Control method and system for steel ladle jacking process, steel ladle jacking system and RH furnace

Abstract

The invention relates to the technical field of steelmaking, and discloses a steel ladle jacking process control method and system, a steel ladle jacking system and an RH furnace. The method comprises the following steps: acquiring an image in a vacuum chamber of the RH furnace in real time; the acquired images are processed so as to identify different steel ladle jacking states; in the process that the steel ladle jacking device is controlled to execute jacking operation according to the initial target stroke, the jacking state of the steel ladle is in the first state; a first target stroke is determined according to the current jacking height of the steel ladle in a first state, the ideal immersion depth of an immersion pipe of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion pipe to the diameter of the steel ladle and the ratio of the inner diameter of the immersion pipe to the diameter of the steel ladle, and the first state is that molten steel begins to appear in the immersion pipe; and the steel ladle jacking device is controlled to execute jacking operation according to the first target stroke. According to the RH furnace steel ladle jacking control system, RH furnace steel ladle jacking can be automatically controlled, and therefore the purpose of'whole-course one-key RH treatment 'is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of steelmaking, and in particular to a control method and system for a ladle lifting process, a ladle lifting system and an RH furnace. Background Art

[0002] When the RH furnace vacuum treats molten steel, it is generally necessary to lift the ladle first so that the immersion tube is immersed in the molten steel in the ladle, and then blow lifting gas from the riser pipe of the immersion tube, turn on the vacuum pump to evacuate, and use the lifting gas to expand due to heat and the pressure difference between the vacuum chamber and the atmosphere to press the molten steel into the vacuum chamber. After the molten steel enters the vacuum chamber, due to its own gravity, it flows back from the downcomer to the ladle. In this way, the molten steel forms a circulation in the ladle and the vacuum chamber, and in this process, the molten steel treatment process that cannot be achieved under normal pressure is completed. In this process, it is very important to control the depth of the immersion tube immersed in the molten steel, that is, the control of the lifting height of the ladle. If the immersion depth of the immersion tube is too large, it may burn the refractory material on the surface of the immersion tube, melt the lifting gas pipeline, and even cause a safety accident of molten steel overflowing from the ladle. If the immersion depth of the immersion tube is too small, the circulation of the molten steel is poor, resulting in a slag suction accident.

[0003] At present, the immersion depth of the immersion pipe is mainly controlled by manual observation and judgment. Although there are some technologies such as laser ranging and radar level meters that can assist workers in operation, these devices have low measurement accuracy in harsh environments and cannot achieve accurate detection, and thus cannot achieve automatic control.

[0004] Ladle lifting control has currently become a key constraint in improving the automation rate of RH furnaces in the industry. Summary of the invention

[0005] The object of the present invention is to provide a control method and system for a ladle lifting process, a ladle lifting system and an RH furnace, which realize automatic control of the ladle lifting of the RH furnace, thereby achieving the purpose of "full-process one-key RH processing".

[0006] In order to achieve the above-mentioned objectives, a first aspect of the present invention provides a control method for a ladle lifting process, the control method comprising: acquiring an image in a vacuum chamber of an RH furnace in real time; and processing the acquired image to identify different ladle lifting states; in a process of controlling a ladle lifting device to perform a lifting operation with an initial target stroke, when the ladle lifting state is in a first state, determining a first target stroke according to a current lifting height of the ladle in the first state, an ideal immersion depth of an immersion tube of the RH furnace, the pressure of the vacuum chamber, a ratio of an outer diameter of the immersion tube to a ladle diameter, and a ratio of an inner diameter of the immersion tube to a ladle diameter, wherein the first state is when molten steel begins to appear in the immersion tube; and controlling the ladle lifting device to perform a lifting operation with the first target stroke.

[0007] Preferably, the control method also includes: in the process of controlling the ladle lifting device to perform a lifting operation with the first target stroke, when the ladle lifting state is in the second state, determining the second target stroke according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion tube of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion tube to the ladle diameter, the ratio of the inner diameter of the immersion tube to the ladle diameter, and the ratio of the vacuum chamber diameter to the ladle diameter, wherein the second state is that the molten steel overflows the riser in the immersion tube; and controlling the ladle lifting device to perform a lifting operation with the second target stroke.

[0008] Preferably, the determining of the second target stroke comprises: determining the current lifting height H of the ladle in the first state according to the current lifting height H of the ladle in the first state; p1 , the ideal immersion depth H1 of the immersion tube of the RH furnace, the pressure p of the vacuum chamber vac , the ratio of the outer diameter of the immersion tube to the diameter of the ladle D1 / D0, the ratio of the inner diameter of the immersion tube to the diameter of the ladle D2 / D0, and the ratio of the vacuum chamber diameter to the ladle diameter D3 / D0, the second target stroke H is determined by the following formula sp2 :

[0009] H sp2 =H p1 +H1+(((p u -p vac )*γ H2O / γ Fe *h u -(L-H1))*D3 2 +2*(L-H1)*D2 2 -2*H1(D1 2 -D2 2 )) / D0 2 ,

[0010] Wherein, L is the height from the lower edge of the immersion tube to the bottom of the vacuum chamber; p u is atmospheric pressure; γ H20 is the density of water; γ Fe is the density of iron; h u It is the height of water column per unit atmospheric pressure.

[0011] Preferably, the determining of the first target stroke comprises: determining the first target stroke according to the current lifting height H of the ladle in the first state p1 , the ideal immersion depth H1 of the immersion tube of the RH furnace, the pressure p of the vacuum chamber vac , the ratio of the outer diameter of the immersion tube to the diameter of the ladle D1 / D0 and the ratio of the inner diameter of the immersion tube to the diameter of the ladle D2 / D0, the first target stroke H is determined by the following formula sp1 :

[0012] H sp1 =H p1 +H1+(L-H1)D2 2 / D0 2 +(p u -p vac )*γ H2O / γ Fe *h u *D2 2 / D0 2 -2H1(D1 2 -D2 2 ) / D0 2 ,

[0013] Wherein, L is the height from the lower edge of the immersion tube to the bottom of the vacuum chamber; p u is atmospheric pressure; γ H20 is the density of water; γ Fe is the density of iron; h u It is the height of water column per unit atmospheric pressure.

[0014] Preferably, the processing of the acquired image comprises: processing the acquired image according to a machine vision model.

[0015] Preferably, the processing of the acquired image according to the machine vision model includes: processing the image acquired during the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke through an adaptive threshold model, a first image detection model or a first image segmentation model to determine whether molten steel begins to appear in the immersion tube of the RH furnace and / or whether the molten steel overflows the riser in the immersion tube.

[0016] Preferably, the processing of the acquired image according to the machine vision model includes: processing the image acquired in the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke through a second image detection model or a second image segmentation model to identify the image area where the immersion tube of the RH furnace is located; and processing the image area where the immersion tube is located through a threshold model, a first image detection model or a first image segmentation model to determine whether molten steel begins to appear in the immersion tube and / or whether the molten steel overflows the riser in the immersion tube.

[0017] Preferably, the control method also includes: in the process of controlling the ladle lifting device to perform the lifting operation with the second target stroke, processing the image acquired in the process through the target detection model to determine whether the ladle lifting state is in a third state, wherein the third state is that the molten steel overflows the vacuum chamber.

[0018] Preferably, the control method further comprises: controlling the ladle lifting device to stop lifting when it is detected that the lifting height of the ladle reaches the second target stroke.

[0019] Through the above technical scheme, the present invention creatively controls the ladle lifting height to characterize the immersion depth of the RH furnace immersion tube into the molten steel, thereby realizing the automatic control of the RH furnace ladle lifting, thereby achieving the purpose of "one-key RH processing throughout the process"; improving the degree of automation of RH furnace production, accelerating the production rhythm, and improving production efficiency; reducing the labor intensity of production workers and protecting the health of workers from harsh working conditions.

[0020] A second aspect of the present invention provides a control system for a ladle lifting process, the control system comprising: an image acquisition device for acquiring images in a vacuum chamber of an RH furnace in real time; an image detection device for processing the acquired images to identify different ladle lifting states; and a control device for performing the following operations: in a process of controlling a ladle lifting device to perform a lifting operation with an initial target stroke, when the ladle lifting state is in a first state, determining a first target stroke according to a current lifting height of the ladle in the first state, an ideal immersion depth of an immersion tube of the RH furnace, the pressure of the vacuum chamber, a ratio of an outer diameter of the immersion tube to a ladle diameter, and a ratio of an inner diameter of the immersion tube to a ladle diameter, wherein the first state is when molten steel begins to appear in the immersion tube; and controlling the ladle lifting device to perform a lifting operation with the first target stroke.

[0021] Preferably, the control device is also used to perform the following operations: in the process of controlling the ladle lifting device to perform a lifting operation with the first target stroke, when the ladle lifting state is in the second state, the second target stroke is determined according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion tube of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion tube to the ladle diameter, the ratio of the inner diameter of the immersion tube to the ladle diameter, and the ratio of the vacuum chamber diameter to the ladle diameter, wherein the second state is that the molten steel overflows the riser in the immersion tube; and the ladle lifting device is controlled to perform a lifting operation with the second target stroke.

[0022] The specific details and benefits of the control system suitable for the ladle lifting process provided by the embodiment of the present invention can be found in the above description of the control method for the ladle lifting process, which will not be repeated here.

[0023] A third aspect of the present invention provides a ladle lifting system, which includes: a ladle lifting device for lifting the ladle; an image acquisition device for real-time acquisition of images in the vacuum chamber of the RH furnace; and a control system for the ladle lifting process, which is used to receive the image and control the operation of the ladle lifting device according to the image.

[0024] A fourth aspect of the present invention provides an RH furnace, comprising: a ladle; and the ladle lifting system.

[0025] A fifth aspect of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the control method of the ladle lifting process is implemented.

[0026] A sixth aspect of the present invention is a computer program product, comprising a computer program, which, when executed by a processor, implements the control method for the ladle lifting process.

[0027] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0029] Figure 1 is a flow chart of a method for controlling a ladle lifting process provided by an embodiment of the present invention;

[0030] Figure 2A-2D They are schematic diagrams of four key states of the ladle lifting process provided by an embodiment of the present invention;

[0031] Figure 3 It is a structural schematic diagram of a control system for lifting a ladle provided in one embodiment of the present invention;

[0032] Figure 4 It is a structural schematic diagram of a ladle lifting system provided by one embodiment of the present invention;

[0033] Figure 5 is a schematic diagram of a control process of ladle lifting provided by an embodiment of the present invention;

[0034] Figure 6 is a schematic structural diagram of a control system for a ladle lifting process provided by an embodiment of the present invention; and

[0035] Figure 7It is a structural schematic diagram of a ladle lifting system provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0036] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0037] Figure 1 FIG. 1 is a flow chart of a method for controlling a ladle lifting process provided by an embodiment of the present invention. Figure 1 As shown, the control method includes: step S101, acquiring an image in a vacuum chamber of an RH furnace in real time; step S102, processing the acquired image to identify different ladle lifting states; step S103, in a process of controlling the ladle lifting device to perform a lifting operation with an initial target stroke, when the ladle lifting state is in a first state, determining a first target stroke according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion tube of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion tube to the ladle diameter, and the ratio of the inner diameter of the immersion tube to the ladle diameter, wherein the first state is that molten steel begins to appear in the immersion tube; and step S104, controlling the ladle lifting device to perform a lifting operation with the first target stroke.

[0038] Since the liquid level and slag thickness of each furnace are different, there is no good technical means to achieve automatic and accurate control of the immersion depth of the immersion tube under high temperature and strong light radiation. From another perspective, the present invention controls the lifting height of the ladle to characterize the immersion depth of the RH furnace immersion tube into the molten steel, thereby realizing automatic control of the RH furnace ladle lifting, thereby achieving the purpose of "full-process one-button RH processing".

[0039] In one embodiment, the control method also includes: in the process of controlling the ladle lifting device to perform a lifting operation with the first target stroke, when the ladle lifting state is in a second state, determining the second target stroke according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion tube of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion tube to the ladle diameter, the ratio of the inner diameter of the immersion tube to the ladle diameter, and the ratio of the vacuum chamber diameter to the ladle diameter, wherein the second state is that the molten steel overflows the riser in the immersion tube; and controlling the ladle lifting device to perform a lifting operation with the second target stroke.

[0040] The core of the automatic control of ladle lifting in the embodiment of the present invention, namely the key state, is described in detail below.

[0041] The first is the preparation stage, such as Figure 2AAs shown, historical image data of the camera device 12 (e.g., camera) in the vacuum chamber 2 during the manual lifting of the ladle 1 is collected, and the key states that need to be detected during the lifting process are determined by analyzing the data. In this embodiment, four key states are determined (in this embodiment):

[0042] P0 (starting state): the ladle begins to rise. Figure 2A shown.

[0043] P1 (first state): molten steel begins to appear in the immersion tube (such as the riser 3 and / or the downcomer 4). Figure 2B shown.

[0044] P2 (second state): molten steel overflows the riser 3, as Figure 2C shown.

[0045] P3 (third state): molten steel overflows the vacuum chamber. Figure 2D shown.

[0046] Among them, the P0 state can be determined according to the ladle lifting start signal, and in order to detect the P1-P3 states, this embodiment can use a trained machine vision model to detect the state of molten steel entering the vacuum chamber as a basis for judgment.

[0047] The step of processing the acquired image includes: processing the acquired image according to a machine vision model.

[0048] In one embodiment, the processing of the acquired image according to the machine vision model includes: processing the image acquired during the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke through a threshold model, a first image detection model or a first image segmentation model to determine whether molten steel begins to appear in the immersion tube of the RH furnace and / or whether the molten steel overflows the riser in the immersion tube.

[0049] The image is directly processed by a threshold model (e.g., an adaptive threshold model based on pixel distribution), a first image detection model, or a first image segmentation model to identify the image region where the immersion tube is located, and then identify the molten steel level to determine whether molten steel begins to appear in the immersion tube and / or whether the molten steel overflows the riser in the immersion tube. However, this method requires processing the entire image, and the processing speed is slow.

[0050] In another embodiment, the processing of the acquired image according to the machine vision model includes: processing the image acquired during the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke through a second image detection model or a second image segmentation model to identify the image area where the immersion tube of the RH furnace is located; and processing the image area where the immersion tube is located through a threshold model, a first image detection model or a first image segmentation model to determine whether molten steel begins to appear in the immersion tube and / or whether the molten steel overflows the riser in the immersion tube.

[0051] like Figure 3 As shown, the image acquisition device 10 can acquire the image information in the vacuum chamber of the RH furnace by any existing method. For example, the image acquisition device 10 receives the image information in the vacuum chamber of the RH furnace transmitted by the image acquisition device. Figure 4 As shown, the image acquisition device may be a camera device 12 for capturing image information in the vacuum chamber of the RH furnace, converting the image information into digital signals for storage, and saving the digital signals in a hard disk recorder 14 .

[0052] like Figure 3 As shown, the image detection device 20 is used to detect the area of ​​interest in the acquired image, such as the immersion tube, the molten steel surface, the bottom of the vacuum chamber, etc., and further identify the lifting state of the ladle based on the area of ​​interest.

[0053] Specifically, the images are screened, analyzed and annotated, and a machine vision model (e.g., an adaptive threshold model based on pixel distribution, an image detection model, an image segmentation model (e.g., a semantic segmentation model), an object detection model, etc.) is trained to identify key states (e.g., a first state, a second state, a third state, etc.). The model parameters of the second image detection model are different from those of the first image detection model, and the model parameters of the second image segmentation model are different from those of the first image segmentation model.

[0054] Semantic segmentation model (M1): a semantic segmentation model for detecting the immersion tube area in the vacuum chamber (any commonly used semantic segmentation model applied to images can be used, and this embodiment can use a lightweight SegFormer model, which can achieve good results).

[0055] Adaptive threshold model based on pixel distribution (M2): An image processing method that directly sets an adaptive pixel threshold to detect the molten steel surface area.

[0056] Target detection model (M3): A target detection model used to detect whether the molten steel level overflows the vacuum chamber.

[0057] like Figure 3 As shown, the control device 30 can be used to execute the above automatic control process, which can be an industrial computer 32, such as Figure 4 As shown, for example, the RH furnace industrial video network and the RH furnace industrial control network can be connected. In addition, an image detection device (such as image detection software) and ladle lifting automatic control software are deployed in the industrial computer. The software obtains images of the camera device 12 and the hard disk recorder 14 in the RH vacuum tank, and obtains the RH furnace body control device (such as PLC) 200 (such as Figure 4 The lifting height and immersion depth of the ladle (as shown) are used to verify the results of various embodiments of the present invention, so that each model can be further modified. Offline, the machine vision model used in the image detection device is trained to detect key states; when running online, the key state of the ladle lifting is identified. When a specific state occurs, the set value of the ladle lifting stroke (i.e., the target stroke) is calculated. After a certain period of time, the ladle is lifted to the set value and the ladle height is maintained, and the vacuum treatment operation is started.

[0058] Based on the above embodiments, the first target stroke is determined as follows: according to the current lifting height H of the ladle in the first state, p1 , the ideal immersion depth H1 of the immersion tube of the RH furnace, the pressure p of the vacuum chamber vac , the ratio of the outer diameter of the immersion tube to the diameter of the ladle D1 / D0 and the ratio of the inner diameter of the immersion tube to the diameter of the ladle D2 / D0, the first target stroke H is determined by the following formula sp1 :

[0059] H sp1 =H p1 +H1+(L-H1)D2 2 / D0 2 +(p u -p vac )*γ H2O / γ Fe *h u *D2 2 / D0 2 -2H1(D1 2 -D2 2 ) / D0 2 , (1)

[0060] Wherein, L is the height from the lower edge of the immersion tube to the bottom of the vacuum chamber; p u is atmospheric pressure; γ H20 is the density of water; γ Fe is the density of iron; h u It is the height of water column per unit atmospheric pressure.

[0061] Based on the above embodiments, the second target stroke is determined as follows: according to the current lifting height H of the ladle in the first state, p1, the ideal immersion depth H1 of the immersion tube of the RH furnace, the pressure p of the vacuum chamber vac , the ratio of the outer diameter of the immersion tube to the diameter of the ladle D1 / D0, the ratio of the inner diameter of the immersion tube to the diameter of the ladle D2 / D0, and the ratio of the vacuum chamber diameter to the ladle diameter D3 / D0, the second target stroke H is determined by the following formula sp2 :

[0062] H sp2 =H p1 +H1+(((p u -p vac )*γ H2O / γ Fe *h u -(L-H1))*D3 2 +2*(L-H1)*D2 2 -2*H1(D1 2 -D2 2 )) / D0 2 , (2)

[0063] Wherein, L is the height from the lower edge of the immersion tube to the bottom of the vacuum chamber; p u is atmospheric pressure; γ H20 is the density of water; γ Fe is the density of iron; h u It is the height of water column per unit atmospheric pressure.

[0064] Specifically, the control process of the ladle lifting process may include the following contents.

[0065] 1: After receiving the ladle lifting start signal sent by the RH furnace body control device (such as PLC), the ladle lifting state is in the initial state (i.e., P0 state), and the real-time image of the camera device (such as an industrial camera) in the vacuum chamber is obtained, and the initial target stroke H is used. default To control the target, the ladle lifting device starts to perform the lifting operation. Among them, the initial target stroke H default It can be set in advance according to the actual situation. At the same time, the first image is taken as P0, and the current time t0 and the ladle lifting height h0 can also be recorded for further correction of the model.

[0066] 2: Use the M1 method to detect the immersion tube area (the immersion tube is divided into an ascending tube and a descending tube, which can be distinguished manually or automatically by the system), and then use the M2 method to detect the molten steel level in the immersion tube area in real time.

[0067] 3: As the ladle is lifted (i.e. at the initial target stroke H defaultIn order to control the target ladle lifting device, when the ladle lifting device starts to perform the lifting operation, when the M2 method detects that molten steel begins to appear in the immersion tube area, it is determined that the ladle lifting state reaches the P1 state, and the new target stroke (i.e., the first target stroke H) is determined according to formula (1). sp1 ), and the first target stroke H sp1 To control the ladle lifting device, perform the lifting operation. At the same time, the image at this time is taken as P1, and the current time t1, ladle lifting height H are recorded. p1 , for further correction of the model. Image P1 can be used to train the M1 and M2 models.

[0068] 4: As the ladle is lifted (based on the first target stroke H sp1 In the process of controlling the ladle lifting device to perform the lifting operation for the control target), when the M2 method detects that the molten steel level in the riser area is basically full, it is determined that the ladle lifting state reaches the P2 state, and a new target stroke (i.e., the second target stroke H) is determined according to formula (2). sp1 ), and the second target stroke H sp2 To control the ladle lifting device, perform the lifting operation. At the same time, the image at this time is taken as P2, and the current time t2, ladle lifting height H are recorded. p2 , the immersion tube depth H1, for further correction of the model. The image P2 can be used to train the M1 and M2 models.

[0069] Based on the above embodiments, the control method also includes: in the process of controlling the ladle lifting device to perform the lifting operation with the second target stroke, processing the image acquired in the process through the target detection model to determine whether the ladle lifting state is in a third state, wherein the third state is that the molten steel overflows the vacuum chamber.

[0070] The target detection model (M3) may be a commonly used target detection model applied to images. In this embodiment, the Yolox model may be used, which can achieve good results.

[0071] On the basis of the above-mentioned embodiments, the control method further comprises: when it is detected that the lifting height of the ladle reaches the second target stroke, controlling the ladle lifting device to stop lifting.

[0072] That is to say, the lifting operation is initially controlled by the initial target stroke, and the initial target stroke can be determined based on actual experience. However, since in the process of ladle lifting, once the first state is reached, the immersion depth is affected by the ladle lifting speed and the vacuum chamber pressure drop rate, if the initial target stroke is always used as the set value to control the ladle to perform the lifting operation, if the lifting speed is too fast and the vacuum chamber pressure drops slowly, it may cause the immersion depth to be too deep and cause an accident. Therefore, in this embodiment, the ladle lifting is gradually controlled to match the immersion depth of the dip tube, the ladle lifting speed and the vacuum chamber pressure drop rate to obtain the best lifting effect.

[0073] Specifically, the control process of the ladle lifting process may include the following contents.

[0074] 1: After receiving the ladle lifting start signal sent by the RH furnace body PLC, the ladle lifting state is in the initial state (i.e., P0 state), and the real-time image of the industrial camera in the vacuum chamber is obtained, and the initial target stroke H is used. default To control the target, the ladle lifting device starts to perform the lifting operation. Among them, the initial target stroke H default It can be set in advance according to the actual situation. At the same time, the first image is taken as P0, and the current time t0 and the ladle lifting height h0 can also be recorded.

[0075] 2: Use the M1 method to detect the immersion tube area (the immersion tube is divided into an ascending tube and a descending tube, which can be distinguished manually or automatically by the system), and then use the M2 method to detect the molten steel level in the immersion tube area in real time.

[0076] 3: As the ladle is lifted (i.e. at the initial target stroke H default In order to control the target ladle lifting device, when the ladle lifting device starts to perform the lifting operation, when the M2 method detects that molten steel begins to appear in the immersion tube area, it is determined that the ladle lifting state reaches the P1 state, and the new target stroke (i.e., the first target stroke H) is determined according to formula (1). sp1 ), and the first target stroke H sp1 To control the ladle lifting device, perform the lifting operation. At the same time, the image at this time is taken as P1, and the current time t1, ladle lifting height H are recorded. p1 . Image P1 can be used to train models M1 and M2.

[0077] 4: As the ladle is lifted (based on the first target stroke H sp1 In the process of controlling the ladle lifting device to perform the lifting operation for the control target), when the M2 method detects that the molten steel level in the riser area is basically full, it is determined that the ladle lifting state reaches the P2 state, and a new target stroke (i.e., the second target stroke H) is determined according to formula (2).sp1 ), and the second target stroke H sp2 To control the ladle lifting device, the lifting operation is performed. At the same time, the image at this time is taken as P2, and the current time t2, ladle lifting height H are recorded. p2 , the immersion tube depth H1. The image P2 can be used to train the M1 and M2 models.

[0078] 5: Use the M3 method to detect in real time whether the molten steel level overflows the vacuum chamber.

[0079] 6: As the ladle is lifted (with the second target stroke H sp2 In order to control the ladle lifting device to perform the lifting operation, the M3 method is used to detect that the molten steel level basically overflows the vacuum chamber, and the ladle lifting state is determined to reach the P3 state. At the same time, the image at this time is taken as P3, and the current time t3, the ladle lifting height H are recorded. p3 , the immersion tube depth H1. The image P3 can be used to train the M3 model.

[0080] 7: When the jacking is in place (i.e. the jacking height reaches the second target stroke H sp2 ) after which the lifting stops.

[0081] Specifically, Figure 3 Take the ladle lifting control system as an example to illustrate.

[0082] The data acquisition device 16 is used to acquire the lifting height of the ladle and the immersion depth of the immersion pipe during the lifting process of the ladle of the RH furnace.

[0083] The image detection device 20 is used to detect the areas that need to be acquired in the image, such as the immersion tube, the molten steel surface, the bottom of the vacuum chamber, etc., to identify the lifting state of the ladle (that is, to identify the key state of the video in the vacuum chamber).

[0084] The control device 30 is used to calculate the target lifting stroke of the ladle in the ladle lifting state in the automatic control mode, and output a control instruction, wherein the control instruction includes the calculated target lifting stroke of the ladle.

[0085] The output device 40 is used to receive the control command (including the ladle lifting target stroke) from the control device 30, and output the control command and the ladle lifting target stroke to the actuator (such as the ladle lifting device) to control the ladle lifting operation. Thus, the ladle lifting device can be automatically controlled to perform the lifting process according to the control command.

[0086] Specifically, Figure 5 Take the example to illustrate the control process of ladle lifting.

[0087] 1: Install a camera in the vacuum tank of the RH furnace to collect images inside the vacuum chamber of the RH furnace.

[0088] 2: Get the image inside the vacuum chamber when the ladle is manually lifted.

[0089] In addition, the lifting height and immersion depth of the ladle at the corresponding moment can also be obtained.

[0090] 3: Filter, analyze and annotate video images and train machine vision models to identify key states.

[0091] In addition, the actual stroke of the ladle lifting can be determined, and the machine vision model can be further corrected by comparing the actual stroke with the target stroke.

[0092] 4: Enable the automatic control mode of ladle lifting and start lifting the ladle.

[0093] 5: Obtain images of the vacuum chamber in real time when the ladle is lifted and identify key states.

[0094] In addition, the corresponding lifting height etc. can also be recorded.

[0095] 6: When the critical state in the ladle lifting process is identified, the new target stroke Hsp required for the ladle lifting is calculated.

[0096] 7: When the ladle lifting height reaches the new target stroke, stop lifting the ladle and maintain the height, and the molten steel is vacuum treated.

[0097] 8: Until the processing is completed, control the ladle to descend.

[0098] It should be noted that the target stroke (for example, the initial target stroke, the first target stroke, the second target stroke, etc.) in each embodiment of the present invention refers to the entire stroke from the start of the lifting of the ladle from the starting state (P0) to the completion of the lifting action.

[0099] The present invention obtains an image in a vacuum chamber, and provides a method and system for automatically controlling the lifting of a ladle by identifying various parts and the molten steel level in the vacuum chamber. The method and system can automatically identify various stages and states in the process of lifting the ladle, and automatically stop lifting the ladle after the ladle is lifted to a suitable position. The entire process does not require human intervention, and can be applied to various conditions and environments such as different molten steel level heights, slag layer thicknesses, and molten steel temperatures. The system has improved applicability and robustness, and is a basic technology for "full-process one-key RH".

[0100] In summary, the present invention creatively controls the ladle lifting height to characterize the immersion depth of the RH furnace immersion tube into the molten steel, thereby realizing automatic control of the RH furnace ladle lifting, thereby achieving the purpose of "one-key RH processing throughout the process"; improving the degree of automation of RH furnace production, accelerating the production rhythm, and improving production efficiency; reducing the labor intensity of production workers and protecting the workers' health from harsh working conditions.

[0101] Figure 6 FIG. 1 is a schematic diagram of the structure of a control system for a ladle lifting process provided by an embodiment of the present invention. Figure 6 As shown, the control system 100 may include: an image acquisition device 10, for acquiring images in the vacuum chamber of the RH furnace in real time; an image detection device 20, for processing the acquired images to identify different ladle lifting states; and a control device 30, for performing the following operations: in the process of controlling the ladle lifting device to perform a lifting operation with an initial target stroke, when the ladle lifting state is in a first state, determining a first target stroke according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion tube of the RH furnace, the ratio of the outer diameter of the immersion tube to the ladle diameter, and the ratio of the inner diameter of the immersion tube to the ladle diameter, wherein the first state is that molten steel begins to appear in the immersion tube; and controlling the ladle lifting device to perform a lifting operation with the first target stroke.

[0102] In one embodiment, the control device is also used to perform the following operations: in the process of controlling the ladle lifting device to perform a lifting operation with the first target stroke, when the ladle lifting state is in a second state, the second target stroke is determined according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion tube of the RH furnace, the ratio of the outer diameter of the immersion tube to the ladle diameter, the ratio of the inner diameter of the immersion tube to the ladle diameter, and the ratio of the vacuum chamber diameter to the ladle diameter, wherein the second state is that the molten steel overflows the riser in the immersion tube; and the ladle lifting device is controlled to perform a lifting operation with the second target stroke.

[0103] The specific details and benefits of the control system suitable for the ladle lifting process provided by the embodiment of the present invention can be found in the above description of the control method for the ladle lifting process, which will not be repeated here.

[0104] An embodiment of the present invention provides a ladle lifting system, the ladle lifting system comprising: a ladle lifting device 50, used to lift the ladle; an image acquisition device (such as a camera 12), used to acquire images in the vacuum chamber of the RH furnace in real time; and a control system 100 for the ladle lifting process, used to receive the images and control the operation of the ladle lifting device 50 according to the images, such as Figure 7 shown.

[0105] In this embodiment, the ladle lifting device 50 is directly controlled by the ladle lifting process control system 100. In another embodiment, the ladle lifting system may further include an RH furnace body control device 200, which receives control instructions from the ladle lifting process control system 100, and then controls the ladle lifting device 50 to perform lifting operations according to the received control instructions. In other words, the ladle lifting device 50 is indirectly controlled by the ladle lifting process control system 100.

[0106] Wherein, the image acquisition device may be a camera device 12, such as Figure 7 shown.

[0107] Specifically, Figure 7 Take the ladle lifting system in the RH furnace as an example to illustrate.

[0108] The camera device 12 is used to capture image information in the vacuum chamber of the RH furnace and convert it into digital signals for storage.

[0109] The control system 100 is used to identify key image states and automatically control the lifting of the ladle.

[0110] The RH furnace body control device 200 is used to receive control instructions (including the ladle lifting target stroke) from the control system 100 and feed back the execution results.

[0111] The ladle lifting device 50 is used to carry and lift the ladle.

[0112] Among them, the camera device 12, the RH furnace body control device 200 and the control system 100 are connected through a communication network, and the RH furnace body control device 200 and the ladle lifting device 50 are connected to exchange data with each other.

[0113] An embodiment of the present invention provides an RH furnace, comprising: a ladle; and the ladle lifting system.

[0114] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the control method of the ladle lifting process is implemented.

[0115] An embodiment of the present invention provides a computer program product, including a computer program, which implements the control method of the ladle lifting process when executed by a processor.

[0116] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, a variety of simple modifications can be made to the technical solution of the present invention, and these simple modifications all belong to the protection scope of the present invention.

[0117] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations.

[0118] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a single-chip microcomputer, a chip or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0119] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0120] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0121] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the present invention, they should also be regarded as the contents disclosed by the present invention.

Claims

1. A method for controlling a ladle lifting process, characterized in that: The control method comprises: Acquire real-time images of the vacuum chamber of the RH furnace; and Processing the acquired images to identify different ladle lifting states; In the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke, when the ladle lifting state is in the first state, the first target stroke is determined according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion pipe of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion pipe to the ladle diameter, and the ratio of the inner diameter of the immersion pipe to the ladle diameter, wherein the first state is that molten steel begins to appear in the immersion pipe; and The ladle lifting device is controlled to perform a lifting operation according to the first target stroke.

2. The control method according to claim 1, characterized in that: The control method further comprises: In the process of controlling the ladle lifting device to perform the lifting operation with the first target stroke, when the ladle lifting state is in the second state, the second target stroke is determined according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion pipe of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion pipe to the diameter of the ladle, the ratio of the inner diameter of the immersion pipe to the diameter of the ladle, and the ratio of the vacuum chamber diameter to the diameter of the ladle, wherein the second state is that the molten steel overflows the riser in the immersion pipe; and The ladle lifting device is controlled to perform a lifting operation according to the second target stroke.

3. The control method according to claim 2, characterized in that: Determining the second target travel distance includes: According to the current lifting height H of the ladle in the first state p1 , the ideal immersion depth H1 of the immersion tube of the RH furnace, the pressure p of the vacuum chamber vac , the ratio of the outer diameter of the immersion tube to the diameter of the ladle D1 / D0, the ratio of the inner diameter of the immersion tube to the diameter of the ladle D2 / D0, and the ratio of the vacuum chamber diameter to the ladle diameter D3 / D0, the second target stroke H is determined by the following formula sp2 : H sp2 =H p1 +H1+(((p u -p vac )*γ H2O / gamma Fe *h u -(L-H1))*D3 2 +2*(L-H1)*D2 2 -2*H1(D1 2 -D2 2 )) / D0 2 , Wherein, L is the height from the lower edge of the immersion tube to the bottom of the vacuum chamber; u is atmospheric pressure; γ H20 is the density of water; γ Fe is the density of iron; h u It is the height of water column per unit atmospheric pressure.

4. The control method according to claim 2, characterized in that: The control method also includes: in the process of controlling the ladle lifting device to perform a lifting operation with the second target stroke, processing the image acquired in the process by the target detection model to determine whether the ladle lifting state is in a third state, wherein the third state is that the molten steel overflows the vacuum chamber.

5. The control method according to claim 2, characterized in that: The control method further comprises: When it is detected that the lifting height of the ladle reaches the second target stroke, the ladle lifting device is controlled to stop lifting.

6. The control method according to claim 1, characterized in that: Determining the first target travel distance includes: According to the current lifting height H of the ladle in the first state p1 , the ideal immersion depth H1 of the immersion tube of the RH furnace, the pressure p of the vacuum chamber vac , the ratio of the outer diameter of the immersion tube to the diameter of the ladle D1 / D0 and the ratio of the inner diameter of the immersion tube to the diameter of the ladle D2 / D0, the first target stroke H is determined by the following formula sp1 :H sp1 =H p1 +H1+(L-H1)D2 2 / D0 2 +(p u -p vac )*γ H2O / γ Fe *h u *D2 2 / D0 2 -2H1(D1 2 -D2 2 ) / D0 2 , Wherein, L is the height from the lower edge of the immersion tube to the bottom of the vacuum chamber; u is atmospheric pressure; γ H20 is the density of water; γ Fe is the density of iron; h u It is the height of water column per unit atmospheric pressure.

7. The control method according to any one of claims 1 to 6, characterized in that: The processing of the acquired image includes: processing the acquired image according to a machine vision model.

8. The control method according to claim 7, characterized in that: The processing of the acquired image according to the machine vision model comprises: The images acquired during the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke are processed by a threshold model, a first image detection model or a first image segmentation model to determine whether molten steel begins to appear in the immersion tube of the RH furnace and / or whether the molten steel overflows the riser in the immersion tube.

9. The control method according to claim 7, characterized in that: The processing of the acquired image according to the machine vision model comprises: Processing an image acquired during the process of controlling the ladle lifting device to perform a lifting operation with the initial target stroke by a second image detection model or a second image segmentation model to identify an image region where the immersion tube of the RH furnace is located; and The image area where the immersion tube is located is processed by a threshold model, a first image detection model or a first image segmentation model to determine whether molten steel begins to appear in the immersion tube and / or whether the molten steel overflows the riser in the immersion tube.

10. A control system for a ladle lifting process, characterized in that: The control system comprises: An image acquisition device, used for acquiring an image in the vacuum chamber of the RH furnace in real time; An image detection device, used to process the acquired image to identify different ladle lifting states; and Control means for performing the following operations: In the process of controlling the ladle lifting device to perform the lifting operation with the initial target stroke, when the ladle lifting state is in the first state, the first target stroke is determined according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion pipe of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion pipe to the ladle diameter, and the ratio of the inner diameter of the immersion pipe to the ladle diameter, wherein the first state is that molten steel begins to appear in the immersion pipe; and The ladle lifting device is controlled to perform a lifting operation according to the first target stroke.

11. The control system according to claim 10, characterized in that: The control device is also used to perform the following operations: In the process of controlling the ladle lifting device to perform the lifting operation with the first target stroke, when the ladle lifting state is in the second state, the second target stroke is determined according to the current lifting height of the ladle in the first state, the ideal immersion depth of the immersion pipe of the RH furnace, the pressure of the vacuum chamber, the ratio of the outer diameter of the immersion pipe to the diameter of the ladle, the ratio of the inner diameter of the immersion pipe to the diameter of the ladle, and the ratio of the vacuum chamber diameter to the diameter of the ladle, wherein the second state is that the molten steel overflows the riser in the immersion pipe; and The ladle lifting device is controlled to perform a lifting operation according to the second target stroke.

12. A ladle lifting system, characterized in that: The ladle lifting system comprises: Ladle lifting device, used for lifting the ladle; An image acquisition device, used for acquiring images in the vacuum chamber of the RH furnace in real time; and The control system for the ladle lifting process according to claim 10 or 11 is used to receive the image and control the operation of the ladle lifting device according to the image.

13. An RH furnace, characterized in that: The RH furnace comprises: a ladle; and the ladle lifting system according to claim 12.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for controlling the ladle lifting process according to any one of claims 1 to 8.

15. A computer program product, characterized in that It comprises a computer program, which, when executed by a processor, implements the control method for the ladle lifting process according to any one of claims 1-8.

Images