Online detection system and method for edge crack defects of strip steel
Through high-speed and high-precision line laser detection technology, the laser camera and mobile device analyze the height data of the edge part of the strip steel is solved, and the problems of low sensitivity and slow processing speed of the strip steel edge crack detection in the existing technology are achieved, achieving high-precision and fast Edge crack defect detection, suitable for high-speed production lines.
Patent Information
- Application Number
- CN202210114935.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-01-31
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2042-01-31
AI Technical Summary
The existing strip side defect detection methods mainly rely on image recognition technology, with low sensitivity and slow processing speed, which cannot meet the detection requirements of strip side crack defects running at high speed. Other types of inspection methods are immature and have not been promoted and applied, resulting in side cracking. Defects cause serious problems in continuous production.
Using high-speed and high-precision line laser detection technology, two laser cameras and mobile devices automatically search the edges to collect and analyze the height data of the strip edges, determine whether there are edge crack defects, calculate the physical shape and position of the defects, and achieve high accuracy. High-density and low-latency scanning, quickly detect edge crack defects with depths greater than or equal to 0.1mm.
It realizes high-precision, fast and accurate detection of strip edges, and can continuously monitor edge crack defects on high-speed production lines, improves detection efficiency and accuracy, and solves the detection problem of edge crack defects on production lines.
Smart Images

Figure CN114406014B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of cold-rolled strip steel production, and more particularly to an online detection system and method for edge crack defects in strip steel. Background Art
[0002] Edge cracking is a type of edge cracking defect caused by uneven microstructure along the edge of the cold-rolled substrate and improperly controlled rolling process parameters. The presence of edge defects can cause severe stress concentration in the strip, leading to various problems during subsequent rolling and continuous conveying. For example, in a continuous annealing unit, strip with edge cracks can easily break within the annealing furnace due to the tension and high temperatures within the furnace, resulting in significant production losses.
[0003] The quality problem of the edge of strip steel in continuous production has always been a concern of steel companies. There are a large number of research results at home and abroad, forming relevant patent technologies. Invention patent CN104985002A discloses a method and device for alarming edge defects of hot-rolled strip steel. Through image recognition, after obtaining an image with edge defects, the edge defects are classified and the severity of the edge defects is judged. Since the detection of this method relies on images, it is difficult to identify those defects that are small and have a small opening because accurate images cannot be obtained. Invention patent CN104897058A discloses a device for detecting edge cracks of strip steel based on a correction CPC system. This patent detects the edge of the strip steel by electromagnetic sensors arranged on the edge of the strip steel. When the induced electromotive force detected by the system shows a sudden and drastic jump, it is determined that there is an edge crack defect at the position on the strip steel that causes the sudden and drastic jump. Since there are many factors that affect the induced electromotive force, it is not reliable to judge the edge crack defect based on the change of electromotive force. Moreover, this detection method is closely related to the relative position between the probe and the strip. It is difficult to ensure the constancy of this distance during the actual detection process.
[0004] Currently, the mainstream technology for detecting strip edge defects uses image recognition technology. However, this method suffers from low sensitivity and slow processing speed for detecting edge cracks in high-speed strip. Other types of inspection methods are immature and have not been widely applied on production lines. However, the presence of edge cracks does pose a serious problem for the continuous production of strip. Therefore, a new technical solution is needed to address this problem. Summary of the Invention
[0005] The purpose of the present invention is to provide an online detection system and method for edge crack defects in strip steel, which solves the problem that the mainstream technology of the current existing methods for detecting edge defects in strip steel is to use image recognition technology. This detection method has low sensitivity and slow processing speed for the detection of edge crack defects in strip steel, and cannot meet the requirements of detecting edge crack defects in high-speed running strip steel. Other types of inspection methods are not promoted and applied on the production line due to their immaturity, but the existence of edge crack defects does bring serious problems to the continuous production of strip steel.
[0006] To achieve the above-mentioned objectives, the present invention provides the following technical solutions: an online detection system and method for edge crack defects in strip steel, comprising: an online detection system for edge crack defects, wherein the online detection system for edge crack defects includes a system control unit, and the system control unit is composed of a strip steel surface data acquisition system, a data transmission module, a data processing module and a defect alarm module.
[0007] As a preferred embodiment of the present invention, the system control unit is used to control the automatic edge search of the mobile devices on both sides of the strip, the data collection of the two laser cameras, the data exchange with the unit, the reception of strip information, operating status and other information sent by the unit, and the sending of detection results.
[0008] As a preferred embodiment of the present invention, the strip surface data acquisition system consists of two sets of high-speed laser cameras with controllers. The two sets of laser cameras are respectively connected to a lateral moving device with PLC control and are arranged vertically on the two sides of the strip. The lateral moving device automatically finds the edge based on the information provided by the laser.
[0009] As a preferred embodiment of the present invention, the function of the data transmission module is to ensure high-speed data transmission between the data acquisition system and the data processing system, and to send the height data collected by the two laser cameras to the data analysis system in groups, with each group containing between 500 and 2000 laser line profiles.
[0010] As a preferred embodiment of the present invention, the data processing module analyzes each contour. For each contour, it first searches for invalid data points on the laser line from the outermost edge. Starting from the first valid data point, it counts the number of points from that point to the last valid data point. The height values corresponding to the valid data points on the contour are recorded to obtain the curve (b). The curve composed of valid data is subjected to moving average noise reduction to obtain the curve shown in (c). The height difference Δh between the highest and lowest points on the curve is calculated.
[0011] As a preferred embodiment of the present invention, the defect alarm system issues an alarm to the operator through the audio system installed in the operating room after receiving the defect information, and displays a two-dimensional image of the current edge crack defect in the human-computer interaction interface next to it, so that the operator can see the morphology and size of the edge crack defect more intuitively and make timely adjustments to the unit operation.
[0012] As a preferred embodiment of the present invention, the online detection method for edge crack defects comprises the following steps:
[0013] Step 1: Based on the strip width specifications produced by the unit, two LJ-V7080 laser cameras were selected, along with a mobile device with an effective travel range, to form a data acquisition system. After an on-site inspection of the installation conditions at the welder outlet of the unit, a gantry-mounted sensor rack was used to confirm the reference height of the laser camera from the strip. The laser camera then collected relative height data of the strip from top to bottom.
[0014] Step 2: The system control unit is installed in the electrical room near the data acquisition system. It receives information such as strip information and operating status from the unit through the data transmission system. After strip welding is completed, it starts automatic edge detection. After edge detection is completed, it performs edge crack detection. When the strip stops, the edge crack detection ends. When an edge crack defect is found, the edge crack defect information is sent to the unit.
[0015] Step 3: During the edge search process, the mobile device determines in real time the percentage of valid data in the height data detected by the laser camera. When the percentage of valid data reaches approximately 50%, the mobile device stops and the edge search is completed.
[0016] Step 4: The system control unit host controls the data acquisition system to collect parameters through the data transmission system, and sends the collected strip height data in groups of 1,000 to the data analysis system;
[0017] Step 5: The data analysis system analyzes the 1,000 height data sent by the system control unit one by one, and regards the demarcation point of the valid data in the height data as the boundary point of the strip. In this way, the position information of the strip edge in the width direction can be accurately obtained. Then, by calculating the peak-to-peak value of the strip edge position in each group of 1,000 data, the maximum depth of the strip edge crack in the current data can be accurately obtained. When the peak-to-peak value is greater than the threshold value of 0.2mm, it is judged that a large edge crack defect exists in this section of the profile. The maximum depth of the edge crack defect is calculated by analyzing and calculating this section of defect data. At the same time, based on the running speed of the strip, the location and width of the edge crack defect are calculated;
[0018] Step 6: The system control unit sends the defect information to the unit and the defect alarm system, and saves the defect data and the analyzed defect information in a txt format in a folder with the current coil number and current time;
[0019] Step 7: After receiving the defect information, the defect alarm system will issue an alarm to the operator through the audio system installed in the operation room, and display a two-dimensional image of the current edge crack defect on the human-computer interaction interface next to it, allowing the operator to more intuitively see the shape and size of the edge crack defect and make timely adjustments to the unit operation;
[0020] Step 8: When the operator queries the edge crack defects of past steel coils, he can enter part of the steel coil number or a certain time period in the human-computer interaction interface to display all the steel coil numbers that meet the conditions. The steel coil numbers of steel coils with edge crack defects are marked in red font, so that the operator can quickly find the defective steel coils; click on any steel coil to display all the defect records of the steel coil, which include the time, location, depth and width of the defect; click on any defect record to display a two-dimensional image of the edge crack defect, so that the operator can more intuitively see the morphology and size of the edge crack defect.
[0021] Compared with the prior art, the present invention has the following beneficial effects:
[0022] The present invention uses high-speed, high-precision line laser detection technology to extract height information data from the strip edge. By analyzing the height information, it is determined whether the strip edge has an edge crack defect. Under the control of the system control unit, the detection system uses a moving device arranged at the strip edge to move two laser cameras arranged at the strip edge to the detection position of the strip edge. After receiving the height information data collected by the laser cameras, the data processing unit analyzes the data one by one. Based on the correlation between the various contour information data, the system determines whether the strip edge has an edge crack defect. When an edge crack defect exceeding the standard is found on the strip edge, the computer automatically retains the data at the edge crack defect, calculates the physical shape of the defect, and provides a two-dimensional morphology of the edge crack defect. The system can achieve high-precision, high-density, and low-latency scanning of both sides of each roll of strip, continuously monitor the surface condition of the strip, and quickly and accurately find edge crack defects with a depth greater than or equal to 0.1 mm on the strip, effectively solving the problem of strip edge crack detection on continuous high-speed strip production lines. The present invention can determine the travel and position of the mobile device according to the width range of the object to be measured. The detection device has the communication function with the upper computer of the unit and can realize the offline query function of the detection data. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 This is a schematic diagram of the edge crack defect detection system of the present invention;
[0024] Figure 2 Schematic diagram of detection by two laser cameras of the present invention;
[0025] Figure 3 Schematic diagram for distinguishing edge crack defects of the present invention;
[0026] Figure 4 This is a data processing flow chart of the present invention. DETAILED DESCRIPTION
[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0028] See also Figure 1-Figure 3 The present invention provides a technical solution: an online detection system and method for edge crack defects in strip steel, including: an online detection system for edge crack defects, the online detection system for edge crack defects including a system control unit, the system control unit consisting of a strip steel surface data acquisition system, a data transmission module, a data processing module and a defect alarm module.
[0029] Further improved, the system control unit is used to control the automatic edge search of the mobile devices on both sides of the strip, the data collection of the two laser cameras, realize data exchange with the unit, receive the strip information, operating status and other information sent by the unit, and send the detection results.
[0030] Further improved, the strip surface data acquisition system consists of two sets of high-speed laser cameras with controllers. The two sets of laser cameras are respectively connected to a lateral moving device with PLC control and are arranged vertically on the two sides of the strip. The lateral moving device automatically finds the edge based on the information provided by the laser.
[0031] Further improved, the function of the data transmission module is to ensure high-speed data transmission between the data acquisition system and the data processing system, and to send the height data collected by the two laser cameras in groups to the data analysis system, with each group containing between 500 and 2000 laser line profiles.
[0032] Further improvements, such as Figure 3 、 4As shown: The data processing module analyzes each contour. For each contour, it first searches for invalid data points on the laser line from the outermost edge of the laser line. Starting from the first valid data point, it counts the number of points from this data point to the last valid data point. The height value corresponding to the number of valid data points on the contour is recorded to obtain the curve (b). The curve composed of the valid data is subjected to moving average noise reduction. After noise reduction, the curve shown in (c) is obtained. The height difference Δh between the highest and lowest points on the curve is calculated. Common edge crack defects are as follows: Figure 3 As shown in the figure, they are three types of defects: opening defect, edge wrinkle and edge non-opening defect. The data processing unit analyzes the data sent by the detection unit one by one to obtain the shape information of these three edge crack defects. The judgment method for these three edge crack defects is as follows:
[0033] Method for judging edge opening defects: compare the current contour with the next contour. When the valid data points in the next contour differ by 3 to 5 from the limited data points in the current contour, record the number of valid height data points of the current contour and continuously compare with the subsequent data. If the valid data points of three consecutive contours decrease, it is judged that an edge crack defect has occurred. After that, store all the data in this group, output the height cloud map formed by this group of data, calculate the minimum and maximum number of points of each contour in this group of data, and multiply the number of points by the point spacing to obtain the length of the edge crack.
[0034] Method for judging edge wrinkle defects: judge whether the Δh in the current profile exceeds the threshold, which is usually defined as 0.05-0.1mm. When 3-5 consecutive profiles exceed the threshold, it is judged that there is a wrinkle defect on the edge. After that, all the data in this group are stored, and the height cloud map formed by this group of data is output. The number of contours with wrinkles in this group of data is calculated. Combined with the current operating speed of the unit, the length of the wrinkle defect can be obtained.
[0035] Judgment and analysis method for non-open edge crack defects: Figure 4 Analyze the defects in (b) to determine whether there is invalid data in the curve. If there is invalid data, record the number of invalid data points in the current curve. If invalid data points appear in 2 to 5 consecutive contours, it is determined that there is a non-open edge crack on the edge. Store all the data in this group and output the height cloud map formed by this group of data. Calculate the number of invalid points in each contour in this group of data. Combined with the current operating speed of the unit, the geometric size of its open defect can be calculated.
[0036] Further improved, the defect alarm system issues an alarm to the operator through the audio system installed in the operating room after receiving the defect information, and displays a two-dimensional image of the current edge crack defect in the human-computer interaction interface nearby, so that the operator can see the shape and size of the edge crack defect more intuitively and make timely adjustments to the unit operation.
[0037] Further improved, the online detection method for edge crack defects comprises the following steps:
[0038] Step 1: Based on the strip width specifications produced by the unit, two LJ-V7080 laser cameras were selected, along with a mobile device with an effective travel range, to form a data acquisition system. After an on-site inspection of the installation conditions at the welder outlet of the unit, a gantry-mounted sensor rack was used to confirm the reference height of the laser camera from the strip. The laser camera then collected relative height data of the strip from top to bottom.
[0039] Step 2: The system control unit is installed in the electrical room near the data acquisition system. It receives information such as strip information and operating status from the unit through the data transmission system. After strip welding is completed, it starts automatic edge detection. After edge detection is completed, it performs edge crack detection. When the strip stops, the edge crack detection ends. When an edge crack defect is found, the edge crack defect information is sent to the unit.
[0040] Step 3: During the edge search process, the mobile device determines in real time the percentage of valid data in the height data detected by the laser camera. When the percentage of valid data reaches approximately 50%, the mobile device stops and the edge search is completed.
[0041] Step 4: The system control unit host controls the data acquisition system to collect parameters through the data transmission system, and sends the collected strip height data in groups of 1,000 to the data analysis system;
[0042] Step 5: The data analysis system analyzes the 1,000 height data sent by the system control unit one by one, and regards the demarcation point of the valid data in the height data as the boundary point of the strip. In this way, the position information of the strip edge in the width direction can be accurately obtained. Then, by calculating the peak-to-peak value of the strip edge position in each group of 1,000 data, the maximum depth of the strip edge crack in the current data can be accurately obtained. When the peak-to-peak value is greater than the threshold value of 0.2mm, it is judged that a large edge crack defect exists in this section of the profile. The maximum depth of the edge crack defect is calculated by analyzing and calculating this section of defect data. At the same time, based on the running speed of the strip, the location and width of the edge crack defect are calculated;
[0043] Step 6: The system control unit sends the defect information to the unit and the defect alarm system, and saves the defect data and the analyzed defect information in a txt format in a folder with the current coil number and current time;
[0044] Step 7: After receiving the defect information, the defect alarm system will issue an alarm to the operator through the audio system installed in the operation room, and display a two-dimensional image of the current edge crack defect on the human-computer interaction interface next to it, allowing the operator to more intuitively see the shape and size of the edge crack defect and make timely adjustments to the unit operation;
[0045] Step 8: When the operator queries the edge crack defects of past steel coils, he can enter part of the steel coil number or a certain time period in the human-computer interaction interface to display all the steel coil numbers that meet the conditions. The steel coil numbers of steel coils with edge crack defects are marked in red font, so that the operator can quickly find the defective steel coils; click on any steel coil to display all the defect records of the steel coil, which include the time, location, depth and width of the defect; click on any defect record to display a two-dimensional image of the edge crack defect, so that the operator can more intuitively see the morphology and size of the edge crack defect.
[0046] The present invention uses high-speed, high-precision line laser detection technology to extract height information data from the strip edge. By analyzing the height information, it is determined whether the strip edge has an edge crack defect. Under the control of the system control unit, the detection system uses a moving device arranged at the strip edge to move two laser cameras arranged at the strip edge to the detection position of the strip edge. After receiving the height information data collected by the laser cameras, the data processing unit analyzes the data one by one. Based on the correlation between the various contour information data, the system determines whether the strip edge has an edge crack defect. When an edge crack defect exceeding the standard is found on the strip edge, the computer automatically retains the data at the edge crack defect, calculates the physical shape of the defect, and provides a two-dimensional morphology of the edge crack defect. The system can achieve high-precision, high-density, and low-latency scanning of both sides of each roll of strip, continuously monitor the surface condition of the strip, and quickly and accurately find edge crack defects with a depth greater than or equal to 0.1 mm on the strip, effectively solving the problem of strip edge crack detection on continuous high-speed strip production lines. The present invention can determine the travel and position of the mobile device according to the width range of the object to be measured. The detection device has the communication function with the upper computer of the unit and can realize the offline query function of the detection data.
[0047] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.
Claims
1. An online detection method for edge crack defects of strip steel, characterized by: An online detection system for edge crack defects of strip steel is used, which includes a system control unit, a data acquisition system, a data transmission module, a data processing module and a defect alarm module. The system control unit is used to control the automatic edge finding of the mobile devices on both sides of the strip, the data collection of the two laser cameras, the data exchange with the unit, the receiving of the strip information and the operation status information from the unit, and the sending of the detection results. The data acquisition system includes two sets of laser cameras with controllers. The two sets of laser cameras are connected to mobile devices with PLC control and arranged vertically on the two sides of the strip. The mobile devices can automatically find the edge according to the information provided by the laser cameras. The function of the data transmission module is to ensure high-speed data transmission between the data acquisition system and the data processing module, and to send the height data collected by the two laser cameras to the data processing module in groups. The number of laser line profiles contained in each group is between 500 and 2000. The data processing module analyzes each contour. For each contour, firstly, invalid data points on the laser line are found from the outermost side of the laser line. Starting from the first valid data point, the number of points from the data point to the last valid data point is calculated. The height values corresponding to the valid data points on the contour are recorded to obtain the b curve. The curve composed of valid data is subjected to moving average noise reduction to obtain the c curve after noise reduction. The height difference Δh between the highest point and the lowest point on the curve is calculated. The method for judging the edge wrinkle defect is to judge whether the Δh in the current profile exceeds the threshold, which is defined as 0.05-0.1mm. When 3-5 consecutive profiles exceed the threshold, it is judged that there is a wrinkle defect on the edge. The judgment method for edge opening defects is: compare the current contour with the next contour. When the valid data points in the next contour differ by 3 to 5 from the limited data points in the current contour, record the number of points of the current contour's valid height data and continuously compare it with the subsequent data. If the number of valid data points in three consecutive contours decreases continuously, it is judged that an edge crack defect has occurred. The edge crack defect online detection method comprises the following steps: Step 1: According to the width specifications of the strip produced by the unit, two LJ-V7080 laser cameras are selected, and a mobile device with an effective stroke is used to form a data acquisition system. The installation conditions of the welding machine outlet of the unit are inspected on site, and the sensor rack is installed in a gantry style to confirm the reference height of the laser camera from the strip. The laser camera collects the relative height data of the strip from top to bottom; Step 2: The system control unit is installed in the electrical room near the data acquisition system installation location. Through the data transmission module, it receives the strip information and operation status information sent by the unit. After the strip welding is completed, it starts to automatically find the edge. After the edge finding is completed, it performs edge crack detection. When the strip stops, the edge crack detection ends. After the edge crack defect is found, the edge crack defect information is sent to the unit. Step 3: During the edge search process of the mobile device, the proportion of valid data in the height data detected by the laser camera is determined in real time. When the proportion of valid data reaches about 50%, the mobile device is stopped and the edge search is completed; Step 4: The system control unit controls the data acquisition system acquisition parameters through the data transmission module, and sends the collected strip height data to the data processing module in groups of 1,000; Step 5: The data processing module analyzes the 1000 height data sent by the system control unit one by one, and regards the demarcation point of the valid data in the height data as the boundary point of the strip, so that the position information of the edge of the strip in the width direction can be accurately obtained, and then by calculating the peak-to-peak value of the edge position of the strip in each group of 1000 data, the maximum depth of the edge crack of the strip in the current data can be accurately obtained. When the peak-to-peak value is greater than the threshold value of 0.2mm, it is judged that there is an edge crack defect. The maximum depth of the edge crack defect is calculated by analyzing this section of defect data, and at the same time, the position of the edge crack defect and the width of the edge crack defect are calculated according to the running speed of the strip; Step 6: The system control unit sends the defect information to the unit and defect alarm module, and saves this defect data and the analyzed defect information in a txt format in a folder with the current steel coil number and current time; Step 7: After receiving the defect information, the defect alarm module sends an alarm to the operator through the audio system installed in the operation room, and displays a two-dimensional image of the current edge crack defect in the human-computer interaction interface next to it, so that the operator can see the shape and size of the edge crack defect more intuitively and make timely adjustments to the unit operation; Step 8: When the operator queries the edge crack defects of the past steel coils, he can enter part of the steel coil number or a certain time period in the human-computer interaction interface to display all the steel coil numbers that meet the conditions. For the steel coils with edge crack defects, the steel coil numbers are marked in red font so that the operator can quickly find the defective steel coils; click on any steel coil to display all the defect records of the steel coil, which include the time, location, depth and width of the defect; Clicking on any defect record will display a two-dimensional image of the edge crack defect, allowing the operator to see the shape and size of the edge crack defect more intuitively.
