Apparatus and method for feedforward control of continuous casting and rolling based on physical properties

Abstract

A method for feedforward control of continuous casting and rolling based on physical properties, according to an embodiment, may comprise the steps of: sensing the content of tramp elements contained in a scrap by applying a current or a magnetic field to the scrap through an electromagnetic sensor; determining an alloy composition ratio of the scrap, as an analysis result, using a deep learning-based image analysis model; and controlling a reduction ratio and a rolling speed of a continuous casting and rolling process on the basis of the content of the tramp elements in the scrap and the alloy composition ratio of the scrap.

Description

Material property-based feedforward continuous rolling control device and method

[0001] The present invention relates to a feed-forward continuous rolling control device and method adaptive to slab components or physical properties.

[0002] Traditional steel production methods have adopted a method of reducing and melting iron ore in a blast furnace before refining it in a Basic Oxygen Furnace (BOF). The molten iron produced in a blast furnace is called pig iron. Pig iron is transferred to a BOF and refined into molten steel. This BOF-BF method adopts a method of reducing iron ore using coke, specifically carbon from the coke or carbon monoxide generated from the coke; however, the use of coke inevitably causes large amounts of CO or CO2 emissions.

[0003] To reduce such carbon emissions, the Electric Arc Furnace (EAF) is being recently developed and introduced. An EAF refers to an electric furnace that melts iron sources (e.g., cold iron sources) using an arc generated from electrode rods. The EAF accepts Hot Briquetted Iron (HBI), scrap, and other auxiliary materials and melts them using the arc from the electrode rods.

[0004] One embodiment of the present invention aims to provide a material property-based feed-forward continuous rolling control apparatus and method for controlling the reduction rate and / or rolling speed of a slab through deep learning-based feed-forward control adaptive to the slab components and / or material properties.

[0005] Among the embodiments, the material property-based feedforward continuous rolling control method may include the steps of: sensing the content of a tramp element contained in the scrap by applying a current or magnetic field to the scrap through an electromagnetic sensor; determining the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model; and controlling the reduction rate and rolling speed of the continuous rolling and rolling processes based on the content of the tramp element in the scrap and the alloy composition ratio of the scrap.

[0006] The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model may include the step of determining the alloy composition ratio of the scrap using a CNN-based deep learning model.

[0007] The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model may include the step of pre-storing data on the alloy composition ratio for each scrap type in a database, the step of estimating the type of scrap through the deep learning-based image analysis model when an image is acquired through a camera, and the step of determining the alloy composition ratio of the scrap by comparing the estimated type of scrap with the alloy composition ratio for each scrap type in the database.

[0008] The method further includes a step of controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel, and the step of controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel may include a step of sampling a portion of the molten steel when tapping the scrap-based molten steel, a step of generating the sampling analysis results by detecting at least one of the temperature or alloy composition of the sampled molten steel through a sensor, and a step of controlling the reduction rate and rolling speed based on the sampling analysis results.

[0009] A first step of controlling the reduction rate and rolling speed of the continuous casting and rolling process based on the content of tramp elements in the scrap and the alloy composition ratio of the scrap obtained using the deep learning-based image analysis model and the electromagnetic sensor, and a second step of controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel can be performed selectively from each other.

[0010] A physical property-based feed-forward continuous rolling control method according to one embodiment may include the steps of: sensing the content of a tramp element contained in the scrap by applying a current or magnetic field to the scrap through an electromagnetic sensor; determining the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model; sampling a portion of the molten steel when tapping the molten steel based on the scrap; generating a sampling analysis result by detecting at least one of the temperature or alloy composition of the sampled molten steel through a sensor; verifying the analysis result of the deep learning-based image analysis model based on the sampling analysis result; and controlling the reduction rate and rolling speed of the continuous rolling and rolling process based on at least one of the content of the tramp element, the analysis result, and the sampling analysis result.

[0011] The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model may include the step of determining the alloy composition ratio of the scrap using a CNN-based deep learning model.

[0012] The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model may include the step of pre-storing data on the alloy composition ratio for each scrap type in a database, the step of estimating the type of scrap through the deep learning-based image analysis model when an image is acquired through a camera, and the step of determining the alloy composition ratio of the scrap by comparing the estimated type of scrap with the alloy composition ratio for each scrap type in the database.

[0013] The step of verifying the analysis results of the deep learning-based image analysis model based on the above sampling analysis results may be performed at regular intervals or selectively under the control of the user.

[0014] Among the embodiments, the physical property-based feedforward continuous rolling control device is a physical property-based feedforward continuous rolling control device that controls the continuous casting and rolling processes through feedforward control adaptive to the physical properties of a slab produced by electric furnace operation by executing program code loaded into one or more memory devices through one or more processors, wherein the program code is executed to sense the content of tramp elements contained in the scrap by applying a current or magnetic field to the scrap through an electromagnetic sensor, determines the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model, and controls the reduction rate and rolling speed of the continuous casting and rolling processes based on the content of tramp elements in the scrap and the alloy composition ratio of the scrap.

[0015] Determining the alloy composition ratio of the scrap through the deep learning-based image analysis model may include determining the alloy composition ratio of the scrap using a CNN-based deep learning model.

[0016] Determining the alloy composition ratio of the scrap through the deep learning-based image analysis model may include pre-storing data on the alloy composition ratio for each scrap type in a database, and when an image is acquired through a camera, estimating the type of scrap through the deep learning-based image analysis model, and comparing the estimated type of scrap with the alloy composition ratio for each scrap type in the database to determine the alloy composition ratio of the scrap.

[0017] The method further includes controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel, and controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel may include sampling a portion of the molten steel when tapping the scrap-based molten steel, detecting at least one of the temperature or alloy composition of the sampled molten steel through a sensor to generate the sampling analysis results, and controlling the reduction rate and rolling speed based on the sampling analysis results.

[0018] Controlling the reduction rate and rolling speed of the continuous casting and rolling process based on the content of tramp elements in the scrap and the alloy composition ratio of the scrap obtained using the deep learning-based image analysis model and the electromagnetic sensor, and controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel can be performed selectively from each other.

[0019] Controlling the reduction rate and rolling speed of the continuous casting and rolling processes through the sampling analysis results of the scrap-based molten steel may further include verifying the analysis results of the deep learning-based image analysis model based on the sampling analysis results.

[0020] A material property-based feed-forward continuous rolling control device and method according to one embodiment of the present invention can control the reduction rate and / or rolling speed of a slab through deep learning-based feed-forward control adaptive to slab components and / or material properties, and enable high-quality, high-efficiency metal processing.

[0021] FIG. 1 schematically shows a material property-based feed-forward continuous rolling control system according to one embodiment of the present invention.

[0022] FIG. 2 is a block diagram of a physical property-based feed-forward continuous rolling control device according to one embodiment of the present invention.

[0023] FIG. 3 is a flowchart of a material property-based feed-forward continuous rolling control method according to one embodiment of the present invention.

[0024] FIG. 4 is a flowchart of a material property-based feed-forward continuous rolling control method according to one embodiment of the present invention.

[0025] FIG. 5 is a drawing for explaining a computing device according to an embodiment of the present invention.

[0026] A material property-based feed-forward continuous rolling control method can sense the content of tramp elements contained in the scrap by applying an electric current or magnetic field to the scrap through an electromagnetic sensor, determine the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model, and control the reduction rate and rolling speed of the continuous rolling and rolling processes based on the content of tramp elements in the scrap and the alloy composition ratio of the scrap.

[0027] Embodiments of the present invention are described below with reference to the attached drawings so that those skilled in the art can easily implement them. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0028] Throughout the specification and claims, when a part is described as "comprising" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components. Terms including ordinal numbers, such as first, second, etc., may be used to describe various components, but said components are not limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another.

[0029] Terms such as "...part," "...unit," and "module" as used in the specification may refer to a unit capable of processing at least one function or operation described in this specification, and may be implemented as hardware or a circuit, software, or a combination of hardware or a circuit and software.

[0030] In addition, at least some of the components or functions of the material property-based feed-forward continuous rolling control device and method according to the embodiments described below may be implemented as a program or software, and the program or software may be stored on a computer-readable medium.

[0031] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0032] FIG. 1 schematically shows a material property-based feed-forward continuous rolling control system according to one embodiment of the present invention.

[0033] Molten steel produced by electric furnace operation contains a large amount of tramp elements leached from scrap.

[0034] Tramp elements include, for example, Cu, Ni, Cr, etc. These metallic elements are contained in large quantities in scrap, so molten steel containing them is immediately manufactured into slabs.

[0035] This has different properties from slabs produced by blast furnace operation. More specifically, slabs produced by electric furnace operation may have relatively higher strength or higher brittleness.

[0036] It is difficult to identify the tramp elements contained in these electric furnace-related slabs. This is because, due to the nature of the scrap, it is not possible to select and use scrap with a constant composition ratio. Therefore, a new operation independent of the slab composition and physical properties is required.

[0037] Referring to FIG. 1, a material property-based feed-forward continuous rolling control system may include a sensor (10), a continuous rolling facility (20), and a material property-based feed-forward continuous rolling control device (100).

[0038] The sensor (10), the continuous casting-rolling equipment (20), and the material-based feed-forward continuous casting-rolling control device (100) can be connected to each other via a wired or wireless network to communicate.

[0039] A material property-based feed-forward continuous rolling control device (100) can control the reduction rate and / or rolling speed in a continuous rolling facility (20) through a feed-forward control method adaptive to the slab components and / or material properties.

[0040] Feed-forward control can be a control method that predicts product quality or physical properties in advance and adjusts process conditions based on this.

[0041] Feed-forward control can predict material properties of metals, such as temperature, stress, compositional distribution, microstructure, and hardness, using data-based models or physics-based simulations.

[0042] Feed-forward control can utilize the correlation between material properties and process conditions (e.g., cooling rate, deformation rate, etc.).

[0043] The continuous rolling facility (20) is equipment that continuously solidifies, shapes, and deforms molten metal in a metal manufacturing process.

[0044] A Continuous Casting Machine (CCM) is equipment that solidifies molten metal (e.g., steel) into the form of a slab, billet, or bloom.

[0045] A rolling mill is equipment that deforms cast semi-finished products (slabs, billets, blooms) to process them into a desired thickness and shape.

[0046] One or more sensors (10) can sense at least one of the temperature of the molten steel and / or the alloy composition. That is, the sensor (10) can obtain temperature information and / or alloy composition information.

[0047] The sensor (10) may include a temperature sensor, an alloy composition sensor, and an electromagnetic sensor. The sensor is at least one sensor that provides various information, and its type is not particularly limited.

[0048] The sensor (10) is placed in the continuous rolling facility (20) and can provide sensing information to the material property-based feed-forward continuous rolling control device (100).

[0049] FIG. 2 is a block diagram of a physical property-based feed-forward continuous rolling control device according to one embodiment of the present invention.

[0050] Referring to FIG. 2, a material-based feed-forward continuous rolling control device (100) according to one embodiment can execute program code or instructions loaded into one or more memory devices through one or more processors.

[0051] For example, the material-based feed-forward continuous rolling control device (100) may be implemented as a computing device (900) as described below in relation to FIG. 5. In this case, one or more processors may correspond to the processor (910) of the computing device (900), and one or more memory devices may correspond to the memory (930) of the computing device (900).

[0052] Program code or instructions can be executed by one or more processors to control continuous casting and rolling processes through feed-forward control that is adaptive to the physical properties of the slab produced by electric furnace operation. In this specification, the term "module" has been used to logically distinguish these functions performed by program code or instructions.

[0053] A material property-based feedforward continuous rolling control device (100) can control the reduction rate and / or rolling speed of the rolling equipment through sampling analysis during the pouring stage of the molten steel (sampling solution).

[0054] A material property-based feed-forward continuous rolling control device (100) can control the reduction rate and / or rolling speed of the rolling equipment by photographing scrap and analyzing the image through a deep learning model (deep learning solution).

[0055] The material property-based feed-forward continuous rolling control device (100) can use the deep learning solution and the sampling solution described above simultaneously or separately.

[0056] When the material property-based feed-forward continuous rolling control device (100) is used simultaneously, the accuracy and reproducibility of the deep learning solution can be verified through sampling analysis.

[0057] When the material property-based feed-forward continuous rolling control device (100) is used optionally at this time, it can skip cumbersome sampling analysis and execute a deep learning solution.

[0058] The material property-based feed-forward continuous rolling control device (100) can execute the simultaneous use or simultaneous use of a deep learning solution and a sampling solution periodically or by manual operation by the user.

[0059] Referring to FIG. 2, a material property-based feed-forward continuous rolling control device (100) may include a sampling module (110), a sensor information receiving module (120), a material property prediction module (130), and a reduction rate and rolling speed control module (140).

[0060] The sampling module (110) can sample some of the molten steel when tapping the scrap-based molten steel.

[0061] The sampling module (110) can move a portion of the molten steel to a sampling process and sample it during the process step of tapping the molten steel.

[0062] The sampling module (110) can sense at least one of the temperature and / or alloy composition of the molten steel using a sensor (10, see FIG. 1). That is, the sampling module (110) can obtain temperature information, alloy composition information, etc. from the molten steel.

[0063] The sensor information receiving module (120) can receive sensor information sensed by the sampling module (110). The sensor information receiving module (120) can receive sensor information directly from the sensor (10).

[0064] The sensor information receiving module (120) can sense the content of tramp elements contained in the scrap by applying an electric current or magnetic field to the scrap through an electromagnetic sensor, and receive information about the content of tramp elements.

[0065] The material property prediction module (130) can sense the content of tramp elements contained in the scrap by applying an electric current or magnetic field to the scrap through an electromagnetic sensor.

[0066] The material property prediction module (130) can determine the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model.

[0067] The material property prediction module (130) stores data on alloy composition ratios by scrap type in advance in a database.

[0068] When an image is acquired through a camera, the material property prediction module (130) can estimate the type of scrap through a deep learning-based image analysis model.

[0069] The material property prediction module (130) can determine the alloy composition ratio of the scrap by comparing the estimated type of scrap with the alloy composition ratio of each scrap type in the database.

[0070] For example, the material property prediction module (130) can determine the alloy composition ratio of the scrap using a CNN-based deep learning model.

[0071] The physical property prediction module (130) can generate a sampling analysis result by detecting at least one of the temperature or alloy composition of the sampled molten steel through a sensor. The sampling analysis result includes sensing information (temperature information, alloy composition information).

[0072] The reduction rate and rolling speed control module (140) can control the reduction rate and rolling speed of the casting and rolling process based on at least one of the tramp element content, deep learning-based analysis results and sampling analysis results.

[0073] In one embodiment, the reduction rate and rolling speed control module (140) can verify the analysis results of a deep learning-based image analysis model based on the sampling analysis results.

[0074] FIG. 3 is a flowchart of a physical property-based feedforward continuous rolling control method according to an embodiment of the present invention. The physical property-based feedforward continuous rolling control method of FIG. 3 can be performed through the physical property-based feedforward continuous rolling control device (100) of FIG. 1.

[0075] In FIG. 3, the material-based feedforward continuous rolling control device (100) can sense the content of tramp elements contained in the scrap by applying a current or magnetic field to the scrap through an electromagnetic sensor (step S310).

[0076] The material property-based feed-forward continuous rolling control device (100) can determine the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model (step S320).

[0077] The material property-based feedforward continuous rolling control device (100) can control the reduction rate and rolling speed of the continuous rolling and rolling process based on the content of the tramp element in the scrap and the alloy composition ratio of the scrap (step S330).

[0078] The material property-based feedforward continuous rolling control device (100) can periodically or selectively control the reduction rate and rolling speed of the continuous rolling and rolling process through sensor-based sampling analysis results of the molten steel generated from scrap (step S340).

[0079] The material property-based feed-forward continuous rolling control device (100) can selectively perform steps S330 and S340, or perform them alternately at a periodic rate.

[0080] FIG. 4 is a flowchart of a physical property-based feedforward continuous rolling control method according to an embodiment of the present invention. The physical property-based feedforward continuous rolling control method of FIG. 4 can be performed through the physical property-based feedforward continuous rolling control device (100) of FIG. 1.

[0081] In FIG. 4, the material-based feedforward continuous rolling control device (100) can sense the content of tramp elements contained in the scrap by applying a current or magnetic field to the scrap through an electromagnetic sensor (step S410).

[0082] The material property-based feed-forward continuous rolling control device (100) can determine the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model (step S420).

[0083] The material property-based feed-forward continuous rolling control device (100) can store data on alloy composition ratios by scrap type in advance in a database.

[0084] When an image is acquired through a camera, the material property-based feed-forward continuous rolling control device (100) can estimate the type of scrap through the deep learning-based image analysis model.

[0085] The alloy composition ratio of the scrap can be determined by comparing the estimated type of scrap with the alloy composition ratio for each scrap type in the database.

[0086] The material property-based feed-forward continuous rolling control device (100) can sample some of the molten steel when tapping the scrap-based molten steel, and detect at least one of the temperature or alloy composition of the sampled molten steel through a sensor to generate a sampling analysis result (step S430).

[0087] The material property-based feed-forward continuous rolling control device (100) can verify the analysis results of the deep learning-based image analysis model based on the sampling analysis results (step S440).

[0088] The material property-based feed-forward continuous rolling control device (100) can perform the step S440) of verifying the analysis results of a deep learning-based image analysis model at regular intervals or selectively perform it according to the control of the user.

[0089] The material property-based feedforward continuous rolling control device (100) can control the reduction rate and rolling speed of the continuous rolling and rolling process based on at least one of the content of the tramp element, analysis results and sampling analysis results (step S450).

[0090] FIG. 5 is a drawing for explaining a computing device according to an embodiment of the present invention.

[0091] Referring to FIG. 5, a material property-based feed-forward continuous rolling control device and method according to embodiments can be implemented using a computing device (900).

[0092] The computing device (900) may include at least one of a processor (910), memory (930), user interface input device (940), user interface output device (950), and storage device (560) that communicate via a bus (920). The computing device (900) may also include a network interface (970) that is electrically connected to a network (90). The network interface (970) may transmit or receive signals to or from other entities via the network (90).

[0093] The processor (910) can be implemented in various types such as an MCU (Micro Controller Unit), AP (Application Processor), CPU (Central Processing Unit), GPU (Graphic Processing Unit), NPU (Neural Processing Unit), etc., and may be any semiconductor device that executes instructions stored in memory (930) or storage device (960). The processor (910) may be configured to implement the functions and methods described above in relation to FIGS. 1 to 4.

[0094] The memory (930) and storage device (960) may include various forms of volatile or non-volatile storage media. For example, the memory may include ROM (read-only memory) (931) and RAM (random access memory) (932). In this embodiment, the memory (930) may be located inside or outside the processor (910), and the memory (930) may be connected to the processor (910) through various known means.

[0095] In some embodiments, at least some configurations or functions of the material property-based feed-forward continuous rolling control device and method according to the embodiments may be implemented as a program or software executed on a computing device (900), and the program or software may be stored on a computer-readable medium.

[0096] In some embodiments, at least some configurations or functions of the material property-based feed-forward continuous rolling control device and method according to the embodiments may be implemented using hardware or circuits of the computing device (900), or may be implemented using separate hardware or circuits that can be electrically connected to the computing device (900).

[0097] Although embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art to which the present invention belongs, utilizing the basic concept of the present invention as defined in the following claims, also fall within the scope of the present invention.

[0098] Material property-based feed-forward continuous rolling control device and method. It is industrially applicable by controlling the reduction rate and / or rolling speed of a slab through feed-forward control adaptive to the slab composition and / or material properties based on deep learning.

Claims

1. A step of sensing the content of tramp elements contained in the scrap by applying an electric current or a magnetic field to the scrap through an electromagnetic sensor; A step of determining the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model; and A material property-based feedforward continuous rolling control method comprising the step of controlling the reduction rate and rolling speed of the continuous rolling and rolling processes based on the content of the tramp element in the scrap and the alloy composition ratio of the scrap.

2. In Paragraph 1, The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model described above is, A method comprising the step of determining the alloy composition ratio of the scrap using a CNN-based deep learning model. Material property-based feed-forward continuous rolling control method.

3. In Paragraph 1, The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model described above is, A step of pre-storing data on alloy composition ratios by scrap type in a database; When an image is acquired through a camera, a step of estimating the type of scrap through the deep learning-based image analysis model; and The method includes the step of determining the alloy composition ratio of the scrap by comparing the estimated scrap type with the alloy composition ratio for each scrap type in the database. Material property-based feed-forward continuous rolling control method.

4. In Paragraph 1, The method further includes a step of controlling the reduction rate and rolling speed of the continuous casting and rolling processes through the sampling analysis results of the scrap-based molten steel. The step of controlling the reduction rate and rolling speed of the continuous casting and rolling processes through the sampling analysis results of the scrap-based molten steel is: A step of sampling a portion of the molten steel when tapping the scrap-based molten steel mentioned above; A step of generating a sampling analysis result by detecting at least one of the temperature or alloy composition of the sampled molten steel through a sensor; and A step comprising controlling the reduction rate and rolling speed based on the above sampling analysis results, Material property-based feed-forward continuous rolling control method.

5. In Paragraph 4, A first step of controlling the reduction rate and rolling speed of the continuous casting and rolling process based on the content of a tramp element in the scrap and the alloy composition ratio of the scrap obtained using the deep learning-based image analysis model and the electromagnetic sensor, and a second step of controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel are characterized by being performed selectively from each other. Material property-based feed-forward continuous rolling control method.

6. A step of sensing the content of tramp elements contained in the scrap by applying an electric current or a magnetic field to the scrap through an electromagnetic sensor; A step of determining the alloy composition ratio of the scrap as an analysis result through a deep learning-based image analysis model; A step of sampling a portion of the molten steel when tapping the scrap-based molten steel mentioned above; A step of generating a sampling analysis result by detecting at least one of the temperature or alloy composition of the sampled molten steel through a sensor; A step of verifying the analysis results of the deep learning-based image analysis model based on the above sampling analysis results; and A physical property-based feedforward continuous rolling control method comprising the step of controlling the reduction rate and rolling speed of the continuous rolling and rolling processes based on at least one of the content of the tramp element, the analysis result, and the sampling analysis result.

7. In Paragraph 6, The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model described above is, A method comprising the step of determining the alloy composition ratio of the scrap using a CNN-based deep learning model. Material property-based feed-forward continuous rolling control method.

8. In Paragraph 6, The step of determining the alloy composition ratio of the scrap through the deep learning-based image analysis model described above is, A step of pre-storing data on alloy composition ratios by scrap type in a database; When an image is acquired through a camera, a step of estimating the type of scrap through the deep learning-based image analysis model; and The method includes the step of determining the alloy composition ratio of the scrap by comparing the estimated scrap type with the alloy composition ratio for each scrap type in the database. Material property-based feed-forward continuous rolling control method.

9. In Paragraph 6, The step of verifying the analysis results of the deep learning-based image analysis model based on the above sampling analysis results is characterized by being performed at regular intervals or selectively under user control. Material property-based feed-forward continuous rolling control method.

10. A material property-based feedforward continuous casting and rolling control device that controls the continuous casting and rolling processes through feedforward control adaptive to the material properties of a slab produced by electric furnace operation by executing program code loaded into one or more memory devices through one or more processors, The above program code is executed, Sensing the content of tramp elements contained within the scrap by applying an electric current or magnetic field to the scrap through an electromagnetic sensor, and The alloy composition ratio of the above scrap is determined as an analysis result through a deep learning-based image analysis model, and A material property-based feedforward continuous rolling control device that controls the reduction rate and rolling speed of the continuous rolling and rolling processes based on the content of the tramp element in the scrap and the alloy composition ratio of the scrap.

11. In Paragraph 10, Determining the alloy composition ratio of the scrap through the deep learning-based image analysis model described above is, A method comprising determining the alloy composition ratio of the scrap using a CNN-based deep learning model. Material property-based feed-forward continuous rolling control device.

12. In Paragraph 10, Determining the alloy composition ratio of the scrap through the deep learning-based image analysis model described above is, Data on alloy composition ratios by scrap type is stored in the database in advance, and When an image is acquired through a camera, the type of scrap is estimated through the deep learning-based image analysis model described above, and Determining the alloy composition ratio of the scrap by comparing the estimated scrap type with the alloy composition ratio for each scrap type in the database. Material property-based feed-forward continuous rolling control device.

13. In Paragraph 10, It further includes controlling the reduction rate and rolling speed of the continuous casting and rolling processes through the sampling analysis results of the scrap-based molten steel. Controlling the reduction rate and rolling speed of the continuous casting and rolling processes through the sampling analysis results of the scrap-based molten steel is, When tapping the above-mentioned scrap-based molten steel, a portion of the molten steel is sampled, and At least one of the temperature or alloy composition of the sampled molten steel is detected through a sensor to generate the sampling analysis result, and Controlling the reduction rate and rolling speed based on the above sampling analysis results, Material property-based feed-forward continuous rolling control device.

14. In Paragraph 13, Controlling the reduction rate and rolling speed of the continuous casting and rolling process based on the content of tramp elements in the scrap and the alloy composition ratio of the scrap obtained using the deep learning-based image analysis model and the electromagnetic sensor, and controlling the reduction rate and rolling speed of the continuous casting and rolling process through the sampling analysis results of the scrap-based molten steel are characterized by being performed selectively from each other. Material property-based feed-forward continuous rolling control device.

15. In Paragraph 13, Controlling the reduction rate and rolling speed of the continuous casting and rolling processes through the sampling analysis results of the scrap-based molten steel is, Further comprising verifying the analysis results of the deep learning-based image analysis model based on the above sampling analysis results. Material property-based feed-forward continuous rolling control device.

Images