TW202317285A - Rolling system and method for detecting steel quality after side guide - Google Patents

Abstract

A rolling system includes a rolling mill, a side guide, a camera, and a computing module. The rolling mill is configured to roll a steel strip. The side guide includes a first side guide and a second side guide. The first side guide and the second side guide are disposed on both sides of the steel strip to maintain the position of the rolled steel strip during transportation. The camera shoots the steel stripe to produce a steel strip image. The computing module determines whether the steel strip is abnormal according to the steel strip image, and a warning message is output if the steel strip is abnormal.

Description

Rolling system and detection method for detecting the quality of edge guide steel strip

This disclosure relates to a rolling system in a steel mill, which can detect the quality of the steel strip in the edge guide in time.

In steelmaking plants, finish rolling is the last rolling step in the hot rolling process. The surface temperature of the steel strip is above 600 degrees, and the material is relatively thin, and the ability to resist bending and buckling decreases. The speed is fast, up to 48 kilometers per hour. If the rolling process causes bad deformation of the material, it will easily affect the movement stability of the steel strip during the finishing rolling process. In severe cases, the bad movement of the steel belt may even easily hit equipment such as side guides, resulting in bad deformation of the steel belt. Therefore, how to timely detect the abnormal condition of the steel strip is a topic concerned by those skilled in the art.

Embodiments of the present disclosure provide a rolling system, including a rolling mill, an edge guide, a camera, and a computing module. Rolling mills are used to roll steel strips. The side guide includes a first side guide and a second side guide. The first side guide and the second side guide are arranged on both sides of the steel strip to maintain the position of the rolled steel strip during conveying. The camera is used to photograph the steel strip to generate a steel strip image. The computing module is connected to the camera through communication to determine whether the steel belt is abnormal according to the steel belt image, and if the steel belt is abnormal, a warning message will be issued.

In some embodiments, the calculation module is also used to display an abnormal list on the user interface, the abnormal list includes the serial number of the steel strip, and the calculation module is used to mark the abnormal steel strip.

In some embodiments, if the steel strip is abnormal, the calculation module is also used to generate a check list. The check list includes the opening of the editor, whether the steel belt has burrs, and whether the steel belt needs to be cut off.

In some embodiments, the computing module is also used to generate instructions to increase the opening between the first side guide and the second side guide.

In some embodiments, the calculation module is further used to classify the abnormal steel strip into one of a plurality of abnormal categories according to the steel strip image, and these abnormal categories include edge filamentous debris, edge damage, and edge rolling foreign matter.

From another point of view, the embodiments of the present disclosure provide the detection method of the above-mentioned rolling system. The detection method includes: photographing the steel strip through a camera to generate a steel strip image; judging whether the steel strip is abnormal according to the steel strip image, and sending a warning message if the steel strip is abnormal.

In some embodiments, the above detection method further includes: displaying an abnormal list on the user interface, the abnormal list includes the serial number of the steel strip, and the calculation module is used to mark the abnormal steel strip.

In some embodiments, the above detection method further includes: if the steel strip is abnormal, generating a checklist, the checklist includes the opening of the braider, whether the steel strip has burrs, and whether the steel strip needs to be cut off.

In some embodiments, the above detection method further includes: if the steel belt is abnormal, generating an instruction to increase the opening between the first side guide and the second side guide.

In some embodiments, the above detection method further includes: classifying the abnormal steel strip into one of a plurality of abnormal categories according to the steel strip image, and these abnormal categories include edge filamentous waste, edge damage and edge rolling foreign body.

The terms "first", "second" and the like used herein do not specifically refer to a sequence or sequence, but are only used to distinguish elements or operations described with the same technical terms.

FIG. 1 is a schematic diagram illustrating a rolling system according to an embodiment. Please refer to FIG. 1 , the rolling system 100 includes a rolling mill 110 , a steel strip 120 , an edge guide 130 , a camera 140 and a computing module 150 . The rolling mill 110 is, for example, a finishing mill, and may also be a roughing mill in some embodiments. The side guide 130 includes a first side guide 131 and a second side guide 132 , which are respectively arranged on the left and right sides of the steel strip 120 in the traveling direction to maintain the position of the rolled steel strip 120 during transport. The first side guide 131 and the second side guide 132 can contact the steel belt 120 or not contact the steel belt 120 , the present disclosure is not limited thereto. The camera 140 includes a photosensitive coupling device (Charge-coupled Device, CCD) sensor, a complementary metal oxide semiconductor (Complementary Metal-Oxide Semiconductor) sensor or other suitable photosensitive elements, in order to photograph the steel strip 120 to produce steel with video. It should be noted that the camera 140 can be a high-speed camera to capture clear images of the steel belt because the steel belt 120 travels at a high speed.

The computing module 150 can be a server, a central console, a personal computer or any electronic device with computing capability. The computing module 150 can communicate with the camera 140 in any wired or wireless way to obtain the steel belt image captured by the camera 140 . The computing module 150 can analyze the obtained steel belt image according to any suitable machine learning algorithm, so as to determine whether the steel belt 120 is abnormal. For example, the machine learning algorithm can be a convolutional neural network. If it is judged that the steel strip 120 is abnormal, the calculation module 150 will output a warning message, which can be represented by any text, symbol, number, voice, sound effect or image. In some embodiments, a warning is set on site. Tower (alarm tower), the above-mentioned warning message can be to make the lights on the warning tower shine, and this disclosure does not limit the form of the warning message. In some embodiments, if it is judged that the steel belt 120 is abnormal, the calculation module 150 can also generate an instruction, which is used to control the motor (not shown) connected to the side guide 130, thereby increasing the first side guide 131 and the second edge guide 132, so as to prevent the steel strip 120 from hitting the first edge guide 131 or the second edge guide 132 to cause damage. In some embodiments, the above commands can also be used to change the clamping force of the edge guide 130 .

In some embodiments, the calculation module 150 can also classify the abnormal steel strip into one of a plurality of abnormal categories according to the steel strip image. For example, please refer to FIG. 2 . FIG. 2 shows steel belt images of different abnormal types. The abnormality category 210 refers to edge filamentary waste, the abnormality category 220 is edge damage, and the abnormality category 230 is foreign matter rolled into the edge. The calculation module 150 can classify according to the machine learning model. These abnormal categories can be classified by experienced personnel in advance, and then the machine learning model can be trained according to the classified steel belt images. However, more or less exception categories can also be distinguished in other embodiments, and the present disclosure is not limited thereto.

Fig. 3 is a structural diagram illustrating a management system according to an embodiment. Please refer to FIG. 3 , where the computing module 150 is divided into a back-end program 320 and a front-end web page 330 from a software perspective. The calculation module 150 can obtain the steel strip image from the camera 140, and obtain relevant information about the steel strip from the production database 310, including steel coil number, steel type, thickness and other related specifications. The back-end program 320 includes programs 321-324, the program 321 is used to collect the steel belt image, the program 322 is used to perform image recognition on the steel belt image to determine whether it is abnormal, and the program 323 is used to classify (that is, generate the above-mentioned abnormal classification) , program 324 is used to control the opening or clamping force of the edge guide. However, the procedures 321 - 324 have been described above, and will not be repeated here.

On the other hand, the computing module 150 can also provide a front-end web page 330 , which provides a user interface for other workers to view. For example, the program 331 can display a warning message on the user interface. The program 332 can display abnormal steel strip images on the user interface, and can also provide information such as the location and size of abnormalities (defects). Program 333 can compile exception information to display an exception list or report. This exception list can be a table with multiple columns and rows. These columns can record the number, thickness, width, steel type and other related specifications of the steel strip, and different columns indicate different steel belt. Program 333 may mark steel strips with exceptions, for example, by marking corresponding columns in the exception list with different colors, fonts or font sizes. In some embodiments, different colors can also be used to distinguish different abnormal categories. Program 334 is used to display a check list, which includes items that need to be confirmed by the inspectors, including the opening of the guide, whether the steel belt has burrs, whether the center of the edge guide is offset, and whether the steel belt needs to be cut off. Those with ordinary knowledge in the field can design a suitable form to include the above information, and the present disclosure does not limit the format of the checklist. In this way, through the front-end web page 330 , relevant personnel can know whether there is any abnormality in the current edge guide, browse abnormal steel strip images, check abnormal information, and perform subsequent checks and processing.

In the prior art, because the temperature of the steel strip after hot rolling is very high, it is necessary to wait for the steel coil to cool for a few days before the problem of the edge guide can be detected by the relevant personnel and cause the edge of the steel strip to be damaged. However, after adopting the method of the above-mentioned embodiment, the abnormal problem of the edge guide can be immediately reflected during the production process of the steel strip, so as to prevent the steel strip from continuing to produce unfavorable deformations such as deformation, turning up, and folding.

Although the present invention has been disclosed above with the embodiments, it is not intended to limit the present invention. Anyone with ordinary knowledge in the technical field may make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention should be defined by the scope of the appended patent application.

100:Rolling system 110: rolling mill 120: steel belt 130: side guide 131: The first side guide 132: The second side guide 140: camera 150: Calculation module 210, 220, 230: Exception categories 310: Production database 320: Backend program 321~324: program 330: front-end webpage 331~334: program

FIG. 1 is a schematic diagram illustrating a rolling system according to an embodiment. Figure 2 shows the steel belt images of different anomaly categories. Fig. 3 is a structural diagram illustrating a management system according to an embodiment.

100:Rolling system

110: rolling mill

120: steel belt

130: side guide

131: The first side guide

132: The second side guide

140: camera

150: Calculation module

Claims (10)

A rolling system comprising: a rolling mill for rolling a steel strip; A side guide, including a first side guide and a second side guide, wherein the first side guide and the second side guide are arranged on both sides of the steel strip to maintain the rolling of the rolled steel strip position at time; a camera for photographing the strip to produce a strip image; and A calculation module, connected to the camera by communication, is used to determine whether the steel belt is abnormal according to the steel belt image, and sends a warning message if the steel belt is abnormal. The rolling system as described in claim 1, wherein the calculation module is also used to display an abnormality list on a user interface, the abnormality list includes the serial number of the steel strip, and the calculation module is used to mark abnormalities of the steel belt. The rolling system as described in claim 2, wherein if the steel strip is abnormal, the calculation module is also used to generate a checklist, the checklist includes the opening of the braider, whether the steel strip has burrs, the Whether the steel strip needs to be cut off. The rolling system as claimed in claim 1, wherein if the steel strip is abnormal, the calculation module is further used to generate an instruction to increase the opening between the first side guide and the second side guide. The rolling system as described in claim 1, wherein the calculation module is further used to classify the abnormal steel strip into one of a plurality of abnormal categories according to the steel strip image, and the abnormal categories include edge filamentous scrap Chips, edge damage, and foreign matter rolled into the edge. A detection method of a rolling system, wherein the rolling system includes a rolling mill and a side guide, the rolling mill is used to roll a steel strip, the side guide includes a first side guide and a second side guide, The first side guide and the second side guide are arranged on both sides of the steel strip to maintain the position of the rolled steel strip during transportation. The detection method includes: filming the strip through a camera to produce a strip image; and According to the image of the steel strip, it is judged whether the steel strip is abnormal, and if the steel strip is abnormal, a warning message is issued. The detection method as described in claim 6, further comprising: An abnormality list is displayed on a user interface, the abnormality list includes the serial number of the steel strip, and the calculation module is used to mark the abnormal steel strip. The detection method as described in claim item 7, further comprising: If the steel strip is abnormal, a checklist is generated, the checklist includes the opening of the braider, whether the steel strip has burrs, and whether the steel strip needs to be cut off. The detection method as described in claim 6, further comprising: If the steel strip is abnormal, an instruction is generated to increase the opening between the first side guide and the second side guide. The detection method as described in claim 6, further comprising: According to the steel strip image, the abnormal steel strip is classified into one of a plurality of abnormal categories, and the abnormal categories include edge filamentous debris, edge damage, and edge rolling foreign matter.

Images