Process control system and its operating method

Abstract

This invention provides a process control system and its operating method that can improve the efficiency and productivity of the etching process. [Solution] The process control system includes a first system that generates thickness information of an internal defect layer contained in a carbon steel product, and a second system that receives the thickness information of the internal defect layer from the first system via a network and controls an etching process to remove at least a portion of the internal defect layer from the carbon steel product using the thickness information of the internal defect layer. The first system provides the second system with a calculation module necessary for the second system to control the etching process, and the second system provides the first system with information necessary for the first system to update the calculation module.

Description

Technical Field

[0001] The present invention relates to a process control system for controlling a process of removing an internal defect layer contained in a carbon steel product and an operation method thereof.

Background Art

[0002] Automobile bearing shells produced using carbon steel products may be subject to continuous and repeated loads on their surfaces. Therefore, strict surface quality is required for carbon steel products used in the production of bearing shells and the like.

[0003] Internal defect layers and the like contained in carbon steel products need to be strictly managed in terms of their thickness, and the internal defect layers can be removed by an etching process or the like. The etching process can be a process of removing the internal defect layer by bringing an etching solution into contact with the carbon steel product.

[0004] Such prior art can be easily understood by referring to Korean Patent Publication No. 10-2019-0124019.

Summary of the Invention

Problems to be Solved by the Invention

[0005] One of the problems to be solved by the technical idea of the present invention is to provide a process control system and an operation method thereof that can improve the efficiency and productivity of an etching process by controlling the etching process according to the thickness of an internal defect layer contained in a carbon steel product.

Means for Solving the Problems

[0006] A process control system according to one embodiment of the present invention includes a first system that generates thickness information of an internal defect layer contained in a carbon steel product, and a second system that receives the thickness information of the internal defect layer from the first system via a network and controls an etching process to remove at least a portion of the internal defect layer from the carbon steel product using the thickness information of the internal defect layer, wherein the first system provides the second system with a calculation module necessary for the second system to control the etching process, and the second system provides the first system with information necessary for the first system to update the calculation module.

[0007] A process control system according to one embodiment of the present invention includes a storage for storing control data necessary for controlling an etching apparatus to remove at least a portion of the internal defect layer contained in a carbon steel product, and a processor for controlling the etching apparatus based on the control data, wherein the carbon steel product includes a first region and a second region different from the first region, the thickness of the internal defect layer contained in the first region and the thickness of the internal defect layer contained in the second region are different from each other, and the control data includes a first transfer speed at which the first region passes through the etching apparatus and a second transfer speed at which the second region passes through the etching apparatus, the first transfer speed and the second transfer speed are different from each other.

[0008] A process control system according to one embodiment of the present invention includes storage for a calculation module that stores information on the thickness of an internal defect layer contained in a carbon steel product based on at least one of the composition, cooling rate, phase fraction, and temperature of the carbon steel product; a communication unit connected to a network; and a processor that transfers the information on the thickness of the internal defect layer and at least one of the control data for controlling the etching process to an external server that controls an etching process to remove at least a portion of the internal defect layer via the communication unit. [Effects of the Invention]

[0009] According to one embodiment of the present invention, a process control system and its operating method can be provided that can improve the productivity and efficiency of an etching process by controlling the etching process based on optimal control data while each region defined in the longitudinal direction of a carbon steel product is in contact with an etching solution.

[0010] Furthermore, according to one embodiment of the present invention, the thickness of the remaining internal defect layer contained in the pickled carbon steel product after the etching process is completed is measured, and a calculation module that generates control data for controlling the etching process using the measured thickness of the remaining internal defect layer is trained, thereby enabling the etching process to be performed based on optimized control data.

[0011] The diverse yet significant advantages and effects of the present invention are not limited to those described above and can be more easily understood in the process of describing specific embodiments of the present invention. [Brief explanation of the drawing]

[0012] [Figure 1] This is a drawing illustrating the manufacturing process of a hot-rolled steel sheet according to one embodiment of the present invention. [Figure 2] This is a simplified drawing of a hot-rolled steel sheet according to one embodiment of the present invention. [Figure 3] This is a block diagram showing a process control system according to one embodiment of the present invention. [Figure 4] This is a block diagram illustrating a process control system according to one embodiment of the present invention. [Figure 5] This is a diagram illustrating the initial learning method for a computation module included in a process control system according to one embodiment of the present invention. [Figure 6] This is a diagram illustrating a learning model included in a process control system according to one embodiment of the present invention. [Figure 7] These are drawings illustrating the operation method of a process control system according to one embodiment of the present invention. [Figure 8]This is a flowchart illustrating the operation method of a process control system according to one embodiment of the present invention. [Figure 9] This is a block diagram illustrating a process control system according to one embodiment of the present invention. [Figure 10] These are drawings illustrating the operation method of a process control system according to one embodiment of the present invention. [Figure 11] This is a diagram illustrating a learning method for a calculation module included in a process control system according to one embodiment of the present invention. [Figure 12] This is a drawing illustrating an etching process according to one embodiment of the present invention. [Figure 13] This graph is provided to illustrate the etching process according to one embodiment of the present invention. [Figure 14] This is a diagram illustrating the initial learning method for a computation module included in a process control system according to one embodiment of the present invention. [Figure 15] This is a diagram illustrating a learning method for a calculation module included in a process control system according to one embodiment of the present invention. [Figure 16] This is a flowchart illustrating the operation method of a process control system according to one embodiment of the present invention. [Figure 17] These are drawings illustrating the operation method of a process control system according to one embodiment of the present invention. [Modes for carrying out the invention]

[0013] In this specification, carbon steel products are products that are alloys of iron and carbon and contain 0.01% to 2.0% carbon, and typically can include hot-rolled products, thick plate products, wire products, etc. For example, the above hot-rolled products can be hot-rolled coils obtained by hot-rolling a slab and wound into a coil shape or hot-rolled steel plates made by cutting into sheet shapes, and can have a thickness of 1 mm to 25 mm. The above thick plate products can be products made by hot-rolling a slab into a plate shape and can have a thickness of 4 mm to 200 mm. The above wire products can be coil-shaped products with a circular cross-section obtained by hot rolling a billet in a hot state and can have a cross-sectional diameter of 3 mm to 100 mm.

[0014] During the etching process for carbon steel products, the process control system of the present invention can control the transfer speed of the carbon steel products to produce pickled carbon steel products (hereinafter referred to as "pickled carbon steel products"), so the surface quality of the carbon steel products can be improved.

[0015] Hereinafter, the technical idea of the present invention will be described centering on hot-rolled steel plates or hot-rolled coils in which the above hot-rolled steel plates are wound into a coil shape. However, the technical idea of the present invention is not limited to hot-rolled steel plates or hot-rolled coils and can be applied to all carbon steel products within the range easily understood by ordinary technicians.

[0016] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same components on the drawings, and duplicate descriptions for the same components are omitted.

[0017] FIG. 1 is a drawing for explaining the manufacturing process of a hot-rolled steel plate according to an embodiment of the present invention.

[0018] Figure 1 is a simplified diagram of a hot-rolled steel sheet production apparatus that produces hot-rolled steel sheets, cools the hot-rolled steel sheets, and winds them to produce hot-rolled coils. Referring to Figure 1, a hot-rolling process can be performed in which a slab heated in a heating furnace is rolled to a predetermined thickness using a roughing mill and a finishing mill. The hot-rolled steel sheets (strips) produced through the hot-rolling process can be transferred to a run-out table (ROT) cooling platform, which is a cooling section. The hot-rolled steel sheets can be cooled by cooling water sprayed from the ROT cooling platform, and the cooling temperature, cooling time, etc., can be determined according to the quality required for the hot-rolled steel sheets.

[0019] After cooling, the hot-rolled steel sheet can be wound into a coil using a winding machine for convenience of storage and / or transport. The hot-rolled coil (HC), formed by winding the hot-rolled steel sheet into a coil, is stacked in a yard, air-cooled, and then shipped. After this, an etching process can be performed to remove internal defect layers and other imperfections contained in the hot-rolled steel sheet. In one embodiment, the manufacturing process of the hot-rolled steel sheet and the etching process for removing internal defect layers and other imperfections, as described with reference to Figure 1, can be carried out by different entities. Also, in one embodiment, the manufacturing process of the hot-rolled steel sheet and the etching process for removing internal defect layers and other imperfections can be carried out by a single entity.

[0020] Figure 2 is a simplified drawing of a hot-rolled steel sheet according to one embodiment of the present invention.

[0021] Referring to Figure 2, a hot-rolled steel sheet can be divided into multiple regions along its length (X-axis direction). For example, a hot-rolled steel sheet may include regions A, B, and C arranged sequentially along its length. In one embodiment shown in Figure 2, region A may be the region that is first wound up in the hot-rolled steel sheet, and region C may be the region that is last wound up. Region B may be the region located between region A and region C in the length direction.

[0022] In one embodiment, the hot-rolled steel sheet may contain surface defects. For example, surface defects may include scale and at least one internal defect layer. Scale may occur during the rolling process and be present on the surface of the material. An internal defect layer may be located below the surface of the material, i.e., inside the material, and can be defined as an internal oxide layer and / or a decarburization layer. An internal oxide layer may occur in the process in which components such as chromium (Cr), manganese (Mn), silicon (Si), zinc (Zn), magnesium (Mg), and aluminum (Al), which have a higher oxygen affinity than iron (Fe), cause oxidation within the base material. A decarburization layer may occur in the process in which carbon in the steel combines with oxygen from the atmosphere and scale and is then released into the atmosphere in gaseous form. The thickness of the internal defect may vary depending on the composition of the hot-rolled steel sheet, the temperature at which the hot-rolled steel sheet is wound into a hot-rolled coil HC, the cooling time after winding, the width and thickness of the hot-rolled steel sheet, and its length. Internal defects such as internal defect layers can be a factor that reduces the durability of products produced using hot-rolled steel sheets.

[0023] For example, the temperature of a wound hot-rolled coil HC can be around 500-700°C, and the wound hot-rolled coil HC can be cooled by air cooling while exposed to the air. In a wound hot-rolled coil HC, regions A and C, which are exposed to the outside, cool relatively quickly, while region B, which is not exposed to the outside, cools relatively slowly. Therefore, the thickness of the internal defect layer in region B of the hot-rolled steel sheet can be greater than the thickness of the internal defect layer in regions A and C, respectively. Consequently, variations in the thickness of the internal defect layer can occur in the regions of the hot-rolled steel sheet. For example, the thickness of the internal defect layer in region A and / or region C may be less than a predetermined standard thickness, while the thickness of the internal defect layer in region B may be greater than the standard thickness.

[0024] Furthermore, the thickness of the internal defect layer can vary depending on the region in the width direction of the hot-rolled steel sheet. For example, regions adjacent to the corners of the hot-rolled steel sheet can cool relatively quickly in the width direction, and differences in the thickness of the internal defect layer depending on the cooling rate can be shown in the width direction as well.

[0025] An etching process to remove at least a portion of the internal defect layer can be carried out by bringing the hot-rolled steel sheet into contact with an etching solution. For example, the etching process can be carried out by transferring the hot-rolled steel sheet while it is immersed in an etching solution contained in an etching tank. Alternatively, the etching process can be carried out by spraying the etching solution onto the surface of the hot-rolled steel sheet in an etching tank, or by brushing the surface of the hot-rolled steel sheet with a brush immersed in the etching solution. The etching process can be at least one of a pickling process, a dry etching process, or a wet etching process.

[0026] As an example, in order to sufficiently remove the internal defect layer contained in the hot-rolled steel sheet during the etching process, the etching process can be carried out by ensuring that the hot-rolled steel sheet is in sufficient contact with the etching solution regardless of the area of ​​the hot-rolled steel sheet. However, the above method may lead to an increase in the etching process time and / or the amount of etching solution introduced into the etching process, potentially resulting in a decrease in productivity.

[0027] According to one embodiment of the present invention, the thickness of the internal defect layer contained in the hot-rolled steel sheet can be calculated and / or measured along the length direction of the hot-rolled steel sheet, and the etching process can be controlled to optimal efficiency according to the thickness information of the internal defect layer. Therefore, productivity can be improved by shortening the etching process time and reducing the amount of etching solution used. In addition, it is possible to reduce the variation in the thickness of the internal defect layer that appears in different regions of the pickled steel sheet after the etching process is completed.

[0028] Figure 3 is a block diagram showing a process control system according to one embodiment of the present invention.

[0029] According to one embodiment of the present invention, the process control system 1 includes a first system SYS1 that generates thickness information of the internal defect layer contained in a hot-rolled coil in which a hot-rolled steel sheet is wound into a coil, and a second system SYS2 that uses the thickness information of the internal defect layer received from the first system SYS1 to control an etching process to remove at least a portion of the internal defect layer from the hot-rolled steel sheet in which the hot-rolled coil is wound. Furthermore, the first system SYS1 can provide the second system SYS2 with a calculation module necessary for the second system SYS2 to control the etching process. Furthermore, the second system SYS2 can acquire and provide the first system SYS1 with information necessary for the first system SYS1 to update the calculation module. The first system SYS1 and the second system SYS2 can communicate with each other via a network 30.

[0030] Referring to Figure 3, a process control system 1 according to one embodiment of the present invention can be operated by a first entity that performs a hot rolling process and a second entity that performs an etching process. In one embodiment, the first entity can produce hot-rolled steel sheets and manufacture hot-rolled coils by winding the hot-rolled steel sheets, and the second entity can receive the hot-rolled coils from the first entity and perform an etching process to remove internal defect layers from the hot-rolled steel sheets. Referring to Figure 3, the first system SYS1 can be operated by the first entity, and the second system SYS2 can be operated by the second entity. However, depending on the embodiment, the first system SYS1 and the second system SYS2 can also be operated by a single entity. In this case, the network 30 can be the internal network of the entity that operates both the first system SYS1 and the second system SYS2.

[0031] Referring to Figure 3, the first system SYS1 includes a first server 10, and the second system SYS2 includes a second server 20. The first server 10 and the second server 20 communicate via a network 30, and the second system SYS2 can control an etching apparatus 22 that removes at least a portion of the internal defect layer of the hot-rolled steel sheet.

[0032] Network 30 can refer to a wired internet network, a wireless internet network, or a wireless local area network (WLAN) such as Wi-Fi (wireless fidelity). The wired internet network or the wireless internet network mentioned above can refer to an open computer network structure that provides the Internet TCP / IP protocol and various services existing at its upper layer, namely HTTP (HyperText Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), SNMP (Simple Network Management Protocol), NFS (Network File Service), NIS (Network Information Service), etc.

[0033] In one embodiment, the first system SYS1 can generate thickness information of the internal defect layer of the hot-rolled steel sheet produced in the hot-rolling process. The first system SYS1 can transfer the internal defect layer thickness information to the second system SYS2 via the network 30. The second system SYS2 can use the internal defect layer thickness information received via the network 30 to control the etching process performed by the etching apparatus 22.

[0034] For example, the first system SYS1 includes a first server 10, which may include a calculation module 11 and a learning model for training the calculation module 11. In one embodiment, the calculation module 11 can generate thickness information of the internal defect layer. The first system SYS1 can generate thickness information by actually measuring the thickness of the internal defect layer of the hot-rolled steel sheet, or by calculating the thickness of the internal defect layer of the hot-rolled steel sheet. In one embodiment, the first system SYS1 can generate thickness information of the internal defect layer by calculating the thickness of the internal defect layer using at least one of the components of the hot-rolled steel sheet, the phase fraction of the hot-rolled steel sheet, and the temperature of the hot-rolled steel sheet.

[0035] The thickness information of the internal defect layer can include the thickness of the internal defect layer contained in the hot-rolled steel sheet in each of several regions defined in the longitudinal direction of the hot-rolled steel sheet. Therefore, the thickness information of the internal defect layer can include the thickness distribution of the internal defect layer appearing along the longitudinal direction of the hot-rolled steel sheet.

[0036] In other embodiments, the calculation module 11 can generate thickness-related information for the internal defect layer. The thickness-related information for the internal defect layer can mean all forms of information obtained by processing the thickness information for the internal defect layer so that the thickness information for the internal defect layer can be extracted. When the first system SYS1 transfers the thickness-related information for the internal defect layer to the second system SYS2 via the network 30, the second system SYS2 can extract the thickness information for the internal bond layer contained in the thickness-related information for the internal defect layer received via the network 30, or calculate the thickness information for the internal defect layer from the thickness-related information, and control the etching process performed by the etching apparatus 22 using the extracted or calculated thickness information for the internal defect layer.

[0037] The calculation module 21 of the second system SYS2 can generate control data necessary to control the etching apparatus 22. For example, the calculation module 21 can generate control data using thickness information of the internal defect layer received from the first system SYS1. The control data may include at least one of the following: the speed at which the etching apparatus 22 transports the hot-rolled steel sheet during the etching process; the concentration of the etching solution that comes into contact with the hot-rolled steel sheet in the etching apparatus 22; the components of the etching solution; the temperature of the etching solution; and whether or not an accelerator is used.

[0038] The calculation module 21 of the second system SYS2 may have received at least one of the calculation modules 11 stored and managed by the first system SYS1. The first system SYS1 can provide the second system SYS2 with modules necessary for controlling the etching apparatus 22 within the calculation module 11. For example, the first system SYS1 can provide the second system SYS2 with modules necessary for controlling the etching apparatus 22 after training it using a learning model 12. The first system SYS1 can provide modules to the second system SYS2 via a transfer method over the network 30. Alternatively, modules trained by the first system SYS1 can also be provided to the second system SYS2 by an operator directly inputting and saving the modules.

[0039] The second system SYS2 can provide the first system SYS1 with the information necessary for the first system SYS1 to update the arithmetic module 11. For example, when the etching apparatus 22 completes the etching process on the hot-rolled steel sheet and produces a pickled steel sheet, the second system SYS2 can provide the first system SYS1 with the thickness of the remaining internal defect layer contained in the pickled steel sheet and at least one of the control data measured by the etching apparatus 22 during the etching process. The first system SYS1 can update the arithmetic module 11 using the thickness of the remaining internal defect layer and at least one of the control data received from the second system SYS2. For example, the arithmetic module 11 can be updated by having the learning model 12 learn the arithmetic module 11 using the thickness of the remaining internal defect layer and at least one of the control data.

[0040] When the arithmetic module 11 is updated, the first system SYS1 can transfer at least one of the updated arithmetic modules 11 to the second system SYS2 via the network 30. The second system SYS2 can then use the arithmetic module 11 received from the first system SYS1 to update the stored arithmetic modules 21. For example, the second system SYS2 can overwrite the existing arithmetic module 21 with the arithmetic module received from the first system SYS1.

[0041] In one embodiment, the calculation module 11 of the first system SYS1 can consist of multiple calculation modules. The calculation module 11 may include a calculation module that generates thickness information of the internal defect layer contained in the hot-rolled steel sheet, and a calculation module that generates control data for controlling the etching apparatus 22. The first system SYS1 can transfer the calculation module that generates the control data to the second system SYS2.

[0042] The learning model 12 can optimize the arithmetic module 11. For example, the learning model 12 can include multiple learning models for training multiple arithmetic modules. The learning model 12 can also train and optimize the arithmetic module 11 using at least one of the remaining internal defect layer thickness and control data received from the second system SYS2. In one embodiment, the learning model 12 can train the arithmetic module 11 so that the remaining internal defect layer thickness calculated by the arithmetic module 11 matches the remaining internal defect layer thickness received from the second system SYS2, or the difference between them is less than or equal to a predetermined reference value.

[0043] The arithmetic module 11 can be implemented in hardware such as circuits, or in software such as source code. The arithmetic module 11 can perform predetermined calculations using input values ​​to generate output values. For example, the learning model 12 can train the arithmetic module 11 by adjusting the weights, coefficients, etc., of the calculations performed by the arithmetic module 11.

[0044] Depending on the embodiment, the second server 20 of the second system SYS2 may further include an additional arithmetic module. The additional arithmetic module may be a separate arithmetic module from the arithmetic module 21 received from the first system SYS1, and may adjust at least a portion of the control data output by the arithmetic module 21 and input it to the etching apparatus 22. For example, the additional arithmetic module may adjust at least a portion of the control data output by the arithmetic module 21 and input it to the etching apparatus 22, taking into account mechanical errors, operational delays, etc., present in the etching apparatus 22. Therefore, if the additional arithmetic module is included in the second server 20, the control data output by the arithmetic module 21 and the control data input to the etching apparatus 22 may differ from each other.

[0045] The additional calculation module can adjust the scale unit of the control data output by the calculation module 21, or convert the control data into a data format that can be input to the etching apparatus 22, so that the control data output by the calculation module 21 is applied to the etching apparatus 22. Depending on the embodiment, the calculation module 21 can also directly adjust the scale unit and data format of the control data and output it to the etching apparatus 22. In this case, the additional calculation module may not be included in the second server 20.

[0046] In one embodiment, the calculation module 21 of the second system SYS2 can receive thickness information of the internal defect layer from the first system SYS1. The thickness information may include the thickness of the internal defect layer calculated and / or measured by the first system SYS1. The second system SYS2 can use the thickness information to determine control data for controlling the etching apparatus 22. In one embodiment, the control data may include the transfer speed at which the etching apparatus 22 transfers the hot-rolled steel sheet, the concentration of the etching solution that comes into contact with the hot-rolled steel sheet in the etching apparatus 22, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used. The etching apparatus 22 can be automatically controlled by the control data output by the calculation module 21.

[0047] For example, the control data input to the etching apparatus 22 can optimize the etching process for each of several regions defined in the longitudinal direction of the hot-rolled steel sheet. For instance, the transfer speed of the hot-rolled steel sheet during the etching process for a portion of the sheet near the edge may differ from the transfer speed of the hot-rolled steel sheet during the etching process for other portions of the sheet far from the edge. For example, the transfer speed of the hot-rolled steel sheet during the etching process for a portion of the sheet near the edge may be faster than the transfer speed of the hot-rolled steel sheet during the etching process for other portions of the sheet far from the edge. In one embodiment, scale present on the surface of the hot-rolled steel sheet can be removed along with the etching process.

[0048] In one embodiment, due to mechanical errors in the etching apparatus 22 and signal delays during the input / output process of control data, the control data output by the calculation module 21 and the control data actually measured by the etching apparatus 22 during the etching process may not match. To illustrate with an example of the control data input to the etching apparatus 22, the transfer speed of the hot-rolled steel sheet input by the calculation module 21 to the etching apparatus 22 may not match the actual transfer speed at which the etching apparatus 22 transfers the hot-rolled steel sheet. In addition, among the control data input to the etching apparatus 22, the concentration of the etching solution and / or the temperature of the etching solution may change during the process in which the hot-rolled steel sheet and the etching solution come into contact.

[0049] In one embodiment of the present invention, the etching apparatus 22 can be controlled taking into account mechanical errors and signal delays. For example, the second system SYS2 can acquire information such as the thickness of the remaining internal defect layer contained in the pickled steel sheet after the etching process is completed, and control data measured by the etching apparatus 22 during the etching process, so that mechanical errors and signal delays are reflected in the control data. The second system SYS2 can transfer the acquired information to the first system SYS1.

[0050] The learning model 12 of the first system SYS1 can be trained to learn at least one of the arithmetic modules 11 using the information transferred by the second system SYS2. Once training of the arithmetic module 11 is complete, the first system SYS1 can transfer at least one of the trained arithmetic modules 11, for example, an arithmetic module that generates control data for controlling the etching apparatus 22, to the second system SYS2. The second system SYS2 can update the previously stored arithmetic modules 21 using the arithmetic modules received from the first system SYS1. Therefore, the arithmetic modules 21 of the second system SYS2 can generate control data taking into account mechanical errors and signal delays, and can control the etching apparatus 22 more accurately.

[0051] Depending on the embodiment, the second system SYS2 may also include a learning model. The learning model included in the second system SYS2 can train the arithmetic module 21 included in the second system SYS2 in a manner similar to the learning model 12 included in the first system SYS1. This will be described later with reference to Figures 14 and 15.

[0052] Alternatively, the second system SYS2 can directly calculate the thickness of the internal defect layer contained in the hot-rolled steel sheet and use this to perform the etching process. In this case, the second system SYS2 can receive information necessary to calculate the thickness of the internal defect layer from the first system SYS1. For example, the first system SYS1 can collect information such as the cooling rate of the hot-rolled steel sheet, the composition of the hot-rolled steel sheet, and the ambient oxygen partial pressure while the hot-rolled steel sheet is cooling, and transfer this information to the second system SYS2. The calculation module 21 of the second system SYS2 can use the variables received from the first system SYS1 to predict the thickness of the internal defect layer contained in the hot-rolled steel sheet and control the etching apparatus 22.

[0053] In one embodiment, as described above, the transfer speed of the hot-rolled steel sheet during the etching process can be adjusted using the calculated thickness of the internal defect layer. For example, the hot-rolled steel sheet can be transferred at a first transfer speed while a first region, where the thickness of the internal defect layer is expected to be relatively small, is in contact with the etching solution. Furthermore, the hot-rolled steel sheet can be transferred at a second transfer speed, which is slower than the first transfer speed, while a second region, where the thickness of the internal defect layer is expected to be relatively large, is in contact with the etching solution.

[0054] In one embodiment of the present invention, operations such as collecting information to predict the internal defect layer of a hot-rolled steel sheet, calculating the internal defect layer, generating control data to control the etching apparatus 22 based on the calculated internal defect layer, and relearning the calculation modules 11 and 21 using control data collected while the etching apparatus 22 performs the etching process can be distributed and implemented across two different systems, SYS1 and SYS2. For example, the above operations can be appropriately distributed and executed between the first system SYS1 and the second system SYS2 as needed. For example, by assigning the above operations to the first system SYS1 and the second system SYS2 according to their performance, the load on systems SYS1 and SYS2 can be efficiently managed, and the entire system 1 can be operated effectively.

[0055] For example, if the performance of the first system SYS1 is significantly better than that of the second system SYS2, the first system SYS1 can handle most of the above operations, and the second system SYS2 can receive the control data generated by the first system SYS1 and control the etching apparatus 22. If the performance of the first system SYS1 and the second system SYS2 are similar, the first system SYS1 can generate thickness information indicating the thickness of the internal defect layer contained in the hot-rolled steel sheet, and the second system SYS2 can generate control data using the internal defect layer thickness information. On the other hand, if the performance of the second system SYS2 is significantly better than that of the first system SYS1, the first system SYS1 can collect the information of the hot-rolled steel sheet necessary to calculate the thickness of the internal defect layer and transmit it to the second system SYS2, and the second system SYS2 can calculate the thickness of the internal defect layer, generate control data and control the etching apparatus 22.

[0056] The information on the hot-rolled steel sheet that the second system SYS2 receives from the first system SYS1 to predict the thickness of the internal defect layer can include information related to the phase fraction of the hot-rolled steel sheet, information related to the temperature of the hot-rolled steel sheet, and information related to the composition of the hot-rolled steel sheet. The information related to the phase fraction can mean all forms of information obtained by processing the above phase fraction information so that the phase fraction information can be extracted, the information related to the temperature of the hot-rolled steel sheet can mean all forms of information obtained by processing the above temperature information so that the temperature information can be extracted, and the information related to the composition of the hot-rolled steel sheet can mean all forms of information obtained by processing the above component information so that the component information can be extracted. Therefore, the second system SYS2 can first calculate the phase fraction, temperature, composition, etc. of the hot-rolled steel sheet using the information received from the first system SYS1, and then use this to predict the thickness of the internal defect layer contained in the hot-rolled steel sheet.

[0057] Figure 4 is a block diagram that simply illustrates a process control system according to one embodiment of the present invention.

[0058] Referring to Figure 4, the process control system 40 according to one embodiment of the present invention may include a first server 41 and a hot-rolled steel sheet production apparatus 45, the first server 41 being able to control and manage the hot-rolled steel sheet production apparatus 45. The hot-rolled steel sheet production apparatus 45 may be a device that produces hot-rolled steel sheets by rolling a slab heated in a heating furnace, and then cools and winds the hot-rolled steel sheets.

[0059] The first server 41 may include a communication unit 42, storage 43, a processor 44, and the like. The communication unit 42 can be connected to the network so that it can communicate with the first server 41. The storage 43 can store data necessary for the operation of the first server 41 and for managing the hot-rolled steel sheet production equipment 45. The processor 44 can control the communication unit 42, storage 43, and the hot-rolled steel sheet production equipment 45, and the like.

[0060] For example, the storage 43 can store calculation modules that perform predetermined calculations. In one embodiment, the calculation module can work in conjunction with the hot-rolled steel sheet production apparatus 45 to generate thickness information of the internal defect layer contained in the hot-rolled steel sheet using information acquired from the hot-rolled steel sheet production apparatus 45. The calculation module can also generate control data necessary for an external system connected via the communication unit 42 and a network to control the etching apparatus. The etching apparatus can be a device that removes at least a portion of the internal defect layer contained in the hot-rolled steel sheet produced by the hot-rolled steel sheet production apparatus 45.

[0061] For example, the control data may include the transfer speed at which the etching apparatus transfers the hot-rolled steel sheet, the temperature, concentration, and components of the etching solution that comes into contact with the hot-rolled steel sheet in the etching apparatus, and whether or not an accelerator is used. As an example, in the internal defect layer thickness information generated by the calculation module, a first region of the hot-rolled steel sheet may have an internal defect layer of a first thickness, and a second region may have an internal defect layer of a second thickness different from the first thickness. The control data includes a first transfer speed at which the hot-rolled steel sheet is transferred in the etching process for the first region, and a second transfer speed at which the hot-rolled steel sheet is transferred in the etching process for the second region, and the first and second transfer speeds may be different from each other. For example, if the first thickness is greater than the second thickness, the first transfer speed may be slower than the second transfer speed. Therefore, a sufficient etching process can be performed on regions with relatively thicker thicknesses.

[0062] In one embodiment, the storage 43 can store the learning model necessary for training the arithmetic module. The learning model is executed by the processor 44, which can optimize the arithmetic module by executing the learning model.

[0063] The processor 44 can be implemented as a CPU, AP, SoC, etc., and can control the storage 43, the communication unit 42, and the hot-rolled steel sheet production equipment 45. For example, the processor 44 can transfer to the external system at least one of the thickness information of the internal defect layer generated by the arithmetic module stored in the storage 43, and the control data necessary for the external system to control the etching apparatus.

[0064] When the processor 44 transfers the thickness information of the internal defect layer to the external system, the external system can directly generate control data for controlling the etching apparatus based on the thickness information of the internal defect layer. In this case, the processor 44 can retrieve a calculation module from the storage 43 that generates control data using the thickness information of the internal defect layer and transfer it to the external system.

[0065] On the other hand, when the processor 44 transfers control data to the external system, the external system can control the etching apparatus using the received control data. If necessary, additional tasks such as converting the format of the control data can be performed by the external system. In this case, the processor 44 does not need to transfer the arithmetic module that generates the control data using the thickness information of the internal defect layer to the external system.

[0066] In one embodiment, the calculation module stored in storage 43 may include a first module that calculates the phase fraction before the hot-rolled steel sheet production apparatus 45 winds the hot-rolled steel sheet. The first module can calculate the phase fraction based on at least one of the process conditions of the hot-rolled steel sheet, such as the cooling rate of the hot-rolled steel sheet, the composition of the hot-rolled steel sheet, and the initial temperature of the hot-rolled steel sheet. The learning model in storage 43 can be trained by comparing the phase fraction actually measured in the hot-rolled steel sheet with the phase fraction predicted by the first module through calculation.

[0067] In one embodiment, the calculation module stored in storage 43 may further include a second module that calculates the temperature change of the hot-rolled steel sheet. The second module can predict the temperature change using at least one of the phase fraction calculated by the first module, the elapsed time after the hot-rolled steel sheet has been coiled, and the components of the hot-rolled steel sheet. The learning model can be trained by comparing the temperature change calculated by the second module with the temperature change actually measured on the hot-rolled steel sheet.

[0068] In one embodiment, the calculation module stored in storage 43 may further include a third module that calculates the thickness of the internal defect layer contained in the hot-rolled steel sheet. For example, the third module can calculate the thickness of the internal defect layer using at least one of the temperature change, the composition of the hot-rolled steel sheet, and the oxygen partial pressure around the hot-rolled steel sheet calculated by the second module. The learning model can be trained by comparing the thickness of the internal defect layer predicted by the third module through calculation with the thickness of the internal defect layer measured from the hot-rolled steel sheet.

[0069] As an example, the third module can calculate the thickness of the internal defect layer in each of several regions defined along the length of the hot-rolled steel sheet. In at least some of these regions, the thickness of the internal defect layer can be calculated to be different from one another.

[0070] In one embodiment, the calculation module stored in the storage 43 may further include a fourth module that calculates the thickness of the internal defect layer to be removed in an etching step that removes at least a portion of the internal defect layer. The fourth module can calculate the thickness of the internal defect layer to be removed in the etching step based on control data input to the etching apparatus to control the etching apparatus that performs the etching step.

[0071] The learning model can be trained by comparing the thickness of the remaining internal defect layer in the pickled steel sheet after the etching process is complete with the thickness of the internal defect layer predicted by the fourth module through calculation. For example, the learning model can compare the difference between the thickness of the internal defect layer calculated by the third module and the thickness of the internal defect layer calculated by the fourth module with the thickness of the remaining internal defect layer in the pickled steel sheet. In order to train the fourth module, the first server 41 can receive the thickness of the remaining internal defect layer measured in the pickled steel sheet by an external system that controls the etching apparatus.

[0072] In one embodiment, the calculation module stored in storage 43 may further include a fifth module that generates control data necessary for an external system to control the etching apparatus. The control data generated by the fifth module may include at least one of the following: the speed at which the etching apparatus transports the hot-rolled steel sheet; the concentration of the etching solution that comes into contact with the hot-rolled steel sheet in the etching apparatus; the components of the etching solution; the temperature of the etching solution; and whether or not an accelerator is used.

[0073] The first server 41 can receive control data from an external system that controls the etching apparatus, so that the learning model can train the fifth module. The control data that the first server 41 receives from the external system can be control data measured by the external system from the etching apparatus during the etching process. The learning model can train the fifth module by comparing the thickness of the internal defect layer calculated by the fourth module using the control data received from the external system with the thickness of the remaining internal defect layer measured from the pickled steel sheet.

[0074] Figure 5 is a diagram illustrating the initial learning method for a computation module included in a process control system according to one embodiment of the present invention.

[0075] Referring to Figure 5, the thickness control system 100 for the internal defect layer of a hot-rolled steel sheet can include a calculation module 110 and a learning model 120. For example, the calculation module 110 can include a thickness information generation module 111 and a control data generation module 112. In one embodiment, the calculation module 110 can include first to fifth modules M1 to M5, the thickness information generation module 111 can include first to third modules M1 to M3, and the control data generation module 112 can include fourth module M4 and fifth module M5. The first to fifth modules M1 to M5 can be implemented in hardware such as circuits or in software such as source code.

[0076] For example, if modules M1 to M5 are implemented in software such as source code, modules M1 to M5 can perform operations to calculate output data using input data. Each operation of modules M1 to M5 can be determined and optimized by the learning model 120.

[0077] The learning model 120 may include first to fourth learning models LM1 to LM4 and a feedback learning model LMT. The first learning model LM1 may be a model for initially learning the first module M1, and the second learning model LM2 may be a model for initially learning the second module M2. Similarly, the third learning model LM3 may be a model for initially learning the third module M3, and the fourth learning model LM4 may be a model for initially learning the fourth module M4. The feedback learning model LMT may be a model for learning at least one of the first to fifth modules M1 to M5 after the etching process is completed. In one embodiment, the feedback learning model LMT may also learn the first to fifth modules M1 to M5 simultaneously.

[0078] The learning model 120 can train the first to fifth modules M1 to M5 using methods such as deviation correction or reinforcement learning. When using deviation correction, the learning model 120 can train the first to fifth modules M1 to M5 by correcting the error between the measured values ​​input from the outside and the calculated values ​​output from the modules. When the learning model 120 uses reinforcement learning, the learning model 120 can be a DNN (deep neural network), but it is not necessarily limited to this example.

[0079] An example of how the learning model 120 uses reinforcement learning to train each module will be described later with reference to Figure 6. The initial learning method of the process control system 100 will be specifically explained below with reference to Figure 5.

[0080] In one embodiment, the first module M1 can calculate the phase fraction before the hot-rolled steel sheet is wound. The first learning model LM1 can receive the first process condition CD1 and the first phase fraction value EPF. The first phase fraction value EPF may include the phase transformation fraction or phase transformation amount actually measured under the first process condition CD1. The first process condition CD1 may include the cooling rate of the hot-rolled steel sheet, the temperature of the hot-rolled steel sheet, and the composition of the hot-rolled steel sheet.

[0081] The first learning model LM1 can input the first process condition CD1 to the first module M1. The first module M1 can calculate the second phase fraction value PPF, which is obtained by calculating the phase fraction of the hot-rolled steel sheet using the first process condition CD1. The second phase fraction value PPF output by the first module M1 can be transmitted to the first learning model LM1.

[0082] The first learning model LM1 can train the first module M1 by comparing the first phase fraction value EPF and the second phase fraction value PPF. For example, if the first phase fraction value EPF and the second phase fraction value PPF do not match, the first learning model LM1 can adjust the weights, coefficients, etc., of the first calculation in the first operation, which calculates the phase fractions when the first process condition CD1 is input from the first module M1. The first learning model LM1 can adjust the weights, coefficients, etc., of the first calculation in the first module M1 so that the first phase fraction value EPF and the second phase fraction value PPF match, or so that the difference between the first phase fraction value EPF and the second phase fraction value PPF is less than or equal to a predetermined value.

[0083] The second learning model LM2 can receive the second process condition CD2, the environmental condition FV representing the surrounding environment, and the first temperature value ET measured from the hot-rolled steel sheet. The second process condition CD2 may include the time at which the temperature of the hot-rolled steel sheet was measured after winding, and the composition of the hot-rolled steel sheet. The environmental condition FV may include a value representing the surrounding environment that affects the rate at which the hot-rolled steel sheet cools after winding. For example, if the surrounding environment is air, the value of environmental condition FV may be "1", if the surrounding environment is wind, the value of environmental condition FV may be "2", and if the surrounding environment is water, the value of environmental condition FV may be "3". The first temperature value ET may include the actual measured temperature of the hot-rolled steel sheet.

[0084] The second module M2 can receive the second process condition CD2 and environmental condition FV from the second learning model LM2. For example, the second module M2 can calculate the temperature of the hot-rolled steel sheet using the second process condition CD2 and environmental condition FV, and can output the second temperature value PT calculated from the temperature of the hot-rolled steel sheet. The second temperature value PT may include values ​​calculated based on the elapsed time after winding, for each temperature in a region defined in the longitudinal direction of the hot-rolled steel sheet. The second module M2 can output the second temperature value PT to the second learning model LM2.

[0085] The second learning model LM2 can train the second module M2 by comparing the first temperature value ET and the second temperature value PT. For example, if the first temperature value ET and the second temperature value PT do not match, or if the difference between the first temperature value ET and the second temperature value PT is greater than a predetermined value, the second learning model LM2 can adjust the weighting values, coefficients, etc., of the second operation used to calculate the second temperature value PT in the second module M2. The second learning model LM2 can train the second module M2 so that the first temperature value ET and the second temperature value PT match, or so that the difference between the first temperature value ET and the second temperature value PT is less than or equal to a predetermined value.

[0086] The third learning model LM3 can receive inputs such as the composition ING of the hot-rolled steel sheet, the oxygen partial pressure OPP around the hot-rolled steel sheet, the second temperature value PT which is the output value of the second module M2, and the first thickness ED of the internal defect layer measured from the hot-rolled steel sheet. The first thickness ED of the internal defect layer can be an actual measured value of the thickness of the internal defect layer of the hot-rolled steel sheet.

[0087] The third module M3 can receive the hot-rolled steel sheet component ING, the oxygen partial pressure OPP around the hot-rolled steel sheet, and the second temperature value PT of the hot-rolled steel sheet as input data from the third learning model LM3. The third module M3 can calculate the second thickness PD of the internal defect layer. The second thickness PD can be the thickness of the internal defect layer that the third module M3 calculates assuming exists in the hot-rolled steel sheet using at least one of the hot-rolled steel sheet component ING, the oxygen partial pressure OPP around the hot-rolled steel sheet, and the second temperature value PT. The second thickness PD of the internal defect layer output by the third module M3 can be transmitted to the third learning model LM3.

[0088] The third learning model LM3 can train the third module M3 by comparing the first thickness ED and the second thickness PD of the internal defect layer. For example, if the first thickness ED and the second thickness PD of the internal defect layer do not match, or if the difference is greater than a predetermined value, the third learning model LM3 can adjust the weighting values, coefficients, etc., of the third operation in the third module M3 that calculates the second thickness PD of the internal defect layer. The third learning model LM3 can train the third module M3 so that the measured value of the first thickness ED and the calculated value of the second thickness PD of the internal defect layer match, or the difference between them is less than or equal to a predetermined value.

[0089] The fourth learning model LM4 can receive the transfer speed PV of the hot-rolled steel sheet, the characteristics AC of the etching solution used in the etching process, and the first thickness EPA of the internal defect layer removed by the etching apparatus. The characteristics AC of the etching solution can include the concentration of the etching solution in contact with the hot-rolled steel sheet, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used. The transfer speed PV of the hot-rolled steel sheet can mean the speed at which the hot-rolled steel sheet moves while in contact with the etching solution. The first thickness EPA of the internal defect layer can mean the thickness of the internal defect layer actually removed in the etching process performed by the transfer speed PV of the hot-rolled steel sheet and the characteristics AC of the etching solution.

[0090] The fourth module M4 can receive the transfer speed PV of the hot-rolled steel sheet and the characteristics AC of the etching solution as input from the fourth learning model LM4. The fourth module M4 can calculate the second thickness PPA1 of the internal defect layer expected to be removed by the etching apparatus using at least one of the transfer speed PV of the hot-rolled steel sheet and the characteristics AC of the etching solution. In one embodiment, the fourth module M4 receives the etching time determined by the transfer speed PV and the characteristics AC of the etching solution as input values, and can perform a calculation to calculate the second thickness PPA1 of the internal defect layer using the above input values. The fourth module M4 can output the second thickness PPA1 of the internal defect layer to the fourth learning model LM4.

[0091] The fourth learning model LM4 can train the fourth module M4 by comparing the second thickness PPA1 of the internal defect layer with the first thickness EPA of the internal defect layer. For example, if the first thickness EPA of the internal defect layer and the second thickness PPA1 of the internal defect layer do not match, or if the difference is greater than a predetermined value, the fourth learning model LM4 can adjust the weighting values, coefficients, etc., of the fourth operation used to calculate the second thickness PPA1 of the internal defect layer in the fourth module M4.

[0092] Module 5, M5, can generate control data for controlling the etching apparatus. For example, Module 5, M5, can find the optimal control data by repeatedly calling Module 4, M4, using an optimization technique (e.g., the golden ratio). Module 5, M5, can select an optimization technique to find the optimal control data or modify an optimization technique.

[0093] Figure 6 is a diagram illustrating a learning model included in a process control system according to one embodiment of the present invention.

[0094] Referring to Figure 6, when a learning model uses reinforcement learning to train each module, the learning model can be implemented using a deep neural network (DNN), etc. The learning model can include an input layer IL, a hidden layer HL, and an output layer OL. For example, multiple nodes included in the input layer IL, the hidden layer HL, and the output layer OL can be fully connected to each other. The input layer IL can include multiple input nodes x1 to xi, and the number of input nodes x1 to xi can correspond to the number of input data. The output layer OL can include multiple output nodes y1 to yj, and the number of output nodes y1 to yj can correspond to the number of output data.

[0095] The hidden layer HL can include the first to third hidden layers HL1 to HL3, and the number of hidden layers HL1 to HL3 can be varied in various ways. For example, the learning model 120 can be trained by adjusting the weight values ​​of each hidden node included in the hidden layer HL.

[0096] Figure 7 is a diagram illustrating the operation method of a process control system according to one embodiment of the present invention.

[0097] Referring to Figure 7, a process control system 200 according to one embodiment of the present invention may include calculation modules 210 to 230. For example, the process control system may be connected via an external system and network that directly controls an etching apparatus for removing at least a portion of the internal defect layer of a hot-rolled steel sheet.

[0098] In one embodiment, calculation modules 210-230 can calculate the thickness of the internal defect layer contained in the hot-rolled steel sheet. In one embodiment, calculating the thickness of the internal defect layer may require the cooling rate of the hot-rolled steel sheet along its length. If the phase transformation is not completed before the hot-rolled steel sheet is wound, heat generation due to the phase transformation after winding may occur. Therefore, the temperature change of the hot-rolled steel sheet after winding can be calculated by considering the phase fraction of the hot-rolled steel sheet before winding.

[0099] In one embodiment, the first module 210 can calculate the phase fraction PPF according to the cooling time using the first process condition CD1. The first process condition CD1 may include the cooling rate of the hot-rolled steel sheet, the temperature of the hot-rolled steel sheet, and the composition of the hot-rolled steel sheet.

[0100] In hot-rolled steel sheets, a phase transformation from austenite to pearlite can occur during the cooling process, and this transformation can result in a rise in temperature due to transformation exothermic reaction. In hot-rolled high-carbon steel, the phase transformation may not be completed during cooling due to a slow pearlite transformation, and further phase transformation may occur in the hot-rolled steel sheet after coiling. As a result, the hot-rolled steel sheet may be exposed to a high-temperature oxidizing atmosphere for a long time after coiling, which can increase the thickness of the internal defect layer.

[0101] The second module 220 can calculate the temperature PT of the hot-rolled steel sheet according to the elapsed time after coiling, using the second process condition CD2 and environmental condition FV, etc. As an example, the second module 220 can calculate the temperature PT in each region along the length direction of the hot-rolled steel sheet. The second process condition CD2 may include the time at which the temperature of the hot-rolled steel sheet was measured after coiling, and the composition of the hot-rolled steel sheet, etc.

[0102] In order to calculate the temperature of the wound hot-rolled steel sheet, the transformation heat generated as described above must also be taken into consideration. Therefore, the second module 220 can receive the phase fraction PPF from the first module 210. Based on the phase fraction PPF, the phase fraction at the time of winding can be determined, and the amount of further phase transformation that occurs in the hot-rolled steel sheet after winding can be determined. Therefore, the second module 220 can calculate the temperature PT of the hot-rolled steel sheet according to the elapsed time after winding, taking into account the amount of further phase transformation that occurs in the wound hot-rolled steel sheet. In one embodiment shown in Figure 7, time t1 can represent the temperature that reflects the transformation heat generated by the phase transformation.

[0103] The third module 230 can calculate the thickness PD of the internal defect layer contained in the hot-rolled steel sheet using the components ING of the hot-rolled steel sheet and the oxygen partial pressure OPP around the hot-rolled steel sheet. In one embodiment, the third module 230 receives the temperature PT corresponding to the elapsed time after winding in each region of the hot-rolled steel sheet from the second module 220 and can calculate the thickness PD of the internal defect layer in each of the above regions of the hot-rolled steel sheet.

[0104] Figure 8 is a flowchart illustrating the operation method of a process control system according to one embodiment of the present invention.

[0105] Referring to Figure 8, the process control system can transfer at least a portion of the calculation module to an external system (S110). As an example, the entity that transfers at least a portion of the calculation module to an external system may be a system that produces hot-rolled steel sheets and winds the hot-rolled steel sheets into hot-rolled coils. The external system that receives at least a portion of the calculation module may be a system that transports the hot-rolled coils and performs an etching process.

[0106] The external system can perform the etching process using the calculation module received in step S110. For example, the calculation module transferred to the external system in step S110 may be a module that generates control data for controlling the etching process. The process control system can receive from the external system the thickness of the remaining internal defect layer contained in the pickled steel sheet after the etching process is completed, and control data measured by the etching apparatus during the etching process (S120). For example, the control data received in step S120 may be the value actually measured by the etching apparatus, and therefore may differ from the value input to the etching apparatus by the external system for controlling the etching apparatus.

[0107] The process control system can use the control data received in step S120 to calculate the thickness of the internal defect layer that is expected to be removed by the etching apparatus (S130). The process control system can also compare the thickness of the internal defect layer calculated in step S130 with the thickness of the remaining internal defect layer received in step S120 and train the calculation module based on the comparison result (S140). The process control system may include a learning model necessary for training the calculation module, and the learning model can adjust weights, coefficients, etc., used in calculations in at least one of the calculation modules.

[0108] Once learning is complete, the process control system can transfer at least one of the calculation modules to the external system (S150). The external system can then update existing calculation modules by overwriting them using the calculation module received from the process control system. Therefore, the external system can perform the etching process in an optimized manner that takes into account the thickness distribution of the internal defect layer contained in the hot-rolled steel sheet.

[0109] Figure 9 is a block diagram that simply illustrates a process control system according to one embodiment of the present invention.

[0110] Referring to Figure 9, the process control system 50 may include a second server 51 and an etching apparatus 55, the second server 51 being able to control and manage the etching apparatus 55. Hot-rolled coils produced from the hot-rolled steel sheet production apparatus 45 described above are transferred to the etching apparatus 55, and after the hot-rolled coils are wound into the form of hot-rolled steel sheets, an etching process is performed on the hot-rolled steel sheets to produce pickled steel sheets.

[0111] The etching apparatus 55 may be at least one of a pickling apparatus, a dry etching apparatus, and a wet etching apparatus. If the etching apparatus 55 is a pickling apparatus, the internal defect layer can be removed by immersing the hot-rolled steel sheet in an acidic etching solution. Alternatively, the internal defect layer can be removed by an etching solution sprayed onto the surface of the hot-rolled steel sheet, or by brushing the surface of the hot-rolled steel sheet with a brush immersed in the etching solution.

[0112] The second server 51 may include a communication unit 52, storage 53, a processor 54, and the like. The communication unit 52 can be connected to the second server 51 to communicate with a network, and for example, the second server 51 can communicate with the first server 41 via the communication unit 52. The first server 41 may be a server that controls the hot-rolled steel sheet production apparatus 45, as previously explained with reference to Figure 4. The storage 53 can store data necessary for the operation of the second server 51 and the control of the etching apparatus 55. The processor 54 can control the communication unit 52, storage 53, and the etching apparatus 55, etc.

[0113] In one embodiment, the storage 53 can store control data necessary for controlling the etching apparatus 55. For example, the control data stored in the storage 53 may be data obtained by the processor 54 by executing an arithmetic module stored in the storage 53, or data received from an external system connected via the communication unit 52. For example, if the control data is stored in the storage 53 without an arithmetic module, the control data may be transferred from an external system that manages the production of hot-rolled steel sheets.

[0114] The storage 53 may include a calculation module that generates control data. The calculation module can perform calculations to control the etching process carried out by the etching apparatus 55. For example, the calculation module can generate control data that allows the etching apparatus 55 to be controlled to efficiently remove internal defect layers, taking into account the thickness distribution of internal defect layers contained in the hot-rolled steel sheet. In one embodiment, the calculation module stored in the storage 53 may be a calculation module received and stored from an external system via the communication unit 52, and the external system may be a system that controls a hot-rolled steel sheet production apparatus. However, the external system is not necessarily limited to a system that controls a hot-rolled steel sheet production apparatus, and calculation modules can be received from various external systems depending on the embodiment.

[0115] As an example, the calculation module can generate control data using the thickness information of the internal defect layer received by the second server 51 via the communication unit 52. The processor 54 can input the thickness information of the internal defect layer into the calculation module stored in the storage 53 and obtain the control data by executing the calculation module. The thickness information of the internal defect layer may include the thickness distribution of the internal defect layer appearing along the length direction of the hot-rolled steel sheet.

[0116] In this embodiment, the calculation module stored in the storage 53 can also generate control data using information about the hot-rolled steel sheet received via the communication unit 52. In this case, the calculation module can calculate the thickness information of the internal defect layer contained in the hot-rolled steel sheet using the phase fraction of the hot-rolled steel sheet, the temperature of the hot-rolled steel sheet, the composition of the hot-rolled steel sheet, etc. The calculation module can also generate control data using the thickness information of the internal defect layer. As an example, the calculation module stored in the storage 53 may include a thickness information generation module that calculates thickness information and a control data generation module that generates control data.

[0117] The control data may include at least one of the following: the transfer speed at which the etching apparatus 55 transfers the hot-rolled steel sheet; the concentration of the etching solution that comes into contact with the hot-rolled steel sheet in the etching apparatus 55; the temperature of the etching solution; the components of the etching solution; and whether or not an accelerator is used. As an example, in the thickness information of the internal defect layer, the thickness of the internal defect layer may appear differently in the first region and the second region of the hot-rolled steel sheet. In this case, the first region and the second region may be regions defined at different positions along the length direction of the hot-rolled steel sheet.

[0118] In the control data, the first transfer speed of the hot-rolled steel sheet while the first region passes through the etching apparatus 55 and comes into contact with the etching solution can be different from the second transfer speed of the hot-rolled steel sheet while the second region passes through the etching apparatus 55 and comes into contact with the etching solution. For example, if the thickness of the internal defect layer in the first region is smaller than the thickness of the internal defect layer in the second region, the first transfer speed can be faster than the second transfer speed.

[0119] The calculation module that generates control data can determine the transfer speed of the hot-rolled steel sheet according to the thickness of the internal defect layer. In one embodiment, the calculation module can determine the transfer speed of the hot-rolled steel sheet by performing a predetermined calculation using the thickness of the internal defect layer. In another embodiment, the calculation module can also compare a predetermined reference thickness with the thickness of the internal defect layer and determine the transfer speed of the hot-rolled steel sheet according to the comparison result.

[0120] In one embodiment, the second server 51 can acquire information necessary to update the arithmetic module from the etching apparatus 55 and feed back the information acquired from the etching apparatus 55 to an external system via the communication unit 52. Here, the external system may be a system that stores and manages the arithmetic module and includes a learning model for updating the arithmetic module. If the learning model for updating the arithmetic module is included in the storage 53, the arithmetic module update operation using the information acquired from the etching apparatus 55 can also be performed by the second server 51.

[0121] For example, the external system may be a system that controls the production process of hot-rolled steel sheets. The external system can update at least some of the calculation modules stored in the external system using information fed back from the second server 51. For example, weight values, coefficients, etc., applied to perform calculations in at least some of the calculation modules stored in the external system can be adjusted.

[0122] Furthermore, the external system can transfer at least one of the updated calculation modules, for example, a calculation module that generates control data, to the second server 51. The second server 51 can then update existing calculation modules stored in storage 53 using the new calculation module received from the external system. For example, the second server 51 can update calculation modules by overwriting existing ones with new ones. Therefore, the etching process can be performed with calculation modules optimized to suit various conditions of the etching apparatus 55 and / or the hot-rolled steel sheet being etched, thereby improving the efficiency and productivity of the etching process.

[0123] In another embodiment, the second server 51 can obtain information necessary to update the arithmetic module from the etching apparatus 55 and directly update the arithmetic module stored in the storage 53 with the information obtained from the etching apparatus 55. Alternatively, as an example, the storage 53 can store only control data for controlling the etching apparatus 55 without a separate arithmetic module. In this case, the second server 51 can feed back the information necessary to update the arithmetic module obtained from the etching apparatus 55 to an external system, and receive the new control data generated by the updated arithmetic module in the external system and store it in the storage 53.

[0124] Figure 10 is a diagram illustrating the operation method of a process control system according to one embodiment of the present invention.

[0125] Referring to Figure 10, the process control system 300 can include calculation modules 310 and 320. For example, calculation modules 310 and 320 can be modules stored and executed in the process control system 300 that controls the etching apparatus. Alternatively, calculation modules 310 and 320 can also be stored in an external system connected to the process control system 300 via a network.

[0126] The calculation modules 310 and 320 can be transferred to and stored in the process control system 300 after learning is completed by a learning model of an external system. In one embodiment, the calculation modules 310 and 320 stored and executed in the process control system 300 that controls the etching apparatus can be modules received from an external system. Referring to the embodiment shown in Figure 5 as an example, among the calculation modules 110 shown in Figure 5, the control data generation module 112 can be transferred to and stored in the process control system 300 as calculation modules 310 and 320.

[0127] The process control system 300 can determine control data for controlling the etching apparatus. The process control system 300 can receive internal defect layer thickness information PD. The internal defect layer thickness information PD received by the process control system 300 can be the thickness of the internal defect layer calculated as being included in the hot-rolled steel sheet, similar to the embodiment described earlier with reference to Figure 9, and may include the thickness distribution of the internal defect layer according to the region of the hot-rolled steel sheet. The process control system 300 can calculate control data PPV2 based on the internal defect layer thickness information PD.

[0128] In one embodiment, the calculation modules 310 and 320 may include a first module 310 and a second module 320, etc. The second module 320 can receive internal defect layer thickness information PD. The internal defect layer thickness information PD is the expected thickness of the internal defect layer present in the hot-rolled steel sheet, and may not exactly match the actual thickness of the internal defect layer present in the hot-rolled steel sheet. The second module 320 can also receive the target thickness value GD of the remaining internal defect layer. The remaining internal defect layer may be the internal defect layer present in the pickled steel sheet after the etching process is completed. For example, if the intention is to completely remove the internal defect layer by the etching process, the target thickness value GD of the remaining internal defect layer may be 0. Depending on the embodiment, the target thickness value GD of the remaining internal defect layer may also be greater than 0. The second module 320 can periodically call the first module 310 to calculate the optimal control data PPV2.

[0129] The second module 320 can output initial values ​​of control data to the first module 310. The initial values ​​of the control data may include the etching solution characteristics AC measured from the tank containing the etching solution in the etching apparatus, and the transfer speed PPV1 of the hot-rolled steel sheet. The initial values ​​of the control data can be determined by arbitrary values ​​or by values ​​measured in real time from the etching apparatus. If the initial values ​​of the control data are determined arbitrarily, the initial values ​​of the control data can be determined by an external system operating the hot-rolled steel sheet production equipment or a process control system operating the etching apparatus. The first module 310 can use at least one of the etching solution characteristics AC and the transfer speed PPV1 to calculate the thickness PPA2 of the internal defect layer that is expected to be removed by the etching apparatus.

[0130] The second module 320 can receive the thickness PPA2 of the internal defect layer from the first module 310. Based on the thickness PPA2 of the internal defect layer, the second module 320 determines whether the etching solution characteristics AC and the arbitrarily set transfer speed PPV1 conform to the optimal control data PPV2. As an example, the second module 320 can compare the difference between the thickness information PD of the internal defect layer and the target thickness GD of the remaining internal defect layer with the thickness PPA2 of the internal defect layer calculated by the first module 310. If the comparison results do not match, the second module 320 can adjust at least one of the etching solution characteristics AC and the transfer speed PPV1 and input it to the first module 310. In the above manner, the second module 320 can change at least one of the etching solution characteristics AC and the transfer speed PPV1 and call the first module 310 until the difference between the thickness information PD of the internal defect layer and the target thickness GD of the remaining internal defect layer matches the thickness PPA2 of the internal defect layer calculated by the first module 310, or until the difference is less than or equal to a predetermined value.

[0131] Table 1 is a table illustrating how the second module 320 generates optimal control data PPV2. For the sake of explanation, it is assumed that the etching solution properties AC in Table 1 are identical. The second module 320 can calculate the target removal thickness GPA, which indicates how much of the internal defect layer needs to be removed, based on the difference between the expected thickness information PD of the internal defect layer present in the hot-rolled steel sheet and the target thickness GD of the remaining internal defect layer.

[0132] The regions X, Y, and Z defined in [Table 1] can be regions defined at different locations along the length of the hot-rolled steel sheet. For example, regions X and Z can be adjacent to the edges of the hot-rolled steel sheet in the length direction compared to region Y. In other words, region Y can be located between regions X and Z in the length direction. Referring to [Table 1], in the first region, the target removal thickness GPA can be determined to be 1 μm (= 3 μm - 2 μm), which is the difference between the thickness information PD of the internal defect layer calculated as existing in the hot-rolled steel sheet and the target thickness GD of the remaining internal defect layer.

[0133] Referring to [Table 1], when the transfer speed PPV1 is 15 mpm, the first module 310 can determine that the thickness PPA2 of the internal defect layer expected to be removed by the etching device can be 2 μm. Since the thickness PPA2 of the internal defect layer expected to be removed by the etching device is greater than the target removal thickness GPA, the second module 320 can increase the transfer speed PPV1 to 20 mpm, which is faster than 15 mpm, and output it to the first module 310. When the transfer speed PPV1 is 20 mpm, if the first module 310 calculates that the thickness PPA2 of the internal defect layer expected to be removed by the etching device is a value different from 1 μm, the second module 320 can change the transfer speed PPV1 to a value different from 20 mpm again and output it to the first module 310. On the other hand, when the transfer speed PPV1 is 20 mpm, if the thickness of the internal defect layer PPA2 calculated by the first module 310 is 1 μm, the thickness of the internal defect layer PPA2 expected to be removed by the etching apparatus matches the target removal thickness GPA, so the second module 320 can determine that the optimal transfer speed for region X of the hot-rolled steel sheet is 20 mpm.

[0134] [Table 1]

[0135] In the embodiment shown in Figure 10, it was explained that the arithmetic modules 310 and 320 of the process control system 300 generate control data, but the system is not necessarily limited to this configuration. Depending on the embodiment, the arithmetic module of an external system connected to the process control system 300 via a network can also generate control data. In this case, the process control system 300 can receive the control data generated by the external system via the network and control the etching apparatus using the received control data.

[0136] Therefore, the process control system 300 can generate control data, including the transfer speed of the hot-rolled steel sheet, based on the thickness information PD of the internal defect layer. As an example, the thickness of the internal defect layer of the hot-rolled steel sheet may differ in multiple regions defined along the length direction as described above. For example, the thickness of the internal defect layer in a region cooled while exposed to the outside may be even smaller than the thickness of the internal defect layer in a region cooled slowly without being exposed to the outside.

[0137] As an example, the etching process can be carried out by ensuring that the hot-rolled steel sheet is in sufficient contact with the etching solution regardless of the area of ​​the sheet, so that the internal defect layer contained in the hot-rolled steel sheet is sufficiently removed. However, the above method may lead to an increase in the etching process time and / or the amount of etching solution introduced into the etching process, resulting in a decrease in productivity.

[0138] According to one embodiment of the present invention, the thickness of the internal defect layer can be calculated along the length direction of the hot-rolled steel sheet, and control data can be calculated that allows the etching process to be controlled to optimal efficiency based on the calculated thickness information of the internal defect layer. Therefore, productivity can be improved by shortening the etching process time and reducing the amount of etching solution used. In addition, it is possible to reduce the variation in the thickness of the internal defect layer that appears along the length direction in the pickled steel sheet after the etching process is completed.

[0139] For example, in the longitudinal direction of a hot-rolled steel sheet, the region where the thickness of the internal defect layer is less than a predetermined reference thickness t can be defined as the first region, and the region where the thickness of the internal defect layer is greater than the reference thickness t can be defined as the second region. In this case, the transport speed of the hot-rolled steel sheet while the first region is in contact with the etching solution can be faster than the transport speed of the hot-rolled steel sheet while the second region is in contact with the etching solution. Alternatively, the transport speeds for the first and second regions can be determined by a calculation that takes the thickness of the internal defect layer as input and determines the transport speed, without setting a separate reference thickness t.

[0140] As a result, the second region can be in contact with the etching solution for a longer period than the first region. Therefore, variations in the thickness of the remaining internal defect layer in the pickled steel sheet after the etching process can be minimized.

[0141] Depending on the embodiment, the reference thickness t can be set to multiple values. For example, if there are two reference thicknesses (t1, t2), the hot-rolled steel sheet can be divided into at least three regions depending on the thickness of the internal defect layer. In other words, if there are n reference thicknesses, the hot-rolled steel sheet can be divided into n+1 regions. If the magnitude of the reference thickness is defined as shown in mathematical formula 1, the transport speed at which the hot-rolled steel sheet is transported while each of the n+1 regions contained in the hot-rolled steel sheet comes into contact with the etching solution can be defined as shown in mathematical formula 2 below. [Mathematical formula 1] t1 <t2<t3<…<tn [Mathematical formula 2] Transfer speed in the first region > Transfer speed in the second region > ... > Transfer speed in the (n+1)th region

[0142] Depending on the embodiment, the etching rate can be adjusted in addition to the transfer speed of the hot-rolled steel sheet. For example, areas with a relatively small thickness of internal defect layer can be brought into contact with an etching solution having a low etching rate, while areas with a relatively large thickness of internal defect layer can be brought into contact with an etching solution having a high etching rate. Alternatively, the etching rate can be adjusted by changing the area of ​​the brush immersed in the etching solution or the pressure of the brush.

[0143] Figure 11 is a diagram illustrating a learning method for a computation module included in a process control system according to one embodiment of the present invention, Figure 12 is a diagram illustrating an etching process according to one embodiment of the present invention, and Figure 13 is a graph provided to illustrate an etching process according to one embodiment of the present invention.

[0144] First, referring to Figure 12, the etching apparatus 500 can carry out the etching process according to the control data PPV2 determined by the embodiment described earlier with reference to Figure 10, thereby removing at least a portion of the internal defect layer of the hot-rolled steel sheet 510. Referring to Figure 12, while the etching apparatus 500 is performing the etching process, measuring instruments MI1 and MI2 can measure the thickness ERD of the remaining internal defect layer in at least a portion of the area of ​​the pickled steel sheet 520 after the etching process is complete, and control data EPV can be collected from the tank TK, etc., where the etching process is performed. As an example, the thickness ERD of the remaining internal defect layer can be measured in multiple areas, and therefore position information PI can also be collected. The control data EPV collected by the first measuring instrument MI1 may have a different value from the control data PPV2 input to the etching apparatus 500 due to mechanical errors, signal delays, process errors, etc., as it is the control data EPV actually measured in the etching apparatus 500, including the tank TK. The first measuring instrument MI1 can acquire at least one of the transfer speed of the hot-rolled steel sheet 510 and the characteristics of the etching solution that comes into contact with the hot-rolled steel sheet 510 in the tank TK.

[0145] Referring to Figure 12, the etching apparatus 500 may include a coiler CL, an uncoiler UCL, a tank TK, a first measuring instrument MI1, a second measuring instrument MI2, etc. The uncoiler UCL can allow a strip of hot-rolled steel sheet 510, wound from a hot-rolled coil HC, to enter the tank TK. The hot-rolled steel sheet 510 can be etched by passing it through an etching solution contained in the tank TK. Alternatively, while the hot-rolled steel sheet 510 is passing through the tank TK, it may also be etched by an etching solution sprayed from around it and / or by a brush immersed in the etching solution.

[0146] While the etching process is underway, the etching apparatus 500 can be controlled by control data PPV2. For example, the transfer speed of the hot-rolled steel sheet 510 may differ in each region along the length of the hot-rolled steel sheet 510 according to the control data PPV2. The first measuring instrument MI1 connected to the tank TK can measure the control data EPV at at least one measurement point while the etching apparatus 500 is performing the etching process. As described above, due to mechanical errors, operational delays, etc., at least a portion of the control data PPV2 input to the etching apparatus 500 and the control data EPV measured by the first measuring instrument MI1 from the etching apparatus 500 may have different values. The first measuring instrument MI1 can transfer the measured control data EPV to the external system 400.

[0147] Referring to Figure 11, the external system 400 can be a system that manages the arithmetic modules 411-415:410 and may include a learning model LMT for training and updating the arithmetic modules 410. For example, the external system 400 can update the arithmetic modules 410 using information collected by the first measuring instrument MI1 and the second measuring instrument MI2 in the etching apparatus 500. By controlling the etching apparatus 500 with the control data generated by the updated arithmetic modules 410, the etching apparatus 500 can be operated in an optimal state.

[0148] In one embodiment shown in Figure 13, the horizontal axis represents the position along the length of the hot-rolled steel sheet, and the vertical axis represents the thickness PD of the internal defect layer and the transport speed. For example, the thickness information PD of the internal defect layer can be a value calculated by at least a part of the calculation module 410 included in the external system 400, assuming its presence in the hot-rolled steel sheet.

[0149] Referring to Figure 13, the thickness PD of the internal defect layer can be calculated to be greater in region Y than in regions X and Z. Regions X, Y, and Z can be regions that appear sequentially along the length direction, similar to those described earlier with reference to Table 1, and regions X and Z can be regions that are closer to the edge of the hot-rolled steel sheet in the length direction compared to region Y. Region Y can be the region between regions X and Z.

[0150] In one embodiment, the calculation module 410 of the external system 400 can calculate the transfer speed OPV to the hot-rolled steel sheet from the thickness PD of the internal defect layer. The transfer speed OPV can be the speed at which the hot-rolled steel sheet 510 is actually transferred from the etching apparatus 500 when the etching process is performed according to the control data PPV2 input to the etching apparatus 500. Since it can be calculated that the thickness of the internal defect layer is greater in region Y than in regions X and Z, the transfer speed OPV while region Y is in contact with the etching solution may be relatively slow so that the internal defect layer in region Y is sufficiently removed.

[0151] The actual transfer speed EPV may be the value actually measured in the etching apparatus 500 that performed the etching process based on the transfer speed OPV. The actual transfer speed EPV may differ from the transfer speed OPV set by the control data PPV2 because it may reflect conditions such as errors and operational delays of the etching apparatus 500.

[0152] As an example, in order to completely remove the internal defect layer, the etching process can be performed by maintaining a uniform transfer speed TPV based on the thickness of the internal defect layer present in the second region D2. However, in this case, the progress speed of the etching process may decrease, potentially reducing productivity. On the other hand, according to one embodiment of the present invention, the etching process can be performed by applying a differently calculated transfer speed OPV to each region of the hot-rolled steel sheet. Therefore, not only can the internal defect layer be efficiently removed, but the time of the etching process and the amount of etching solution introduced into the etching process can also be reduced. Thus, the productivity of the etching process can be improved.

[0153] Referring again to Figure 12, before the etching apparatus 500 winds up the pickled steel sheet 520 etched using the coiler CL to manufacture it into a pickled coil PC, the second measuring instrument MI2 can measure the position information PI of the pickled steel sheet 520 and the thickness ERD of the remaining internal defect layer at the position corresponding to the position information PI.

[0154] In the example, at least one measurement position can be designated on the etched pickled steel sheet 520, and a sample can be taken by cutting a portion of the pickled steel sheet 520 at the measurement position. By observing the cross-section of the sample taken from the pickled steel sheet 520 with a microscope, the thickness of the remaining internal defect layer can be measured. The measurement position can correspond to the position information PI of the pickled steel sheet 520, and the thickness of the remaining internal defect layer can correspond to the thickness ERD of the remaining internal defect layer at the position corresponding to the position information PI.

[0155] Referring again to Figure 11, the learning model 420 can receive the thickness ERD of the remaining internal defect layer measured by the second instrument MI2 of the etching apparatus 500 at at least one measurement position on the pickled steel sheet 520. The fifth module 415 can receive measured control data EPV from the etching apparatus 500 while the etching process is being carried out. For example, the fifth module 415 can receive measured control data EPV via a network from a process control system including the etching apparatus 500.

[0156] The fifth module 415 can output the control data EPV measured by the fourth module 414. Based on the measured control data EPV, the fourth module 414 can calculate the thickness PPA2 of the internal defect layer that is expected to be removed by the etching apparatus 500. The fifth module 415 can receive the internal defect layer thickness PPA2 from the fourth module 414. Based on the internal defect layer thickness PPA2 that is expected to be removed by the etching apparatus 500, the fifth module 415 can calculate the expected thickness PRD of the remaining internal defect layer. The expected thickness PRD of the remaining internal defect layer can be the thickness of the internal defect layer that is expected to remain in the pickled steel sheet 520 after the etching process has been performed according to the measured control data EPV.

[0157] The learning model LMT can receive the predicted thickness PRD of the internal defect layer from the fifth module 415. The learning model LMT can compare the actual thickness ERD of the remaining internal defect layer measured from the pickled steel sheet 520 with the predicted thickness PRD of the remaining internal defect layer. The learning model LMT can train at least one of the calculation modules 411 to 415 so that the ERD of the remaining internal defect layer measured from the pickled steel sheet 520 matches the predicted thickness PRD of the remaining internal defect layer, or the difference between them is less than or equal to a predetermined value. For example, the learning model LMT can train all of the calculation modules 411 to 415 simultaneously, or it can selectively train only specific models.

[0158] Depending on the embodiment, the process control system 50 in Figure 9, which includes a server and an etching apparatus, may also include a learning model. The learning model included in the process control system 50 in Figure 9 can be trained on the arithmetic module included in the process control system 50 in Figure 9. Below, a method for training the learning model of the process control system 50 in Figure 9 on the arithmetic module included in the process control system 50 in Figure 9 will be described with reference to Figures 14 and 15.

[0159] Figure 14 is a diagram illustrating the initial learning method for a computation module included in a process control system according to one embodiment of the present invention.

[0160] Referring to Figure 14, the process control system 600 may include a calculation module 610 and a learning model 620, and the calculation module 610 and the learning model 620 may be stored in storage. For example, the calculation module 610 may be a control data generation module. The learning model 620 may include a first learning model LM1 and a feedback learning model LMT. The first learning model LM1 may be a model for initially training the first module M1, and the feedback learning model LMT may be a model for training at least one of the first module M1 and the second module M2 after the etching process is completed. In one embodiment, the feedback learning model LMT may also train the first module M1 and the second module M2 simultaneously. In some embodiments, a single learning model may perform all the training of the first module M1 and the second module M2 without distinguishing between the first learning model LM1 and the feedback learning model LMT.

[0161] The first learning model LM1 can receive the transfer speed PV of the hot-rolled steel sheet, the characteristics AC of the etching solution used in the etching process, and the first thickness EPA of the internal defect layer removed by the etching apparatus. The characteristics AC of the etching solution can include the concentration of the etching solution in contact with the hot-rolled steel sheet, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used. The transfer speed PV of the hot-rolled steel sheet can mean the speed at which the hot-rolled steel sheet moves while in contact with the etching solution. The first thickness EPA of the internal defect layer can mean the thickness of the internal defect layer actually removed in the etching process performed by the transfer speed PV of the hot-rolled steel sheet and the characteristics AC of the etching solution.

[0162] The first module M1 can receive the transfer speed PV of the hot-rolled steel sheet and the characteristics AC of the etching solution as input from the first learning model LM1. The first module M1 can calculate the second thickness PPA1 of the internal defect layer expected to be removed by the etching apparatus using at least one of the transfer speed PV of the hot-rolled steel sheet and the characteristics AC of the etching solution. In one embodiment, the first module M1 receives the etching time determined by the transfer speed PV and the characteristics AC of the etching solution as input values, and can perform a calculation to calculate the second thickness PPA1 of the internal defect layer using the above input values. The first module M1 can output the second thickness PPA1 of the internal defect layer to the first learning model LM1.

[0163] The first learning model LM1 can train the first module M1 by comparing the second thickness PPA1 of the internal defect layer with the first thickness EPA of the internal defect layer. For example, if the first thickness EPA of the internal defect layer and the second thickness PPA1 of the internal defect layer do not match, or if the difference is greater than a predetermined value, the first learning model LM1 can adjust the weighting values, coefficients, etc., of the first operation in the first module M1 that calculates the second thickness PPA1 of the internal defect layer.

[0164] The second module M2 can generate control data for controlling the etching apparatus. For example, the second module M2 can find the optimal control data by repeatedly calling the first module M1 using an optimization technique (e.g., the golden ratio method). The second module M2 can select an optimization technique to find the optimal control data, or it can modify an optimization technique.

[0165] Figure 15 is a diagram illustrating a learning method for a calculation module included in a process control system according to one embodiment of the present invention.

[0166] Referring to Figure 15, the process control system 700 can be a system that manages the arithmetic modules 711, 712, and 710, and can include a learning model LMT for training the arithmetic modules 710. For example, the process control system 700 can train the arithmetic modules 710 using information collected by the first measuring instrument MI1 and the second measuring instrument MI2 in the etching apparatus 500. By controlling the etching apparatus 500 with the control data generated by the trained arithmetic modules 710, the etching apparatus 500 can be operated in an optimal state.

[0167] The learning model 720 can receive the thickness ERD of the remaining internal defect layer measured by the second measuring instrument MI2 of the etching apparatus 500 at at least one measurement position of the pickled steel sheet 520. The second module 712 can receive the measured control data EPV from the etching apparatus 500 while the etching process is being carried out.

[0168] The second module 712 can output the control data EPV measured by the first module 711. Based on the measured control data EPV, the first module 711 can calculate the thickness PPA2 of the internal defect layer that is expected to be removed by the etching apparatus 500. The second module 712 can receive the thickness PPA2 of the internal defect layer from the first module 711. Based on the thickness PPA2 of the internal defect layer that is expected to be removed by the etching apparatus 500, the second module 712 can calculate the expected thickness PRD of the remaining internal defect layer. The expected thickness PRD of the remaining internal defect layer can be the thickness of the internal defect layer that is expected to remain in the pickled steel sheet 520 after the etching process has been performed according to the measured control data EPV.

[0169] The learning model LMT can receive the predicted thickness PRD of the internal defect layer from the second module 712. The learning model LMT can compare the actual thickness ERD of the remaining internal defect layer measured from the pickled steel sheet 520 with the predicted thickness PRD of the remaining internal defect layer. The learning model LMT can train at least one of the calculation modules 711 and 712 so that the ERD of the remaining internal defect layer measured from the pickled steel sheet 520 matches the predicted thickness PRD of the remaining internal defect layer, or the difference between them is less than or equal to a predetermined value. For example, the learning model LMT can train all of the calculation modules 712 and 712 simultaneously, or it can selectively train only specific models.

[0170] Figure 16 is a flowchart illustrating the operation method of a process control system according to one embodiment of the present invention.

[0171] Referring to Figure 16, the process control system controlling the etching process can receive a calculation module from an external system (S210). The calculation module received in S210 can generate control data for controlling the etching process. The process control system can perform the etching process using the control data generated by the calculation module (S220). The etching process can remove at least a portion of the internal defect layer contained in the hot-rolled steel sheet.

[0172] The process control system can transfer control data actually measured by the etching apparatus performing the etching process, and the thickness of the remaining internal defect layer contained in the pickled steel sheet after the etching process is completed, to an external system (S230). The control data transferred to the external system in step S230 can be the values ​​actually measured from the etching apparatus. The thickness of the remaining internal defect layer can be measured by taking a sample from a portion of the pickled steel sheet and examining it under a microscope, or by other methods.

[0173] The external system can train the calculation module using the control data received in step S230 and the remaining thickness of the internal defect layer. Once training is complete, the process control system can receive at least one of the trained calculation modules from the external system (S240). The process control system can update an existing calculation module with the calculation module received in step S240 (S250). Therefore, the etching process can be accurately controlled, reflecting process errors that may occur in the etching equipment, etc.

[0174] For example, the external system could mean a system that manages and controls the production process of hot-rolled steel sheets. This is an illustrative example, and the claims are not limited thereto.

[0175] Figure 17 is a diagram illustrating the operation method of a process control system according to one embodiment of the present invention.

[0176] Referring to Figure 17, the first system SYS1 for manufacturing hot-rolled steel sheets can generate internal defect layer thickness information PD (S310). The internal defect layer thickness information PD generated in step S310 can be the thickness of the internal defect layer calculated based on at least one of the following: temperature measured in each region of the hot-rolled steel sheet defined in the longitudinal direction based on the elapsed time after winding, the composition of the hot-rolled steel sheet, and the oxygen partial pressure around the hot-rolled steel sheet. The first system SYS1 can transfer the internal defect layer thickness information PD to the second system SYS2 via a network (S320).

[0177] The second system SYS2 can generate control data PPV for controlling the etching process of a hot-rolled steel sheet using the thickness information PD of the internal defect layer (S330). The second system SYS2 can perform an etching process to remove at least a portion of the internal defect layer of the hot-rolled steel sheet using the control data PPV (S340). During the etching process, the second system SYS2 can acquire position information PI that points to a specific location on the hot-rolled steel sheet, the thickness ERD of the remaining internal defect layer measured at the position corresponding to the position information PI, and control data EPV (S350). The control data EPV acquired in step S350 is data measured from the etching apparatus during the etching process and may not match the control data PPV generated in step S330 due to various factors such as process errors and signal delays. The second system SYS2 can transfer the information acquired in step S350 to the first system SYS1 (S360).

[0178] The first system SYS1 can calculate the predicted thickness PRD of the remaining internal defect layer using the information received in step S360 (S370). The first system SYS1 can compare the predicted thickness PRD of the remaining internal defect layer calculated in step S370 with the thickness ERD of the remaining internal defect layer measured at at least one location (S380). The first system SYS1 can train at least one of the calculation modules included in the first system SYS1 so that the predicted thickness PRD of the remaining internal defect layer calculated at a specific location on the hot-rolled steel sheet matches the measured thickness ERD of the remaining internal defect layer, or the difference between them is less than or equal to a predetermined value (S390).

[0179] The first system SYS1 can transfer at least one of the learned arithmetic modules to the second system SYS2 (S400). The second system SYS2 can use the arithmetic module received from the first system SYS1 to update the arithmetic modules that were previously stored in the second system SYS2 (S410).

[0180] The present invention is not limited by the embodiments described above or the accompanying drawings, but is limited by the claims provided. Therefore, various forms of substitution, modification, and alteration are possible by persons with ordinary skill in the art, without departing from the technical idea of ​​the present invention as described in the claims, and these also fall within the scope of the present invention.

Claims

1. A process control system, A first system for generating thickness information of internal defect layers contained in carbon steel products, A second system receives thickness information of the internal defect layer from the first system via a network and controls an etching process to remove at least a portion of the internal defect layer from the carbon steel product using the thickness information of the internal defect layer. Includes, The first system provides the second system with a calculation module necessary for the second system to control the etching process. The second system is a process control system that provides the first system with information necessary for the first system to update the calculation module.

2. The process control system according to claim 1, wherein the first system provides the second system with a calculation module that determines at least one of the following in the etching process: the transfer speed of the carbon steel product, the concentration of the etching solution that comes into contact with the carbon steel product in the etching process, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used.

3. The process control system according to claim 1, wherein the second system provides the first system with at least one of the thickness of the remaining internal defect layer contained in the pickled carbon steel product after the etching process is completed, and control data measured by the etching apparatus performing the etching process.

4. The process control system according to claim 1, wherein the second system controls the etching process by adjusting at least one of the following: the transfer speed at which the etching apparatus performing the etching process transfers the carbon steel product, the concentration of the etching solution that comes into contact with the carbon steel product in the etching apparatus, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used.

5. The carbon steel product includes a first region in which the internal defect layer has a first thickness, and a second region in which the internal defect layer has a second thickness different from the first thickness. The process control system according to claim 4, wherein the second system transports the carbon steel product at a first transport speed while the first region is in contact with the etching solution, and transports the carbon steel product at a second transport speed different from the first transport speed while the second region is in contact with the etching solution.

6. The first thickness is smaller than the second thickness. The process control system according to claim 5, wherein the first transfer speed is faster than the second transfer speed.

7. The process control system according to claim 1, wherein the etching process is at least one of a pickling process, a dry etching process, and a wet etching process.

8. The process control system according to claim 1, wherein the first system stores the calculation module and the learning model for training the calculation module.

9. The process control system according to claim 1, wherein the first system measures the thickness of the internal defect layer contained in the carbon steel product and generates thickness information of the internal defect layer.

10. The process control system according to claim 9, wherein the first system measures the thickness of the internal defect layer in each of a plurality of regions defined in the longitudinal direction of the carbon steel product.

11. The process control system according to claim 1, wherein the first system calculates the thickness of the internal defect layer using at least one of the phase fraction of the carbon steel product, the composition of the carbon steel product, and the temperature of the carbon steel product.

12. The process control system according to claim 11, wherein the calculation module calculates the thickness of the internal defect layer in each of a plurality of regions defined in the longitudinal direction of the carbon steel product.

13. A process control system, A storage device for storing control data necessary for controlling an etching apparatus that removes at least a portion of the internal defect layer contained in carbon steel products, A processor that controls the etching apparatus based on the control data. Includes, The carbon steel product includes a first region and a second region different from the first region, wherein the thickness of the internal defect layer included in the first region and the thickness of the internal defect layer included in the second region are different from each other. A process control system in which the control data includes a first transfer speed at which the first region passes through the etching apparatus and a second transfer speed at which the second region passes through the etching apparatus, wherein the first transfer speed and the second transfer speed are different from each other.

14. The process control system according to claim 13, wherein the storage stores the arithmetic module that generates the control data.

15. It further includes a communication unit connected to a network, The process control system according to claim 13, wherein the processor receives the arithmetic module that generates the control data via the communication unit and stores it in the storage.

16. It further includes a communication unit connected to a network, The process control system according to claim 13, wherein the processor receives information related to the thickness of the internal defect layer via the communication unit, and inputs the information related to the thickness of the internal defect layer to a calculation module that generates the control data to obtain the control data.

17. It further includes a communication unit connected to a network, The process control system according to claim 13, wherein the processor receives information about the carbon steel product via the communication unit, inputs the information about the carbon steel product to a calculation module that generates the control data, and obtains the control data.

18. The process control system according to claim 17, wherein the information on the carbon steel product includes at least one of phase fraction-related information on the carbon steel product, temperature-related information on the carbon steel product, and component-related information on the carbon steel product.

19. It further includes a communication unit connected to a network, The process control system according to claim 13, wherein the processor receives the control data via the communication unit and stores it in the storage.

20. The process control system according to claim 13, wherein the control data further includes at least one of the concentration of the etching solution that comes into contact with the carbon steel product in the etching apparatus, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used.

21. The process control system according to claim 13, wherein the etching apparatus is at least one of a pickling apparatus, a dry etching apparatus, and a wet etching apparatus.

22. The process control system according to claim 13, wherein the thickness of the internal defect layer included in the first region is smaller than the thickness of the internal defect layer included in the second region, and the first transfer speed is faster than the second transfer speed.

23. The processor receives the transfer speed of the carbon steel product, the characteristics of the etching solution used in the etching process, and the first thickness of the internal defect layer removed by the etching apparatus. The storage includes a calculation module that calculates the second thickness of the internal defect layer expected to be removed by the etching apparatus using at least one of the transfer speed of the carbon steel product and the properties of the etching solution. The process control system according to claim 13, wherein the processor compares the second thickness of the internal defect layer with the first thickness of the internal defect layer to train the arithmetic module.

24. The storage includes a calculation module that receives control data measured from the etching apparatus and calculates the expected thickness of the remaining internal defect layer that is expected to be removed by the etching apparatus based on the measured control data. The process control system according to claim 13, wherein the processor compares the expected thickness of the remaining internal defect layer with the thickness of the remaining internal defect layer measured from the pickled carbon steel product after the etching process is completed, thereby training the calculation module.

25. A process control system, A storage system for storing a calculation module that generates thickness information of internal defect layers contained in a carbon steel product based on at least one of the carbon steel product's composition, cooling rate, phase fraction, and temperature, A communication unit connected to the network, A processor that, via the communication unit, transfers thickness information of the internal defect layer and at least one of control data for controlling the etching process to an external server that controls an etching process to remove at least a portion of the internal defect layer. A process control system including...

26. The process control system according to claim 25, wherein the storage stores a learning model for training the arithmetic module.

27. The process control system according to claim 25, wherein the calculation module includes a first module for calculating the phase fraction before winding the carbon steel product.

28. The process control system according to claim 27, wherein the processor compares the phase fraction measured in the carbon steel product with the phase fraction calculated by the first module from the temperature of the carbon steel product and at least one of the components of the carbon steel product to cause the first module to learn.

29. The process control system according to claim 27, wherein the calculation module further includes a second module for calculating the temperature change of the carbon steel product.

30. The process control system according to claim 29, wherein the processor causes the second module to learn by comparing the temperature change calculated by the second module using at least one of the phase fraction calculated before winding the carbon steel product, the elapsed time after winding the carbon steel product, and the components of the carbon steel product with the temperature change measured from the carbon steel product.

31. The process control system according to claim 29, wherein the calculation module further includes a third module for calculating the thickness of the internal defect layer contained in the carbon steel product.

32. The process control system according to claim 31, wherein the processor causes the third module to learn by comparing the thickness of the internal defect layer calculated by the third module using at least one of the temperature change calculated according to the elapsed time after winding the carbon steel product, the composition of the carbon steel product, and the oxygen partial pressure around the carbon steel product with the thickness of the internal defect layer measured from the carbon steel product.

33. The process control system according to claim 31, wherein the third module calculates the thickness of the internal defect layer in each of a plurality of regions divided along the length direction of the carbon steel product.

34. The process control system according to claim 31, wherein the calculation module further includes a fourth module that calculates the thickness of the internal defect layer to be removed in the etching step based on the control data.

35. The process control system according to claim 34, wherein the processor compares the thickness of the remaining internal defect layer contained in the pickled carbon steel product after the etching process is completed with the thickness of the internal defect layer calculated by the fourth module to cause the fourth module to learn.

36. The process control system according to claim 34, wherein the calculation module further includes a fifth module for generating the control data.

37. The process control system according to claim 36, wherein the processor uses the control data measured by the external server from the etching apparatus performing the etching process to compare the thickness of the internal defect layer calculated by the fourth module with the thickness of the remaining internal defect layer measured from the pickled carbon steel product after the etching process is completed, thereby training the calculation module.

38. The process control system according to claim 25, wherein the control data includes at least one of the transfer speed of the carbon steel product in the etching process, the concentration of the etching solution that comes into contact with the carbon steel product in the etching process, the temperature of the etching solution, the components of the etching solution, and whether or not an accelerator is used.

39. The carbon steel product includes a first region in which the internal defect layer has a first thickness, and a second region in which the internal defect layer has a second thickness different from the first thickness. The process control system according to claim 25, wherein the control data includes a first transfer speed at which the first region is transferred and a second transfer speed at which the second region is transferred in the etching process, and the first transfer speed and the second transfer speed are different from each other.

40. The process control system according to claim 25, wherein the etching process is at least one of a pickling process, a dry etching process, and a wet etching process.

Images