Information processing device

Abstract

This information processing device comprises a control unit. The control unit acquires information relating to defect positions obtained from a surface of a steel sheet before performing a prescribed step, and, on the basis of the information relating to the defect positions, creates a defect map indicating defect positions on each steel sheet in a step preceding or following the prescribed step, and / or creates an overlay map in which defect positions across a plurality of steel sheets in the step preceding or following the prescribed step are aggregated.

Description

Information processing device

[0001] The present invention relates to an information processing device.

[0002] In the cold rolling process of steel sheets, defects are sometimes detected using optical defect inspection devices or eddy current sensors before the cold rolling process in order to improve productivity and yield and reduce breakage. This makes it possible to check the occurrence of defects before the sheet is threaded in the cold rolling process, and to take appropriate measures to prevent breakage. Therefore, it can be said that defect detection before the cold rolling process contributes to reducing breakage in the cold rolling process.

[0003] An optical defect inspection device is a device that optically detects defects present on a steel plate surface by irradiating the surface with light and capturing an image of the reflected light. This inspection method often uses a line light source and a line sensor to capture images of both specularly reflected light and diffusely reflected light (see, for example, Patent Document 1).

[0004] Recently, a surface inspection device using quasi-specular reflected light has been developed, which shifts the angle of reflection by about 2 to 3 degrees from the specular reflection angle to reduce the influence of reflected light from the texture of the steel sheet and improve the accuracy of defect detection (see, for example, Patent Document 2).

[0005] In an eddy current sensor, an AC voltage is applied to an excitation coil to generate eddy currents in the test material, such as magnetic metal materials, while the material is being transported. These eddy currents then induce currents in two detection coils aligned perpendicular to the direction of transport of the test material, which are then detected. Defect detection is performed based on the differential signal between these detection signals.

[0006] The difference between optical defect inspection devices and eddy current sensors is that the former detects defects from the appearance, while the latter detects defects from changes in the electric field. Eddy current sensors also have the advantage of being able to detect unopened defects that are not visible from the outside (see, for example, Patent Document 3).

[0007] Furthermore, a technology has been developed that uses a defect inspection device to display defect locations and specify cut-off locations for steel sheets. This technology makes it possible to determine processing methods before production, contributing to improved production efficiency and yield (see, for example, Patent Document 4).

[0008] Japanese Patent Application Laid-Open No. 2008-145373 Japanese Patent Application Laid-Open No. 2021-169949 Japanese Patent Application Laid-Open No. 2006-284191 Special Publication No. 2023-531259

[0009] With conventional defect detection methods, it was necessary to check defect information for each steel plate (each coil), but this provided useful data for advance confirmation of defect location and severity during the cold rolling process. In particular, since the system for advance display of defect location was established, it has become possible to consider how to deal with defects before production, which has reduced the time required to slow down and consider how to deal with defects in order to check for defects, thereby contributing to improved production efficiency.

[0010] However, conventional defect detection methods have not been able to be used to identify and address the causes of defects that occur continuously, for example, because it is difficult to use them for trend analysis of defects that frequently occur at specific locations in the steelmaking process, which is a process before the cold rolling process, or in the hot rolling process. Similarly, conventional defect detection methods have also been difficult to use for correcting defects in the annealing process and finishing process, which are processes after the cold rolling process. Therefore, there has been a need to develop a system that can confirm the locations of defects in multiple steel plates, including the latest defect occurrence status.

[0011] The present invention has been made in consideration of the above, and aims to provide an information processing device that can check the defect location of a steel plate before a specified process and can take measures in advance to deal with defects in the steel plate that occur after the specified process.

[0012] In order to solve the above-mentioned problems and achieve the object, the information processing device of the present invention is equipped with a control unit, which acquires information regarding defect positions obtained from the surface of a steel plate before a specified process is performed, and based on the information regarding the defect positions, creates at least one of a defect map showing the defect positions for each steel plate in a process before or after the specified process, and a superposition map that aggregates defect positions across multiple steel plates in a process before or after the specified process.

[0013] In addition, the information processing device of the present invention is the above invention, wherein the defect map is formed by dividing the steel plate into a grid and each grid is colored according to the size of the defect within the grid, and the superposition map is formed by dividing the steel plate into a grid and each grid is colored according to the frequency of defect occurrence within the grid.

[0014] In the information processing apparatus according to the present invention, in the above invention, the defect map and the superposition map include positions of at least one of predetermined harmful defects and harmless defects.

[0015] In the information processing device according to the present invention, in the above invention, the predetermined process is a cold rolling process.

[0016] In addition, in the information processing device of the present invention, in the above invention, the control unit creates the superposition map by converting the defect positions of the steel plate in the specified process into defect positions of the steel plate in the previous process and aggregating them, and provides the created superposition map to a monitoring device that monitors the previous process.

[0017] In addition, in the information processing device of the present invention, in the above invention, the control unit identifies at least one of an abnormality in the equipment of the upstream process and an operational peculiarity based on the defect location included in the superposition map, and provides the identified abnormality to a monitoring device that monitors the upstream process.

[0018] In addition, in the information processing device of the present invention, in the above invention, the control unit creates the superposition map by converting and aggregating the defect positions of the steel plate in the specified process into defect positions of the steel plate in the subsequent process, and provides the created superposition map to a monitoring device that monitors the subsequent process.

[0019] In addition, in the information processing device of the present invention, in the above invention, the control unit identifies the processing content to be performed in the subsequent process based on the defect position included in the superposition map, and provides it to a monitoring device that monitors the subsequent process.

[0020] According to the information processing device of the present invention, it is possible to check the defect location of a steel plate before a predetermined process, and to take measures in advance against defects that occur in the steel plate after the predetermined process.

[0021] FIG. 1 is a diagram showing an example of an annealing line in which an eddy current sensor is installed. FIG. 2 is a diagram showing an example of the configuration of an eddy current sensor. FIG. 3 is a diagram showing an example of equipment used in defect detection using an eddy current sensor. FIG. 4 is a diagram showing an example of a pickling line in which an optical defect inspection device is installed. FIG. 5 is a diagram showing an example of equipment including an information processing device according to an embodiment. FIG. 6 is a diagram showing an example of defects detected by the optical defect inspection device. FIG. 7 is a diagram showing an example of a superposition map created by the information processing device according to an embodiment. FIG. 8 is a diagram showing an example of multiple defect maps that are the basis of the superposition map of FIG. 7. FIG. 9 is an example of an superposition map created based on defect information acquired by an optical defect inspection device, according to an embodiment of the present invention. FIG. 10 is an example of an superposition map created when frequent scabs occur at a predetermined position on a hot-rolled steel sheet, according to an embodiment of the present invention. FIG. 11 is an example of an superposition map created after reviewing the slab placement position of FIG. 10. FIG. 12 is an example of an embodiment of the present invention, showing a month in which frequent acid dripping occurred in the pickling line and an superposition map for the previous month.

[0022] An information processing device according to an embodiment of the present invention will be described with reference to the drawings. Note that the components in the following embodiments include those that are easily replaceable by those skilled in the art, or those that are substantially identical. Furthermore, in the following description, descriptions of identical or overlapping parts will be omitted or simplified as appropriate. Furthermore, in each of the drawings referred to below, identical or overlapping parts are assigned the same reference numerals.

[0023] For convenience of explanation, the steel sheet in this embodiment is assumed to be manufactured through various processes including a steelmaking process, a hot rolling process, a pickling process, a first cold rolling process, an intermediate annealing process, a second cold rolling process, a finish annealing process, and a finishing process. However, processes other than those described above may be included, or certain processes among those described above may not be included.

[0024] Before describing the specific contents of the present invention, the background to the invention and the usefulness of the defect map and superposition map of the present invention in the pre-process and post-process of a predetermined process (in this embodiment, the cold rolling process) will be described.

[0025] The inventors conducted the following studies when conceiving the present invention. For example, scabbing, which is a defect that occurs on the surface of a steel sheet, poses a risk of breaking the steel sheet or damaging equipment during the cold rolling process. Therefore, when scabbing occurs, it is necessary to roll at a slower speed, or to repair the defected area or perform an empty run. While these measures can reduce the risk of breaking the steel sheet, they also reduce the efficiency of the rolling line and result in a loss of yield at the scabbing cut-off area.

[0026] Therefore, reducing scabbing leads to improved production capacity in the cold rolling process.In addition, even for defects such as acid marks and streaks that do not pose a risk of breaking the steel sheet or damaging equipment in the cold rolling process, there are cases where the relevant parts are discarded as having poor appearance when the product is made.

[0027] To reduce scabbing, it is necessary to optimize or correct operational conditions in the steelmaking process and hot rolling process, which are processes preceding (upstream of) the cold rolling process. Furthermore, it is also necessary to identify the cause of scabbing and implement countermeasures based on information such as the location and frequency of scabbing confirmed in the cold rolling process. However, in the past, trend analysis of scabbing location was performed by analyzing reports for each steel plate (each coil), making it difficult to accurately identify scabbing locations or analyze scabbing trends across multiple steel plates.

[0028] Therefore, in order to solve the above-mentioned problems, the present invention considers that it is necessary to visualize the tendency of scab formation by mapping the frequency of scab formation and defect size at the hot-rolled steel sheet position obtained by, for example, an eddy current sensor or an optical defect inspection device through data analysis. Furthermore, optical defect inspection devices, in particular, can detect not only harmful defects such as scabs and edge cracks that show surface irregularities, but also defects such as patterns that are harmless in cold rolling but will result in a product being rejected. In this case, as with scabs, it can be used to identify the occurrence of defects in processes prior to the installation of the optical defect inspection device and to identify the equipment that causes the defects.

[0029] Furthermore, the present invention also enables a feedforward mechanism for transmitting defect information to a downstream process, similar to the feedback to a upstream process in the cold rolling process. Therefore, by utilizing the present invention, it is possible to not only treat defective parts, but also to check the occurrence status of defects that occur consecutively at a specific position, and to check them in a downstream process (downstream process).

[0030] Both eddy current sensors and optical defect inspection devices can identify the location of defects in the longitudinal and transverse directions, as well as on the front and back sides. Therefore, by linking the accumulated data on steel plate threading performance, cutoff position, and cutoff length in each process, it is possible to convert the defect location not only in the process itself but also in a specific process. This also makes it possible to provide feedback and feedforward of defect locations and defect occurrence status to other processes.

[0031] In particular, defects that cause fractures during the cold rolling process are almost always caused by the upstream steelmaking or hot rolling processes, making it meaningful to identify the location of defects in the upstream processes. Even if a defect does not affect cold rolling, if it results in a poor appearance or irregularities on the steel plate surface, it can affect the lamination of the steel plate, making it impossible to ship the product. To identify the process and causes of defects, it is important to analyze the tendency of defect locations in multiple steel plates.

[0032] As described above, there are two methods for inspecting defects occurring on the steel sheet surface: using an eddy current sensor and an optical defect inspection device. Therefore, when referring to information detected by different inspection methods, it is possible to refer to information not only from the process immediately before cold rolling, but also from several processes prior to the cold rolling process. It is also possible to cover defect occurrence conditions even when the detection situation differs due to different inspection methods. Furthermore, even if the steel sheet elongates after passing through the rolling process and its longitudinal position changes, the defect position can be easily confirmed in a subsequent process, making it possible to take measures to reduce risks during cold rolling, such as cutting off the defective portion. The present invention will be described in detail below.

[0033] 1 shows an example of an intermediate annealing line for an intermediate annealing process provided between a first cold rolling process and a second cold rolling process, and an eddy current sensor is installed in the intermediate annealing line. The intermediate annealing line includes a payoff reel 1, an entry shear 2, a welding machine 3, a notcher 4, a deflation roll 5, a bridle roll 6, an annealing furnace 7, and a water cooling device 8. The intermediate annealing line also includes a pickling tank 9, a pickling tank brush 10, a rinsing dryer 11, a rinsing tank 12, a dryer 13, a bridle roll 14, a welding point detector (WPD) 15, an exit shear 16, and a tension reel 17.

[0034] 2, the eddy current sensor 18 applies a magnetic field to the steel sheet, and detects the presence or absence of defects in the steel sheet and the length (size) of the defects based on changes in the electric field. The eddy current sensor 18 can also determine the level of the defect (for example, large, medium, small, etc.) based on the length of the detected defect.

[0035] 2, the eddy current sensor 18 includes an E-shaped core 181, an exciting coil 182, a first detection coil 183, and a second detection coil 184. The eddy current sensor 18 is installed near the bridle roll 6, for example, as shown in FIG. 3, but may be installed in another location.

[0036] The eddy current sensor 18 may be installed at a position where it can inspect the steel sheet surface after the first cold rolling process and before the second cold rolling process, and although it is installed upstream of the annealing furnace 7 in Fig. 1, the installation position is not limited as long as it is within the intermediate annealing line. In this embodiment, the predetermined process is the second cold rolling process.

[0037] 3, the electrical signal obtained by the eddy current sensor 18 is transmitted to a flaw detection panel 19, which determines the presence or absence of defects and their lengths. Information regarding the presence or absence of defects and their lengths determined by the flaw detection panel 19 (hereinafter referred to as "defect information") is transmitted to a group of personal computers 20 (control PC, inspection PC), and further transmitted via a control panel 21 to a process computer 22 together with steel plate information.

[0038] Subsequently, as will be described later, a defect map and a superposition map are created in the information processing device 24. The map data are then transmitted to a front-end process monitoring device 25 that monitors front-end processes such as the steelmaking process, hot rolling process, and first cold rolling process, and a back-end process monitoring device 26 that monitors back-end processes such as the finish annealing process and the finishing process. This makes it possible to perform computer-based analysis and check defect information in front-end and back-end processes, in addition to the annealing line where the eddy current sensor 18 is installed.

[0039] 4 shows an example of a pickling line for a pickling process provided between a hot rolling process and a first cold rolling process, and an optical defect inspection device is installed in the pickling line. This pickling line includes a bridle roll 14, a weld point detector (WPD) 15, an outlet shear 16, and a tension reel 17.

[0040] The optical defect inspection device 23 captures an image of the steel sheet and detects the presence or absence of defects, the type of defects, and the degree of defects based on the appearance of the defects on the steel sheet surface. This optical defect inspection device 23 is composed of an imaging device 231 and an image processing device 232, as shown in Figure 5, for example.

[0041] The imaging device 231 is, for example, a camera equipped with a CCD (Charge Coupled Device) sensor or a CMOS (Complementary Metal Oxide Semiconductor) sensor. This imaging device 231 captures an image of reflected light from a steel sheet irradiated with light from, for example, a light source (not shown). The imaging device 231 may be installed in a position where it can inspect the steel sheet surface after the hot rolling process and before the first cold rolling process. In FIG. 4 , the imaging device 231 is installed near the bridle roll 14, but the installation position is not limited as long as it is within the pickling line. In this embodiment, the predetermined process is the first cold rolling process.

[0042] The image processing device 232 is realized by, for example, a general-purpose computer such as a workstation or a personal computer. The image captured by the imaging device 231 is transmitted to this image processing device 232. The image processing device 232 determines that portions (pixels) of the transmitted image that exceed a predetermined threshold value, such as a brightness difference, are defects, and distinguishes between harmful and harmless defects based on a plurality of predetermined feature quantities. Examples of "harmful defects" include scabs and cracked edges. Examples of "harmless defects" include patterns and streaks.

[0043] [Creation of Defect Map and Superposition Map] The information processing device 24 is realized by, for example, a general-purpose computer such as a workstation or a personal computer. The information processing device 24 includes a control unit including a processor such as a CPU (Central Processing Unit) and a memory (main storage unit) such as a RAM (Random Access Memory) and a ROM (Read Only Memory). As shown in FIG. 5 , the control unit functions as a defect determination unit 241 and a map creation unit 242.

[0044] The defect determination unit 241 performs defect determination for defects determined as harmful defects by the optical defect inspection device 23. The defect determination unit 241 determines whether a defect is harmful or harmless and determines the degree (type) of a harmful defect, for example, using defect determination logic created based on past defect appearance information. This defect determination logic can be constructed using machine learning techniques such as neural networks, decision trees, random forests, support vector regression, and gradient boosting.

[0045] The map creation unit 242 creates a defect map and a superposition map. Here, when defect detection is performed using the eddy current sensor 18, the map creation unit 242 creates the defect map and the superposition map based on defect information acquired from the process computer 22. When defect detection is performed using the optical defect inspection device 23, the map creation unit 242 creates the defect map and the superposition map based on the determination result of the defect determination unit 241.

[0046] When defect detection is performed using the optical defect inspection device 23, the map creation unit 242 performs, for example, the following processing. First, the map creation unit 242 obtains information on the defect determination results (hereinafter referred to as "defect information") from the defect determination unit 241. This defect information includes information on defect positions obtained from the surface of the steel sheet before a predetermined process is performed, a determination of whether each defect is harmful or harmless, and information on the degree (type) of the defect if it is harmful. Furthermore, the "predetermined process" is, for example, the first cold rolling process or the second cold rolling process. Furthermore, the "information on defect positions" is information that has been subjected to image processing by the image processing device 232 described above.

[0047] Next, the map creation unit 242 creates at least one of a defect map and a superposition map based on the defect information. A "defect map" is a map that shows the defect positions for each steel plate in a process preceding or following a predetermined process. Specifically, as described below, this defect map is a map in which the steel plate is divided into a grid, and each grid is colored, for example, according to the size of the defect within the grid (see FIG. 8). Furthermore, a "superposition map" is a map that aggregates the defect positions across multiple steel plates in a process preceding or following a predetermined process. Specifically, as described below, this superposition map is a map in which the steel plate is divided into a grid, and each grid is colored, for example, according to the frequency of defect occurrence within the grid (see FIG. 7).

[0048] Furthermore, the "preceding process of the predetermined process" refers to, for example, a steelmaking process, a hot rolling process, a pickling process, etc., when the predetermined process is the first cold rolling process. Furthermore, the "post-preceding process of the predetermined process" refers to, for example, a finish annealing process, a refinement process, etc., when the predetermined process is the second cold rolling process. Furthermore, when defect detection is performed using the optical defect inspection device 23, the defect map and the superposition map include the positions of at least one of predetermined harmful defects and harmless defects.

[0049] The map creating unit 242 may create a superimposed map by aggregating a plurality of defect maps created in advance, or may create a superimposed map directly from defect information.

[0050] The map creation unit 242 creates a superimposition map by converting the defect positions of the steel plate in a predetermined process into defect positions of the steel plate in the previous process and aggregating them, and provides the created superimposition map to the previous process monitoring device 25 that monitors the previous process. This makes it possible to check defect information in the process previous to the predetermined process. Note that "converting to defect positions of the steel plate in the previous process" can be, for example, the following processing when the predetermined process is the first cold rolling process and the target previous process is a hot rolling process.

[0051] (1) Based on the changes in the reduction rate and plate thickness in the hot rolling process, the defect positions are converted into those of the steel plate before, during, and after hot rolling. (2) Because the OP side (operator side) and DR side (drive side (equipment side)) are interchanged at the width direction end in the previous process, and the leading end and the trailing end are interchanged, the defect positions are converted based on the OP, DR, leading end, and trailing end positions in the previous process.

[0052] Furthermore, the map creation unit 242 may identify at least one of an abnormality in equipment in a process upstream of a specified process and an operational peculiarity based on the defect locations included in the superimposed map created as described above, and provide the identified abnormality to the upstream process monitoring device 25 that monitors the upstream process.

[0053] Furthermore, the map creation unit 242 creates a superposition map by converting the defect positions of the steel plate in a predetermined process into defect positions of the steel plate in a subsequent process and aggregating them, and provides the created superposition map to the subsequent process monitoring device 26 that monitors the subsequent process. This makes it possible to take measures such as removing the defective portions in the process subsequent to the predetermined process. Note that "converting into defect positions of the steel plate in a subsequent process" can include the following processing when, for example, the predetermined process is the second cold rolling process and the target subsequent process is a finishing process.

[0054] (1) Based on the change in the reduction rate and the plate thickness in the second cold rolling process, the defect positions in the steel plate after the second cold rolling are converted. (2) Because the OP side and DR side at the width direction end portion are interchanged and the leading end and the trailing end are interchanged in the subsequent process, the defect positions are converted based on the OP, DR, leading end, and trailing end positions in the subsequent process.

[0055] Furthermore, the map creation unit 242 may identify the processing content to be performed in the subsequent process (e.g., removal of the defective portion) based on the defect positions included in the superposition map created as described above, and provide the identified processing content to the subsequent process monitoring device 26 that monitors the subsequent process.

[0056] In this way, the information processing device 24 performs mapping using data from the eddy current sensor 18 and optical defect inspection device 23, etc., which are introduced before a specified process, thereby making it possible to adjust the operating conditions in the process before or after the specified process.

[0057] Figure 6 shows an example of a defect detected by the optical defect inspection device 23 shown in Figure 4. In defect detection using the optical defect inspection device 23, machine learning is used to determine whether a defect is harmful or harmless, and the degree (type) of the harmful defect is determined. The defect is determined as large, medium, or small in order of the risk of breakage during rolling and the need for maintenance. In addition, in defect detection using the optical defect inspection device 23, if a defect is determined to be harmless based on its appearance, it is determined to be a harmless defect. In the figure, (a) shows a harmful defect with a large degree, (b) shows a harmful defect with a medium degree, (c) shows a harmful defect with a small degree, and (d) shows a harmless defect.

[0058] The eddy current sensor 18 also measures the length of the defect from changes in the electric field and determines the extent of the defect based on the length information, and is designed so that no information is provided for defects shorter than a certain length.

[0059] In this way, by being able to check the degree of defects in advance, it is possible to take measures to prevent breakage in the cold rolling process, which is a predetermined process, and it is also possible to consider treatment methods in advance before defects are confirmed in the subsequent annealing process and finishing process. Furthermore, in the case of the optical defect inspection device 23, it is possible to check the appearance of defects, including harmless defects, in advance, making it possible to more accurately consider treatment methods.

[0060] Furthermore, the defect information detected by the eddy current sensor 18 and the optical defect inspection device 23 is utilized to create a defect map and a superposition map, as described above. Furthermore, for a line using the optical defect inspection device 23, it is possible to map even harmless defects, making it possible to investigate defects that may cause breakage in the cold rolling process, as well as to evaluate the occurrence of defects that result in rejection due to problems such as appearance.

[0061] Fig. 7 shows an example of a superposition map created by the information processing device 24. Fig. 8 shows an example of a plurality of defect maps that are the basis for the superposition map of Fig. 7. As described above, the superposition map and defect map are created by converting, for example, the positions of defects before the cold rolling process into the positions of defects during hot rolling, based on defect information on the steel sheet surface detected by the optical defect inspection device 23 and eddy current sensor 18 located before the cold rolling process. Furthermore, the superposition map and defect map use color coding to map the frequency and size of defects.

[0062] By overlaying information on multiple steel plates, as in the superposition map shown in Figure 7, it is possible to confirm the location, frequency, and severity of defects across multiple steel plates. In addition, by breaking down the superposition map into information for each steel plate, as in the defect map shown in Figure 8, it is possible to investigate the presence or absence and characteristics of defects in specific steel plates. This makes it possible to use the map to investigate the operating conditions of steel plates in which defects have occurred, analyze the tendency of defect occurrence, and investigate equipment abnormalities.

[0063] In the superimposed map and the defect map, only harmful defects may be color-coded, only harmless defects may be color-coded, or both harmful and harmless defects may be color-coded. Criteria for color-coding the defect map include, for example, the degree of defect (defect type), the size of the defect area, and the defect length. Criteria for color-coding the superimposed map include the frequency of defects occurring across multiple steel plates. When multiple defect maps are superimposed to create the superimposed map, if defects exist at the same position, the color of the corresponding position on the superimposed map is made darker.

[0064] According to the information processing device of the embodiment described above, it is possible to check the defect location of the steel plate before a specified process, and to take measures in advance against defects in the steel plate that occur after the specified process.

[0065] That is, the information processing device according to the embodiment visualizes the location and severity of defects based on the defect information acquired by the eddy current sensor 18 and the optical defect inspection device 23. This can be utilized to take measures against the cause of defect occurrence in the pre-process and to treat the defective portion in the post-process.

[0066] Furthermore, the information processing device according to the embodiment can confirm the location of defects before cold rolling, and by removing the defective portions in advance, it is possible to reduce breakages in the cold rolling process. Furthermore, the information processing device according to the embodiment can analyze the location, degree, and form of defects, thereby making it possible to identify the cause of defect occurrence and take measures against the defects.

[0067] (Embodiment 1: Feedback to Previous Process) Embodiment 1 of the present invention will be described with reference to Fig. 9. The figure shows an example of a superposition map created based on defect information acquired by an optical defect inspection device, organized by heating furnaces A and B and the month in which heating furnaces A and B were operated in the hot rolling process.

[0068] 9, the vertical axis indicates the operation month of heating furnaces A and B, and the horizontal axis indicates the type of heating furnace and the front and back sides of the steel plate. In each superimposed map, the vertical axis indicates the width direction position of the steel plate, and the horizontal axis indicates the longitudinal direction position of the steel plate.

[0069] In each superimposed map, the size of each grid is 50 mm in the width direction and 25 mm in the length direction. ２ In the above cases, paint in red and ２ More than 25mm ２ If the total defect area is less than 0.1 mm, the defect area is colored in a gradation from light blue to blue in ascending order of defect area. ２ If it is less than this, it is judged to have no defects and is not painted.

[0070] In the superimposed map shown in Figure 9, defects were detected at the leading and trailing ends of the front surface of the hot-rolled steel sheet and at the leading and trailing ends of the back surface of the hot-rolled steel sheet in both heating furnaces A and B. Meanwhile, apart from these trends, it can be seen that defects have specifically been occurring in the longitudinal center on the DR side of the back surface of the hot-rolled steel sheet in heating furnace B since February 2023.

[0071] In this case, it is possible to confirm the specificity of the heating furnace. In other words, since the equipment specifications differ for each heating furnace, it is possible to identify the equipment causing the defect from the location where the defect occurred and determine the cause of the defect. In addition to collecting defect samples, which has been done conventionally, by performing comparison using an overlay map such as that shown in Figure 9, it is possible to identify the cause of the defect more quickly and accurately.

[0072] (Example 2: Feedback to Previous Process) Example 2 of the present invention will be described with reference to Fig. 10 and Fig. 11. Fig. 10 shows an example of an overlay map when scabs frequently occurred at a position 900 m from the front end of the back surface of a hot-rolled steel sheet in a specific heating furnace in the past for hot rolling.

[0073] As is clear from the superimposed map shown in Figure 10, defects are concentrated at a position 900 m from the leading edge of the back surface of the hot-rolled steel sheet. In this case, by checking the defect occurrence positions in advance using the superimposed map, it is possible to take measures to remove defects that could cause breakage or line trouble.

[0074] Furthermore, before cold rolling, the locations of defects vary for each steel plate due to the effects of truncation in the previous process and the processes through which the steel plate passes, but by referring to the superposition map created by converting these into defect locations during hot rolling, it was confirmed that defects occur at specific locations. Then, by comparing the defects that occur at these locations with the equipment locations in the hot rolling process, it was confirmed that there is one piece of equipment that is causing the defects.

[0075] There are various causes of scabs, but when defects occur continuously across the width as in this case, it is thought that they are caused by collisions with equipment in the hot rolling process, such as skids. Therefore, the results of this analysis were applied to the hot rolling process, which is the upstream process, and the slab placement position in the heating furnace was reviewed to reduce equipment interference.

[0076] Figure 11 shows the superimposed map created after reviewing the slab placement position compared to Figure 10. It can be seen that reviewing the slab placement position has reduced the occurrence of defects. In this way, by quickly detecting the occurrence of defects using the superimposed map, appropriate countermeasures can be taken. Furthermore, by taking countermeasures against defects using the superimposed map, it is possible to suppress the occurrence of line troubles such as steel plate cutoff losses before the cold rolling process and breakage during the cold rolling process, thereby contributing to improved production capacity.

[0077] (Example 3: Feedforward to downstream processes) Fig. 12 shows an overlay map of the month in which many appearance defects occurred in the pickling line (see (b)) and the month before that (see (a)). In the overlay map in the same figure, the defect positions are converted into defect positions in the pickling line. By referring to this overlay map, it is possible to identify the equipment in the pickling line where the appearance defects occurred.

[0078] Although cosmetic defects (e.g., acid dripping) do not cause losses due to breakage or truncation during the cold rolling process, they may result in the product being unshippable. As shown in Figure 12(b), when examining the overlaid map for the month in which cosmetic defects occurred, it can be seen that the number of detected defects increases across the entire upper portion of the overlaid map.

[0079] Therefore, it was determined that the cause of the appearance defect was on the dry roll side of the pickling line, and corrective measures were taken to prevent the cause of the appearance defect. In this way, when analysis is performed using a superimposed map that uses defect information from an optical defect inspection system, it is possible to follow up on the occurrence of defects that do not affect the cold rolling process, allowing for prompt action to be taken.

[0080] Although the information processing device according to the present invention has been specifically described above using the preferred embodiment and examples, the gist of the present invention is not limited to these descriptions and should be broadly interpreted based on the claims. Furthermore, it goes without saying that various changes and modifications based on these descriptions are also included in the gist of the present invention.

[0081] For example, in the information processing device according to the embodiment, the predetermined process is assumed to be a cold rolling process, but the predetermined process may be a finish annealing process (AL). For example, if a defect (flaw) is detected in a surface inspection after the finish annealing process, the corresponding portion may be removed in a subsequent process (e.g., a finishing process), and as a measure to address the root cause, the superposition map may be used to check whether there is a defect at the same position in the previous process, and the defect caused by the previous process may be corrected.

[0082] REFERENCE SIGNS LIST 1 Payoff reel 2 Entry shear 3 Welding machine 4 Notcher 5 Defleur 6 Bridle roll 7 Annealing furnace 8 Water cooling device 9 Pickling tank 10 Pickling tank brush 11 Rinsing dryer 12 Rinsing tank 13 Dryer 14 Bridle roll 15 Welding point detector (WPD) 16 Exit shear 17 Tension reel 18 Eddy current sensor 181 E-core 182 Excitation coil 183 First detection coil 184 Second detection coil 19 Flaw detector 20 Personal computers 21 Control panel 22 Process computer 23 Optical defect inspection device 231 Imaging device 232 Image processing device 24 Information processing device 241 Defect determination unit 242 Map creation unit 25 Pre-process monitoring device 26 Post-process monitoring device

Claims

1. An information processing device comprising a control unit, which acquires information relating to defect positions obtained from the surface of a steel plate before a specified process is performed, and based on the information relating to the defect positions, creates at least one of a defect map showing the defect positions for each steel plate in a process preceding or following the specified process, and a superposition map that aggregates defect positions across multiple steel plates in a process preceding or following the specified process.

2. The information processing device according to claim 1, wherein the defect map is obtained by dividing the steel plate into a grid and coloring each grid according to the size of the defects within the grid, and the superposition map is obtained by dividing the steel plate into a grid and coloring each grid according to the frequency of defects within the grid.

3. The information processing apparatus according to claim 1, wherein the defect map and the superposition map include the positions of at least one of predetermined harmful defects and harmless defects.

4. The information processing device according to claim 1, wherein the predetermined process is a cold rolling process.

5. The information processing device according to claim 1, wherein the control unit creates the superposition map by converting the defect positions of the steel plate in the specified process into defect positions of the steel plate in the previous process and aggregating them, and provides the created superposition map to a monitoring device that monitors the previous process.

6. The information processing device according to claim 1, wherein the control unit identifies at least one of an abnormality in the equipment of the upstream process and an operational peculiarity based on the defect location included in the superposition map, and provides the identified abnormality to a monitoring device that monitors the upstream process.

7. The information processing device according to claim 1, wherein the control unit creates the superposition map by converting the defect positions of the steel plate in the specified process into defect positions of the steel plate in the subsequent process and aggregating them, and provides the created superposition map to a monitoring device that monitors the subsequent process.

8. The information processing device according to claim 1, wherein the control unit identifies the processing content to be performed in the post-process based on the defect position included in the superposition map, and provides the identified processing content to a monitoring device that monitors the post-process.