On-line detection method and device for slab strip casting blank and rolling blank defects

Abstract

A plate strip casting blank rolling blank defect online detection method comprises the steps that a rolling blank surface image detection device is installed in an outlet area of a reversible rolling mill or a roughing mill, rolling blank surface images of different thicknesses are collected online, and the positions and sizes of defects generated on the rolling blank surface of each pass are recognized and analyzed through an image mode; and judging whether the area where each defect source is generated is in the rolling area or the smelting area, and gradually restoring the distribution and size of the defects in the original strip casting blank according to the images of the connected rolling passes in sequence for the defects in the smelting area.

Description

TECHNICAL FIELD

[0001] The present application relates to a kind of strip casting billet rolling billet defect online detection method and device, belong to steel metallurgy in strip casting billet rolling billet quality detection field. BACKGROUND

[0002] In the production process of continuous casting billet, due to the influence of process and equipment, cracks, scratches and pinholes and other defects are inevitable, and serious defects will have adverse effects on the next rolling process. The surface quality and internal quality of the casting billet are one of the important indicators of the quality of the casting billet, and the nature and degree of the surface defects of the casting billet are one of the important bases for judging whether the casting billet is qualified, and the internal defects of the casting billet are also one of the important bases for judging whether the casting billet is qualified, so it is necessary to quantitatively detect the surface and internal of the casting billet.

[0003] In recent years, the high-temperature continuous casting billet hot charging and direct rolling technology developed is a new system process, which directly affects the integration degree of steelmaking-continuous casting-hot rolling process, and its economic benefits are directly reflected in saving energy, reducing production cost, shortening production cycle and improving product quality. The supporting technologies for realizing high-temperature continuous casting billet hot charging and direct rolling include many aspects, including high-temperature defect-free continuous casting billet production technology, so the online detection technology of high-temperature continuous casting billet surface defects and internal defects is particularly important. At present, the practical online detection method of high-temperature metal casting billet surface defects is the optical detection method of high-temperature casting billet surface defects by machine vision, which has the advantages of non-contact, fast response and less influence of high-temperature metal, and has developed rapidly in recent years and been widely applied. However, when using optical method to detect high-temperature casting billet surface defects, it cannot detect subcutaneous shallow defects. There is no good online detection method for subcutaneous shallow defects and internal defects, and only online sampling inspection can be used, typical of CN202311575615, which discloses a continuous casting billet internal defect detection method, i.e. cutting billet sample and sending to laboratory for inspection. However, this sampling inspection method has the problem of missing detection, and it is difficult to detect the whole billet comprehensively. Therefore, it is necessary to provide a strip casting billet rolling billet defect online detection method and device to overcome the above defects. SUMMARY

[0004] The purpose of the present application is to provide a strip casting billet rolling billet defect online detection method and device, which can detect the type, depth and size of the internal defects of the strip casting billet, solve the problem of missing detection of internal defects of the strip casting billet which cannot be solved by current online sampling inspection, and further save energy, reduce production cost, shorten production cycle and improve product quality.

[0005] The technical solution of the present application is: a strip casting billet rolling billet defect online detection method, characterized by comprising the following steps: 1) Install high-temperature rolling stock surface image detection device at the exit area of reversible mill or rough rolling mill; 2) Collect rolling stock surface images of different thicknesses in each pass in line; 3) Use image pattern recognition processing algorithm to analyze and judge the defect position and size generated on the surface of rolling stock in each pass; 4) Judge the defect type and the area where the defect is generated, whether in the rolling area or before rolling; 5) For defects determined to be generated in the continuous casting area, the three-dimensional distribution position and size of the defects in the original slab are recovered from the images of the connected rolling passes in sequence; similarly, for the defects in the intermediate slab in the secondary reheating rolling, the three-dimensional distribution position and size of the defects in the original intermediate slab are recovered in sequence; at the same time, the position of the rolling defects generated in each pass is also obtained.

[0006] In order to distinguish the surface defects of the slab, a slab surface image detection device is installed on the strip continuous casting production line, and the slab surface detection image is sent to the rolling stock image processing computer for internal defect detection of the slab.

[0007] A database for analyzing slab and rolling stock defects, an AI analysis tool, and a defect sample training self-correcting learning tool are constructed for analyzing the slab and rolling stock defects.

[0008] The steps for recovering the three-dimensional distribution position and size of the defects in the original slab from the images of the connected rolling passes in sequence for the defects generated in the smelting area are as follows: 1) Establish a deformation database corresponding to different rolling pressures for different steel types and different defects; 2) Calculate the expected deformation range of each current defect after rolling, compare the defect images of the rolling stock in different subsequent passes, and judge whether the defects will occur in the thickness area of the subsequent rolling stock or new types of defects will occur; 3) Check and judge which defects are caused by the slab; 4) For defects caused by the slab, analyze and compare the changes of each defect in different rolling stock thickness images to determine the distribution of the defects in different rolling stock thicknesses, and compare the changes of different rolling stock thickness images to determine the type, size, and three-dimensional position of the defects in the rolling stock by sequentially comparing the rolling stock images from thin to thick or from thick to thin, or repeatedly comparing the rolling stock images of different passes; 5) Obtain the defect distribution in the original slab space by sequential recursion.

[0009] In order to better detect the internal defects of the slab, the two-dimensional and three-dimensional information of the slab and rolling stock defects is obtained by combining other detection methods, and the defect type, size, and three-dimensional position in the rolling stock are comprehensively judged. Other detection techniques include laser detection, three-dimensional detection, etc.

[0010] In order to better detect the internal defects of the casting blank, the high-temperature rolling blank surface image detection device is installed at the entrance area of the reversible rolling mill or the rough rolling mill, and the exit of the finishing mill, and the detection image data is sent to the rolling blank image processing computer.

[0011] In order to better detect the internal defects of the casting blank, the high-temperature rolling blank surface image detection device is installed at the entrance area of the reversible rolling mill or the rough rolling mill, and the exit of the finishing mill, and the detection image data is sent to the rolling blank image processing computer.

[0012] The casting blank defect detection result is sent to the steelmaking and casting process, so as to provide a basis for improving the process.

[0013] The application also provides a plate strip casting blank defect online detection device, which mainly comprises a camera and a protection device, characterized in that the protection device is provided with a gas blowing device, and the blowing gas is used for cleaning the surface of the rolling blank to be detected.

[0014] The plate strip casting blank defect online detection device is characterized in that a barrier layer is installed between the surface detection device and the rolling mill to prevent the cooling water and steam splashed from the rolling mill. The application has the beneficial effects that the water vapor barrier layer and the gas blowing device at the front end of the detection device can effectively remove the water film and dust on the surface of the rolling blank, effectively improve the detection range of the surface of the rolling blank, comprehensively display the internal defects of the plate strip casting blank online, provide a basis for improving the continuous casting process, and provide a basis for distinguishing the responsibility of the rolling steel and smelting product quality defects. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 The embodiment of the application is a plate continuous casting blank and rolling blank rolling defect online detection method and device installation schematic diagram.

[0016] Figure 2 The embodiment of the application is a plate continuous casting blank and rolling blank defect detection analysis comparison schematic diagram.

[0017] Figure 3 The embodiment of the application is a plate continuous casting blank and rolling blank rolling defect image detection analysis comparison schematic diagram. DETAILED DESCRIPTION

[0018] In order to better understand the application, the content of the application will be further illustrated below in combination with the embodiments, but the content of the application is not limited to the following embodiments. Those skilled in the art can make various modifications or modifications to the application, and these equivalent forms are also within the scope of the claims of the present application.

[0019] The following are the labels in the attached diagram: 1-Reversible rolling mill; 2-Blank; 3-Rolling mill cooling water vapor baffle; 4-Blank upper surface inspection camera gas purging device; 5-Upper surface billet CCD camera inspection device cabin; 6-Roller table; 7-Rolled strip billet; 8-Defect.

[0020] Installation diagram of the detection device provided in this embodiment of the invention Figure 1 As shown in the defect comparison diagram, Figure 2 , Figure 3 As shown.

[0021] Figure 1 A CCD camera is installed at the exit of the reversible rolling mill 1. A cooling water vapor barrier 3 is installed between the rolling mill 1 and the CCD camera housing 5. A gas cleaning nozzle 4 is installed at the inlet of the CCD camera housing 5. A surface billet inspection device is located below the rolls, but it is not shown in the diagram. The billet passes through the reversible rolling mill multiple times, and the CCD camera captures surface images of billets of different thicknesses. The images are then analyzed to identify various defects and determine which parts are defects caused by the plate / strip casting.

[0022] Figure 2 The image shows the distribution of surface defects on the slab or strip billet detected for the first time, corresponding to the defects in the slab or strip casting. Figure 2 b is the first time a defect appears on the surface of the rolled billet. This defect is caused by... Figure 2 It is made by rolling out internal defects in a billet or cast billet. Figure 2 The internal defect of a, with coordinates (X1, Y1, Z1) in step 8-1, first appeared during a certain rolling pass. Figure 2 The defects on the surface of the b-rolled billet correspond to coordinates (X1, Y1) in 8-2.

[0023] Figure 3 middle Figure 3 a, Figure 3 b、 Figure 3 c is a schematic diagram of image detection of rolled slabs or rolled products of different thicknesses from the same cast slab. When a defect first appears during a certain rolling pass, it can be determined that the defect is located at a height below the thickness of the previous pass. Then, based on the defect changes shown in the images of subsequent rolling passes, this defect change is compared with a defect change pattern database to determine the size of the defect, verify its type, and determine whether it is a defect caused by the slab or strip cast slab. For example: Figure 3 a, Figure 3 b、 Figure 3 If the black streak in c is identified as an inclusion, it can be determined to be a defect in the strip casting, and the location and size of the inclusion in the strip casting can be deduced.

[0024] The technical solution of the present invention has the following advantages over the prior art: The comprehensive on-line display of the internal defect distribution of the slab provides basis for improvement of the continuous casting process; The comprehensive on-line display of the defect position of the rolled slab provides basis for improvement of the rolling process The comprehensive on-line display of the defect position of the rolled slab provides basis for improvement of the rolling process

[0025] The contents not described in detail in the specification belong to the prior art known to the person skilled in the art.

Claims

1. A method for on-line detection of defects in a slab-cast strip process, characterized in that, It comprises the following steps: 1) installing a high-temperature rolling stock surface image detection device at the exit area of a reversible rolling mill or rough rolling mill; 2) collecting rolling stock surface images of different thicknesses in each pass in real time; 3) using image pattern recognition processing to analyze and determine the position and size of defects on the surface of rolling stock in each pass; 4) determining the type of each defect and whether the defect is generated in the rolling area or the smelting area; 5) for defects determined to be generated in the smelting area, sequentially restoring the three-dimensional distribution position and size of the defects in the original casting blank from the images of each rolling pass in sequence; similarly, for the second heating and rolling intermediate blank, the three-dimensional distribution position and size of the defects in the original intermediate blank are also sequentially restored; at the same time, the position of the rolling defects of the rolling stock generated in each pass is also obtained.

2. The strip cast billet rolling defect on-line detection method according to claim 1, characterized in that, A casting blank surface image detection device is installed on a continuous casting production line, and the casting blank surface detection image is sent to a rolling stock image processing computer for casting blank and rolling stock defect detection.

3. The strip cast billet rolling defect on-line detection method according to claim 1, characterized in that, A database for analyzing casting blank and rolling stock defects, an AI analysis tool, and a defect training self-correcting learning tool are constructed for analyzing casting blank and rolling stock defects.

4. The method according to claim 1 or 3, characterized in that: The specific steps are as follows: 1) establishing a deformation database corresponding to different rolling pressures for different steel grades and different defects; 2) calculating the deformation range of each defect after rolling, comparing the defect images of rolling stock in different subsequent passes, and determining whether the defects will occur in the thickness area of the subsequent passes or new types of defects will occur; 3) checking and determining which defects are caused by the casting blank; 4) for defects caused by the casting blank, analyzing and comparing the changes of each defect in different rolling stock thickness images to determine the distribution of the defects in the rolling stock thickness, and comparing the image changes according to the rolling stock thickness to sequentially determine the type, size, and three-dimensional position of the defects in the rolling stock from thin to thick or from thick to thin, or repeatedly comparing the relative rolling stock images of different passes to determine the type, size, and three-dimensional position of the defects in the rolling stock; 5) sequentially obtaining the distribution of defects in the three-dimensional space of the original casting blank.

5. The strip cast billet rolling defect on-line detection method according to claim 1, characterized in that, Two-dimensional and three-dimensional information of casting blank and rolling stock defects is obtained by combining other detection methods to comprehensively determine the type, size, and three-dimensional position of the defects in the rolling stock. Other detection techniques include laser detection and three-dimensional detection.

6. The strip cast billet rolling defects on-line detection method according to claim 1, characterized in that, A high-temperature rolling stock surface image detection device is installed at the entrance area of the reversible rolling mill or rough rolling mill and the exit area of the finishing mill, and the detected images are sent to a casting blank defect recognition processing computer.

7. The strip cast billet rolling defects on-line detection method according to claim 1, characterized in that, The defect data detected by various rolling stock defect detection instruments installed in the rolling area is sent to the casting blank recognition processing computer.

8. The strip cast billet rolling defects on-line detection method according to claim 1, characterized in that, The casting blank defect detection results are sent to the steelmaking and casting processes.

9. A strip casting and rolling defect on-line detection device, the detection device mainly comprising a camera, a protection device, characterized in that: The protection device of the image detection instrument is provided with a gas blowing device, and the blowing gas is used to clean the surface of the rolling stock to be detected.

10. The apparatus according to claim 8, wherein A barrier layer is installed between the reversible rolling mill and the rolling stock surface image detection device to prevent cooling water and steam sprayed from the rolling mill.

Images