A cold-rolled steel plate defect intelligent detection method and device

By conducting defect detection and data analysis on the key production links of cold-rolled steel plates, and establishing defect models, the problem of lack of meticulous analysis in the existing technology is solved, efficient quality control and intelligent inspection are achieved, and product quality and production stability are improved. .

CN118520994BActive Publication Date: 2024-11-08SHANDONG XINTAI NEW MATERIAL TECH CO LTD
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202410575797.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-10
Publication Date
2024-11-08
Estimated Expiration
2044-05-10

AI Technical Summary

Technical Problem

The existing technology lacks meticulous production link analysis in steel plate defect detection, resulting in unreliable quality control and inability to promptly discover problems and bottlenecks in the production process, affecting product quality and safety.

Method used

By counting and detecting defect data in the first rolling link, second annealing link and third pickling treatment link of cold-rolled steel plate, a regression analysis method is used to establish a correlation model between defects and production data, obtain defect threshold data, and conduct real-time operation Detect and filter, identify and prompt abnormal links.

Benefits of technology

It has achieved efficient quality control in the production process of cold-rolled steel plates, timely discover and solve defects, improve product quality and production efficiency, reduce quality risks and cost losses, and promote the scientific and intelligent quality management.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118520994B_ABST
    Figure CN118520994B_ABST
Patent Text Reader

Abstract

The present invention discloses a cold-rolled steel plate defect intelligent detection method and device, which relates to the technical field of steel plate defect detection, including S1, counting the production links of cold-rolled steel plates; S2, detecting the production links of cold-rolled steel plates; S3, screening the defective production links of cold-rolled steel plates. The first rolling link, the second annealing link and the third pickling treatment link are statistically detected, and the corresponding defect detection results are obtained, and finally the defective production links of the cold-rolled steel plates are screened and prompted, and the production data and real-time detection results of the cold-rolled steel plates are comprehensively analyzed. By comparing with the defect threshold data of the cold-rolled steel plates, abnormal links can be identified and automatically prompted, which helps to take measures in time to prevent the further continuation of quality problems, improve the quality of cold-rolled steel plates, and further reduce quality risks and cost losses, thereby ensuring the effectiveness and sustainability of intelligent detection of cold-rolled steel plate defects.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of steel plate defect detection, and in particular to an intelligent defect detection method and device for cold-rolled steel plates. Background Art

[0002] In the production process of cold-rolled steel sheets, due to the complex process and various factors, some defects often occur. These defects include but are not limited to surface flaws, internal inclusions, pores, cracks, etc., which may reduce the quality of cold-rolled steel sheets, affect the service life of the product, and may cause safety hazards in subsequent processing and use.

[0003] Prior art, such as the invention patent with announcement number: CN112949108B, is a full-process prediction method and a graphical user interface device for hot-rolled high-strength steel plate shape defects. The method includes: detecting and counting the feedback on strip production in the most recent preset time period, screening out the specifications and models of strip that are prone to strip bending during production, and obtaining the strip parameters and production process parameters of the strip from the finishing exit to the stripping process through the specifications and models; establishing a strip bending prediction model for the strip from the finishing exit to the stripping process based on the strip parameters and production process parameters, and obtaining the predicted value of the strip bending after stripping of the specification model; measuring the actual bending value of the strip of the specification model through a sampling detection method, and verifying and correcting the bending prediction model.

[0004] Combined with the above scheme, it is found that in the current defect detection of steel plates, there is no detailed analysis of the production links of steel plates, and the quality control process is not reliable enough, which may lead to potential problems in the production process and steel plate quality problems involved in the production not being discovered in time, and the production process cannot be efficiently analyzed, and problems and bottlenecks in the steel plate production process cannot be discovered in time, thus making it impossible to carry out effective quality control. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and device for intelligently detecting defects of cold-rolled steel plates, which can effectively solve the problems involved in the above-mentioned background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a method for intelligent defect detection of cold-rolled steel plates, including counting the production links of cold-rolled steel plates, including a first rolling link, a second annealing link and a third pickling treatment link, and analyzing to obtain defect threshold data of the cold-rolled steel plates.

[0007] The production links of the cold-rolled steel sheet are inspected, including the first rolling link inspection, the second annealing link inspection and the third pickling treatment link inspection, and the defect detection results of the first rolling link, the second annealing link and the third pickling treatment link are obtained.

[0008] According to the defect threshold data of the cold-rolled steel sheet and according to the defect detection results of the first rolling link, the second annealing link and the third pickling treatment link, the defective production links of the cold-rolled steel sheet are screened and prompts are given.

[0009] Furthermore, the analysis obtains defect threshold data of the cold-rolled steel plate, and the specific analysis process is: extracting historical production data and historical defect data of the first rolling link, the second annealing link, and the third pickling treatment link of the cold-rolled steel plate, using a regression analysis method to establish a correlation model between defects and cold-rolled steel plate production data, and obtaining the defect threshold data of the cold-rolled steel plate, wherein the defect threshold data of the cold-rolled steel plate includes the defect characterization threshold of the first rolling link, the defect characterization threshold of the second annealing link, and the defect characterization threshold of the third pickling treatment link.

[0010] Further, the defect detection results of the first rolling link, the second annealing link and the third pickling link respectively include: the first rolling link defect characterization value of the cold-rolled steel plate, the second annealing link defect characterization value of the cold-rolled steel plate and the third pickling link defect characterization value of the cold-rolled steel plate;

[0011] The first rolling link defect characterization value of the cold-rolled steel plate is a quantitative result obtained by detecting and analyzing the first rolling link of the cold-rolled steel plate and is used to evaluate the degree of defects of the cold-rolled steel plate in the first rolling link.

[0012] The second annealing stage defect characterization value of the cold-rolled steel sheet represents a quantitative result obtained by detecting and analyzing the second annealing stage of the cold-rolled steel sheet, which is used to evaluate the defect degree of the cold-rolled steel sheet in the second annealing stage.

[0013] The defect characterization value of the third pickling treatment link of the cold-rolled steel sheet represents a quantitative result obtained by detecting and analyzing the third pickling treatment link of the cold-rolled steel sheet, which is used to evaluate the degree of defects of the cold-rolled steel sheet in the third pickling treatment link.

[0014] Furthermore, the first rolling link detection is used to detect the first rolling link of the cold-rolled steel plate to obtain the defect detection result of the first rolling link. The specific analysis process is: collecting the image of the first rolling link termination time point of the cold-rolled steel plate, and comparing it with the reference rolling image of the cold-rolled steel plate stored in the cloud database, thereby obtaining the number of roller marks and the number of depressions of the cold-rolled steel plate.

[0015] Several screening points are randomly arranged, and the thickness of each screening point of the cold-rolled steel plate at the starting time point and the ending time point of the first rolling link is counted. The thickness deviation value of each screening point of the cold-rolled steel plate in the first rolling link is obtained through difference processing, and according to the number of roller marks and the number of depressions of the cold-rolled steel plate, the defect characterization value of the first rolling link of the cold-rolled steel plate is obtained after processing, and used as the defect detection result of the first rolling link.

[0016] Furthermore, the second annealing link is detected to detect the second annealing link of the cold-rolled steel plate and obtain the defect detection result of the second annealing link. The specific analysis process is: a number of monitoring time points are arranged in the second annealing link, and the annealing temperature of the cold-rolled steel plate is monitored at each monitoring time point, thereby constructing a temperature-time variation curve of the second annealing link, and integrating it to obtain the annealing time of the cold-rolled steel plate in the second annealing link.

[0017] The annealing temperature at the start and end times of the second annealing step and the duration of the second annealing step are extracted to obtain the cooling rate of the cold-rolled steel plate in the second annealing step. Based on the annealing time of the cold-rolled steel plate in the second annealing step, the defect characterization value of the second annealing step of the cold-rolled steel plate is obtained after processing and used as the defect detection result of the second annealing step.

[0018] Furthermore, the third pickling treatment link is detected to detect the third pickling treatment link of the cold-rolled steel plate and obtain the defect detection result of the third pickling treatment link. The specific analysis process is: after the third pickling treatment link, several detection areas are arranged on the cold-rolled steel plate, and the surface roughness and surface defect density of the cold-rolled steel plate are monitored in each detection area. After processing, the defect characterization value of the third pickling treatment link of the cold-rolled steel plate is obtained, and it is used as the defect detection result of the third pickling treatment link.

[0019] Furthermore, the defective production links of the cold-rolled steel plates are screened and prompts are given, and the specific process is: comparing the defect characterization value of the first rolling link of the cold-rolled steel plate with the defect characterization threshold of the first rolling link; if the defect characterization value of the first rolling link of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the first rolling link, the first rolling link is marked as a defective link and a prompt is given.

[0020] The defect characterization value of the second annealing stage of the cold-rolled steel plate is compared with the defect characterization threshold of the second annealing stage. If the defect characterization value of the second annealing stage of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the second annealing stage, the second annealing stage is marked as a defective stage and a prompt is given.

[0021] The defect characterization value of the third pickling treatment link of the cold-rolled steel plate is compared with the defect characterization threshold of the third pickling treatment link. If the defect characterization value of the third pickling treatment link of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the third pickling treatment link, the third pickling treatment link is marked as a defective link and a prompt is given.

[0022] Furthermore, the first rolling step defect characterization value of the cold-rolled steel sheet is specifically analyzed under the following conditions:

[0023]

[0024] In the formula, represents the defect characterization value of the first rolling link of cold-rolled steel sheet, α 1j represents the thickness deviation value of the jth screening point in the first rolling link, Δα1 represents the set defined thickness deviation value, β2 represents the number of roller marks of the cold-rolled steel plate, ε1 represents the correction factor corresponding to the set thickness deviation value, ε2 represents the defect influence factor corresponding to the set single roller mark, j represents the number of each screening point, j=1, 2, 3, ..., m, m represents the total number of screening points.

[0025] Furthermore, the defect characterization value of the second annealing step of the cold-rolled steel sheet is specifically analyzed under the following conditions:

[0026]

[0027] In the formula, γ represents the defect characterization value of the second annealing stage of the cold-rolled steel sheet, v represents the cooling rate of the cold-rolled steel sheet in the second annealing stage, T represents the annealing time of the cold-rolled steel sheet in the second annealing stage, Δv represents the set reference cooling rate, ΔT represents the set reference annealing time, τ1 represents the correction factor corresponding to the set cooling rate, and τ2 represents the correction factor corresponding to the set annealing time.

[0028] The second aspect of the present invention also provides an intelligent defect detection device for cold-rolled steel plates, characterized in that it includes: a processor, and a memory and a network interface connected to the processor; the network interface is connected to a non-volatile memory in a server; when the processor is running, it calls a computer program from the non-volatile memory through the network interface, and runs the computer program through the memory to execute the above-mentioned intelligent defect detection method for cold-rolled steel plates.

[0029] The present invention has the following beneficial effects:

[0030] (1) The present invention detects by counting the first rolling link, the second annealing link and the third pickling treatment link, and obtains the corresponding defect detection results, and finally screens and prompts the defective production links of the cold-rolled steel sheet. The production data and real-time detection results of the cold-rolled steel sheet are comprehensively analyzed, which helps to timely discover and solve the defect problems in the production links, improve the production quality of the cold-rolled steel sheet, and help to further reduce quality risks and cost losses.

[0031] (2) The present invention obtains the defect threshold data of the cold-rolled steel plate by analysis, which provides an important basis for the subsequent intelligent defect detection. By establishing a correlation model between defects and production data, potential problems in the production process of cold-rolled steel plates can be identified more accurately, which helps to provide early warning and timely adjust the production links, thereby improving product quality and production efficiency, and realizing intelligent defect detection of cold-rolled steel plates.

[0032] (3) The present invention obtains the defect characterization value of the first rolling link, the defect characterization value of the second annealing link of the cold-rolled steel plate, and the defect characterization value of the third pickling treatment link of the cold-rolled steel plate by analyzing, which helps to achieve efficient quality control and defect monitoring of the cold-rolled steel plate production process. By comparing with the defect threshold data of the cold-rolled steel plate, the abnormal link can be identified and automatically prompted, which helps to take timely measures to prevent the further continuation of quality problems and provide reliable protection for the production quality of cold-rolled steel plates.

[0033] (4) The present invention helps to take timely measures to deal with and improve the production links with quality problems by identifying them, so as to avoid further expansion of quality defects. It can effectively improve the controllability and stability of the cold-rolled steel plate production process and promote the scientific and intelligent quality management.

[0034] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a schematic diagram of the method flow of the present invention.

[0036] Figure 2 This is the temperature-time variation curve of the second annealing step of the present invention. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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 creative work are within the scope of protection of the present invention.

[0038] See also Figure 1 As shown, an embodiment of the present invention provides a technical solution for a cold-rolled steel plate defect intelligent detection method: a cold-rolled steel plate defect intelligent detection method, including S1, statistics of the production links of the cold-rolled steel plate: statistics of the production links of the cold-rolled steel plate, including a first rolling link, a second annealing link and a third pickling treatment link, and analysis to obtain the defect threshold data of the cold-rolled steel plate.

[0039] S2. Inspection of production links of cold-rolled steel sheets: Inspection of production links of cold-rolled steel sheets, including the first rolling link inspection, the second annealing link inspection and the third pickling treatment link inspection, obtains defect detection results of the first rolling link, the second annealing link and the third pickling treatment link.

[0040] S3. Screening defective production links of cold-rolled steel sheets: Based on the defect threshold data of the cold-rolled steel sheets and based on the defect detection results of the first rolling link, the second annealing link and the third pickling treatment link, screen the defective production links of the cold-rolled steel sheets and give prompts.

[0041] Specifically, the defect threshold data of the cold-rolled steel plate is analyzed, and the specific analysis process is: extracting the historical production data and historical defect data of the first rolling link, the second annealing link and the third pickling treatment link of the cold-rolled steel plate, using the regression analysis method to establish a correlation model between defects and cold-rolled steel plate production data, and obtaining the defect threshold data of the cold-rolled steel plate, wherein the defect threshold data of the cold-rolled steel plate includes the defect characterization threshold of the first rolling link, the defect characterization threshold of the second annealing link and the defect characterization threshold of the third pickling treatment link.

[0042] It should be noted that the first rolling link, the second annealing link and the third pickling link of cold-rolled steel sheets are the key links in the production process of cold-rolled steel sheets. Many defects of cold-rolled steel sheets are often generated in these three links, such as surface flaws, internal inclusions, etc. Therefore, monitoring and analyzing these links can detect and solve potential quality problems early and avoid unexpected situations in subsequent processing and use. It has an important impact on the quality and performance of the final product.

[0043] It should be noted that after extracting the historical production data and historical defect data of the first rolling stage, the second annealing stage, and the third pickling stage of the cold-rolled steel sheet, data preprocessing was first performed, including data cleaning, missing value processing, and outlier processing, to ensure the accuracy and completeness of the data. Then, the correlation model between defects and cold-rolled steel sheet production data was established using regression analysis methods.

[0044] For example, for the first rolling process, the historical production data of steel plates, including key parameters such as rolling speed, rolling temperature, and rolling pressure, were collected and sorted, and the historical defect data of steel plates, such as surface roughness and surface defect density, were collected. Then, a regression model was established using the multivariate linear regression analysis method, with these production parameters as independent variables and the surface quality data of steel plates as dependent variables, to explore the correlation between production parameters and surface quality of steel plates. Through the analysis and verification of the regression model, the defect characterization threshold of the first rolling process was obtained.

[0045] Exemplarily, for the second annealing step, data such as annealing temperature, annealing time, cooling rate, and historical defect data of the steel plate, such as pores, inclusions, internal cracks, etc., are collected. The correlation model between production parameters and the internal quality of the steel plate is established by regression analysis method, and the defect characterization threshold of the second annealing step is obtained. Exemplarily, for the third pickling treatment step, historical production data of pickling treatment, such as pickling temperature, pickling time, acid concentration, etc., and historical defect data such as surface cleanliness and corrosion rate of the steel plate are collected. The correlation model between production parameters and surface quality is established by regression analysis method, and the defect characterization threshold of the third pickling treatment step is obtained, which provides strong support for the optimization and quality control of the production process.

[0046] Specifically, the defect detection results of the first rolling link, the second annealing link and the third pickling treatment link respectively include: the defect characterization value of the first rolling link of the cold-rolled steel plate, the defect characterization value of the second annealing link of the cold-rolled steel plate and the defect characterization value of the third pickling treatment link of the cold-rolled steel plate.

[0047] The first rolling link defect characterization value of the cold-rolled steel plate is a quantitative result obtained by detecting and analyzing the first rolling link of the cold-rolled steel plate and is used to evaluate the degree of defects of the cold-rolled steel plate in the first rolling link.

[0048] The second annealing stage defect characterization value of the cold-rolled steel sheet represents a quantitative result obtained by detecting and analyzing the second annealing stage of the cold-rolled steel sheet, which is used to evaluate the defect degree of the cold-rolled steel sheet in the second annealing stage.

[0049] The defect characterization value of the third pickling treatment link of the cold-rolled steel sheet represents a quantitative result obtained by detecting and analyzing the third pickling treatment link of the cold-rolled steel sheet, which is used to evaluate the degree of defects of the cold-rolled steel sheet in the third pickling treatment link.

[0050] Specifically, an image of the termination time point of the first rolling stage of the cold-rolled steel plate is collected and compared with the reference rolling image of the cold-rolled steel plate stored in the cloud database, thereby obtaining the number of roller marks and the number of depressions of the cold-rolled steel plate.

[0051] It should be noted that cloud database is a database service based on cloud computing technology. It deploys the database on a cloud server and provides data storage, management and access services through the Internet. Users can connect to the cloud database through the network and store, back up, restore and share data without having to build and maintain their own database server.

[0052] Several screening points are randomly arranged, and the thickness of each screening point of the cold-rolled steel plate at the starting time point and the ending time point of the first rolling link is counted. The thickness deviation value of each screening point of the cold-rolled steel plate in the first rolling link is obtained through difference processing, and according to the number of roller marks and the number of depressions of the cold-rolled steel plate, the defect characterization value of the first rolling link of the cold-rolled steel plate is obtained after processing, and used as the defect detection result of the first rolling link.

[0053] It should be noted that in the first rolling stage, the steel plate may produce surface defects such as roll marks and depressions when it is rolled by the cold rolling mill rolls. These defects can be collected and recorded by timely image acquisition, thereby helping to timely discover and solve possible problems in the first rolling stage, and take corresponding measures to reduce the defect rate and improve the quality of cold-rolled steel plates.

[0054] It should be noted that after being rolled by the first rolling mill, the thickness of the cold-rolled steel sheet will gradually decrease. This is due to the pressure of the cold rolling mill rollers, which squeezes the steel sheet when it passes through the rollers, causing the thickness of the steel sheet to continue to decrease. In the normal production process, in order to ensure that the thickness of the cold-rolled steel sheet can meet the product specification requirements and ensure the stability of product quality, the thickness change in the first rolling link is usually smooth, continuous and accurate, without drastic fluctuations. Therefore, in the first rolling link, the thickness of the cold-rolled steel sheet can be finely monitored, which helps to ensure the dimensional accuracy and quality consistency of the cold-rolled steel sheet.

[0055] It should be noted that the thickness deviation value is an important indicator reflecting the thickness control accuracy and stability of the cold-rolled steel plate in the first rolling link. In the first rolling link, the thickness deviation value is too small or increased, which may mean that the roller over-extrudes or over-relaxes the steel plate, thereby increasing the generation of roller marks and depressions. At the same time, the number of roller marks and depressions can indirectly reflect the thickness deviation value of the cold-rolled steel plate. For example, if there are more uneven wear or damage on the roller surface of the cold rolling mill, irregular marks and depressions may appear on the surface of the steel plate, thereby affecting the thickness measurement result of the steel plate. In addition, the increase in the number of roller marks and depressions may affect each other. More roller marks may cause the surface of the steel plate to be uneven, thereby increasing the generation rate of depressions; while more depressions may aggravate the wear of the rollers and increase the number of roller marks. Therefore, the thickness deviation value, the number of roller marks and the number of depressions of the cold-rolled steel plate in the first rolling link affect each other, and together reflect the quality status and process stability of the cold-rolled steel plate in the first rolling link.

[0056] Specifically, the second annealing link is detected to detect the second annealing link of the cold-rolled steel plate and obtain the defect detection result of the second annealing link. The specific analysis process is: a number of monitoring time points are arranged in the second annealing link, and the annealing temperature of the cold-rolled steel plate is monitored at each monitoring time point, thereby constructing the temperature-time variation curve of the second annealing link, and integrating it to obtain the annealing time of the cold-rolled steel plate in the second annealing link.

[0057] The annealing temperature at the start and end times of the second annealing step and the duration of the second annealing step are extracted to obtain the cooling rate of the cold-rolled steel plate in the second annealing step. Based on the annealing time of the cold-rolled steel plate in the second annealing step, the defect characterization value of the second annealing step of the cold-rolled steel plate is obtained after processing and used as the defect detection result of the second annealing step.

[0058] like Figure 2 As shown, Figure 2 The temperature-time variation curve of the second annealing step, where the horizontal axis represents the monitoring time point in minutes. The vertical axis represents the temperature in degrees Celsius (℃). The temperature-time variation curve of the second annealing step reflects the trend of temperature change over time. At the beginning, the temperature gradually rises. The temperature change rate is relatively fast in this stage until the set annealing temperature is reached. After reaching the annealing temperature, the temperature will remain stable for a period of time, so that the temperature distribution and organizational structure inside the steel plate can be uniformly adjusted, thereby revealing the law and characteristics of temperature change in the second annealing step.

[0059] It should be noted that the specific calculation formula for the annealing time of the cold-rolled steel sheet in the second annealing step is: Wherein T represents the annealing time of the cold-rolled steel sheet in the second annealing stage, W(i) represents the annealing temperature of the second annealing stage of the cold-rolled steel sheet at the i-th monitoring time point, i represents the number of each monitoring time point, i=1,2,3,...,a, a represents the total number of monitoring time points.

[0060] It should be noted that the annealing time refers to the total time length of the annealing treatment performed in the second annealing step.

[0061] It should be noted that the cooling rate refers to the cooling speed of the steel plate from high temperature to room temperature. A faster cooling rate may cause uneven cooling of the steel plate during the annealing process, increase the risk of internal stress and deformation, and thus affect the quality and performance of the steel plate. And changes in the cooling rate may affect the annealing time of the steel plate. For example, a faster cooling rate will cause the steel plate to cool below the target temperature faster during the annealing process, which may shorten the annealing time. On the contrary, a slower cooling rate will cause the steel plate to cool below the target temperature more slowly during the annealing process, which may extend the annealing time.

[0062] Specifically, the third pickling treatment link is detected, which is used to detect the third pickling treatment link of the cold-rolled steel plate and obtain the defect detection result of the third pickling treatment link. The specific analysis process is: after the third pickling treatment link, several detection areas are arranged on the cold-rolled steel plate, and the surface roughness and surface defect density of the cold-rolled steel plate are monitored in each detection area. After processing, the defect characterization value of the third pickling treatment link of the cold-rolled steel plate is obtained, and it is used as the defect detection result of the third pickling treatment link.

[0063] It should be noted that the defect characterization value of the third pickling treatment link of the cold-rolled steel sheet represents the quantitative result obtained by analyzing the surface roughness and surface defect density of the cold-rolled steel sheet in each detection area, which is used to describe the degree of defects of the steel sheet in the third pickling treatment link. Not only can the surface of the cold-rolled steel sheet be imaged and analyzed by using high-resolution imaging technology, such as scanning electron microscopy (SEM) or atomic force microscopy (AFM). These technologies can provide detailed information on surface morphology, surface roughness, surface defects, etc. By processing and analyzing the image, the defect characterization value of the third pickling treatment link of the cold-rolled steel sheet can be obtained. In this embodiment, it is obtained by the following calculation method, and the specific calculation method is as follows:

[0064]

[0065] Where, δ represents the defect characterization value of the third pickling treatment of cold-rolled steel sheets, C r represents the surface roughness of the cold-rolled steel sheet in the rth detection area, ρ rrepresents the surface defect density of the cold-rolled steel plate in the rth inspection area, ΔC represents the set defined surface roughness, Δρ represents the set defined surface defect density, σ1 represents the correction factor corresponding to the set surface roughness, σ2 represents the correction factor corresponding to the set surface defect density of the cold-rolled steel plate, r represents the number of each inspection area, r=1,2,3,...,d, d represents the total number of inspection areas.

[0066] It should be noted that surface roughness is a measure of the unevenness of the surface. A surface roughness meter or surface profiler can be used to scan and measure the surface of the cold-rolled steel plate, and generate a surface profile or surface roughness parameters such as Ra, Rz, etc. The surface roughness of the test area can be obtained by statistical analysis or averaging of the measurement results. After pickling, the roughness of the steel plate surface is usually reduced, showing a smoother and flatter surface. Surface roughness is usually expressed as Ra value (average roughness) in micrometers (μm) or inches (i nch).

[0067] It should be noted that the surface defect density refers to the number of surface defects per unit area. To detect the surface defect density, it is usually necessary to observe the surface of the cold-rolled steel plate with a camera or other equipment, and then determine the number of defects per unit area through counting or image processing technology to obtain the surface defect density. Pickling can remove oxides, rust and other impurities on the surface of the steel plate, thereby reducing the surface defect density and improving the surface quality.

[0068] It should be noted that the surface roughness and the surface defect density may show a negative correlation, that is, the smaller the surface roughness, the lower the surface defect density may be.

[0069] It should be noted that the defined surface roughness refers to a set of predefined standards or specifications for the surface roughness of cold-rolled steel plates in applications. In the embodiment, Ra (average roughness) surface unevenness, concave-convexity and surface quality are used, which is usually in the range of 0.1μm to 3.2μm. In this embodiment, the defined surface roughness value is set to 0.5μm.

[0070] It should be noted that the defined surface defect density refers to a set of predefined standards or specifications for surface defects of cold-rolled steel plates in applications. In this embodiment, the number of surface defects shall not exceed 5 per square centimeter, so the defined surface defect density is set to 5 per square centimeter.

[0071] It should be noted that the correction factor of surface roughness is usually the result of calibration or correction based on the characteristics of the measuring equipment, environmental conditions, and measuring methods. The correction factor can be determined by measuring a series of standard samples and comparing them with reference samples of known accuracy and surface quality. For example, the surface roughness correction factor of the cold-rolled steel sheet is specified as 0.95 in this embodiment, which means that when actually measuring the surface roughness, the measurement result should be multiplied by 0.95 for correction.

[0072] It should be noted that the correction factor of the surface defect density can be obtained based on standardized testing methods and processes, as well as statistical analysis of the test results of a series of standard samples. The correction factor is usually used to correct the differences between different testing equipment or operators, and takes into account environmental factors during measurement. For example, if the surface defect density correction factor of the cold-rolled steel sheet is specified as 1.1 in this embodiment, the measurement result of the surface defect density should be multiplied by 1.1 for correction.

[0073] Specifically, the defective production links of the cold-rolled steel plates are screened and prompts are given. The specific process is: compare the defect characterization value of the first rolling link of the cold-rolled steel plate with the defect characterization threshold of the first rolling link. If the defect characterization value of the first rolling link of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the first rolling link, the first rolling link is marked as a defective link and a prompt is given.

[0074] The defect characterization value of the second annealing stage of the cold-rolled steel plate is compared with the defect characterization threshold of the second annealing stage. If the defect characterization value of the second annealing stage of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the second annealing stage, the second annealing stage is marked as a defective stage and a prompt is given.

[0075] The defect characterization value of the third pickling treatment link of the cold-rolled steel plate is compared with the defect characterization threshold of the third pickling treatment link. If the defect characterization value of the third pickling treatment link of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the third pickling treatment link, the third pickling treatment link is marked as a defective link and a prompt is given.

[0076] It should be noted that when the data processing system detects a possible defect in a certain link, it will mark the link as a defective link according to the preset prompt mechanism, and the generated prompt information will be displayed in real time through display screens, alarms and other hardware devices. It can also be transmitted to relevant operators or monitoring centers through the network or communication equipment so that timely measures can be taken to deal with it.

[0077] It should be noted that by comparing the defect characterization values ​​of cold-rolled steel plates in each production link with the defect characterization threshold data, it is possible to quickly and accurately identify production links with quality problems, and take timely measures to deal with and improve them to avoid further expansion of quality defects. This can effectively improve the controllability and stability of the cold-rolled steel plate production process and promote scientific and intelligent quality management.

[0078] Specifically, the defect characterization value of the first rolling link of the cold-rolled steel plate can be obtained not only by pattern recognition and analysis using historical data, but also by the following calculation method. The specific calculation method is as follows:

[0079]

[0080] In the formula, represents the defect characterization value of the first rolling link of the cold-rolled steel plate, α1j represents the thickness deviation value of the jth screening point in the first rolling link, Δα1 represents the set defined thickness deviation value, β2 represents the number of roller marks of the cold-rolled steel plate, ε1 represents the correction factor corresponding to the set thickness deviation value, ε2 represents the defect influence factor corresponding to the set single roller mark, j represents the number of each screening point, j=1,2,3,...,m, m represents the total number of screening points.

[0081] Several screening points are randomly arranged, and the thickness of each screening point of the cold-rolled steel plate at the starting time point and the ending time point of the first rolling link is counted. The difference processing is performed to obtain the thickness deviation value of each screening point in the first rolling link, and according to the number of roller marks and the number of depressions of the cold-rolled steel plate, the defect characterization value of the first rolling link of the cold-rolled steel plate is obtained after processing, and used as the defect detection result of the first rolling link.

[0082] It should be noted that the defined thickness deviation value is determined by measuring the thickness data of the cold-rolled steel sheet during the first rolling process according to the production process requirements and relevant standards. For example, if the thickness deviation value of the cold-rolled steel sheet during the first rolling process is required to be no more than plus or minus 0.02 mm in this embodiment, it means that at the end of the first rolling process, the defined thickness deviation value measured should not exceed 0.02 mm.

[0083] It should be noted that the correction factor corresponding to the thickness deviation value requires a series of experiments to simulate the thickness measurement process under different production conditions and examine the impact on the measurement precision and accuracy. According to the experimental results, the degree of influence of various factors on the thickness measurement is evaluated and the correction factor corresponding to the thickness deviation value is determined based on production data analysis and experimental research. This factor is usually a coefficient used to correct the measurement results to more accurately reflect the actual situation. For example, according to production data analysis and experimental research, it is determined that under specific production conditions, the measured thickness deviation value is 10% higher than the actual value. Then the correction factor corresponding to the thickness deviation value is 0.9. If the measured thickness deviation value is 0.2 mm, then the corrected thickness deviation value is 0.2 mm multiplied by 0.9, that is, 0.18 mm.

[0084] It should be noted that the defect impact factor corresponding to a single roller mark is usually formulated based on experimental data analysis and process experience. The defect impact factor corresponding to a single roller mark is a coefficient used to quantify the degree of influence of the number of roller marks on the quality of cold-rolled steel sheets. For example, according to experimental data analysis and process experience, if the number of roller marks increases by 1 per square meter, it will have a 0.05 unit impact on the surface quality of the cold-rolled steel sheet. Then the defect impact factor corresponding to a single roller mark is set to 0.05.

[0085] Specifically, the defect characterization value of the second annealing step of the cold-rolled steel sheet can be obtained not only by pattern recognition and analysis using historical data, but also by the following calculation method. The specific calculation method is as follows:

[0086]

[0087] In the formula, γ represents the defect characterization value of the second annealing stage of the cold-rolled steel sheet, v represents the cooling rate of the cold-rolled steel sheet in the second annealing stage, T represents the annealing time of the cold-rolled steel sheet in the second annealing stage, Δv represents the set reference cooling rate, ΔT represents the set reference annealing time, τ1 represents the correction factor corresponding to the set cooling rate, and τ2 represents the correction factor corresponding to the set annealing time.

[0088] In a specific embodiment, as shown in Table 1, Table 1 is example data of various correction factors in the defect characterization value of the second annealing step of the cold-rolled steel plate, which lists the correction factors corresponding to the cooling rate and the correction factors corresponding to the annealing time.

[0089] Table 1 Example data of various correction factors in the defect characterization value of the second annealing stage of cold-rolled steel sheets

[0090]

[0091]

[0092] It should be noted that by adjusting the correction factor corresponding to the cooling rate and the correction factor corresponding to the annealing time, the actual situation can be reflected more accurately, thereby improving the accuracy and reliability of defect detection. The defect characterization value of the second annealing stage of the cold-rolled steel plate obtained reflects the quality status of the cold-rolled steel plate in the second annealing stage.

[0093] It should be noted that the set reference cooling rate is usually determined by process experience, experimental data and relevant standards, and is used to evaluate and control the cooling rate of the cold-rolled steel plate. For example, in this embodiment, the reference cooling rate should be controlled at about 10 degrees Celsius per minute.

[0094] It should be noted that the reference annealing time is a standard or reference value set during the production process of cold-rolled steel sheets, which is used to guide and control the time of the annealing process. For example, the reference annealing time of the cold-rolled steel sheets in this embodiment is set to 3 hours.

[0095] It should be noted that the correction factor corresponding to the cooling rate refers to the value of adjusting the actual cooling rate. The correction factor corresponding to the cooling rate is calculated by simulating the annealing process of the cold-rolled steel sheet using numerical simulation software or finite element analysis. In the simulation, the influence of different cooling rates on the microstructure and performance of the steel sheet can be considered, and then the correction factors under different cooling rates can be derived to consider the changes in influencing factors, so as to more accurately reflect the actual situation of the cold-rolled steel sheet. For example, if the cooling rate of the cold-rolled steel sheet increases by 20%, the correction factor corresponding to the increase in cooling rate can be set to 1.2.

[0096] It should be noted that the correction factor corresponding to the annealing time is used to adjust the actual measured annealing time, and is used to correct the actual annealing time of the cold-rolled steel sheet. It is usually obtained based on experimental data or simulation calculations. For example, when the annealing time increases, the correction factor corresponding to the increase in annealing time can be set to 0.9. In this way, in practical applications, multiplying by the correction factor can adjust the actual measured value to a more accurate predicted value to take into account the actual impact of the annealing time.

[0097] The second aspect of the present invention also provides an intelligent defect detection device for cold-rolled steel plates, characterized in that it includes: a processor, and a memory and a network interface connected to the processor; the network interface is connected to a non-volatile memory in a server; when the processor is running, it calls a computer program from the non-volatile memory through the network interface, and runs the computer program through the memory to execute the above-mentioned intelligent defect detection method for cold-rolled steel plates.

[0098] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0099] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A cold-rolled steel sheet defect intelligent detection method, characterized in that: include: Statistics are collected on the production links of cold-rolled steel sheets, including the first rolling link, the second annealing link, and the third pickling link, and defect threshold data of the cold-rolled steel sheets are obtained through analysis; Testing the production links of the cold-rolled steel sheet, including the first rolling link test, the second annealing link test and the third pickling link test, and obtaining defect detection results of the first rolling link, the second annealing link and the third pickling link; According to the defect threshold data of the cold-rolled steel sheet and according to the defect detection results of the first rolling link, the second annealing link and the third pickling treatment link, the defective production links of the cold-rolled steel sheet are screened and prompted; The defect detection results of the first rolling link, the second annealing link, and the third pickling link respectively include: a defect characterization value of the first rolling link of the cold-rolled steel plate, a defect characterization value of the second annealing link of the cold-rolled steel plate, and a defect characterization value of the third pickling link of the cold-rolled steel plate; The first rolling link defect characterization value of the cold-rolled steel plate is a quantitative result obtained by detecting and analyzing the first rolling link of the cold-rolled steel plate and used to evaluate the degree of defects of the cold-rolled steel plate in the first rolling link; The second annealing stage defect characterization value of the cold-rolled steel sheet represents a quantitative result obtained by detecting and analyzing the second annealing stage of the cold-rolled steel sheet, which is used to evaluate the degree of defects of the cold-rolled steel sheet in the second annealing stage; The third pickling treatment link defect characterization value of the cold-rolled steel sheet represents a quantitative result obtained by detecting and analyzing the third pickling treatment link of the cold-rolled steel sheet, which is used to evaluate the degree of defects of the cold-rolled steel sheet in the third pickling treatment link; The first rolling link detection is used to detect the first rolling link of the cold-rolled steel plate to obtain the defect detection result of the first rolling link. The specific analysis process is: Collecting an image of the cold-rolled steel plate at the end time of the first rolling step, and comparing it with a reference rolling image of the cold-rolled steel plate stored in a cloud database, thereby obtaining the number of roller marks and the number of depressions of the cold-rolled steel plate; A number of screening points are randomly arranged, and the thickness of each screening point of the cold-rolled steel plate at the start time point and the end time point of the first rolling link is counted. The thickness deviation value of each screening point of the cold-rolled steel plate in the first rolling link is obtained through difference processing, and according to the number of roller marks and the number of depressions of the cold-rolled steel plate, the defect characterization value of the first rolling link of the cold-rolled steel plate is obtained through processing, and used as the defect detection result of the first rolling link; The second annealing step is tested to detect the second annealing step of the cold-rolled steel sheet and obtain the defect detection result of the second annealing step. The specific analysis process is as follows: Arrange a plurality of monitoring time points in the second annealing step, monitor the annealing temperature of the cold-rolled steel sheet at each monitoring time point, thereby constructing a temperature-time variation curve of the second annealing step, and integrate the curve to obtain the annealing time of the cold-rolled steel sheet in the second annealing step; Extract the annealing temperature at the start and end time of the second annealing step and the duration of the second annealing step to obtain the cooling rate of the cold-rolled steel sheet in the second annealing step, and obtain the defect characterization value of the second annealing step of the cold-rolled steel sheet according to the annealing time of the cold-rolled steel sheet in the second annealing step, and use it as the defect detection result of the second annealing step; The third pickling treatment link is used to detect the third pickling treatment link of the cold-rolled steel sheet to obtain the defect detection result of the third pickling treatment link. The specific analysis process is as follows: After the third pickling treatment step, several detection areas are arranged on the cold-rolled steel plate, and the surface roughness and surface defect density of the cold-rolled steel plate are monitored in each detection area. After processing, the defect characterization value of the third pickling treatment step of the cold-rolled steel plate is obtained, and used as the defect detection result of the third pickling treatment step; The first rolling step defect characterization value of the cold-rolled steel sheet is specifically analyzed under the following conditions: ; In the formula, Indicates the defect characterization value of the first rolling link of cold-rolled steel sheet, It represents the thickness deviation value of the jth screening point in the first rolling link, Indicates the set thickness deviation value. Indicates the number of roller marks on cold-rolled steel sheets. Indicates the correction factor corresponding to the set thickness deviation value. represents the defect influencing factor corresponding to a set single roll mark, j represents the number of each screening point, , m represents the total number of sieve core points; The second annealing step defect characterization value of the cold-rolled steel sheet, the specific analysis conditions are: ; In the formula, Indicates the defect characterization value of the second annealing step of cold-rolled steel sheet, Indicates the cooling rate of the cold-rolled steel sheet in the second annealing step, Indicates the annealing time of the cold-rolled steel sheet in the second annealing step, represents the set reference cooling rate, Indicates the set reference annealing time, Indicates the correction factor corresponding to the set cooling rate, Indicates the correction factor corresponding to the set annealing time.

2. The cold-rolled steel sheet defect intelligent detection method according to claim 1, characterized in that: The analysis obtains the defect threshold data of the cold-rolled steel sheet, and the specific analysis process is as follows: The historical production data and historical defect data of the first rolling link, the second annealing link and the third pickling treatment link of the cold-rolled steel plate are extracted, and a correlation model between defects and the production data of the cold-rolled steel plate is established using a regression analysis method to obtain the defect threshold data of the cold-rolled steel plate. The defect threshold data of the cold-rolled steel plate includes the defect characterization threshold of the first rolling link, the defect characterization threshold of the second annealing link and the defect characterization threshold of the third pickling treatment link.

3. The cold-rolled steel sheet defect intelligent detection method according to claim 1, characterized in that: The specific process of screening the defects of cold-rolled steel sheets and providing prompts is as follows: The defect characterization value of the first rolling link of the cold-rolled steel plate is compared with the defect characterization threshold of the first rolling link. If the defect characterization value of the first rolling link of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the first rolling link, the first rolling link is marked as a defective link and a prompt is given; The defect characterization value of the second annealing link of the cold-rolled steel sheet is compared with the defect characterization threshold of the second annealing link. If the defect characterization value of the second annealing link of the cold-rolled steel sheet is higher than or equal to the defect characterization threshold of the second annealing link, the second annealing link is marked as a defect link and a prompt is given; The defect characterization value of the third pickling treatment link of the cold-rolled steel plate is compared with the defect characterization threshold of the third pickling treatment link. If the defect characterization value of the third pickling treatment link of the cold-rolled steel plate is higher than or equal to the defect characterization threshold of the third pickling treatment link, the third pickling treatment link is marked as a defective link and a prompt is given.

4. A device for applying a cold-rolled steel sheet defect intelligent detection method as claimed in any one of claims 1 to 3, characterized in that: include: A processor, and memory and network interfaces connected to the processor; The network interface is connected to a non-volatile memory in the server; When running, the processor retrieves a computer program from the non-volatile memory through the network interface, and runs the computer program through the memory to execute the method described in any one of claims 1 to 3.

Citation Information

Cited By

  • Abnormal result tracing method and system for steel material inspection and detection process

    CN122966070A