liquid level data corresponding to the gas flow data further includes:

[0028] Acquire gas input parameters of the VD furnace by means of a gas flow meter disposed at a gas input port of the VD furnace, and determine gas flow data in the VD furnace within a preset time period based on the gas input parameters;

[0029] The liquid level data in the VD furnace within a preset time period is obtained by a liquid level pressure regulating sensor arranged in the VD furnace.

[0030] Optionally, the method further comprises:

[0031] Acquire actual liquid level data corresponding to the target gas flow rate, use the target gas flow rate as the gas flow rate data, use the actual liquid level data as the liquid level data, and use the target gas flow rate and the actual liquid level data corresponding to the target gas flow rate as training data;

[0032] The target prediction model is trained based on the training data to obtain an updated model, and the updated model is fitted to obtain prediction line data of the gas flow data and the liquid level data.

[0033] In the second aspect, the present application provides a VD furnace liquid level control system, the system comprising: a computer, a field control box, an electrical cabinet, a display terminal and a VD furnace, wherein the computer is signal-connected to the field control box, the electrical cabinet and the display terminal, the field control box is signal-connected to the electrical cabinet and the liquid level pressure regulating sensor in the VD furnace respectively, the display terminal receives image data sent by a preset camera in the VD furnace, the electrical cabinet is gas-interactively connected to the VD furnace, and the computer implements the above-mentioned method when executing the computer program in the computer.

[0034] In a third aspect, the present application provides a terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above method when executing the computer program.

[0035] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0036] Compared with the prior art, the embodiments of the present application have the following beneficial effects: by obtaining the gas flow data and the liquid level data corresponding to the gas flow data in the VD furnace within a preset time period; training the preset deep learning model based on the gas flow data and the liquid level data to obtain the target prediction model; obtaining the predicted liquid level under the preset gas flow in the predicted blowing process based on the target prediction model, and determining the prediction line data of the preset gas flow and the predicted liquid level; obtaining the control information for the VD furnace liquid level, and determining the target gas flow corresponding to the control information according to the prediction line data and the control information, and completing the control of the VD furnace liquid level based on the target gas flow. The prediction of the liquid level rise height is achieved, the uniformity of the stirring of the molten steel is improved, and the liquid level height in the ladle is precisely controlled to avoid the uncertainty of the refiner's operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 It is a flow chart of a VD furnace liquid level control method provided by the present invention;

[0039] Figure 2 This is an application scenario diagram of a VD furnace liquid level control method provided by the present invention;

[0040] Figure 3 It is a structural schematic diagram of a VD furnace liquid level control system provided in one embodiment of the present application;

[0041] Figure 4 is a structural block diagram of an electronic device in an embodiment;

[0042] Figure 5 is a block diagram of a computer storage medium in another embodiment. DETAILED DESCRIPTION

[0043] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0044] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0045] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0046] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0047] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0048] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0049] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0050] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0051] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0052] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting" depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]" depending on the context.

[0053] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0054] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0055] like Figure 1 As shown, the embodiment of the present application provides a VD furnace liquid level control method, comprising the following steps:

[0056] S101, obtaining gas flow data and liquid level data corresponding to the gas flow data in a VD furnace within a preset time period;

[0057] S102, training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model;

[0058] S103, fitting the target prediction model to obtain prediction line data of the gas flow data and the liquid level data;

[0059] S104, obtaining control information for the VD furnace liquid level, and determining a target gas flow corresponding to the control information according to the prediction line data and the control information, and completing the control of the VD furnace liquid level based on the target gas flow.

[0060] By obtaining the gas flow data in the VD furnace within a preset time period and the liquid level data corresponding to the gas flow data; training the preset deep learning model based on the gas flow data and the liquid level data to obtain the target prediction model; obtaining the predicted liquid level under the preset gas flow in the predicted blowing process based on the target prediction model, and determining the prediction line data of the preset gas flow and the predicted liquid level; obtaining the control information for the VD furnace liquid level, and determining the target gas flow corresponding to the control information according to the prediction line data and the control information, and completing the control of the VD furnace liquid level based on the target gas flow. The prediction of the liquid level rise height is achieved, the uniformity of the stirring of the molten steel is improved, and the liquid level height in the ladle is precisely controlled to avoid the uncertainty of the refining operation.

[0061] In a possible implementation, the step of training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model includes:

[0062] Based on the gas flow data and the liquid level data, a gradient descent optimization algorithm is used to train the preset deep learning model until the training of the preset number of iterations is completed to obtain a target prediction model, wherein the mathematical representation of the gradient descent optimization algorithm is:

[0063]

[0064] Among them, x t represents the gradient at time step t, a represents the learning rate, Represents the loss function J(x) for parameter x t The gradient of Bt represents the small batch consisting of training data sample indexes. Represents x t-1 The gradient at Represents the mini-batch stochastic gradient obtained by repeated sampling.

[0065] Exemplarily, the principle formula of the gradient descent algorithm is calculated according to the relationship shown in Formula 1:

[0066]

[0067] In formula (1), x trepresents the gradient at time step t, a represents the learning rate, Represents the loss function J(x) for parameter x t The gradient of J(x t ) is the changing trend and speed at the current point, and the time step t is set to 0 before the iteration starts.

[0068] Based on formula (1), the small batch stochastic gradient descent algorithm formula 2 is selected.

[0069]

[0070] In the formula (2), t represents the time step, Bt represents the small batch consisting of the training data sample index, |B| represents the size of the batch, that is, the number of samples in the small batch, Represents x t-1 The gradient at Represents the mini-batch stochastic gradient obtained by repeated sampling.

[0071] Substituting formula (2) into formula (1) yields formula (3):

[0072]

[0073] For example, we use the built-in optm module and troch.nn module of pytorch. In the torch.nn module, we use nn.Linar to define a linear layer of a neural network, perform a linear transformation on the input features, and use the loss function Loss as the square difference between the model output and the target value. We use the stochastic gradient descent optimizer SGD provided by optim to update the parameters, and set the training iteration cycle to 19,000.

[0074] In a possible implementation manner, the step of acquiring control information for the VD furnace liquid level and determining a target gas flow rate corresponding to the control information according to the prediction line data and the control information includes:

[0075] Sending the predicted line data to a preset receiving end and obtaining user request information for the VD furnace liquid level;

[0076] The user request information is used as the control information, and a target gas flow rate corresponding to the control information is determined according to the prediction line data and the control information.

[0077] Exemplarily, the trained data model is fitted to generate a straight line (prediction line data), wherein the x-axis represents the blowing rate and the y-axis represents the liquid level.

[0078] Exemplarily, the prediction line data is sent to a preset receiving end so that professionals can estimate the current molten steel height and the degree of stirring of the molten steel according to the prediction line data, and further make real-time flow adjustments.

[0079] In a possible implementation manner, the step of acquiring control information for the VD furnace liquid level and determining a target gas flow rate corresponding to the control information according to the prediction line data and the control information further includes:

[0080] Acquiring threshold data of the VD furnace liquid level and actual liquid level data of the VD furnace liquid level;

[0081] Matching the threshold data with the actual liquid level data, if the threshold data is greater than the actual liquid level data, generating control information for the VD furnace liquid level based on the threshold data, and determining a target gas flow corresponding to the threshold data according to the prediction line data and the threshold data;

[0082] If the threshold data is less than or equal to the actual liquid level data, an alarm message is generated, and the alarm message is sent according to a preset rule to obtain feedback information for the alarm message;

[0083] The feedback information is used as control information for the VD furnace liquid level, and a target gas flow rate corresponding to the feedback information is determined according to the prediction line data and the control information.

[0084] For example, if there is no on-site worker to perform manual control, in order to ensure the best effect in the bottom blowing stage, under the premise of ensuring no overflow, the target gas flow corresponding to the control information is determined with the highest liquid level as the target to achieve intelligent control.

[0085] In a possible implementation manner, the step of acquiring gas flow data in the VD furnace within a preset time period and liquid level data corresponding to the gas flow data includes:

[0086] Acquire the operating parameters of the gas control device corresponding to the VD furnace within a preset time period, and determine the gas flow data in the VD furnace within the preset time period based on the operating parameters;

[0087] Acquire image data in the VD furnace within a preset time period, and enhance the image data using a DDE image algorithm to obtain a target image;

[0088] Liquid level data corresponding to the gas flow data within a preset time period is determined based on the target image.

[0089] In a possible implementation manner, the step of acquiring gas flow data in the VD furnace within a preset time period and liquid level data corresponding to the gas flow data further includes:

[0090] Acquire gas input parameters of the VD furnace by means of a gas flow meter disposed at a gas input port of the VD furnace, and determine gas flow data in the VD furnace within a preset time period based on the gas input parameters;

[0091] The liquid level data in the VD furnace within a preset time period is obtained by a liquid level pressure regulating sensor arranged in the VD furnace.

[0092] In a possible implementation, the method further includes:

[0093] Acquire actual liquid level data corresponding to the target gas flow rate, use the target gas flow rate as the gas flow rate data, use the actual liquid level data as the liquid level data, and use the target gas flow rate and the actual liquid level data corresponding to the target gas flow rate as training data;

[0094] The target prediction model is trained based on the training data to obtain an updated model, and the updated model is fitted to obtain the gas flow data and the liquid level data to obtain the prediction line data.

[0095] In a possible implementation, Figure 2 As shown, the application process of a VD furnace liquid level control method is: obtain the predicted value, control the program according to the predicted value, obtain the predicted information and flow signal, control the gas cabinet to adjust the flow according to the flow signal and the response result of the field control box to the predicted information, start blowing, and monitor through the camera, and display the detection information on the main display station (display). When the liquid level is higher than the liquid level threshold, an alarm is generated and the pressure is reduced. According to the pressure reduction information, it is determined whether the model in the control program needs to be updated. If it does not need to be updated, the blowing is terminated.

[0096] In a possible implementation, Figure 3 As shown, the application provides a VD furnace liquid level control system, the system comprising: a computer, a field control box, an electrical cabinet, a display terminal and a VD furnace, wherein the computer is signal-connected to the field control box, the electrical cabinet and the display terminal, the field control box is signal-connected to the electrical cabinet and the liquid level pressure regulating sensor in the VD furnace respectively, the display terminal receives image data sent by a preset camera in the VD furnace, the electrical cabinet is gas-interactively connected to the VD furnace, and the computer implements the above-mentioned method when executing the computer program in the computer.

[0097] For example, the camera in the furnace is used to monitor the fluctuation of the slag surface in the furnace in real time. Through the gradient descent optimization algorithm provided by the deep learning framework Pytorch, a large amount of data on the gas flow rate and the liquid level height in the ladle over a period of time in actual production is used as the data source for data training. Through the trained data, the corresponding molten steel height under different blowing rates can be predicted. By establishing a signal connection between the training program in the computer and the field control box, gas control cabinet, main display station, water-cooled camera, and regulating valve, the automatic control of the liquid level and the adaptive prediction of the liquid level function are realized.

[0098] Exemplarily, the VD furnace is also provided with a liquid level pressure regulating sensor. When the liquid level is too high, the liquid level pressure regulating sensor sends an alarm signal to the on-site control box, which converts the digital signal into a control signal and sends it to the computer. After receiving the signal, the computer converts the signal into a computer instruction and generates a current slag overflow alarm log. The log content includes the alarm time, blowing flow rate, and liquid level height. The target prediction model combines the maximum value of the liquid level for self-learning optimization, selects the best gas flow rate, and sends the flow information to the on-site control box for emergency flow adjustment to avoid molten steel overflow.

[0099] For example, in the middle stage of vacuum refining, as the blowing rate increases, the internal pressure of the vacuum ladle gradually increases, the stirring energy in the ladle increases, the molecules in the molten steel constantly collide violently, a deoxidation reaction occurs in the molten steel, and the emulsification zone on the liquid level quickly approaches the upper edge of the ladle. When the emulsification zone rises to reach the critical liquid level, the sensor above the critical liquid level is triggered, and the vacuum pressure regulating valve and the gas control cabinet are notified that air-breaking means are needed to press the slag to bring the liquid level back down.

[0100] In a possible implementation, Figure 4 As shown, an embodiment of the present application provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: obtaining gas flow data and liquid level data corresponding to the gas flow data in a VD furnace within a preset time period; training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model; fitting the target prediction model to obtain prediction line data of the gas flow data and the liquid level data; obtaining control information for the liquid level of the VD furnace, and determining a target gas flow corresponding to the control information based on the prediction line data and the control information, and completing control of the liquid level of the VD furnace based on the target gas flow.

[0101] In a possible implementation, Figure 5As shown, an embodiment of the present application provides a computer-readable storage medium 400, and the computer-readable storage medium stores a computer program 411. When the computer program is executed by a processor, the following steps are implemented: obtaining gas flow data in a VD furnace within a preset time period and liquid level data corresponding to the gas flow data; training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model; fitting the target prediction model to obtain prediction line data of the gas flow data and the liquid level data; obtaining control information for the liquid level of the VD furnace, and determining the target gas flow corresponding to the control information according to the prediction line data and the control information, and completing the control of the liquid level of the VD furnace based on the target gas flow.

[0102] If the integrated unit is implemented in the form of 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 present application implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the camera / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0103] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0104] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0105] In the embodiments provided in the present application, it should be understood that the disclosed devices / network equipment / terminal equipment and methods can be implemented in other ways. For example, the device / network equipment / terminal equipment embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0106] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0107] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A VD furnace liquid level control method, characterized in that: The following steps are involved: Acquire gas flow data and liquid level data corresponding to the gas flow data in the VD furnace within a preset time period; Training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model; Fitting the target prediction model to obtain prediction line data of the gas flow data and the liquid level data; The control information for the liquid level of the VD furnace is obtained, and the target gas flow corresponding to the control information is determined according to the prediction line data and the control information, and the control of the liquid level of the VD furnace is completed based on the target gas flow.

2. A VD furnace liquid level control method as claimed in claim 1, characterized in that: The step of training a preset deep learning model based on the gas flow data and the liquid level data to obtain a target prediction model includes: Based on the gas flow data and the liquid level data, a gradient descent optimization algorithm is used to train the preset deep learning model until the training of the preset number of iterations is completed to obtain a target prediction model, wherein the mathematical representation of the gradient descent optimization algorithm is: Among them, x t represents the gradient at time step t, a represents the learning rate, Represents the loss function J(x) for parameter x t The gradient of Bt represents the small batch consisting of training data sample indexes. Represents x t-1 The gradient at Represents the mini-batch stochastic gradient obtained by repeated sampling.

3. A VD furnace liquid level control method as claimed in claim 1, characterized in that: The step of obtaining control information for the VD furnace liquid level and determining a target gas flow rate corresponding to the control information according to the prediction line data and the control information includes: Sending the predicted line data to a preset receiving end and obtaining user request information for the VD furnace liquid level; The user request information is used as the control information, and a target gas flow rate corresponding to the control information is determined according to the prediction line data and the control information.

4. A VD furnace liquid level control method as claimed in claim 1, characterized in that: The step of obtaining control information for the VD furnace liquid level and determining a target gas flow rate corresponding to the control information according to the prediction line data and the control information also includes: Acquiring threshold data of the VD furnace liquid level and actual liquid level data of the VD furnace liquid level; Matching the threshold data with the actual liquid level data, if the threshold data is greater than the actual liquid level data, generating control information for the VD furnace liquid level based on the threshold data, and determining a target gas flow corresponding to the threshold data according to the prediction line data and the threshold data; If the threshold data is less than or equal to the actual liquid level data, an alarm message is generated, and the alarm message is sent according to a preset rule to obtain feedback information for the alarm message; The feedback information is used as control information for the VD furnace liquid level, and a target gas flow rate corresponding to the feedback information is determined according to the prediction line data and the control information.

5. A VD furnace liquid level control method as claimed in claim 1, characterized in that: The step of obtaining gas flow data in the VD furnace within a preset time period and liquid level data corresponding to the gas flow data comprises: Acquire the operating parameters of the gas control device corresponding to the VD furnace within a preset time period, and determine the gas flow data in the VD furnace within the preset time period based on the operating parameters; Acquire image data in the VD furnace within a preset time period, and enhance the image data using a DDE image algorithm to obtain a target image; Liquid level data corresponding to the gas flow data within a preset time period is determined based on the target image.

6. A VD furnace liquid level control method as claimed in claim 1, characterized in that: The step of obtaining gas flow data in the VD furnace within a preset time period and liquid level data corresponding to the gas flow data also includes: Acquire gas input parameters of the VD furnace by means of a gas flow meter disposed at a gas input port of the VD furnace, and determine gas flow data in the VD furnace within a preset time period based on the gas input parameters; The liquid level data in the VD furnace within a preset time period is obtained by a liquid level pressure regulating sensor arranged in the VD furnace.

7. A VD furnace liquid level control method as claimed in claim 1, characterized in that: The method further comprises: Acquire actual liquid level data corresponding to the target gas flow rate, use the target gas flow rate as the gas flow rate data, use the actual liquid level data as the liquid level data, and use the target gas flow rate and the actual liquid level data corresponding to the target gas flow rate as training data; The target prediction model is trained based on the training data to obtain an updated model, and the updated model is fitted to obtain prediction line data of the gas flow data and the liquid level data.

8. A VD furnace liquid level control system, characterized in that: The system includes: a computer, a field control box, an electrical cabinet, a display terminal and a VD furnace, wherein the computer is signal-connected to the field control box, the electrical cabinet and the display terminal, the field control box is signal-connected to the electrical cabinet and the liquid level pressure regulating sensor in the VD furnace respectively, the display terminal receives image data sent by a preset camera in the VD furnace, the electrical cabinet is gas-interactively connected to the VD furnace, and the computer implements the method according to any one of claims 1 to 7 when executing the computer program in the computer.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images