Online analysis method, apparatus, equipment and storage medium for alkaline washing quality of silicon steel strip

Abstract

This invention relates to the field of silicon steel production technology, specifically disclosing an online method, apparatus, equipment, and storage medium for alkaline washing quality analysis of silicon steel strip. The method includes the following steps: S1, acquiring linear quality distribution information in real time through a detection head; S2, analyzing whether there are any anomalies in the linear quality distribution information based on a first preset quality threshold; S3, when an anomaly is present, acquiring regional alkaline washing quality distribution information through the detection head; S4, acquiring all abnormal location points of the regional alkaline washing quality distribution information based on a second preset quality threshold; S5, inferring the anomaly type based on the distribution of all abnormal location points and the corresponding quality information. This method achieves automated detection and analysis of alkaline washing quality, improves the control level of the production process, and overcomes the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

Description

Technical Field

[0001] This application relates to the field of silicon steel production technology, and more specifically, to a method, apparatus, equipment, and storage medium for online analysis of the quality of silicon steel strip after alkaline washing. Background Technology

[0002] In the silicon steel production industry, the processing flow plays a decisive role in product quality and performance. Core steps such as normalizing pickling, cold rolling, alkaline washing, decarburizing annealing, coating, and high-temperature annealing are interconnected. Among them, the alkaline washing step is particularly critical, aiming to remove rolling oil stains, iron powder, and other residues from the surface of the silicon steel strip, providing a good foundation for subsequent processes.

[0003] Traditional alkaline washing processes typically involve four steps: First, alkaline solution is sprayed through a nozzle to wash the surface of the silicon steel strip, initially removing rolling oil stains; next, the alkaline solution nozzle and brush rollers work together to deeply wash the surface of the silicon steel strip, thoroughly removing residual oil stains; then, hot water spraying and a combination of hot water and brush rollers are used to clean the residual alkaline solution from the surface of the silicon steel strip; finally, the surface moisture of the silicon steel strip is dried with hot air.

[0004] However, inadequate alkaline washing often occurs during actual production. The traditional approach to this problem is to rely on manual analysis of the alkaline washing quality of the silicon steel strip surface to identify the problematic area and make adjustments or repairs. This method is not only time-consuming and labor-intensive, but also highly dependent on the experience and skill level of the workers.

[0005] There is currently no effective technical solution to the above problems. Summary of the Invention

[0006] The purpose of this application is to provide an online method, apparatus, equipment and storage medium for alkaline washing quality analysis of silicon steel strip, so as to realize the automated detection and analysis of alkaline washing quality and overcome the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

[0007] Firstly, this application provides an online quality analysis method for alkaline washing of silicon steel strip, used for quality monitoring in alkaline washing equipment for silicon steel strip. The output end of the alkaline washing equipment is equipped with a quality detection component, which includes multiple linearly arranged detection heads. The detection heads are used to collect quality information at corresponding locations. The method includes the following steps: S1. Obtain linear quality distribution information in real time through the detection head. The linear quality distribution information includes the quality information corresponding to each position point that is linearly distributed perpendicular to the length direction of the silicon steel strip. S2. Analyze whether there are any anomalies in the linear quality distribution information based on the first preset quality threshold; S3. When an anomaly is present, the detection head is used to obtain the regional alkaline washing quality distribution information, which includes the quality information of each location point in the preset rectangular window area. S4. Based on the second preset quality threshold, obtain all abnormal location points of the alkaline washing quality distribution information in the region; S5. Infer the anomaly type based on the distribution of all abnormal locations and the corresponding quality information.

[0008] The online analysis method for alkaline washing quality of silicon steel strip in this application achieves online detection of alkaline washing quality by acquiring linear quality distribution information through linear scanning of the detection head. When an alkaline washing quality abnormality is initially determined, the detection head performs regional scanning to acquire regional alkaline washing quality distribution information. Based on the distribution of abnormal locations in the regional alkaline washing quality distribution information and the corresponding quality information, the abnormality type causing the alkaline washing quality problem is inferred. This realizes automated detection and analysis of alkaline washing quality, improves the control level of the production process, and overcomes the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

[0009] The online quality analysis method for alkaline washing of silicon steel strip, wherein step S5 includes: S51. Extract the number, location coordinates, shape and orientation of all abnormal location points to obtain distribution parameters, and construct the feature vector to be identified based on the distribution parameters and the quality information; S52. Calculate the similarity between the feature vector to be identified and the feature vector corresponding to each preset anomaly type in the pre-built anomaly type knowledge base, and obtain multiple similarity values. The anomaly type knowledge base includes multiple preset anomaly types, and each preset anomaly type corresponds to at least one feature vector. S53. The preset anomaly type with the highest similarity among the similarity values ​​is used as the inferred anomaly type.

[0010] This series of steps works in tandem to transform the original anomaly location distribution and quality information into specific anomaly type identifiers, thereby solving the problem of how to accurately, efficiently, and objectively infer specific anomaly types based on this information. Compared to manual judgment, this scheme avoids subjectivity and reliance on experience through standardized feature extraction and quantified similarity calculation, improving the accuracy and efficiency of inference and reducing dependence on human experience.

[0011] The aforementioned online quality analysis method for alkaline washing of silicon steel strip includes, in which the quality information includes information on oil residue and information on alkali residue.

[0012] This clear definition of quality information enables the online analysis method for alkaline washing quality of silicon steel strip in this application to more comprehensively and accurately assess the quality of alkaline washing.

[0013] The online quality analysis method for alkaline washing of silicon steel strip is described above, wherein the second preset quality threshold is less than the first preset quality threshold.

[0014] The online analysis method for alkaline washing quality of silicon steel strip is described above, wherein the abnormality type is one or more of the following: insufficient alkaline washing solution concentration, wear of alkaline washing brush roller, and wear of clean water brush roller.

[0015] The online quality analysis method for alkaline washing of silicon steel strip includes generating alkali replenishment information based on the quality information of all abnormal locations when the abnormality type only includes insufficient concentration of alkaline washing solution; and generating warning information when the abnormality type includes wear of alkaline washing brush roller or wear of clean water brush roller.

[0016] In the aforementioned online quality analysis method for alkaline washing of silicon steel strip, the distribution density of the detection head in the quality detection component is less than the distribution density on both sides.

[0017] Secondly, this application also provides an online quality analysis device for alkaline washing of silicon steel strip, used for quality monitoring of alkaline washing equipment for silicon steel strip. The output end of the alkaline washing equipment for silicon steel strip is equipped with a quality detection component, which includes multiple linearly arranged detection heads. The detection heads are used to collect quality information at corresponding positions. The device includes: The first acquisition module is used to acquire linear quality distribution information in real time through the detection head. The linear quality distribution information includes the quality information corresponding to each position point that is linearly distributed perpendicular to the length direction of the silicon steel strip. An anomaly detection module is used to analyze whether there are any anomalies in the linear quality distribution information based on a first preset quality threshold. The second acquisition module is used to acquire regional alkaline washing quality distribution information by detecting the detection head when an anomaly is present. The regional alkaline washing quality distribution information includes the quality information of each location point in a preset rectangular window area. Anomaly location module is used to obtain all abnormal location points of the alkaline washing quality distribution information in the region based on a second preset quality threshold. The anomaly inference module is used to infer the anomaly type based on the distribution of all anomaly locations and the corresponding quality information.

[0018] The online quality analysis device for alkaline washing of silicon steel strip in this application realizes online detection of alkaline washing quality by acquiring linear quality distribution information through linear scanning of the detection head. When an alkaline washing quality abnormality is initially determined, the device performs regional scanning based on the detection head to acquire regional alkaline washing quality distribution information. The distribution of abnormal locations in the regional alkaline washing quality distribution information and the corresponding quality information are used to infer the type of abnormality causing the alkaline washing quality problem. This realizes automated detection and analysis of alkaline washing quality, improves the control level of the production process, and overcomes the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

[0019] Thirdly, this application also provides an electronic device, including a processor and a memory, the memory storing computer-readable instructions that, when executed by the processor, perform the steps of the method provided in the first aspect above.

[0020] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the steps of the method provided in the first aspect above.

[0021] As can be seen from the above, this application provides an online analysis method, apparatus, equipment, and storage medium for the alkaline washing quality of silicon steel strip. The online analysis method for alkaline washing quality of silicon steel strip in this application achieves online detection of alkaline washing quality by acquiring linear quality distribution information through linear scanning of the detection head. When an alkaline washing quality anomaly is initially determined, the detection head performs regional scanning to acquire regional alkaline washing quality distribution information. The distribution of abnormal locations in the regional alkaline washing quality distribution information and the corresponding quality information are used to infer the type of anomaly causing the alkaline washing quality problem. This achieves automated detection and analysis of alkaline washing quality, improves the control level of the production process, and overcomes the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience. Attached Figure Description

[0022] Figure 1 A flowchart of the online analysis method for alkaline washing quality of silicon steel strip provided in the embodiments of this application.

[0023] Figure 2 This is a schematic diagram of the output end of a silicon steel strip alkaline washing equipment.

[0024] Figure 3 This is a schematic diagram of the online quality analysis device for alkaline washing of silicon steel strip provided in an embodiment of this application.

[0025] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0026] Reference numerals: 101, Alkali washing equipment for silicon steel strip; 102, Quality detection component; 201, First acquisition module; 202, Anomaly detection module; 203, Second acquisition module; 204, Anomaly location module; 205, Anomaly inference module; 301, Processor; 302, Memory; 303, Communication bus. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0028] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0029] Firstly, please refer to Figure 1 and Figure 2 This application provides an online quality analysis method for alkaline washing of silicon steel strip, used for quality monitoring in an alkaline washing equipment 101. The output end of the alkaline washing equipment 101 is equipped with a quality detection component 102, which includes multiple linearly arranged detection heads. These detection heads are used to collect quality information at corresponding locations. The method includes the following steps: S1. The linear quality distribution information is obtained in real time through the detection head. The linear quality distribution information includes the quality information corresponding to each position point that is linearly distributed perpendicular to the length direction of the silicon steel strip. S2. Analyze whether there are any anomalies in the linear quality distribution information based on the first preset quality threshold; S3. When an anomaly is detected, the regional alkaline washing quality distribution information is obtained through the detection head. The regional alkaline washing quality distribution information includes the quality information of each location point in the preset rectangular window area. S4. Obtain all abnormal location points of the regional alkaline washing quality distribution information based on the second preset quality threshold; S5. Infer the anomaly type based on the distribution of all abnormal locations and the corresponding quality information.

[0030] Specifically, the quality detection component 102 is set at the output end of the silicon steel strip alkaline washing equipment 101. The component includes multiple detection heads, which are arranged along a straight line parallel to the width direction of the silicon steel strip, for parallel collection of quality information of multiple position points on the surface width direction of the silicon steel strip.

[0031] More specifically, in the actual production process, steps S1 and S2 are processes of online acquisition and analysis of linear quality distribution information based on a pre-set sampling interval. They are used to preliminarily determine whether there is an abnormality in the previous alkaline washing process to determine whether step S3 needs to be triggered to determine the abnormality. Among them, step S1 realizes a rapid scan of the quality in the width direction of the silicon steel strip. The linear quality distribution information it acquires provides an overview of the current surface quality of the silicon steel strip, which can be used for a rapid preliminary assessment of the alkaline washing effect.

[0032] More specifically, step S2 provides a preliminary anomaly screening mechanism, which determines whether there is a quality anomaly in the current scan line by comparing the magnitude of the first preset quality threshold with the quality information of each location point in the linear quality distribution information.

[0033] More specifically, step S3 is triggered only when step S2 determines that an anomaly exists. It utilizes the detection head to acquire mass distribution information within a defined rectangular area, which includes the mass information of each location point within that rectangular area. In practice, step S3 can construct mass distribution information by continuously triggering the detection head to collect multiple linear mass distribution information during the silicon steel strip conveying process, thereby achieving scanning of a two-dimensional area when an anomaly exists.

[0034] More specifically, in step S4, the second preset quality threshold may be the same as or different from the first preset quality threshold, both of which are set according to the quality detection requirements. The objective of step S4 is to identify and obtain all location points within the rectangular window area that do not meet the requirements of the second preset quality threshold by comparing the second preset quality threshold with the quality information, thereby determining the abnormal location points and their corresponding quality information.

[0035] More specifically, step S5 is used to analyze and infer the specific cause type of the quality abnormality based on the spatial distribution characteristics of all abnormal location points identified in step S4 and the quality information corresponding to these location points.

[0036] More specifically, the online quality analysis method for alkaline washing of silicon steel strip in this application embodiment achieves automated and online monitoring of the alkaline washing quality of silicon steel strip by setting a quality detection component 102 at the output end of the alkaline washing equipment. Step S1 uses real-time acquisition of linear quality distribution information along the width direction of the silicon steel strip, and then compares the first preset quality threshold with the linear quality distribution information to quickly and preliminarily assess the alkaline washing effect, rapidly identifying whether the current alkaline washing quality is abnormal. When an abnormality occurs, a more detailed scan is triggered to obtain more detailed two-dimensional quality data than the linear scan, providing richer information for subsequent anomaly analysis. Subsequently, step S4 uses a second preset quality threshold to compare the quality information of all points within the rectangular window area to identify and filter all specific points with unqualified quality as abnormal points. Finally, step S5 analyzes the spatial arrangement of these abnormal points and their corresponding quality information to infer the specific cause type of the alkaline washing quality problem. This inference process transforms the detection results into operable diagnostic information, helping to take timely corrective measures, improving the control level of the production process, and overcoming the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

[0037] The online analysis method for alkaline washing quality of silicon steel strip in this application realizes online detection of alkaline washing quality by obtaining linear quality distribution information through linear scanning of the detection head. When an alkaline washing quality abnormality is initially determined, the detection head performs regional scanning to obtain regional alkaline washing quality distribution information. The distribution of abnormal locations in the regional alkaline washing quality distribution information and the corresponding quality information are used to infer the type of abnormality causing the alkaline washing quality problem. This realizes automated detection and analysis of alkaline washing quality, improves the control level of the production process, and overcomes the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

[0038] In some preferred embodiments, step S5 includes: S51. Extract the number, location coordinates, shape and orientation of all abnormal location points to obtain distribution parameters. Based on the distribution parameters and quality information, construct the feature vector to be identified. S52. Calculate the similarity between the feature vector to be identified and the feature vector corresponding to each preset anomaly type in the pre-built anomaly type knowledge base, and obtain multiple similarity values. The anomaly type knowledge base includes multiple preset anomaly types, and each preset anomaly type corresponds to at least one feature vector. S53. The preset anomaly type with the highest similarity among the similarity values ​​is used as the inferred anomaly type.

[0039] Specifically, step S51 involves extracting the quantity, coordinates, shape, and orientation of all abnormal location points to obtain distribution parameters. Based on these distribution parameters and quality information, a feature vector to be identified is constructed. This can be achieved using image processing techniques or spatial analysis algorithms. For example, connected component analysis can be used to obtain the shape and size of the abnormal region, centroid calculation can be used to obtain the location coordinates, principal component analysis can be used to obtain the orientation information, and the number of pixels can be obtained by counting the number of pixels. Distribution parameters may include the area, perimeter, aspect ratio, orientation angle, centroid coordinates, and dispersion of the abnormal region. These parameters and quality information are integrated into a multi-dimensional vector, which serves as the feature vector to be identified. This step transforms the raw, discrete abnormal location point data into a structured, computable feature representation, laying the foundation for subsequent automated identification.

[0040] More specifically, step S52 refers to calculating the similarity between the feature vector to be identified and the feature vector corresponding to each preset anomaly type in the pre-built anomaly type knowledge base, obtaining multiple similarity values. Various vector similarity calculation methods can be used, such as Euclidean distance, cosine similarity, and Manhattan distance, to calculate the similarity or distance between two feature vectors. The anomaly type knowledge base pre-stores typical feature vectors for different types of alkali washing anomalies (e.g., strip anomalies, block anomalies, dot anomalies, edge anomalies, etc.). These typical feature vectors can be constructed through analysis, clustering, or expert annotation of historical anomaly data. This step, through quantitative comparison, measures the degree of matching between the currently detected anomaly features and the typical features of known anomaly types.

[0041] More specifically, step S53 refers to selecting the preset anomaly type with the highest similarity among the similarity values ​​as the inferred anomaly type. This can be achieved using a comparison algorithm. For example, after calculating the similarity values ​​between the feature vector to be identified and the feature vectors of all preset anomaly types in the knowledge base, the preset anomaly type with the highest similarity value is selected as the final inference result.

[0042] More specifically, steps S51, S52, and S53 together constitute a feature-matching-based anomaly type inference process. This series of steps works collaboratively to transform the original anomaly location distribution and quality information into specific anomaly type identifiers, thereby solving the problem of how to accurately, efficiently, and objectively infer specific anomaly types based on this information. Compared to manual judgment, this scheme avoids subjectivity and experience dependence through standardized feature extraction and quantified similarity calculation, improving the accuracy and efficiency of inference and reducing reliance on human experience.

[0043] In some preferred embodiments, the quality information includes information on oil residue and information on alkali residue.

[0044] Specifically, quality information refers to data used to evaluate the alkaline washing effect of silicon steel strip. As mentioned above, the alkaline washing process of the silicon steel strip alkaline washing equipment 101 actually includes alkaline washing liquid rinsing, hot water rinsing, and drying. Therefore, abnormal alkaline washing quality can manifest as oil residue or alkaline residue. If the collected quality information only includes information on one type of residue, it cannot fully reflect the alkaline washing effect. For example, only detecting the amount of oil residue may not detect the problem of excessive alkaline residue; conversely, the same applies. Therefore, the quality information obtained by the online analysis method for alkaline washing quality of silicon steel strip in this embodiment includes information on the amount of oil residue and the amount of alkaline residue, so as to simultaneously monitor the two main types of residue on the surface of the silicon steel strip. This also facilitates step S5 in inferring the type of abnormality by referring to more types of data. That is, when step S5 infers the type of abnormality based on the distribution of all abnormal locations and the corresponding information on the amount of oil residue and the amount of alkaline residue, it can more accurately distinguish different types of alkaline washing problems. This clear definition of the quality information content enables the online analysis method for alkaline washing quality of silicon steel strip in this embodiment to more comprehensively and accurately evaluate the alkaline washing quality.

[0045] More specifically, the process of analyzing anomalies based on the first preset quality threshold in step S2 and screening anomaly locations based on the second preset quality threshold in step S4 requires simultaneous consideration of oil residue threshold and alkali residue threshold.

[0046] In some preferred embodiments, the second preset quality threshold is less than the first preset quality threshold.

[0047] Specifically, the first preset quality threshold is used to quickly determine whether there is a preliminary anomaly during the linear scanning phase. Once a preliminary anomaly is detected, the system performs a more detailed area scan. The second preset quality threshold is used to identify abnormal location points during the area scanning phase. Since quality information includes oil residue and alkali residue information, a higher quality value indicates a worse alkaline washing quality. The method in this application, by setting the second preset quality threshold to be smaller than the first preset quality threshold, can identify more location points whose quality deviates from the normal range within the area scan, even if the deviation is insufficient to trigger an alarm during the linear scanning phase. These additionally identified abnormal location points, although their deviation may be small, are crucial for subsequent anomaly type inference due to their distribution, quantity, shape, and corresponding quality information within the area. Obtaining more data on abnormal location points provides richer sample information, making the analysis based on the distribution and quality information of abnormal location points more accurate and reliable, thereby improving the accuracy of anomaly type inference.

[0048] More specifically, the second preset quality threshold is preferably set to 80-90% of the first preset quality threshold.

[0049] In some preferred embodiments, the abnormality type is one or more of the following: insufficient concentration of alkaline washing solution, wear of alkaline washing brush roller, and wear of clean water brush roller.

[0050] Specifically, insufficient concentration of alkaline washing solution will reduce its oil removal capacity, resulting in blocky, sheet-like, or large-area oil residue on the surface of the silicon steel strip output from the output end of the silicon steel strip alkaline washing equipment 101. This will increase the amount of oil residue in the quality information of adjacent locations in the regional alkaline washing quality distribution information. Wear of alkaline washing brush rollers and clean water brush rollers is generally manifested as circumferential brush bristle wear, which will increase the amount of oil residue or alkali residue in multiple locations arranged along the length of the silicon steel strip in the regional alkaline washing quality distribution information.

[0051] More specifically, these anomaly types are the most common causes of gap quality abnormalities in the silicon steel strip alkaline washing equipment 101, and they produce alkaline washing quality distribution information with relatively significant regional differences. By limiting the specific scope of the anomaly types, the inference results have clear process significance and can directly guide subsequent fault diagnosis and handling. This allows step S5 to infer the anomaly type based on the distribution of these anomaly locations and the corresponding quality information. For example, if the anomaly mainly manifests as a general high level of alkaline residue in blocky areas of the silicon steel strip, step S5 may infer that the anomaly type is insufficient alkaline washing solution concentration. If the anomaly manifests as strip-shaped oil residue along the length of the silicon steel strip and is located at a specific transverse position of the silicon steel strip, the system may infer that the anomaly type is alkaline washing brush roller wear. If the anomaly manifests as strip-shaped alkaline residue along the length of the silicon steel strip, the system may infer that the anomaly type is clean water brush roller wear. This clear anomaly type inference provides a basis for subsequent targeted measures, improving the efficiency and accuracy of fault diagnosis.

[0052] It should be noted that the abnormality type is one or more of the following: insufficient concentration of alkaline washing solution, wear of alkaline washing brush roller, and wear of clean water brush roller. That is, the abnormality type preset in the abnormality type knowledge base in step S52 can be one or more of the following: insufficient concentration of alkaline washing solution, wear of alkaline washing brush roller, and wear of clean water brush roller. For example, the abnormality type corresponding to a certain feature vector is a composite abnormality of insufficient concentration of alkaline washing solution and wear of alkaline washing brush roller.

[0053] In some preferred embodiments, when the only abnormality type is insufficient alkaline washing solution concentration, alkaline replenishment information is generated based on the quality information of all abnormal locations; when the abnormality type includes wear of alkaline washing brush roller or wear of clean water brush roller, a warning message is generated.

[0054] Specifically, in actual production, if the anomaly type only involves insufficient alkaline washing solution concentration, operators can replenish alkali online to adjust the concentration and ensure production continuity. However, if the anomaly type includes wear on the alkaline washing brush roller or the clean water brush roller, a shutdown for replacement is required. Therefore, the above handling method provides two different response mechanisms based on the inferred anomaly type. When the inferred anomaly type only involves insufficient alkaline washing solution concentration, it indicates that no shutdown maintenance is currently required. Therefore, the solution uses the quality information from all anomaly locations to calculate the amount of alkali to be replenished and generates alkali replenishment information. The quality information reflects the alkaline washing effect, while the quality information from the anomaly locations indicates the degree of alkaline insufficiency. By processing this information, the degree of alkali deficiency can be quantified, thereby generating alkali replenishment information. This information can be directly used to control the automated system to add alkali to the alkaline solution tank or to prompt operators to add alkali, thereby quickly adjusting the alkaline concentration and restoring normal alkaline washing effects. When the inferred anomaly type includes wear on the alkaline washing brush roller or the clean water brush roller, it indicates that a shutdown maintenance is required to resolve the alkaline washing quality issue. Therefore, the solution generates a warning message in this case. This message directly alerts the operator, prompting them to stop the machine and replace the corresponding brush roller. This response mechanism prevents the continued production of defective products despite brush roller wear, addressing the problem at its root by promptly replacing parts.

[0055] More specifically, the online analysis method for alkaline washing quality of silicon steel strip in this application distinguishes different types of anomalies and takes corresponding automated or prompting measures. This solution can address common problems in the alkaline washing process, improve the efficiency and accuracy of fault handling, reduce the need for manual intervention, and ensure the alkaline washing quality of silicon steel strip.

[0056] In some preferred embodiments, the step of generating alkali replenishment information based on the quality information of all abnormal location points includes: A1. Based on the quality information of each abnormal location, estimate the amount of alkali missing at each abnormal location. The amount of alkali missing is the amount of alkali required to make the location reach the preset alkali washing quality standard. A2. Add up the alkali loss corresponding to all abnormal locations to obtain the total alkali loss. A3. Calculate the alkali replenishment information based on the total alkali loss and alkali replenishment coefficient.

[0057] Specifically, in step A1, the quality information collected at each abnormal location, such as the amount of oil or alkali residue, is compared with a preset alkaline washing quality standard. The degree of deviation between the quality information and the standard is used to estimate the amount of alkali solution needed to be added at that location to achieve the cleaning standard. This estimation process can be based on a pre-established relationship model or lookup table between the quality information and the required amount of alkali solution.

[0058] More specifically, step A2 sums up the estimated amount of alkali loss at all locations identified as abnormal within the preset rectangular window area to obtain a total amount representing the overall degree of inadequacy in the cleaning of the area.

[0059] More specifically, in step A3, the final mass of alkali to be replenished to the alkali solution tank is calculated. During the calculation, the total amount of alkali loss is multiplied by an alkali replenishment coefficient. This coefficient can be dynamically adjusted based on historical alkali washing data and real-time alkali washing parameters to compensate for losses or errors during the alkali washing process. The value of this coefficient is adjusted in real-time based on historical production data and current process parameters, such as silicon steel strip speed, alkali temperature, current alkali concentration, and alkali volume in the tank. The purpose of this adjustment coefficient is to more accurately reflect potential alkali losses, reaction consumption, or measurement errors in actual production, ensuring that the calculated replenishment amount effectively improves the quality of the alkali washing process.

[0060] In some preferred embodiments, the detection head acquires quality information based on infrared spectroscopy, laser scattering, laser Raman spectroscopy, or laser-induced breakdown spectroscopy.

[0061] Specifically, the methods described above are non-contact optical detection techniques that can analyze the material composition or distribution on the surface of silicon steel strips.

[0062] More specifically, infrared spectroscopy identifies and quantifies the chemical bond information of organic and inorganic substances. Laser scattering detects surface particles or roughness. Laser Raman spectroscopy provides information on the molecular structure of substances. Laser-induced breakdown spectroscopy (LIBS) generates plasma through laser ablation and analyzes its emission spectrum for elemental analysis, detecting residual alkali metals or hydrocarbons in oil on the surface of silicon steel strips, quantifying the amount of alkali residue and oil residue. In this embodiment, the detection head preferably acquires quality information based on the laser-induced breakdown spectroscopy method, ensuring the accuracy and timeliness of quality information acquisition. In summary, the detection head preferably uses a pulsed laser, combined with a corresponding spectrometer to acquire spectral line intensity data, and converts the spectral line intensity data into specific values ​​for oil residue and alkali residue through a pre-established calibration curve. These values ​​are output as quality information for subsequent online analysis steps.

[0063] In some preferred embodiments, the number of detection points corresponding to the preset rectangular window area is a preset number, the two sides of which are adjusted in real time based on the edge position of the silicon steel strip, and the length is adjusted and set based on the distance between the two sides and the preset number.

[0064] More specifically, the two sides of the preset rectangular window area are adjusted in real time based on the edge position of the silicon steel strip. This means that the side position of the window will dynamically adjust as the actual edge position of the silicon steel strip changes. This ensures that the window area always covers the actual width range of the silicon steel strip, including the edge area. The length of the preset rectangular window area is adjusted based on the distance between the two sides and a preset number, determining the size of the window area along the width direction of the silicon steel strip, and this size is then correlated with the number of data points to be collected.

[0065] More specifically, the online analysis method for alkaline washing quality of silicon steel strip in this application adjusts the side position of the window in real time to ensure that the window always covers the effective range of the silicon steel strip, avoiding the collection of invalid data outside the silicon steel strip or the omission of key data at the edge of the silicon steel strip. This improves the targeting and effectiveness of regional quality distribution information collection, supports the subsequent acquisition and analysis of abnormal location points, and provides a reliable data foundation for subsequently acquiring all abnormal location points based on a second preset quality threshold.

[0066] In some preferred embodiments, the distribution density of the detection head in the middle of the mass detection assembly 102 is less than the distribution density on both sides.

[0067] Specifically, the distribution density of the detection heads along the width of the silicon steel strip affects the precision of data acquisition. In actual production, since both alkaline washing solution and hot water are sprayed onto the surface of the silicon steel strip through nozzles for cleaning, both the alkaline washing solution and clean water flow to both sides of the silicon steel strip's width. Simultaneously, burrs are more likely to appear at the edges of the silicon steel strip, and the locations of alkaline washing quality abnormalities are generally concentrated at the two edges. Therefore, the online quality analysis method for alkaline washing of silicon steel strip in this application sets the distribution density of the detection heads in the middle to be less than that on both sides. This allows the system to collect quality information from the edge areas of the silicon steel strip more densely, obtaining more detailed quality data for the edges and improving the detection capability and positioning accuracy of quality problems in the edge areas. At the same time, reducing the detection density in the middle area optimizes the amount of data collected while ensuring coverage of the main areas, reducing the data processing burden. This distribution method adjusts the allocation of data acquisition resources according to the different probabilities of quality problems occurring in different areas of the silicon steel strip, improving the effectiveness of the entire online quality analysis method.

[0068] Secondly, please refer to Figure 3Some embodiments of this application also provide an online quality analysis device for alkaline washing of silicon steel strip, used for quality monitoring of alkaline washing equipment 101. The output end of alkaline washing equipment 101 is provided with a quality detection component 102, which includes multiple linearly arranged detection heads. The detection heads are used to collect quality information at corresponding locations. The device includes: The first acquisition module 201 is used to acquire linear quality distribution information in real time through the detection head. The linear quality distribution information includes the quality information corresponding to each position point that is linearly distributed perpendicular to the length direction of the silicon steel strip. Anomaly detection module 202 is used to analyze whether there are anomalies in the linear quality distribution information based on a first preset quality threshold; The second acquisition module 203 is used to acquire regional alkaline washing quality distribution information by detecting the detection head when an anomaly is present. The regional alkaline washing quality distribution information includes the quality information of each location point in a preset rectangular window area. Anomaly location module 204 is used to obtain all abnormal location points of regional alkaline washing quality distribution information based on a second preset quality threshold. The anomaly inference module 205 is used to infer the anomaly type based on the distribution of all anomaly locations and the corresponding quality information.

[0069] The online quality analysis device for alkaline washing of silicon steel strip in this application embodiment realizes online detection of alkaline washing quality by obtaining linear quality distribution information through linear scanning of the detection head. When an alkaline washing quality abnormality is initially determined, the device performs regional scanning based on the detection head to obtain regional alkaline washing quality distribution information. The distribution of abnormal locations in the regional alkaline washing quality distribution information and the corresponding quality information are used to infer the type of abnormality causing the alkaline washing quality problem. This realizes automated detection and analysis of alkaline washing quality, improves the control level of the production process, and overcomes the limitations of traditional manual analysis, which is characterized by low efficiency and reliance on experience.

[0070] In some preferred embodiments, the online quality analysis device for alkaline washing of silicon steel strip in this application is used to perform the online quality analysis method for alkaline washing of silicon steel strip provided in the first aspect above.

[0071] Thirdly, please refer to Figure 4 Some embodiments of this application also provide a schematic diagram of the structure of an electronic device. This application provides an electronic device including: a processor 301 and a memory 302. The processor 301 and the memory 302 are interconnected and communicate with each other through a communication bus 303 and / or other forms of connection mechanism (not shown). The memory 302 stores computer-readable instructions that can be executed by the processor 301. When the electronic device is running, the processor 301 executes the computer-readable instructions to execute the method in any optional implementation of the above embodiments.

[0072] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program performs the method in any optional implementation of the above embodiments. The computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0073] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0074] Furthermore, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0075] Furthermore, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0076] In this document, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, without necessarily requiring or implying any such actual relationship or order between these entities or operations.

[0077] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. An online quality analysis method for alkaline washing of silicon steel strip, used for quality monitoring in alkaline washing equipment for silicon steel strip, characterized in that, The output end of the silicon steel strip alkaline washing equipment is equipped with a quality detection component, which includes multiple linearly arranged detection heads. Each detection head is used to collect quality information at a corresponding location. The method includes the following steps: S1. Obtain linear quality distribution information in real time through the detection head. The linear quality distribution information includes the quality information corresponding to each position point that is linearly distributed perpendicular to the length direction of the silicon steel strip. S2. Analyze whether there are any anomalies in the linear quality distribution information based on the first preset quality threshold; S3. When an anomaly is present, the detection head is used to obtain the regional alkaline washing quality distribution information, which includes the quality information of each location point in the preset rectangular window area. S4. Based on the second preset quality threshold, obtain all abnormal location points of the alkaline washing quality distribution information in the region; S5. Infer the anomaly type based on the distribution of all abnormal locations and the corresponding quality information; Step S5 includes: S51. Extract the number, location coordinates, shape and orientation of all abnormal location points to obtain distribution parameters, and construct the feature vector to be identified based on the distribution parameters and the quality information; S52. Calculate the similarity between the feature vector to be identified and the feature vector corresponding to each preset anomaly type in the pre-built anomaly type knowledge base, and obtain multiple similarity values. The anomaly type knowledge base includes multiple preset anomaly types, and each preset anomaly type corresponds to at least one feature vector. S53. The preset anomaly type with the highest similarity among the similarity values ​​is used as the inferred anomaly type; The quality information includes information on oil residue and alkali residue; The abnormality type is one or more of the following: insufficient alkaline washing solution concentration, wear of alkaline washing brush roller, and wear of clean water brush roller. When the abnormality type only includes insufficient alkaline washing solution concentration, alkaline replenishment information is generated based on the quality information of all abnormal location points. When the abnormality type includes wear of alkaline washing brush roller or wear of clean water brush roller, a warning message is generated.

2. The online quality analysis method for alkaline washing of silicon steel strip according to claim 1, characterized in that, The second preset quality threshold is less than the first preset quality threshold.

3. The online quality analysis method for alkaline washing of silicon steel strip according to claim 1, characterized in that, The distribution density of the detection head in the middle of the mass detection component is lower than that on both sides.

4. An online quality analysis device for alkaline washing of silicon steel strip, used for quality monitoring of alkaline washing equipment for silicon steel strip, characterized in that, The device is used to perform the online quality analysis method for alkaline washing of silicon steel strip as described in any one of claims 1-3. The output end of the alkaline washing equipment for silicon steel strip is equipped with a quality detection component, which includes multiple linearly arranged detection heads. Each detection head is used to collect quality information at a corresponding location. The device includes: The first acquisition module is used to acquire linear quality distribution information in real time through the detection head. The linear quality distribution information includes the quality information corresponding to each position point that is linearly distributed perpendicular to the length direction of the silicon steel strip. An anomaly detection module is used to analyze whether there are any anomalies in the linear quality distribution information based on a first preset quality threshold. The second acquisition module is used to acquire regional alkaline washing quality distribution information by detecting the detection head when an anomaly is present. The regional alkaline washing quality distribution information includes the quality information of each location point in a preset rectangular window area. Anomaly location module is used to obtain all abnormal location points of the alkaline washing quality distribution information in the region based on a second preset quality threshold. The anomaly inference module is used to infer the anomaly type based on the distribution of all anomaly locations and the corresponding quality information.

5. An electronic device, characterized in that, It includes a processor and a memory, the memory storing computer-readable instructions, which, when executed by the processor, perform the steps of the method as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it performs the steps of the method as described in any one of claims 1-3.

Images