Modularized electrolytic steel plate production line control method

Abstract

The present invention relates to the technical field of steel plate production line control, and in particular to a modular electrolytic steel plate production line control method. The method comprises the following steps: obtaining steel plate basic data and production line initial data, and performing initial equipment configuration according to the steel plate basic data and production line initial data to obtain equipment initial configuration data; obtaining steel plate image data, and performing steel plate surface cleanliness detection on the steel plate image data to obtain steel plate surface cleanliness data; performing electrolysis process control operations according to the equipment initial configuration data and steel plate surface cleanliness data to obtain electrolysis result data; performing electrolyte quantity detection according to the electrolysis result data to obtain electrolyte quantity detection data. The present invention analyzes the cleanliness, thickness and uniformity of the steel plate surface in real time, automatically adjusts the electrolysis process parameters, and ensures the high efficiency and quality stability of the electrolysis process.

Description

Technical Field

[0001] The present invention relates to the technical field of steel plate production line control, and in particular to a modular electrolytic steel plate production line control method. Background Art

[0002] With the increasing demand for high-performance materials in modern manufacturing, steel plate surface treatment technology plays a vital role in industrial production. The surface quality of steel plates not only directly affects the appearance of the product, but also is related to its important characteristics such as corrosion resistance and mechanical strength. As a commonly used surface treatment technology, electrolytic process can effectively remove oxides and pollutants on the surface of steel plates and improve the surface finish of steel plates. However, the traditional electrolytic process has many challenges in terms of control accuracy, process parameter adjustment and quality inspection.

[0003] The current electrolytic treatment process of steel plates mostly relies on manual experience to set parameters, lacking accurate cleanliness detection methods and effective process parameter control mechanisms, resulting in problems such as uneven surface treatment and inconsistent thickness during the electrolysis process, affecting product quality. In addition, traditional surface cleanliness detection methods mostly rely on a single visual detection method, which makes it difficult to evaluate the distribution of pollutants on the steel plate surface, the thickness of the oxide layer, and the uniformity after treatment. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a modular electrolytic steel plate production line control method to solve at least one of the above technical problems.

[0005] The present application provides a modular electrolytic steel plate production line control method, comprising the following steps:

[0006] Step S1: acquiring steel plate basic data and production line initial data, and performing initial equipment configuration according to the steel plate basic data and production line initial data to obtain equipment initial configuration data;

[0007] Step S2: acquiring steel plate image data, and performing steel plate surface cleanliness detection on the steel plate image data to obtain steel plate surface cleanliness data;

[0008] Step S3: performing electrolysis process control operations according to the equipment initial configuration data and the steel plate surface cleanliness data to obtain electrolysis result data;

[0009] Step S4: Perform electrolyte quantity detection according to the electrolysis result data to obtain electrolyte quantity detection data.

[0010] In the present invention, by obtaining the basic data of the steel plate and the initial data of the production line, it is possible to realize the automated and precise configuration of the equipment. Through automated cleanliness detection, the pollution problem on the surface of the steel plate can be timely discovered and solved before the electrolytic treatment, ensuring that the surface of the steel plate is sufficiently clean before entering the electrolytic cell, and reducing the production of unqualified products. According to the cleanliness detection data and the initial configuration data of the equipment, the production line can dynamically adjust the electrolytic process parameters (such as current, voltage, etc.) to realize electrolytic treatment for different steel plates. Through automated electrolyte quantity detection, the surface quality of the steel plate after electrolysis can be quickly and accurately evaluated, reducing the errors caused by manual inspection and improving the detection efficiency. The test results can be fed back to the front-end process to optimize the production process, form a closed-loop feedback system, continuously improve the production process, and ensure that the final product quality meets the standards.

[0011] Preferably, step S1 specifically includes:

[0012] Obtain basic data of steel plates and initial data of production lines;

[0013] Generate equipment configuration data according to the basic data of the steel plate to obtain equipment configuration data;

[0014] Generate production line electrolysis parameters according to equipment configuration data and production line initial data to obtain production line electrolysis parameter data;

[0015] The equipment settings are initialized according to the electrolysis parameter data of the production line and the initial data of the production line to obtain the initial configuration data of the equipment.

[0016] In the present invention, by obtaining the basic data of the steel plate (such as size, thickness, material, surface state, etc.) and the initial data of the production line (such as equipment status, electrolyte state, ambient temperature, etc.), it is ensured that each batch of steel plates can be processed according to its unique characteristics. According to the basic data of the steel plate, the system automatically generates equipment configuration data to ensure that the equipment parameters (such as current, voltage, flow rate, etc.) can accurately match the characteristics of the steel plate. By combining the equipment configuration data and the initial data of the production line, the generated production line electrolysis parameters can accurately control the key parameters involved in the electrolysis process, such as electrolyte concentration, temperature, voltage and current, etc., to ensure that the electrolysis process can be carried out in the best state. Through the automated equipment setting initialization, the equipment can self-adjust and enter the working state according to the generated electrolysis parameters and the initial data of the production line in a short time. Compared with traditional manual initialization, automated configuration greatly shortens the production preparation time and improves the working efficiency of the production line.

[0017] Preferably, the steel plate image data includes first steel plate image data, second steel plate image data and third steel plate image data, and step S2 is specifically:

[0018] Capturing the steel plate image by a visible light acquisition device to obtain first steel plate image data;

[0019] The steel plate image is collected by ultraviolet light collection equipment to obtain second steel plate image data;

[0020] The steel plate image is collected by infrared light collection equipment to obtain the third steel plate image data;

[0021] Performing a first steel plate surface cleanliness detection on the first steel plate image data to obtain first steel plate surface cleanliness data;

[0022] Performing a second steel plate surface cleanliness detection on the second steel plate image data to obtain second steel plate surface cleanliness data;

[0023] Performing a third steel plate surface cleanliness detection on the third steel plate image data to obtain third steel plate surface cleanliness data, wherein the first steel plate surface cleanliness detection, the second steel plate surface cleanliness detection and the third steel plate surface cleanliness detection are different steel plate surface cleanliness detection methods;

[0024] A multi-dimensional cleanliness test is performed based on the first steel plate surface cleanliness data, the second steel plate surface cleanliness data and the third steel plate surface cleanliness data to obtain the steel plate surface cleanliness data.

[0025] In the present invention, by collecting steel plate image data under different spectra through visible light, ultraviolet light and infrared light equipment, different features of the steel plate surface can be captured. Visible light can detect physical defects and surface stains, ultraviolet light can identify organic pollutants and grease residues, and infrared light can detect thermal anomalies, such as moisture or oxide layers. The data of the three spectra are complementary, and various defects on the surface of the steel plate can be identified more accurately. Compared with the detection method using only a single spectrum, multi-spectral detection can reduce omissions and significantly improve the comprehensiveness and accuracy of detection. By using multiple detection methods in parallel, the missed detections and false detections caused by a single detection method can be effectively reduced. For example, ultraviolet light can make up for the shortcomings of visible light in detecting organic residues, and infrared light can detect temperature differences and find tiny defects that are difficult to detect with other spectra.

[0026] Preferably, the first steel plate surface cleanliness detection is specifically as follows:

[0027] Performing contrast enhancement on the first steel plate image data to obtain steel plate visible light image enhancement data;

[0028] Performing steel plate region segmentation on the steel plate visible light image enhancement data to obtain steel plate visible light image segmentation data;

[0029] Surface texture feature extraction, surface color feature extraction and local grayscale feature extraction are performed on the visible light image segmentation data of the steel plate to obtain surface color feature data of the steel plate image, surface color feature data of the steel plate image and grayscale feature data of the steel plate respectively;

[0030] Extract irregular surface features from the color feature data of the steel plate image surface and the grayscale feature data of the steel plate to obtain irregular surface feature data of the steel plate;

[0031] Defect contour detection is performed based on the irregular feature data of the steel plate surface and the color feature data of the steel plate image surface to obtain the first steel plate surface cleanliness data.

[0032] In the present invention, by contrast enhancement, the texture, color and grayscale differences on the surface of the steel plate can be strengthened, and subtle surface defects such as stains and scratches can be highlighted. Through the segmentation algorithm, the steel plate is separated from the background, so that subsequent feature extraction and detection are only focused on the surface of the steel plate, avoiding the interference of background noise. Accurate regional segmentation ensures the accuracy of the detection algorithm. By extracting surface texture, color and grayscale features, the state of the steel plate surface can be fully described. Texture features can detect subtle scratches or corrosion, color features can identify stains or uneven coatings, and grayscale features can detect local smoothness or unevenness. Through irregular surface feature extraction, irregular shapes on the surface of the steel plate can be effectively identified, including protrusions, depressions, scratches, rust, etc. Defect contour detection based on irregular surface feature data and color feature data can accurately outline the contours of surface defects, detect surface contamination, scratches, pits and other problems, and help evaluate the scope and severity of defects through contour information. By combining irregular feature data and color feature data, a more comprehensive cleanliness assessment result can be generated, ensuring that cleanliness detection is not limited to surface stains, but also includes morphological irregularities and color anomalies.

[0033] Preferably, the local grayscale feature extraction is specifically:

[0034] Performing pixel domain relationship processing on the visible light image segmentation data of the steel plate to obtain pixel domain relationship data;

[0035] The co-occurrence matrix is ​​calculated based on the pixel domain relationship data and the steel plate visible light image segmentation data to obtain the co-occurrence matrix data;

[0036] Gradient contrast calculation is performed based on the co-occurrence matrix data to obtain local gradient contrast data;

[0037] The brightness contrast is calculated according to the co-occurrence matrix data to obtain the local brightness contrast data;

[0038] Perform feature fusion based on local gradient contrast data and local brightness contrast data to obtain preliminary local grayscale feature data;

[0039] The local structure tensor is calculated based on the pixel domain relationship data and the steel plate visible light image segmentation data to obtain the image structure tensor feature data;

[0040] Feature splicing is performed based on the preliminary local grayscale feature data and the image structure tensor feature data to obtain the grayscale feature data of the steel plate.

[0041] In the present invention, by analyzing the grayscale relationship between each pixel and its neighboring pixels, the local features on the surface of the steel plate can be captured more carefully. The processing of the pixel neighborhood relationship makes the system more sensitive to local anomalies and subtle defects (such as scratches or tiny contaminants), which can improve the detection accuracy and avoid omissions. Through the calculation of the co-occurrence matrix, the spatial relationship between the grayscale of pixels in the image can be quantified to describe the texture and structural information of the surface of the steel plate. The co-occurrence matrix can capture detailed texture information, which is particularly effective for areas with small surface texture changes or subtle surface defects. Gradient contrast can effectively detect the edges, subtle protrusions or depressions on the surface of the steel plate by capturing the direction and amplitude of grayscale changes, ensuring that the detection of surface changes is more sensitive. Brightness contrast is used to capture local brightness changes, and can effectively detect uneven brightness areas on the surface of the steel plate, such as the transition between smooth and rough areas, or brightness anomalies caused by stains and scratches. Through the calculation of the structural tensor, the local geometric structure information of the steel plate surface, especially the main direction and intensity of the grayscale change, can be captured, which is of great help to detect regular or irregular morphological changes on the surface, and can identify morphological defects such as surface protrusions and depressions. Feature stitching enhances the classification and identification capabilities of surface defects by combining multi-dimensional features, especially when processing steel plates with complex surface textures and structures, which can reduce missed detections and false detections.

[0042] Preferably, the surface cleanliness detection of the second steel plate is specifically as follows:

[0043] Performing fluorescence reaction enhancement on the second steel plate image data to obtain steel plate ultraviolet light image enhancement data;

[0044] Performing fluorescence region segmentation on the steel plate ultraviolet image enhancement data to obtain steel plate ultraviolet image segmentation data;

[0045] Extract pollutant spectral features from the steel plate ultraviolet image segmentation data to obtain pollutant spectral feature data;

[0046] Performing fluorescence intensity analysis on pollutant spectral characteristic data to obtain pollutant fluorescence intensity data;

[0047] Calculate the pollutant area based on the pollutant spectral characteristic data to obtain the pollutant area data;

[0048] Cleanliness calculation is performed based on the pollutant fluorescence intensity data and the pollutant area data to obtain the second steel plate surface cleanliness data.

[0049] The fluorescence reaction enhancement technology in the present invention can strengthen the organic pollution reaction under ultraviolet light, such as oil, fatty acids, chemical residues and other substances, making them more obvious in the ultraviolet image. Fluorescence area segmentation can automatically identify and segment the pollutant area on the surface of the steel plate, separate the pollutants from the background area, ensure that the subsequent feature extraction focuses on the truly polluted area, and reduce the interference of non-polluted areas on the analysis. Through spectral feature extraction, the composition and properties of the pollutants can be further analyzed. Different organic matter and chemical residues will produce different spectral reactions under ultraviolet light. Spectral feature extraction can help identify the type of pollutant and provide more detailed pollutant information. The fluorescence intensity is related to the concentration of the pollutant. Through fluorescence intensity analysis, the concentration or thickness of the pollutant can be quantitatively evaluated. By calculating the area of ​​the pollutant on the surface of the steel plate, the distribution of the pollutant can be understood.

[0050] Preferably, the surface cleanliness detection of the third steel plate is specifically as follows:

[0051] Performing thermal image enhancement on the image data of the third steel plate to obtain an infrared image enhanced image of the steel plate;

[0052] Perform steel plate thermal area segmentation on the steel plate infrared image enhancement image to obtain steel plate infrared image segmentation data;

[0053] Perform thermal anomaly detection on the infrared image segmentation data of the steel plate to obtain thermal anomaly data of the steel plate image;

[0054] Perform moisture distribution detection on the infrared image segmentation data of the steel plate to obtain moisture distribution data of the steel plate image;

[0055] Perform oxide layer detection according to the infrared image segmentation data of the steel plate to obtain oxide layer data of the steel plate image;

[0056] Thermal feature extraction of surface pollutants is performed based on thermal anomaly data of the steel plate image, moisture distribution data of the steel plate image, and oxide layer data of the steel plate image to obtain thermal feature data of pollutants on the steel plate surface;

[0057] The surface cleanliness is calculated based on the thermal characteristic data of the contaminants on the steel plate surface to obtain the surface cleanliness data of the third steel plate.

[0058] In the present invention, by enhancing the infrared image, the visibility of the thermal features on the surface of the steel plate, especially the tiny thermal anomalies, cold areas or hot areas, can be improved. By segmenting the hot areas of the infrared image, the thermal anomaly areas can be effectively distinguished from the normal areas. Thermal anomaly detection can capture the thermal conduction anomalies on the surface of the steel plate, such as cracks, local contamination or temperature differences caused by uneven surface coating. Infrared imaging can identify the moisture distribution on the surface and inside of the steel plate, especially trace moisture or wet areas. The oxide layer usually shows specific thermal diffusion and thermal reflection characteristics in infrared imaging, and the oxide layer detection can accurately identify the oxidized area on the surface of the steel plate. Different types of pollutants (such as oil, moisture, oxides) have different thermal characteristics in infrared images. By analyzing the thermal features, these pollutants can be distinguished and their distribution and properties can be accurately identified. Thermal feature extraction can help the system accurately capture the thermal anomaly areas of pollutants, so that the detection process is not only limited to visible pollutants, but also can detect surface problems caused by thermal conduction anomalies, such as coating defects or moisture penetration.

[0059] Preferably, the oxide layer detection is specifically:

[0060] The temperature gradient is calculated according to the infrared image segmentation data of the steel plate to obtain the infrared temperature gradient data of the steel plate;

[0061] Conduct thermal diffusion simulation on the infrared temperature gradient data of the steel plate to obtain thermal diffusion simulation data of the steel plate;

[0062] Adjust the thermal diffusion coefficient according to the steel plate thermal diffusion simulation data to obtain the steel plate thermal diffusion coefficient adjustment data;

[0063] The temperature stability feature is extracted according to the steel plate thermal diffusion coefficient adjustment data to obtain the temperature stability data;

[0064] The temperature time series feature extraction is performed according to the steel plate thermal diffusion coefficient adjustment data to obtain the temperature time series feature data;

[0065] Calculate the temperature difference according to the temperature stability data and the temperature time series characteristic data to obtain the temperature difference characteristic data;

[0066] Segment the thermal anomaly area according to the temperature difference characteristic data to obtain the oxide layer segmentation data;

[0067] Marking the oxide layer region according to the oxide layer segmentation data to obtain oxide layer region marking data;

[0068] The oxide layer thickness is estimated based on the oxide layer area annotation data according to the temperature stability data and the temperature time series characteristic data, and the oxide layer data of the steel plate image is obtained.

[0069] In the present invention, by calculating the temperature gradient of the infrared image of the steel plate, the area with temperature difference on the surface can be effectively identified, and the location of thermal anomaly or oxide layer can be preliminarily determined. The temperature gradient calculation can capture the slight changes in the temperature of the steel plate surface, especially the difference in heat conduction caused by the oxide layer, which helps to improve the sensitivity of oxide layer detection. Through thermal diffusion simulation, the heat conduction process of different areas on the surface of the steel plate can be reconstructed, so as to analyze the diffusion characteristics of heat in different materials. By adaptively adjusting the thermal diffusion coefficient, the thermal diffusion behavior of different materials can be simulated more accurately. The oxide layer usually has different thermal diffusion coefficients, and the adjusted data helps to detect the existence of the oxide layer more accurately. The temperature stability feature extraction can identify which areas on the surface of the steel plate have a small temperature change after thermal diffusion. These areas usually have an oxide layer because of its poor thermal conductivity and slow temperature change. By extracting the temperature time series features, the temperature changes of different areas in different time periods can be analyzed. Due to poor heat conduction, the oxide layer shows a slower temperature change, and the oxide layer is further identified from the time dimension. The temperature difference calculation can more accurately determine the difference between the oxide layer and the substrate by comparing the temperature changes in different areas.

[0070] Preferably, step S3 specifically includes:

[0071] When it is determined that the surface cleanliness data of the steel plate is less than the preset cleanliness threshold data, a steel plate to-be-cleaned signal is sent to the control platform;

[0072] When it is determined that the surface cleanliness data of the steel plate is greater than or equal to the preset cleanliness threshold data, process parameter mapping is performed according to the initial configuration data of the equipment to obtain process parameter mapping data of the electrolysis equipment;

[0073] The electrolysis process control operation is performed according to the electrolysis equipment process parameter mapping data to obtain the electrolysis result data.

[0074] In the present invention, the system automatically detects the surface cleanliness of the steel plate and compares it with the preset threshold value without manual intervention. Through the preset cleanliness threshold standard, the system can strictly control the surface quality of the steel plate entering the electrolysis process, ensure that each steel plate meets the cleanliness requirements, and reduce the impact of surface residues on the amount of electrolyte. When the cleanliness is lower than the threshold, the system will send a steel plate to be cleaned signal to prevent steel plates that do not meet the cleanliness standards from entering subsequent processes, thereby reducing the defective rate and rework rate. When the surface cleanliness of the steel plate reaches the preset standard, the system can automatically map the electrolysis process parameters suitable for the current steel plate based on the initial configuration data of the equipment. Process parameter mapping ensures that the electrolysis treatment of each steel plate is based on its unique properties, such as material, thickness and cleanliness, to maximize the accuracy of the electrolysis process. According to the mapped electrolysis process parameters, the system can automatically control the electrolysis process to ensure the accuracy of the steel plate surface treatment process. The automated control process reduces manual intervention and ensures the stability and efficiency of production.

[0075] Preferably, step S4 is specifically:

[0076] Collect electrolytic surface images according to electrolytic result data to obtain electrolytic steel plate image data;

[0077] The electrolytic thickness data of the steel plate is collected by non-contact thickness measuring equipment to obtain the electrolytic thickness data;

[0078] The surface uniformity is evaluated based on the electrolytic steel plate image data and electrolytic thickness data to obtain electrolyte quantity detection data.

[0079] In the present invention, by collecting surface images after electrolysis, defects, flaws or unevenness on the surface of the electrolytic steel plate, such as scratches, color difference, roughness, etc., can be detected in real time. High-resolution images can accurately capture tiny surface changes and improve the accuracy of surface quality assessment. Non-contact thickness measurement equipment can accurately collect thickness data of electrolytic steel plates and avoid errors that may be introduced by contact measurement equipment. By combining the surface image and thickness data of the electrolytic steel plate, the surface uniformity of the steel plate can be evaluated. The uniformity evaluation can detect problems such as uneven thickness of the surface coating or electrolytic layer, uneven surface, excessive or insufficient treatment of local areas, etc., to ensure that the quality of the steel plate after electrolysis meets the standards.

[0080] The beneficial effect of the present invention is that by acquiring the basic data of the steel plate (such as size, thickness, material, etc.) and the initial data of the production line (such as equipment status, environmental parameters, etc.), the system can automatically generate the initial configuration data of the equipment adapted to the current production needs. This configuration method based on real-time data avoids the errors caused by the traditional manual adjustment of the equipment and ensures the optimal operation state of the equipment. By combining the image data of three different spectra of visible light, ultraviolet light and infrared light, multi-dimensional detection of the surface cleanliness of the steel plate is realized. Visible light images can identify scratches, stains and rust on the surface; ultraviolet light can identify grease and organic residues; infrared light can detect oxide layers, thermal anomalies and moisture residues. Multi-spectral detection ensures the comprehensiveness and accuracy of cleanliness evaluation and reduces missed detection and false detection. By combining the initial configuration data of the equipment with the surface cleanliness data of the steel plate, personalized electrolysis process parameters are generated. According to the cleanliness state of the steel plate, the system can adjust the electrolysis parameters (such as current, voltage, time, etc.) in real time to ensure that each steel plate can obtain the best electrolysis treatment effect. The electrolyte quantity detection data can be fed back to the control system in real time. Through the data-driven process adjustment mechanism, the system can self-optimize the electrolysis process based on the currently detected surface uniformity and thickness information. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting implementations made with reference to the following drawings:

[0082] Figure 1 A flowchart showing a method for controlling a modular electrolytic steel plate production line according to an embodiment of the present invention is provided;

[0083] Figure 2 A flowchart showing a method for initial configuration of equipment of an electrolytic steel plate production line according to an embodiment of the present invention is provided;

[0084] Figure 3 A flow chart showing the steps of a method for detecting the cleanliness of a steel plate surface according to an embodiment is shown;

[0085] Figure 4 A flow chart showing the steps of an electrolysis process control method according to an embodiment;

[0086] Figure 5 A flowchart of a method for detecting electrolyte amount according to an embodiment is shown. DETAILED DESCRIPTION

[0087] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are 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 technicians in this field without creative work are within the scope of protection of the present invention.

[0088] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0089] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0090] See also Figures 1 to 5 The present application provides a modular electrolytic steel plate production line control method, comprising the following steps:

[0091] Step S1: acquiring steel plate basic data and production line initial data, and performing initial equipment configuration according to the steel plate basic data and production line initial data to obtain equipment initial configuration data;

[0092] Specifically, extract the basic data of steel plates from the steel plate raw material database, including steel plate thickness, material, size, etc. Extract the initial data of the production line from the production line control system, such as the current status of the equipment, availability, temperature, humidity, etc. Match the basic data of the steel plate with the initial data of the production line to generate matching parameters, including current, voltage, production speed, etc. Automatically adjust the production line equipment according to the matching parameters, generate an equipment configuration plan, and ensure that the initial state of the equipment meets the steel plate processing requirements. Perform equipment calibration, update equipment parameters according to the equipment configuration plan, and record initial configuration data. Perform self-inspection on the equipment to ensure that the equipment is in normal status and generate an equipment self-inspection report. Output the initial configuration data of the equipment, including specific electrolytic cell parameters, flow rate, temperature control parameters, etc.

[0093] Step S2: acquiring steel plate image data, and performing steel plate surface cleanliness detection on the steel plate image data to obtain steel plate surface cleanliness data;

[0094] Specifically, an industrial camera is used to collect image data of the steel plate surface to ensure that the image covers the key parts of the steel plate. The collected image data is preprocessed, including denoising, grayscale adjustment, edge enhancement, etc., to obtain standardized image data. The standardized image data is divided into blocks, and the smoothness, reflectivity and stains of each area are analyzed. By comparing the brightness value of each area with the reference brightness standard, possible stains or grease areas are identified. Further feature analysis is performed on the suspected stain areas in the image to extract the area, shape, color and other features of the stains. The feature data is compared with the standard cleanliness model to determine the surface cleanliness and generate a steel plate cleanliness score. The steel plate surface cleanliness data is output, including the cleanliness level, the area to be processed and the cleaning recommendations.

[0095] Step S3: performing electrolysis process control operations according to the equipment initial configuration data and the steel plate surface cleanliness data to obtain electrolysis result data;

[0096] Specifically, according to the initial configuration data of the equipment in step S1, the electrolysis equipment is started, and the initial current, voltage and other parameters are set. According to the cleanliness data generated in step S2, the parameters of the electrolysis process (such as electrolyte concentration, current density, etc.) are adjusted to improve the electrolysis efficiency. The current, voltage and temperature changes during the electrolysis process are monitored in real time, and the real-time electrolysis data are recorded. The equipment parameters are dynamically adjusted according to the electrolysis reaction speed and the changes in the steel plate surface to ensure the uniformity of the electrolysis process. The consumption of the electrolyte is calculated in real time, and the replenishment rate and concentration of the electrolyte are adjusted to maintain process stability. The bubble and sediment data generated during the electrolysis process are regularly collected to analyze the effectiveness of the electrolysis reaction and the generation of by-products. The electrolysis result data, including electrolysis time, electrolysis efficiency, degree of surface modification of the steel plate, etc., are recorded and output.

[0097] Step S4: Perform electrolyte quantity detection according to the electrolysis result data to obtain electrolyte quantity detection data.

[0098] Specifically, collect surface image data of the steel plate after electrolysis and conduct preliminary surface inspection. Compare the surface image data before and after electrolysis to identify surface changes, such as surface finish, oxide deposition, etc. Perform physical property tests on the steel plate after electrolysis, and collect data such as hardness, corrosion resistance, and conductivity as evaluation indicators of the electrolyte amount. Calculate the modification efficiency, surface uniformity, and qualified rate of the steel plate based on the electrolysis result data and physical property test data. Use data analysis methods to evaluate abnormal conditions that occur during the electrolysis process and give suggestions for improvement. Output electrolyte amount detection data, including specific performance indicators of the steel plate after electrolysis, qualified grades, and recommendations on whether re-electrolysis is needed. Summarize all data to generate a complete electrolysis operation report to provide a reference for subsequent process optimization.

[0099] In the present invention, by obtaining the basic data of the steel plate and the initial data of the production line, it is possible to realize the automated and precise configuration of the equipment. Through automated cleanliness detection, the pollution problem on the surface of the steel plate can be timely discovered and solved before the electrolytic treatment, ensuring that the surface of the steel plate is sufficiently clean before entering the electrolytic cell, and reducing the production of unqualified products. According to the cleanliness detection data and the initial configuration data of the equipment, the production line can dynamically adjust the electrolytic process parameters (such as current, voltage, etc.) to realize electrolytic treatment for different steel plates. Through automated electrolyte quantity detection, the surface quality of the steel plate after electrolysis can be quickly and accurately evaluated, reducing the errors caused by manual inspection and improving the detection efficiency. The test results can be fed back to the front-end process to optimize the production process, form a closed-loop feedback system, continuously improve the production process, and ensure that the final product quality meets the standards.

[0100] Preferably, step S1 specifically includes:

[0101] Step S11: Acquire basic data of steel plates and initial data of production lines;

[0102] Specifically, extract the basic information of the steel plate from the steel plate database, including structural parameters such as steel plate thickness, material, length and width. Retrieve the surface treatment history of the steel plate, obtain relevant surface cleanliness, oxide layer thickness and other data, and generate steel plate surface status data. Obtain the initial parameters of the production line from the production line database, including equipment status, current environmental conditions (such as temperature, humidity, etc.), equipment availability, etc. Obtain the current configuration data of the electrolytic equipment, including equipment working status data such as electrolyte concentration, voltage, current, etc.

[0103] Step S12: generating equipment configuration data according to the basic data of the steel plate to obtain equipment configuration data;

[0104] Specifically, according to the physical parameters such as the thickness and material of the steel plate, the appropriate current density range is calculated to generate a preliminary current setting value. Combined with the surface state data of the steel plate, the electrolysis time and temperature are determined to generate the electrolysis time and temperature setting values. The conductivity of the steel plate is calculated to determine the voltage setting value and generate a preliminary voltage configuration. The current setting value, voltage setting value, time and temperature setting value are summarized to generate equipment configuration data. In the process of generating equipment configuration data, the electrolysis process is adapted and adjusted according to the different surface characteristics of the steel plate to ensure that the process meets production requirements.

[0105] Step S13: generating production line electrolysis parameters according to the equipment configuration data and the production line initial data to obtain production line electrolysis parameter data;

[0106] Specifically, according to the current and voltage setting values ​​in the equipment configuration data, the current distribution of the electrolytic cell is calculated to generate current distribution data. Combined with the equipment status and environmental data in the initial data of the production line, the flow rate of the electrolyte is adjusted to generate electrolyte flow rate parameters. According to the temperature, humidity and other environmental conditions of the production line, the cooling or heating power required in the electrolysis process is calculated to generate temperature control parameters. The data such as current distribution, electrolyte flow rate, and temperature control parameters are summarized to generate production line electrolysis parameter data. By estimating the power consumption and efficiency in the electrolysis process, the electrolysis time is optimized and the production line electrolysis parameters are updated.

[0107] Step S14: Initialize the equipment settings according to the production line electrolysis parameter data and the production line initial data to obtain the equipment initial configuration data.

[0108] Specifically, input the electrolysis parameter data of the production line into the electrolysis equipment, initialize the equipment parameters, and set the initial current, voltage, and electrolyte flow rate. Check the equipment self-test data to ensure that the electrolysis equipment and other auxiliary equipment are in normal working condition. According to the initial data of the production line, start the temperature control equipment, set the temperature adjustment range, and initialize the temperature control. Check the liquid level and flow rate of the electrolytic cell to ensure the fluidity of the electrolyte supply and discharge system, and initialize the liquid flow control equipment. Perform unified electrolysis parameter settings for the entire production line equipment to ensure that the equipment parameters of each link meet the requirements of the electrolysis process and generate the initial configuration data of the equipment. Record the initial configuration data of the equipment to the production line control system so that subsequent steps can be adjusted and monitored in real time.

[0109] In the present invention, by obtaining the basic data of the steel plate (such as size, thickness, material, surface state, etc.) and the initial data of the production line (such as equipment status, electrolyte state, ambient temperature, etc.), it is ensured that each batch of steel plates can be processed according to its unique characteristics. According to the basic data of the steel plate, the system automatically generates equipment configuration data to ensure that the equipment parameters (such as current, voltage, flow rate, etc.) can accurately match the characteristics of the steel plate. By combining the equipment configuration data and the initial data of the production line, the generated production line electrolysis parameters can accurately control the key parameters involved in the electrolysis process, such as electrolyte concentration, temperature, voltage and current, etc., to ensure that the electrolysis process can be carried out in the best state. Through the automated equipment setting initialization, the equipment can self-adjust and enter the working state according to the generated electrolysis parameters and the initial data of the production line in a short time. Compared with traditional manual initialization, automated configuration greatly shortens the production preparation time and improves the working efficiency of the production line.

[0110] Preferably, the steel plate image data includes first steel plate image data, second steel plate image data and third steel plate image data, and step S2 is specifically:

[0111] Step S21: collecting steel plate images by using a visible light collection device to obtain first steel plate image data;

[0112] Specifically, a visible light camera device is used to perform a full coverage scan of the steel plate surface to collect high-resolution visible light images. The collected images are preprocessed, including grayscale adjustment, image denoising, and edge enhancement, to ensure image clarity and detail retention. The preprocessed images are divided into blocks to ensure that each area can be analyzed separately. The first steel plate image data is output, which includes image features of physical defects on the steel plate surface (such as scratches, stains, cracks, etc.).

[0113] Step S22: collecting the steel plate image by ultraviolet light collection equipment to obtain second steel plate image data;

[0114] Specifically, a UV camera is used to irradiate the steel plate with UV light, focusing on collecting oil stains, oxide layers and other non-visible light pollutants on the surface of the steel plate. The UV image is preprocessed to filter out noise and enhance the characteristic areas such as stains and coatings under UV light. The polluted areas that appear under UV light are identified through image segmentation and feature extraction algorithms. The second steel plate image data is output, including the characteristics of the pollutants on the steel plate surface under UV light.

[0115] Step S23: collecting the steel plate image by infrared light collection equipment to obtain third steel plate image data;

[0116] Specifically, an infrared camera is used to collect thermal image data of the steel plate surface, focusing on identifying cracks, minor defects or thermal damage on the steel plate surface caused by thermal effects. Thermal image analysis is performed on the infrared image, including temperature field uniformity and hot spot detection, to identify abnormal areas. The image is segmented by temperature thresholds to extract data features of thermal abnormal areas. The third steel plate image data is output, including defect features caused by thermal effects on the steel plate surface.

[0117] Step S24: performing a first steel plate surface cleanliness detection on the first steel plate image data to obtain first steel plate surface cleanliness data;

[0118] Specifically, according to the surface features in the visible light image, physical defects such as surface finish, stains, scratches, etc. are analyzed by comparing with the standard cleanliness model (the standard cleanliness model is constructed by historical steel plate image data with sufficient cleanliness, and the model construction method includes deep learning algorithm and feature-based regression model). The surface reflectivity and brightness changes are calculated to determine whether there is obvious dirt or foreign matter. The surface defect area is matched with the cleanliness standard model to generate the first steel plate surface cleanliness score, and the first steel plate surface cleanliness data is output.

[0119] Step S25: performing a second steel plate surface cleanliness detection on the second steel plate image data to obtain second steel plate surface cleanliness data;

[0120] Specifically, the pollutant distribution area in the UV image is analyzed, and grease, chemical pollutants, etc. are identified by comparing the pollution characteristics under UV light. The distribution and severity of different pollutants are evaluated by the change in spectral reflectance of the UV image. The cleanliness detection score under UV light is generated, and the surface cleanliness data of the second steel plate is output.

[0121] Step S26: performing a third steel plate surface cleanliness detection on the third steel plate image data to obtain third steel plate surface cleanliness data, wherein the first steel plate surface cleanliness detection, the second steel plate surface cleanliness detection and the third steel plate surface cleanliness detection are different steel plate surface cleanliness detection methods;

[0122] Specifically, based on the thermal anomaly area in the infrared image, detect whether the steel plate has uneven thermal effects caused by contaminants. Identify thermal damage or cracks caused by insufficient cleaning and generate relevant cleanliness scores. Output the third steel plate surface cleanliness data, covering the surface unevenness caused by thermal effects and damage detection results.

[0123] Step S27: performing multi-dimensional cleanliness detection according to the first steel plate surface cleanliness data, the second steel plate surface cleanliness data and the third steel plate surface cleanliness data to obtain the steel plate surface cleanliness data.

[0124] Specifically, a multi-dimensional cleanliness model for the steel plate surface is constructed by integrating the surface cleanliness data of visible light, ultraviolet light, and infrared light. The weights of the cleanliness scores of each dimension are calculated, and the cleanliness scores are generated based on the test results under different light sources. The spatial distribution of various surface defects, stains, and contaminants is detected to determine whether the steel plate meets the cleanliness requirements. The final steel plate surface cleanliness data is output, covering the comprehensive cleanliness of the surface physical, chemical, and thermal effects.

[0125] In the present invention, by collecting steel plate image data under different spectra through visible light, ultraviolet light and infrared light equipment, different features of the steel plate surface can be captured. Visible light can detect physical defects and surface stains, ultraviolet light can identify organic pollutants and grease residues, and infrared light can detect thermal anomalies, such as moisture or oxide layers. The data of the three spectra are complementary, and various defects on the surface of the steel plate can be identified more accurately. Compared with the detection method using only a single spectrum, multi-spectral detection can reduce omissions and significantly improve the comprehensiveness and accuracy of detection. By using multiple detection methods in parallel, the missed detections and false detections caused by a single detection method can be effectively reduced. For example, ultraviolet light can make up for the shortcomings of visible light in detecting organic residues, and infrared light can detect temperature differences and find tiny defects that are difficult to detect with other spectra.

[0126] Preferably, the first steel plate surface cleanliness detection is specifically as follows:

[0127] Performing contrast enhancement on the first steel plate image data to obtain steel plate visible light image enhancement data;

[0128] Specifically, the brightness of the original visible light image data is first adjusted, and the visibility of the low-brightness areas in the image is improved through equalization. The contrast stretching or histogram equalization method is used to increase the difference between high-brightness and low-brightness areas in the image to ensure that surface details (such as stains, scratches, etc.) are more obvious. The edge of the image is sharpened to enhance the edge features in the image to ensure that the uneven surface areas of the steel plate can be clearly displayed. The processed steel plate visible light image enhancement data is output for subsequent image segmentation and feature extraction.

[0129] Performing steel plate region segmentation on the steel plate visible light image enhancement data to obtain steel plate visible light image segmentation data;

[0130] Specifically, by setting the brightness threshold, the steel plate area is separated from the background, irrelevant information is removed, and the main part of the steel plate is extracted. The edge detection method is used to identify the boundaries and internal details of the steel plate to ensure accurate surface segmentation. Through morphological processing (such as corrosion and expansion operations), isolated pixels and noise areas are eliminated during the segmentation process. The visible light image segmentation data of the steel plate is generated, and the area and surface details of the steel plate are marked to provide a basis for the next step of feature extraction.

[0131] Surface texture feature extraction, surface color feature extraction and local grayscale feature extraction are performed on the visible light image segmentation data of the steel plate to obtain surface color feature data of the steel plate image, surface color feature data of the steel plate image and grayscale feature data of the steel plate respectively;

[0132] Specifically, the texture changes of the local area of ​​the steel plate surface are analyzed, and the local texture pattern is calculated using a sliding window. By analyzing the concave and convex features of the surface, the irregular texture information of the surface is extracted to obtain the surface texture feature data of the steel plate.

[0133] Convert image color space (such as RGB to HSV) to extract color features of steel plate surface. Analyze the color distribution in each segmented area, identify color abnormal areas such as surface contaminants or oxide layers, and generate color feature data of steel plate surface.

[0134] In the segmented steel plate image, the grayscale distribution of each area is analyzed, and the statistical characteristics such as the average value and variance of the local grayscale are calculated. The surface brightness uneven area is extracted through contrast analysis to obtain the grayscale feature data of the steel plate.

[0135] Extract irregular surface features from the color feature data of the steel plate image surface and the grayscale feature data of the steel plate to obtain irregular surface feature data of the steel plate;

[0136] Specifically, the surface color feature data and grayscale feature data are fused, and the local change rate of color and grayscale is combined to identify surface irregular areas. The fused feature data is analyzed to detect surface abnormalities, such as scratches, depressions or convex areas. Based on the local surface feature changes, the surface irregular feature data of the steel plate is generated for further defect analysis.

[0137] Defect contour detection is performed based on the irregular feature data of the steel plate surface and the color feature data of the steel plate image surface to obtain the first steel plate surface cleanliness data.

[0138] Specifically, based on the irregular feature data and color feature data, an edge detection algorithm is used to extract the outline of the surface defect, including the boundary shape and size of the defect. The area, shape and distribution of the defect area are calculated to determine the severity of the surface defect. The defect area is compared with the preset cleanliness standard to evaluate the cleanliness of the steel plate surface and generate a cleanliness score. The first steel plate surface cleanliness data is output, including the outline of the steel plate surface defect, the cleanliness score and a recommendation on whether further processing is required.

[0139] In the present invention, by contrast enhancement, the texture, color and grayscale differences on the surface of the steel plate can be strengthened, and subtle surface defects such as stains and scratches can be highlighted. Through the segmentation algorithm, the steel plate is separated from the background, so that subsequent feature extraction and detection are only focused on the surface of the steel plate, avoiding the interference of background noise. Accurate regional segmentation ensures the accuracy of the detection algorithm. By extracting surface texture, color and grayscale features, the state of the steel plate surface can be fully described. Texture features can detect subtle scratches or corrosion, color features can identify stains or uneven coatings, and grayscale features can detect local smoothness or unevenness. Through irregular surface feature extraction, irregular shapes on the surface of the steel plate can be effectively identified, including protrusions, depressions, scratches, rust, etc. Defect contour detection based on irregular surface feature data and color feature data can accurately outline the contours of surface defects, detect surface contamination, scratches, pits and other problems, and help evaluate the scope and severity of defects through contour information. By combining irregular feature data and color feature data, a more comprehensive cleanliness assessment result can be generated, ensuring that cleanliness detection is not limited to surface stains, but also includes morphological irregularities and color anomalies.

[0140] Preferably, the local grayscale feature extraction is specifically:

[0141] Performing pixel domain relationship processing on the visible light image segmentation data of the steel plate to obtain pixel domain relationship data;

[0142] Specifically, the visible light image of the steel plate is segmented according to predefined rules, and the pixel data of each block area in the image is extracted. For each pixel, the gray value changes of the surrounding adjacent pixels are analyzed to generate a pixel domain relationship matrix. The average gray value difference value of the neighborhood of each pixel is calculated, and the gray value change trend in its neighborhood is determined to obtain the pixel domain relationship data. The pixel domain relationship data of each block area is output to indicate the degree of mutual influence between pixels.

[0143] The co-occurrence matrix is ​​calculated based on the pixel domain relationship data and the steel plate visible light image segmentation data to obtain the co-occurrence matrix data;

[0144] Specifically, for each block area, the co-occurrence matrix is ​​calculated using the pixel domain relationship data and segmentation data to record the frequency of occurrence of pixel grayscale value pairs. In the co-occurrence matrix, the grayscale relationship of each pair of pixels is counted, the spatial dependence between pixel pairs is calculated, and the correlation degree at different grayscale levels is obtained. The calculation of the co-occurrence matrix includes the calculation of the grayscale relationship of pixel pairs in the horizontal, vertical and diagonal directions. The co-occurrence matrix data is output to reflect the statistical information of the local grayscale changes of the image.

[0145] Gradient contrast calculation is performed based on the co-occurrence matrix data to obtain local gradient contrast data;

[0146] Specifically, the spatial distribution information of the pixel grayscale value is extracted from the co-occurrence matrix, and the local gradient change of each pixel is calculated. For the co-occurrence matrix data in different directions, the gradient difference between the pixels is calculated to evaluate the local edge sharpness of the image. The calculation of gradient contrast takes into account the amplitude of the grayscale change of adjacent pixels to reflect the local detail changes of the image. The local gradient contrast data is output to indicate the grayscale gradient change of the local area in the image.

[0147] The brightness contrast is calculated according to the co-occurrence matrix data to obtain the local brightness contrast data;

[0148] Specifically, the pixel grayscale relationship in the co-occurrence matrix is ​​used to calculate the brightness difference of the local area of ​​the image. The brightness values ​​of each block area in the image are counted to calculate the local brightness contrast. Brightness contrast is used to evaluate the brightness and darkness changes of the image in different areas, and is especially suitable for detecting shadows and light source reflection areas in the image. The local brightness contrast data is output to reflect the light and dark contrast characteristics of the image.

[0149] Perform feature fusion based on local gradient contrast data and local brightness contrast data to obtain preliminary local grayscale feature data;

[0150] Specifically, the local gradient contrast data and brightness contrast data are normalized to ensure the consistency of data scale. The two types of contrast features are fused by weighted averaging or other fusion methods to generate preliminary local grayscale feature data. Feature fusion considers the weight distribution of different features in local areas to enhance the expression ability of local grayscale information of the image. The preliminary local grayscale feature data is output to reflect the changes in gradient and brightness of the image.

[0151] The local structure tensor is calculated based on the pixel domain relationship data and the steel plate visible light image segmentation data to obtain the image structure tensor feature data;

[0152] Specifically, the pixel domain relationship data is used to calculate the structure tensor of each pixel in the image to describe its local geometric structure. The calculation of the structure tensor takes into account the gradient changes and local symmetry of the pixel in different directions, reflecting the local structural characteristics. The structure tensor is feature decomposed to extract the local main direction information of the image and identify the main edges and textures in the image. The image structure tensor feature data is output to represent the geometric structure characteristics of the image in the local area.

[0153] Feature splicing is performed based on the preliminary local grayscale feature data and the image structure tensor feature data to obtain the grayscale feature data of the steel plate.

[0154] Specifically, the preliminary local grayscale feature data is spliced ​​with the image structure tensor feature data to ensure that the grayscale features and structural features of each pixel are retained at the same time. The spliced ​​feature data is optimized to remove redundant information to ensure the compactness and representativeness of the features. Finally, the global grayscale feature data of the steel plate is generated for subsequent analysis and detection of the surface cleanliness of the steel plate. The grayscale feature data of the steel plate is output, which contains multi-dimensional local grayscale and structural feature information.

[0155] In the present invention, by analyzing the grayscale relationship between each pixel and its neighboring pixels, the local features on the surface of the steel plate can be captured more carefully. The processing of the pixel neighborhood relationship makes the system more sensitive to local anomalies and subtle defects (such as scratches or tiny contaminants), which can improve the detection accuracy and avoid omissions. Through the calculation of the co-occurrence matrix, the spatial relationship between the grayscale of pixels in the image can be quantified to describe the texture and structural information of the surface of the steel plate. The co-occurrence matrix can capture detailed texture information, which is particularly effective for areas with small surface texture changes or subtle surface defects. Gradient contrast can effectively detect the edges, subtle protrusions or depressions on the surface of the steel plate by capturing the direction and amplitude of grayscale changes, ensuring that the detection of surface changes is more sensitive. Brightness contrast is used to capture local brightness changes, and can effectively detect uneven brightness areas on the surface of the steel plate, such as the transition between smooth and rough areas, or brightness anomalies caused by stains and scratches. Through the calculation of the structural tensor, the local geometric structure information of the steel plate surface, especially the main direction and intensity of the grayscale change, can be captured, which is of great help to detect regular or irregular morphological changes on the surface, and can identify morphological defects such as surface protrusions and depressions. Feature stitching enhances the classification and identification capabilities of surface defects by combining multi-dimensional features, especially when processing steel plates with complex surface textures and structures, which can reduce missed detections and false detections.

[0156] Preferably, the surface cleanliness detection of the second steel plate is specifically as follows:

[0157] Performing fluorescence reaction enhancement on the second steel plate image data to obtain steel plate ultraviolet light image enhancement data;

[0158] Specifically, the fluorescence reaction on the surface of the steel plate is stimulated by ultraviolet light irradiation, and the fluorescence signal reflected by the surface pollutants (such as grease, chemical residues) is detected. The original ultraviolet image is subjected to fluorescence enhancement processing to amplify the brightness and contrast of the fluorescence signal to ensure clear display of the pollutant area. The image enhancement algorithm is used to amplify the weak fluorescence signal, reduce noise interference, and highlight the fluorescence reaction area. The steel plate ultraviolet image enhancement data is output, including the enhanced pollutant fluorescence area.

[0159] The overall brightness distribution of the image is adjusted through techniques such as histogram equalization or logarithmic transformation. Since the fluorescence signal is relatively weak, the focus is on increasing the brightness of the darker areas, so that the fluorescent reaction area is more prominent. The contrast of the fluorescence signal is enhanced by contrast stretching technology, which increases the difference between areas with strong and weak fluorescence intensity, making the fluorescent pollutant area more obvious. This is achieved through methods such as linear contrast stretching and CLAHE (contrast limited adaptive histogram equalization).

[0160] For different types of fluorescence reactions (caused by different components of pollutants), each channel in the image may contain different fluorescence signals. These channels are enhanced separately to make the fluorescence reaction of each pollutant more significant. The enhanced channels are resynthesized to generate a global UV fluorescence image that integrates the fluorescence information of various pollutants.

[0161] Performing fluorescence region segmentation on the steel plate ultraviolet image enhancement data to obtain steel plate ultraviolet image segmentation data;

[0162] Specifically, based on the fluorescence brightness threshold, the fluorescent reaction area in the UV image is automatically segmented to distinguish the fluorescent signal from the background area. The edge detection method is used to identify the boundary of the fluorescent area to ensure the accuracy of the segmentation. Through morphological processing methods (such as expansion and corrosion operations), the segmentation results are optimized to eliminate small isolated points and noise. The UV image segmentation data of the steel plate is output, which contains a clear fluorescent contaminant area.

[0163] Extract pollutant spectral features from the steel plate ultraviolet image segmentation data to obtain pollutant spectral feature data;

[0164] Specifically, for the fluorescent pollutant area, extract its spectral reflection characteristics under ultraviolet light, and analyze the fluorescence response of different pollutants at different wavelengths. Calculate the spectral distribution of the fluorescence signal, record the fluorescence intensity of different wavelengths, and generate the spectral characteristic map of the pollutant. Compare the spectral characteristics of different areas, classify and identify various pollutants, and generate corresponding spectral characteristic data. Output the pollutant spectral characteristic data to reflect the type of pollutant and its spectral response characteristics.

[0165] Performing fluorescence intensity analysis on pollutant spectral characteristic data to obtain pollutant fluorescence intensity data;

[0166] Specifically, the fluorescence intensity values ​​at different wavelengths in the pollutant area are calculated to evaluate the concentration and coverage of the pollutants. According to the peak intensity in the spectral feature data, the reaction intensity of the pollutants under ultraviolet light is analyzed to generate a fluorescence intensity graph. For different types of pollutants, their fluorescence intensities are calculated separately, and the intensity values ​​in different areas are normalized. The pollutant fluorescence intensity data is output as an indicator to measure the concentration and coverage of pollutants.

[0167] Calculate the pollutant area based on the pollutant spectral characteristic data to obtain the pollutant area data;

[0168] Specifically, the spatial distribution of pollutants in the image is identified through spectral feature data, and the area of ​​pollutants in the fluorescent area is calculated. The image segmentation results are used to accumulate the pollutant area pixel by pixel to ensure accurate calculation. The areas of different pollutants are classified and counted to obtain the coverage area data of various pollutants. The pollutant area data is output to indicate the spatial distribution range of each pollutant.

[0169] Cleanliness calculation is performed based on the pollutant fluorescence intensity data and the pollutant area data to obtain the second steel plate surface cleanliness data.

[0170] Specifically, the fluorescence intensity data and the contaminant area data are comprehensively analyzed to calculate the concentration and coverage of the contaminants on the steel plate surface. The cleanliness standard is set, and the actual contaminant intensity and area are compared with the standard value to generate a cleanliness score. The overall cleanliness level of the contaminants is determined by combining the weights of the fluorescence intensity and area. The second steel plate surface cleanliness data is output, including a comprehensive cleanliness score and a recommended cleaning treatment method.

[0171] The fluorescence reaction enhancement technology in the present invention can strengthen the organic pollution reaction under ultraviolet light, such as oil, fatty acids, chemical residues and other substances, making them more obvious in the ultraviolet image. Fluorescence area segmentation can automatically identify and segment the pollutant area on the surface of the steel plate, separate the pollutants from the background area, ensure that the subsequent feature extraction focuses on the truly polluted area, and reduce the interference of non-polluted areas on the analysis. Through spectral feature extraction, the composition and properties of the pollutants can be further analyzed. Different organic matter and chemical residues will produce different spectral reactions under ultraviolet light. Spectral feature extraction can help identify the type of pollutant and provide more detailed pollutant information. The fluorescence intensity is related to the concentration of the pollutant. Through fluorescence intensity analysis, the concentration or thickness of the pollutant can be quantitatively evaluated. By calculating the area of ​​the pollutant on the surface of the steel plate, the distribution of the pollutant can be understood.

[0172] Preferably, the surface cleanliness detection of the third steel plate is specifically as follows:

[0173] Performing thermal image enhancement on the image data of the third steel plate to obtain an infrared image enhanced image of the steel plate;

[0174] Specifically, infrared light camera equipment is used to collect thermal images of the steel plate surface to generate preliminary infrared light image data. The preliminary image data is enhanced, including improving the contrast and sharpness of the image and enhancing the details of the hot area. Low-temperature noise in the image is filtered out to ensure that the thermal anomaly features in the high-temperature area are clear. The steel plate infrared light image enhancement image is output to ensure that the features of the hot area are significant for subsequent analysis.

[0175] Perform steel plate thermal area segmentation on the steel plate infrared image enhancement image to obtain steel plate infrared image segmentation data;

[0176] Specifically, according to the heat distribution in different areas of the image, the threshold segmentation method is used to distinguish the high temperature area from the background area. The region growing algorithm is used to refine the boundaries of the hot area to ensure the segmentation accuracy. For different temperature areas on the surface of the steel plate, the abnormal high temperature or low temperature areas are marked. The infrared image segmentation data of the steel plate is output, including the segmentation results of the heat distribution area on the surface of the steel plate.

[0177] Perform thermal anomaly detection on the infrared image segmentation data of the steel plate to obtain thermal anomaly data of the steel plate image;

[0178] Specifically, for the segmented hot areas, the uniformity of temperature distribution is detected to identify local abnormal hot spots. The temperature gradient analysis is used to determine the scope and intensity of the thermal anomaly area and generate thermal anomaly detection results. The detected thermal anomaly data is compared with the standard temperature distribution model to evaluate whether there are thermal anomalies caused by insufficient cleaning or contaminants. The thermal anomaly data of the steel plate image is output to identify the thermal defect areas on the steel plate surface.

[0179] Perform moisture distribution detection on the infrared image segmentation data of the steel plate to obtain moisture distribution data of the steel plate image;

[0180] Specifically, the temperature difference in the infrared image is used to analyze the moisture distribution in different areas of the steel plate surface and identify wet or dry areas. The multi-band thermal analysis method is used to detect the thermal response of moisture and extract the characteristics of moisture distribution. Combined with the physical properties of the steel plate surface, the absorption and reflection characteristics of moisture under infrared light are analyzed to generate a moisture distribution map. The moisture distribution data of the steel plate image is output to represent the spatial distribution of moisture on the steel plate surface.

[0181] Perform oxide layer detection according to the infrared image segmentation data of the steel plate to obtain oxide layer data of the steel plate image;

[0182] Specifically, the temperature response differences of different areas on the steel plate surface are analyzed to detect the presence of the oxide layer. The thickness and distribution of the oxide layer are determined based on the change in temperature gradient. Infrared spectroscopy is used to identify the unique spectral response of the oxide layer and distinguish the oxide layer area from other pollutant areas. The oxide layer data of the steel plate image is output to reflect the distribution and thickness of the oxide on the steel plate surface.

[0183] Thermal feature extraction of surface pollutants is performed based on thermal anomaly data of the steel plate image, moisture distribution data of the steel plate image, and oxide layer data of the steel plate image to obtain thermal feature data of pollutants on the steel plate surface;

[0184] Specifically, the thermal anomaly data, moisture distribution data and oxide layer data are integrated to analyze the differences in thermal responses of different pollutants. The features of pollutants in thermal images are extracted, including thermal diffusion, thermal reflection and thermal absorption characteristics. For different types of pollutants, thermal feature description data is generated to distinguish grease, chemical residues, etc. The thermal feature data of pollutants on the surface of steel plates is output, including the thermal response feature information of pollutants.

[0185] The surface cleanliness is calculated based on the thermal characteristic data of the contaminants on the steel plate surface to obtain the surface cleanliness data of the third steel plate.

[0186] Specifically, the thermal characteristic data of the surface contaminants are compared with the preset cleanliness standards to evaluate the severity of the contaminants. The cleanliness of the steel plate surface is comprehensively evaluated by combining the thermal anomaly, moisture distribution and oxide layer data to generate a cleanliness score. The score weight is adjusted according to the thermal characteristics of different contaminants to ensure the accuracy of the cleanliness calculation. The third steel plate surface cleanliness data is output, including the thermal response cleanliness score and contaminant distribution information.

[0187] In the present invention, by enhancing the infrared image, the visibility of the thermal features on the surface of the steel plate, especially the tiny thermal anomalies, cold areas or hot areas, can be improved. By segmenting the hot areas of the infrared image, the thermal anomaly areas can be effectively distinguished from the normal areas. Thermal anomaly detection can capture the thermal conduction anomalies on the surface of the steel plate, such as cracks, local contamination or temperature differences caused by uneven surface coating. Infrared imaging can identify the moisture distribution on the surface and inside of the steel plate, especially trace moisture or wet areas. The oxide layer usually shows specific thermal diffusion and thermal reflection characteristics in infrared imaging, and the oxide layer detection can accurately identify the oxidized area on the surface of the steel plate. Different types of pollutants (such as oil, moisture, oxides) have different thermal characteristics in infrared images. By analyzing the thermal features, these pollutants can be distinguished and their distribution and properties can be accurately identified. Thermal feature extraction can help the system accurately capture the thermal anomaly areas of pollutants, so that the detection process is not only limited to visible pollutants, but also can detect surface problems caused by thermal conduction anomalies, such as coating defects or moisture penetration.

[0188] Preferably, the oxide layer detection is specifically:

[0189] The temperature gradient is calculated according to the infrared image segmentation data of the steel plate to obtain the infrared temperature gradient data of the steel plate;

[0190] Specifically, the temperature distribution information of the steel plate surface is extracted from the infrared image segmentation data, and the temperature gradient of each area is calculated. 1.2 The temperature difference of adjacent pixels is used to calculate the local temperature change rate and generate temperature gradient data. According to the different material properties of the steel plate surface, the temperature gradient calculation model is adjusted to improve the accuracy. The infrared temperature gradient data of the steel plate is output to indicate the temperature change of different areas on the steel plate surface.

[0191] Conduct thermal diffusion simulation on the infrared temperature gradient data of the steel plate to obtain thermal diffusion simulation data of the steel plate;

[0192] Specifically, based on the temperature gradient data, the heat diffusion process on the steel plate surface is simulated to predict the heat transfer path from the high temperature area to the low temperature area. The time step and space step are used to simulate the heat diffusion behavior and generate the heat diffusion dynamic data. In the heat diffusion simulation process, the thermal conductivity and surface characteristics of the steel plate material are considered and the diffusion model is adjusted. The steel plate heat diffusion simulation data is output to reflect the heat diffusion characteristics on the steel plate surface.

[0193] Adjust the thermal diffusion coefficient according to the steel plate thermal diffusion simulation data to obtain the steel plate thermal diffusion coefficient adjustment data;

[0194] Specifically, the thermal diffusion rate of different regions in the thermal diffusion simulation data is analyzed, and the thermal diffusion coefficient of each region is adjusted. The thermal diffusion coefficient is calibrated by experimental data to ensure the consistency between the simulation results and the actual situation. The diffusion coefficient is optimized using numerical methods to enhance the simulation accuracy and output the thermal diffusion coefficient adjustment data of the steel plate.

[0195] The temperature stability feature is extracted according to the steel plate thermal diffusion coefficient adjustment data to obtain the temperature stability data;

[0196] Specifically, based on the adjusted thermal diffusion coefficient data, the stability of the steel plate surface temperature is analyzed to evaluate the temperature change amplitude of each area after thermal diffusion. The temperature stability characteristics are extracted to identify areas with smaller temperature fluctuations and mark them as temperature stable areas. The temperature stability data is output to reflect the temperature stability differences of each area on the steel plate surface.

[0197] The temperature time series feature extraction is performed according to the steel plate thermal diffusion coefficient adjustment data to obtain the temperature time series feature data;

[0198] Specifically, the temperature change time series data of different areas on the steel plate surface are recorded, and the evolution of the temperature in each area over time is analyzed. Through the time series analysis method, the key features of the temperature change, including the rate of temperature rise and fall, are extracted. The temperature time series feature data is generated to represent the time dynamic characteristics of the steel plate during the thermal diffusion process.

[0199] Calculate the temperature difference according to the temperature stability data and the temperature time series characteristic data to obtain the temperature difference characteristic data;

[0200] Specifically, the temperature difference of each area is calculated by combining the temperature stability and time series characteristics to identify the thermal anomaly area. The rate of change of local temperature difference is used to determine the thermal non-uniformity on the surface of the steel plate. The temperature difference characteristic data is output to reflect the temperature change amplitude and degree of difference in different areas.

[0201] Segment the thermal anomaly area according to the temperature difference characteristic data to obtain the oxide layer segmentation data;

[0202] Specifically, the temperature difference feature data is used to segment the area with abnormal temperature on the steel plate surface and mark the abnormal thermal area. The high temperature difference area is distinguished from other normal areas, and the area with oxide layer is identified. The oxide layer segmentation data is output to represent the spatial distribution of the oxide layer on the steel plate surface.

[0203] Marking the oxide layer region according to the oxide layer segmentation data to obtain oxide layer region marking data;

[0204] Specifically, the segmented oxide layer area is annotated to mark the range and boundary of the oxide layer. The annotation accuracy of the oxide layer area is further optimized based on the infrared image and temperature difference data. The oxide layer area annotation data is output to reflect the specific location and distribution of the oxide layer on the surface of the steel plate.

[0205] The oxide layer thickness is estimated based on the oxide layer area annotation data according to the temperature stability data and the temperature time series characteristic data, and the oxide layer data of the steel plate image is obtained.

[0206] Specifically, the thickness of the oxide layer is estimated using the temperature stability data and the temperature time series characteristic data. The formation depth of the oxide layer is estimated based on the temperature change of the oxide layer area. The oxide layer data of the steel plate image is output, including the thickness and spatial distribution information of the oxide layer.

[0207] In the present invention, by calculating the temperature gradient of the infrared image of the steel plate, the area with temperature difference on the surface can be effectively identified, and the location of thermal anomaly or oxide layer can be preliminarily determined. The temperature gradient calculation can capture the slight changes in the temperature of the steel plate surface, especially the difference in heat conduction caused by the oxide layer, which helps to improve the sensitivity of oxide layer detection. Through thermal diffusion simulation, the heat conduction process of different areas on the surface of the steel plate can be reconstructed, so as to analyze the diffusion characteristics of heat in different materials. By adaptively adjusting the thermal diffusion coefficient, the thermal diffusion behavior of different materials can be simulated more accurately. The oxide layer usually has different thermal diffusion coefficients, and the adjusted data helps to detect the existence of the oxide layer more accurately. The temperature stability feature extraction can identify which areas on the surface of the steel plate have a small temperature change after thermal diffusion. These areas usually have an oxide layer because of its poor thermal conductivity and slow temperature change. By extracting the temperature time series features, the temperature changes of different areas in different time periods can be analyzed. Due to poor heat conduction, the oxide layer shows a slower temperature change, and the oxide layer is further identified from the time dimension. The temperature difference calculation can more accurately determine the difference between the oxide layer and the substrate by comparing the temperature changes in different areas.

[0208] Preferably, step S3 is specifically:

[0209] Step S31: When it is determined that the steel plate surface cleanliness data is less than the preset cleanliness threshold data, a steel plate to be cleaned signal is sent to the control platform;

[0210] Specifically, the surface cleanliness data of the steel plate is read and compared with the preset cleanliness threshold to determine whether the cleaning standard is met. When the surface cleanliness data of the steel plate is lower than the threshold, the system records that the steel plate needs further cleaning. A signal to be cleaned is generated, which includes the unique identification of the steel plate and the area information to be cleaned. The signal to be cleaned is sent to the cleaning equipment through the control platform to indicate the start of the cleaning operation.

[0211] Step S32: when it is determined that the surface cleanliness data of the steel plate is greater than or equal to the preset cleanliness threshold data, process parameter mapping is performed according to the initial configuration data of the equipment to obtain process parameter mapping data of the electrolysis equipment;

[0212] Specifically, the initial configuration data of the equipment is read, including equipment operating parameters such as current, voltage, temperature and flow rate. Based on the initial configuration data of the equipment, the process parameters are mapped. The mapping process includes adjusting the electrolysis current, time and other process conditions according to the surface cleanliness data of the steel plate. Combined with the material, thickness and other characteristics of the steel plate, the electrolysis process is dynamically adjusted to generate the electrolysis equipment process parameter mapping data suitable for the current steel plate processing. The electrolysis equipment process parameter mapping data is output to ensure that the electrolysis equipment operates under the best conditions.

[0213] Step S33: Perform electrolysis process control operations according to the electrolysis equipment process parameter mapping data to obtain electrolysis result data.

[0214] Specifically, the process parameter mapping data is input into the electrolysis equipment to automatically adjust the current, voltage, temperature and electrolyte flow rate during the electrolysis process. The key parameters of the electrolysis process, including reaction rate, electrolysis efficiency, etc., are monitored in real time, and the temperature and current fluctuations during the process are recorded. For steel plate surfaces of different cleanliness, the duration and reaction intensity of the electrolysis process are dynamically adjusted to ensure uniform treatment of the steel plate surface. The electrolysis result data is output, including the surface quality of the steel plate after electrolysis, the reaction efficiency, and whether the quality requirements are met.

[0215] In the present invention, the system automatically detects the surface cleanliness of the steel plate and compares it with the preset threshold value without manual intervention. Through the preset cleanliness threshold standard, the system can strictly control the surface quality of the steel plate entering the electrolysis process, ensure that each steel plate meets the cleanliness requirements, and reduce the impact of surface residues on the amount of electrolyte. When the cleanliness is lower than the threshold, the system will send a steel plate to be cleaned signal to prevent steel plates that do not meet the cleanliness standards from entering subsequent processes, thereby reducing the defective rate and rework rate. When the surface cleanliness of the steel plate reaches the preset standard, the system can automatically map the electrolysis process parameters suitable for the current steel plate based on the initial configuration data of the equipment. Process parameter mapping ensures that the electrolysis treatment of each steel plate is based on its unique properties, such as material, thickness and cleanliness, to maximize the accuracy of the electrolysis process. According to the mapped electrolysis process parameters, the system can automatically control the electrolysis process to ensure the accuracy of the steel plate surface treatment process. The automated control process reduces manual intervention and ensures the stability and efficiency of production.

[0216] Preferably, step S4 is specifically:

[0217] Step S41: collecting electrolytic surface images according to electrolytic result data to obtain electrolytic steel plate image data;

[0218] Specifically, high-resolution industrial camera equipment is used to capture images of the surface of the electrolytic steel plate, covering the key areas of the steel plate to ensure the clarity and details of the image. The captured images are preprocessed, including noise filtering, brightness adjustment and edge enhancement, to obtain standardized electrolytic surface images. For images captured at different angles and light sources, image alignment and fusion are performed to ensure image integrity. The electrolytic steel plate image data is output to reflect the surface structure and processing uniformity after electrolysis.

[0219] Step S42: collecting steel plate electrolytic thickness data by non-contact thickness measuring equipment to obtain electrolytic thickness data;

[0220] Specifically, a laser or ultrasonic non-contact thickness measuring device is used to scan the electrolyzed steel plate and collect the thickness data of the steel plate in real time. The thickness of different areas is measured in detail to ensure the accuracy and consistency of the thickness data. The thickness measurement results are corrected to remove equipment errors and environmental interference to obtain accurate thickness values. The electrolysis thickness data is output to indicate the thickness distribution of the surface of the steel plate after electrolysis.

[0221] Step S43: Surface uniformity evaluation is performed based on the electrolytic steel plate image data and the electrolytic thickness data to obtain electrolyte quantity detection data.

[0222] Specifically, image analysis is performed on the electrolytic steel plate image data to identify whether there are defects, bubbles, cracks or uneven electrolysis on the surface. Based on the electrolysis thickness data, the thickness difference in different areas of the steel plate is calculated to evaluate the thickness consistency after electrolysis. The image data is combined with the thickness data for analysis to identify problems in the electrolysis process (such as excessive local thickness or uneven surface treatment). A surface uniformity score is generated to evaluate the overall quality after electrolysis based on the thickness distribution and surface quality. The electrolyte quantity detection data is output, including the judgment results of surface uniformity, thickness consistency, and whether further processing is required.

[0223] In the present invention, by collecting surface images after electrolysis, defects, flaws or unevenness on the surface of the electrolytic steel plate, such as scratches, color difference, roughness, etc., can be detected in real time. High-resolution images can accurately capture tiny surface changes and improve the accuracy of surface quality assessment. Non-contact thickness measurement equipment can accurately collect thickness data of electrolytic steel plates and avoid errors that may be introduced by contact measurement equipment. By combining the surface image and thickness data of the electrolytic steel plate, the surface uniformity of the steel plate can be evaluated. The uniformity evaluation can detect problems such as uneven thickness of the surface coating or electrolytic layer, uneven surface, excessive or insufficient treatment of local areas, etc., to ensure that the quality of the steel plate after electrolysis meets the standards.

[0224] The beneficial effect of the present invention is that by acquiring the basic data of the steel plate (such as size, thickness, material, etc.) and the initial data of the production line (such as equipment status, environmental parameters, etc.), the system can automatically generate the initial configuration data of the equipment adapted to the current production needs. This configuration method based on real-time data avoids the errors caused by the traditional manual adjustment of the equipment and ensures the optimal operation state of the equipment. By combining the image data of three different spectra of visible light, ultraviolet light and infrared light, multi-dimensional detection of the surface cleanliness of the steel plate is realized. Visible light images can identify scratches, stains and rust on the surface; ultraviolet light can identify grease and organic residues; infrared light can detect oxide layers, thermal anomalies and moisture residues. Multi-spectral detection ensures the comprehensiveness and accuracy of cleanliness evaluation and reduces missed detection and false detection. By combining the initial configuration data of the equipment with the surface cleanliness data of the steel plate, personalized electrolysis process parameters are generated. According to the cleanliness state of the steel plate, the system can adjust the electrolysis parameters (such as current, voltage, time, etc.) in real time to ensure that each steel plate can obtain the best electrolysis treatment effect. The electrolyte quantity detection data can be fed back to the control system in real time. Through the data-driven process adjustment mechanism, the system can self-optimize the electrolysis process based on the currently detected surface uniformity and thickness information.

[0225] Therefore, from any point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the attached application documents rather than the above description, and it is intended that all changes falling within the meaning and scope of equivalent elements of the application documents are included in the present invention.

[0226] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A modular electrolytic steel plate production line control method, characterized in that: The following steps are involved: Step S1: acquiring steel plate basic data and production line initial data, and performing initial equipment configuration according to the steel plate basic data and production line initial data to obtain equipment initial configuration data; Step S2: acquiring steel plate image data, and performing steel plate surface cleanliness detection on the steel plate image data to obtain steel plate surface cleanliness data; Step S3: performing electrolysis process control operations according to the equipment initial configuration data and the steel plate surface cleanliness data to obtain electrolysis result data; Step S4: performing electrolyte quantity detection according to the electrolysis result data to obtain electrolyte quantity detection data; The steel plate image data includes first steel plate image data, second steel plate image data and third steel plate image data, and step S2 is specifically as follows: Capturing the steel plate image by a visible light acquisition device to obtain first steel plate image data; The steel plate image is collected by ultraviolet light collection equipment to obtain second steel plate image data; The steel plate image is collected by infrared light collection equipment to obtain the third steel plate image data; Performing a first steel plate surface cleanliness detection on the first steel plate image data to obtain first steel plate surface cleanliness data; Performing a second steel plate surface cleanliness detection on the second steel plate image data to obtain second steel plate surface cleanliness data; Performing a third steel plate surface cleanliness detection on the third steel plate image data to obtain third steel plate surface cleanliness data, wherein the first steel plate surface cleanliness detection, the second steel plate surface cleanliness detection and the third steel plate surface cleanliness detection are different steel plate surface cleanliness detection methods; Performing multi-dimensional cleanliness detection according to the first steel plate surface cleanliness data, the second steel plate surface cleanliness data, and the third steel plate surface cleanliness data to obtain the steel plate surface cleanliness data; The first steel plate surface cleanliness test is specifically as follows: Performing contrast enhancement on the first steel plate image data to obtain steel plate visible light image enhancement data; Performing steel plate region segmentation on the steel plate visible light image enhancement data to obtain steel plate visible light image segmentation data; Surface texture feature extraction, surface color feature extraction and local grayscale feature extraction are performed on the visible light image segmentation data of the steel plate to obtain surface color feature data of the steel plate image, surface color feature data of the steel plate image and grayscale feature data of the steel plate respectively; Extract irregular surface features from the color feature data of the steel plate image surface and the grayscale feature data of the steel plate to obtain irregular surface feature data of the steel plate; Defect contour detection is performed based on the irregular feature data of the steel plate surface and the color feature data of the steel plate image surface to obtain the first steel plate surface cleanliness data; The local grayscale feature extraction is specifically as follows: Performing pixel domain relationship processing on the visible light image segmentation data of the steel plate to obtain pixel domain relationship data; The co-occurrence matrix is ​​calculated based on the pixel domain relationship data and the steel plate visible light image segmentation data to obtain the co-occurrence matrix data; Gradient contrast calculation is performed based on the co-occurrence matrix data to obtain local gradient contrast data; The brightness contrast is calculated according to the co-occurrence matrix data to obtain the local brightness contrast data; Perform feature fusion based on local gradient contrast data and local brightness contrast data to obtain preliminary local grayscale feature data; The local structure tensor is calculated based on the pixel domain relationship data and the steel plate visible light image segmentation data to obtain the image structure tensor feature data; Feature splicing is performed based on the preliminary local grayscale feature data and the image structure tensor feature data to obtain the grayscale feature data of the steel plate.

2. The method according to claim 1, characterized in that Step S1 is specifically as follows: Obtain basic data of steel plates and initial data of production lines; Generate equipment configuration data according to the basic data of the steel plate to obtain equipment configuration data; Generate production line electrolysis parameters according to equipment configuration data and production line initial data to obtain production line electrolysis parameter data; The equipment settings are initialized according to the electrolysis parameter data of the production line and the initial data of the production line to obtain the initial configuration data of the equipment.

3. The method according to claim 1, characterized in that The specific test of the surface cleanliness of the second steel plate is as follows: Performing fluorescence reaction enhancement on the second steel plate image data to obtain steel plate ultraviolet light image enhancement data; Performing fluorescence region segmentation on the steel plate ultraviolet image enhancement data to obtain steel plate ultraviolet image segmentation data; Extract pollutant spectral features from the steel plate ultraviolet image segmentation data to obtain pollutant spectral feature data; Performing fluorescence intensity analysis on pollutant spectral characteristic data to obtain pollutant fluorescence intensity data; Calculate the pollutant area based on the pollutant spectral characteristic data to obtain the pollutant area data; Cleanliness calculation is performed based on the pollutant fluorescence intensity data and the pollutant area data to obtain the second steel plate surface cleanliness data.

4. The method according to claim 1, characterized in that: The specific test of the surface cleanliness of the third steel plate is as follows: Performing thermal image enhancement on the image data of the third steel plate to obtain an infrared image enhanced image of the steel plate; Perform steel plate thermal area segmentation on the steel plate infrared image enhancement image to obtain steel plate infrared image segmentation data; Perform thermal anomaly detection on the infrared image segmentation data of the steel plate to obtain thermal anomaly data of the steel plate image; Perform moisture distribution detection on the infrared image segmentation data of the steel plate to obtain moisture distribution data of the steel plate image; Perform oxide layer detection according to the infrared image segmentation data of the steel plate to obtain oxide layer data of the steel plate image; Thermal feature extraction of surface pollutants is performed based on thermal anomaly data of the steel plate image, moisture distribution data of the steel plate image, and oxide layer data of the steel plate image to obtain thermal feature data of pollutants on the steel plate surface; The surface cleanliness is calculated based on the thermal characteristic data of the contaminants on the steel plate surface to obtain the surface cleanliness data of the third steel plate.

5. The method according to claim 4, characterized in that The specific detection of oxide layer is as follows: The temperature gradient is calculated according to the infrared image segmentation data of the steel plate to obtain the infrared temperature gradient data of the steel plate; Conduct thermal diffusion simulation on the infrared temperature gradient data of the steel plate to obtain thermal diffusion simulation data of the steel plate; Adjust the thermal diffusion coefficient according to the steel plate thermal diffusion simulation data to obtain the steel plate thermal diffusion coefficient adjustment data; The temperature stability feature is extracted according to the steel plate thermal diffusion coefficient adjustment data to obtain the temperature stability data; The temperature time series feature extraction is performed according to the steel plate thermal diffusion coefficient adjustment data to obtain the temperature time series feature data; Calculate the temperature difference according to the temperature stability data and the temperature time series characteristic data to obtain the temperature difference characteristic data; Segment the thermal anomaly area according to the temperature difference characteristic data to obtain the oxide layer segmentation data; Marking the oxide layer region according to the oxide layer segmentation data to obtain oxide layer region marking data; The oxide layer thickness is estimated based on the oxide layer area annotation data according to the temperature stability data and the temperature time series characteristic data, and the oxide layer data of the steel plate image is obtained.

6. The method according to claim 1, characterized in that Step S3 is specifically as follows: When it is determined that the surface cleanliness data of the steel plate is less than the preset cleanliness threshold data, a steel plate to-be-cleaned signal is sent to the control platform; When it is determined that the surface cleanliness data of the steel plate is greater than or equal to the preset cleanliness threshold data, process parameter mapping is performed according to the initial configuration data of the equipment to obtain process parameter mapping data of the electrolysis equipment; The electrolysis process control operation is performed according to the electrolysis equipment process parameter mapping data to obtain the electrolysis result data.

7. The method according to claim 1, characterized in that Step S4 is specifically as follows: Collect electrolytic surface images according to electrolytic result data to obtain electrolytic steel plate image data; The electrolytic thickness data of the steel plate is collected by non-contact thickness measuring equipment to obtain the electrolytic thickness data; The surface uniformity is evaluated based on the electrolytic steel plate image data and electrolytic thickness data to obtain electrolyte quantity detection data.

Images