Full-process intelligent monitoring and early warning method and equipment for laser decoiling blanking line

Abstract

The invention discloses a full-process intelligent monitoring and early warning method and device for a laser decoiling blanking line. The method comprises the steps that a target coiled material is prepared to be fed and uncoiled; collecting real-time operation data and coiled material state data; synchronously correcting the target coiled material in combination with the real-time operation data and the coiled material state data; acquiring surface state information; constructing a bare board surface defect distribution diagram according to the board surface state information; the target coiled material is controlled to enter a single-face film laminating machine to be subjected to film laminating treatment, and an electrical response signal is collected; identifying micro-bubble defect information in the film coating area based on the electrical response signal, and generating a digital film coating defect map by combining the micro-bubble defect information and the bare board surface defect distribution map; constructing an air escape path according to the digital film defect map; and linkage early warning information is generated through the digital film covering defect map, and a laser cutting path adjusting instruction is generated to conduct laser cutting on the target coiled material. The operation stability of the laser decoiling blanking line can be effectively improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of laser uncoiling, leveling and blanking line monitoring, and particularly relate to a full-process intelligent monitoring and early warning method and device for a laser uncoiling, leveling and blanking line. BACKGROUND

[0002] With the continuous advancement of modern equipment manufacturing industry towards high precision, high efficiency and multi-variety flexibility, the application value of laser uncoiling, leveling and blanking production line in the field of steel plate processing is increasingly prominent. Such production line undertakes the continuous links of plate uncoiling, leveling, positioning, laser cutting and blanking conveying, and its running stability and quality consistency are directly related to the downstream processing rhythm, product size precision and overall line synergy efficiency. Therefore, monitoring and risk identification of the laser uncoiling, leveling and blanking line has become a key link to guarantee production quality and equipment efficiency.

[0003] At present, the state monitoring of the laser uncoiling, leveling and blanking line mainly depends on the local detection modules of each device, such as collecting current fluctuation, motion feedback signal, cutting head power state, speed and position information of the feeding mechanism, etc. through the electrical control unit of the equipment end, to realize the judgment of the state of a single process or a single device. However, such monitoring method is usually centered on an independent controller, and there is a lack of data linkage between monitoring nodes, making it difficult to realize the correlation analysis of continuous characteristics. Especially in the film coating process, due to the existence of defects such as micro-bubbles, air retention or local uneven film layer, the traditional monitoring system is difficult to identify these film coating defects and their spatial distribution in a timely and accurate manner, resulting in that the actual defect situation of the plate surface cannot be fully considered in the cutting path planning and power adjustment of the downstream laser cutting. The above technical architecture often causes small defects to cause local cutting deviation, cutting edge quality reduction or local damage accumulation in the cutting process, thereby making it difficult to expose potential faults in a timely manner, thereby affecting the cutting quality stability and the continuity of the line operation. SUMMARY

[0004] Embodiments of the present application provide a full-process intelligent monitoring and early warning method and device for a laser uncoiling, leveling and blanking line, for improving the stability of the laser uncoiling, leveling and blanking line operation.

[0005] To achieve the above-mentioned purpose, the embodiments of the present application adopt the following technical solutions: In a first aspect, a full-process intelligent monitoring and early warning method for a laser uncoiling, leveling and blanking line is provided, applied to a laser uncoiling, leveling and blanking monitoring system for a laser uncoiling, leveling and blanking line including a leveling machine, and the method comprises: performing a feeding and uncoiling preparation step on a target coil; when detecting that the target coil after the preparation step passes through the leveling machine, collecting real-time operation data of the laser uncoiling, leveling and blanking line and coil state data of the target coil; synchronize the coating position and the feeding path of the target coil based on the real-time running data and the coil state data; When the target coil is processed by the correction, the plate surface state information of the corrected target coil is obtained; According to the plate surface state information, a bare plate surface defect distribution map of the target coil is constructed, and whether the target coil meets the coating condition is determined according to the bare plate surface defect distribution map; If the condition is met, the target coil is controlled to enter the single-sided coating machine for coating processing, and the electrical response signal of the target coil is collected through the multi-type sensor arranged after the coating section; Based on the electrical response signal, the micro-bubble defect information in the coating area is identified, and a digital coating defect map is generated based on the micro-bubble defect information and the bare plate surface defect distribution map; According to the digital coating defect map, an air escape path on the surface of the target coil is constructed; Through the digital coating defect map, linkage early warning information is generated, and based on the linkage early warning information and the air escape path, a laser cutting path adjustment instruction is generated to perform laser cutting on the target coil.

[0006] In another possible implementation of the first aspect, the method further includes: After the laser cutting is completed, the robot is controlled to grab the cut target coil for stacking processing, and the waste of the cut target coil is transported to a waste collection box through a belt conveying device; After the stacking process of the target coil is completed, a cycle control program of the laser cutting and flattening line is started to realize the continuous processing flow of a new target coil.

[0007] In another possible implementation of the first aspect, the real-time running data includes feeding roller speed and film material tension data, the coil state data includes coil actual speed, coil edge trajectory data and coil stress distribution data, and the synchronization correction of the coating position and the feeding path of the target coil based on the real-time running data and the coil state data includes: According to the coil edge trajectory data, a polynomial fitting is performed to obtain a coil edge fitting result; The transverse alignment deviation of the target coil is determined through the coil edge fitting result; The center line of the target coil is calculated based on the coil edge trajectory data, and the deviation distance and the center deviation fluctuation frequency are calculated between the center line and the reference center line; The film material tension fluctuation frequency is calculated based on the film material tension data, and the deviation reason is determined through the center deviation fluctuation frequency and the film material tension fluctuation frequency; The target coil is synchronized corrected in the coating position based on the deviation distance and the deviation reason; The feeding path is corrected by combining the stress distribution data of the rolled material and the lateral alignment deviation.

[0008] In another possible implementation of the first aspect, the correction of the feeding path by combining the coil stress distribution data and the lateral alignment deviation includes: The lateral offset trajectory of the target roll material is determined based on the lateral alignment deviation; By superimposing the stress distribution data of the target roll material with the lateral offset trajectory, it can be determined whether there is an offset aggravation segment corresponding to a high stress area. If there is a high-stress region corresponding to an aggravated offset segment, it is determined to be an active offset for residual stress release; The offset aggravation section is compensated by a preset reverse compensation amount to adjust the feeding path and correct the target roll material.

[0009] In another possible implementation of the first aspect, the step of identifying microbubble defect information within the coating area based on electrical response signals and generating a digital coating defect map by combining the microbubble defect information and the bare board surface defect distribution map includes: The electrical response signal is normalized. A two-dimensional phase space of electrical characteristics is constructed, and the normalized eddy current response signal is used as the horizontal axis of the phase space of electrical characteristics and the normalized capacitance response signal is used as the vertical axis of the phase space of electrical characteristics. The electrical response signal is mapped onto the phase space of electrical characteristics. Within the period window of any composite harmonic signal, the electrical response signals collected under the same time reference are plotted as a closed trajectory loop to obtain the electrical signal trajectory. Trajectory distortion analysis is performed on the electrical signal trajectory to obtain distortion characteristic parameters; The similarity index is calculated based on the distortion feature parameters and the baseline trajectory; When the similarity index is lower than the preset defect threshold, it is determined that there is a microbubble defect in the corresponding coating area, and the microbubble defect is classified according to the type based on the distortion feature parameters to obtain microbubble defect information. The spatial location of microbubble defects in the coated area is determined by microbubble defect information, and the spatial location of microbubble defects in the coated area is matched with the defect distribution map on the bare plate surface to form the same spatial reference system; Under the same spatial reference frame, the microbubble defect information is superimposed on the defect distribution map of the bare plate surface, and the rendering is performed in a graded manner according to the type of microbubble defect. When the defects of the coated microbubble overlap with the defects of the bare plate surface in spatial location, it is marked as a high-risk overlapping area, and the composite risk index is calculated. A digital coating defect map is obtained by combining composite risk indicators.

[0010] In another possible implementation of the first aspect, the method of classifying microbubble defects based on distortion feature parameters to obtain microbubble defect information includes: The amplitude distortion parameter and the phase distortion parameter are determined by the distortion characteristic parameters. The amplitude distortion parameter includes the compressibility in the eddy current axis direction and the compressibility in the capacitance axis direction, and the phase distortion parameter includes the phase rotation angle. When the compressibility in the capacitor axis direction exceeds the first capacitor compression threshold and the compressibility in the eddy current axis direction is less than the first eddy current compression threshold, the microbubble defect type is determined to be an air-type microbubble. When the compressibility of the capacitor axis exceeds the second capacitor compression threshold and the phase rotation angle exceeds the phase rotation threshold, the microbubble defect type is determined to be a water vapor microbubble. When the eddy current axial compressibility exceeds the third eddy current compressibility threshold and the capacitance axial compressibility is less than the second capacitance compressibility threshold, the microbubble defect type is determined to be microbubble induced by under-film metal microparticles.

[0011] In another possible implementation of the first aspect, the multi-type sensor array includes eddy current sensors and capacitive sensors, prior to identifying microbubble defect information within the coating area based on the electrical response signal, comprising: Composite harmonic signals are applied to the eddy current probe coil in the eddy current sensor and the measurement circuit in the capacitance sensor, respectively. Obtain the time reference pulse provided by the main encoder of the laser leveling and blanking line; When the time reference pulse is triggered, the eddy current response signal of the eddy current sensor and the capacitance response signal of the capacitance sensor are acquired. The eddy current response signal and the capacitance response signal are used as electrical response signals under the same time reference.

[0012] In another possible implementation of the first aspect, constructing the air escape path on the surface of the target roll material based on the digital coating defect map includes: Each microbubble defect type in the digital coating defect map is mapped to an air resistance value, and the air resistance values ​​are integrated into an air resistance matrix for the target roll surface; The connectivity of the target roll material surface is analyzed based on the air resistance matrix, and the target roll material surface is divided into air escape regions using the region growing algorithm. An air connectivity region is determined when the distance between any two air escape regions is less than a preset connectivity threshold. For any air-connected domain, the defective boundary points of the air-connected domain are used as candidate nodes, and the adjacency relationship between the candidate nodes is established using a graph structure. Based on the adjacency relationship, the shortest path algorithm is used to screen the path with the least resistance from the air connectivity domain to obtain local air escape candidate paths; Construct a global air resistance map of the target roll material based on local candidate paths; The path with minimum air resistance is determined by combining the global air resistance map with the path stitching algorithm of the resistance gradient. The path of least air resistance is taken as the air escape path.

[0013] Secondly, this application provides a machine-readable storage medium storing instructions that cause a machine to execute the aforementioned intelligent monitoring and early warning method for the entire process of a laser leveling and blanking line.

[0014] Thirdly, this application provides a fully intelligent monitoring and early warning system for a laser leveling and blanking line, including: The memory is configured to store instructions; and The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the aforementioned intelligent monitoring and early warning method for the entire process of a laser leveling and blanking line.

[0015] The above technical solution firstly ensures that the target roll material is in the correct initial state and position before entering the processing flow by performing feeding and unwinding preparation steps. This lays the foundation for subsequent leveling, laminating, and cutting processes, effectively avoiding problems caused by improper initial positioning or preparation. When the target roll material passes through the leveling machine after the preparation step, acquiring key real-time data on the equipment and materials during the leveling process helps to promptly identify and correct potential problems in the leveling stage, improving production stability. By synchronously correcting the laminating position and feeding path of the target roll material, the accurate position of the roll material during the laminating process can be ensured, avoiding uneven laminating or laminating failures caused by positional deviations, improving laminating quality and accuracy, and reducing production costs. By acquiring information on the surface condition of the sheet material, defects on the roll material surface can be detected in a timely manner. Early identification of surface defects and corresponding measures can effectively reduce the generation of defective products caused by surface defects, improving the product qualification rate. By constructing a defect distribution map, the distribution of surface defects of the roll material can be intuitively understood, avoiding laminating roll material with serious defects, saving material and time costs, and improving laminating quality. When the target roll material enters the single-sided laminator for lamination, and electrical response signals are collected, changes during the lamination process can be monitored in real time, allowing for timely detection of problems. Based on the electrical response signals, microbubble defects can be identified, accurately detecting hidden defects such as microbubbles within the lamination area. This information is then integrated with bare board defect information to form a comprehensive digital lamination defect map, improving lamination quality. By constructing air escape paths, laser cutting paths can be rationally planned, reducing factors during the cutting process and improving cutting accuracy. Laser cutting path adjustment commands are generated by linking early warning information and air escape paths to perform laser cutting on the target roll material. This achieves linkage between lamination defect information and downstream laser cutting, dynamically optimizing the cutting path, reducing cutting defects and material waste, and improving cutting accuracy and production line stability.

[0016] Other features and advantages of the embodiments of this application will be described in detail in the following detailed description section. Attached Figure Description

[0017] Figure 1 A flowchart illustrating a fully intelligent monitoring and early warning method for a laser leveling and blanking line, provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a digital coating defect map provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only for illustration and explanation of the embodiments of this application and are not intended to limit the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0019] It should be noted that if the embodiments of this application involve directional indicators (such as up, down, left, right, front, back, etc.), the directional indicators are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicators will also change accordingly.

[0020] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0021] Figure 1 This illustration schematically shows a process diagram of a fully intelligent monitoring and early warning method for a laser leveling and blanking line according to an embodiment of this application. Figure 1 As shown in the embodiment of this application, a full-process intelligent monitoring and early warning method for a laser leveling and blanking line is provided. The method is applied to a laser leveling and blanking monitoring system, which is used in a laser leveling and blanking line. The laser leveling and blanking line includes a leveling machine. The method may include the following steps.

[0022] S110. Preparation steps for feeding and unwinding the target roll material; S120. When the target roll material after the preparation step is detected to pass through the leveling machine, collect the real-time running data of the laser leveling and blanking line and the roll material status data of the target roll material. S130. Synchronously correct the lamination position and feeding path of the target roll material by combining real-time operation data and roll material status data. S140. After the target roll material has undergone the correction process, obtain the surface condition information of the corrected target roll material. S150. Construct a bare surface defect distribution map of the target roll material based on the surface condition information of the board material, and determine whether the target roll material meets the coating conditions based on the bare surface defect distribution map. S160. If the conditions are met, the target roll material is controlled to enter the single-sided laminating machine for lamination, and the electrical response signal of the target roll material is collected by multiple types of sensors set after the lamination section. S170. Identify microbubble defect information in the coating area based on electrical response signals, and generate a digital coating defect map by combining microbubble defect information and bare board surface defect distribution map. S180. Construct the air escape path on the surface of the target roll material based on the digital coating defect map; S190. Generate linkage early warning information through a digital coating defect map, and generate laser cutting path adjustment instructions based on the linkage early warning information and air escape path to perform laser cutting on the target roll material.

[0023] The process involves feeding and unwinding the target roll material, which refers to metal, plastic, or composite material rolls to be processed on the production line. In this embodiment, the laser leveling and unwinding line includes a storage rack, a feeding trolley, an unwinder, a heavy-duty precision leveler, a single-sided laminator, a film roll unwinding rack, a servo feeder, a length controller, a laser marking machine, a laser cutting unit, and a robot gripping device. Specifically, it includes processes such as feeding and unwinding preparation, sheet leveling and lamination, laser cutting, finished product stacking and waste disposal, and automatic circulation and roll changing. The feeding and uncoiling preparation steps involve transporting the stainless steel coil to be processed to the main shaft of the uncoiler via a feeding trolley, and completing centering and mechanical locking to prevent deviation or shaking during operation. Next, the pressure arm is activated to press the coil firmly, ensuring it will not bounce or loosen when the strapping is untied. Then, the strapping is unlocked, putting the coil head in a traction-ready state. Next, the traction mechanism guides the head to the take-up machine, ensuring the take-up machine and uncoiler operate synchronously to ensure stable tension and no tearing or folding during the unwinding process. Finally, the head is sent to the shearing position, where the head shearing device removes irregular edges, resulting in a neat rectangular shape that meets the feeding requirements of subsequent processes.

[0024] After the target roll material completes the feeding and unwinding preparation steps, the system detects that the roll material has entered the leveling machine area. At this time, the laser leveling and blanking monitoring system, through its set sensors and data acquisition devices, collects real-time operating status data of the laser leveling and blanking line and the roll material status data of the target roll material. The operating status data includes, but is not limited to, equipment operating data of each device in the laser leveling and blanking line, such as current, tension, speed, pressure, and related encoder pulse information; the roll material status data includes the surface flatness, thickness, tension, and slight edge warping information of the target roll material, which describes the physical state of the roll material during the process of passing through the leveling machine. By synchronously acquiring the above equipment operating data and roll material status data, the status of the target roll material during the leveling process can be monitored, thereby effectively ensuring the accuracy of roll material processing and the continuous and stable operation of the production line.

[0025] Subsequently, the coating position and feeding path of the target roll are synchronously corrected using real-time operational data and roll status data. Real-time operational data includes the rotational speed of the feeding rollers and film tension data, reflecting the dynamic operating status of the production line equipment and changes in roll traction force. Roll status data includes the actual roll speed, roll edge trajectory data, and roll stress distribution data, reflecting potential offsets, warping, or stress concentrations that may occur during roll movement. Specifically, polynomial fitting is performed based on the roll edge trajectory data to obtain the roll edge fitting result, thereby determining the lateral alignment deviation of the roll. Furthermore, the roll centerline is calculated based on the roll edge trajectory and compared with a reference centerline to obtain the center deviation distance and its fluctuation frequency. The film tension fluctuation frequency is calculated using film tension data, and the specific cause of the roll deviation is determined by analyzing the relationship between the center deviation fluctuation frequency and the film tension fluctuation frequency. Then, the coating position of the target roll is synchronously corrected in real-time based on the deviation distance and the cause of the deviation, ensuring that the coating position is consistent with the roll movement state. In addition, the feeding path is optimized and corrected by combining the stress distribution data of the roll material. Specifically, the lateral offset trajectory of the roll material is determined based on the lateral alignment deviation. The lateral offset trajectory is superimposed and analyzed with the stress distribution data of the roll material to determine whether there is an offset aggravation segment corresponding to the high stress area. If there is, it is determined to be an active offset caused by the release of residual stress. The offset aggravation segment is compensated in reverse using a preset reverse compensation amount, thereby adjusting the feeding path and realizing the control of the lateral position and stress distribution of the roll material throughout the entire coating process.

[0026] After the target roll material undergoes synchronous correction of the feeding path and coating position, the system performs real-time detection and collection of the surface condition of the target roll material to obtain specific surface characteristic data of the target roll material, including information such as the edge position, thickness uniformity, surface flatness, and stress distribution of the roll material.

[0027] Next, after acquiring the surface condition information of the target roll material, the collected surface image data and height field data are preprocessed. Then, based on the preprocessed surface condition information, a bare surface defect distribution map of the target roll material is constructed. This map characterizes the spatial location, size, and type of various surface defects along the length and width of the roll material. Specifically, defect features are extracted from the preprocessed surface condition information. Algorithms such as edge detection and grayscale gradient analysis are used to identify bare surface defects such as scratches, pits, oxide spots, particle residues, and roller marks. The spatial coordinates, area, shape, and depth information of each defect are obtained. Subsequently, the extracted defect information is mapped to the roll material's two-dimensional coordinate system, and its position is calibrated along the length and width directions to form a two-dimensional coordinate system containing defect location, type, and severity level. Finally, a bare surface defect distribution map of the target roll material is generated based on this two-dimensional coordinate system. Furthermore, the surface defect distribution map of the bare board is used to determine whether the target roll material meets the lamination conditions. The area, depth, number, and distribution density of defects such as scratches, pits, roller marks, particle residue, and oxide spots recorded in the surface defect distribution map are quantitatively analyzed, and each defect indicator is compared against preset process thresholds. If the defect depth exceeds the maximum allowable fluctuation threshold for lamination, the area of ​​a single defect exceeds the upper limit of the lamination area, the number of defects per unit area exceeds the preset density threshold, or the defects are concentrated in sensitive locations such as critical lamination areas or laser cutting path areas, then the target roll material is determined not to meet the lamination conditions. The preset process thresholds can be set according to customer specifications.

[0028] When the target roll material is determined to meet the lamination process requirements based on the bare board surface defect distribution map, the control system sends an execution command to the lamination device to drive the target roll material into the single-sided laminator and complete the lamination process. After the target roll material passes through the lamination unit, multiple types of online sensors installed after the lamination section perform real-time detection on the surface of the laminated roll material. These sensors include at least one of capacitive sensors, eddy current sensors, and photoelectric sensors, used to collect the electrical response signals of the target roll material in the lamination area. The electrical response signals can reflect the changes in dielectric constant, conductivity differences, electromagnetic coupling characteristics, and thickness uniformity of the lamination layer in local areas, thereby characterizing whether there are bubbles, voids, incomplete adhesion, wrinkles, or other lamination quality defects in the lamination area.

[0029] Subsequently, microbubble defect information within the coated area is identified based on the electrical response signals. A digital coated defect map is generated by combining the microbubble defect information with the bare board surface defect distribution map. First, the collected electrical response signals of various types, such as eddy current and capacitance, are normalized to construct a two-dimensional electrical characteristic phase space with the normalized eddy current response signal as the horizontal axis and the normalized capacitance response signal as the vertical axis. Electrical response data within the composite harmonic cycle obtained under the same time reference are plotted as closed trajectory loops to form electrical signal trajectories. Trajectory distortion analysis is performed on the electrical signal trajectories to obtain distortion characteristic parameters. The similarity index between the distortion characteristic parameters and the baseline trajectory is used to determine whether the local electrical characteristics of the coated area are abnormal. When the similarity index is lower than a preset defect threshold, microbubble defects are identified at the corresponding location. The microbubble defects are then classified according to the distortion characteristic parameters to obtain microbubble defect information. Furthermore, microbubble defect information is mapped to the spatial coordinate system of the coated area and aligned with the bare board surface defect distribution map to form the same spatial reference system. Based on this, the microbubble defect information is superimposed onto the bare board surface defect distribution map, and different types of microbubble defects are rendered hierarchically. When coated microbubble defects and bare board surface defects overlap spatially, the area is marked as a high-risk overlap area, and the corresponding composite risk index is calculated. Thus, a digital coated defect map is formed based on the fusion of multi-source defect information, which is used to achieve a visual presentation of coated quality and the optimization of subsequent control strategies.

[0030] After acquiring a digital map of the coating defects of the target roll material, an air escape path construction step is further performed to identify the spatial distribution characteristics of microbubble defects in the film bonding interface and establish air escape paths on the surface of the target roll material. Specifically, each microbubble defect type in the digital coating defect map is mapped to an air resistance value, and all resistance values ​​are integrated to form an air resistance matrix on the surface of the target roll material. Subsequently, based on this resistance matrix, connectivity analysis is performed on the roll material surface, and a region growing algorithm is used to divide the surface into air escape regions. When the distance between any two air escape regions is less than a preset connectivity threshold, they are determined to constitute an air connectivity region, and the defect boundary points of this connectivity region are used as candidate nodes to construct a graph structure to establish adjacency relationships between candidate nodes. Further, based on this adjacency relationship, a shortest path algorithm is used to select the local escape path with the minimum air resistance from the air connectivity region, and a global air resistance map of the target roll material is constructed based on this. Based on the global graph, the system uses a path stitching algorithm based on resistance gradients to determine the path with the minimum air resistance and uses this path as the air escape path on the surface of the target roll material. In this embodiment, a digital coating defect map is used to calculate the air resistance matrix of the target roll surface based on defect type and spatial distribution, and an air escape path is constructed based on this matrix. The air escape path reflects the local connectivity between the film and the substrate and the risk of gas retention. When the air resistance in a certain area is high, it means that there is a high risk of microbubble aggregation or film instability in that area. If the laser cutting path directly crosses this area, it may lead to obstruction of cutting energy transfer, deterioration of cutting edge quality, or slag accumulation. Therefore, when generating linkage early warning information, this application spatially matches the air escape path with the cutting path, avoids or compensates for high-risk areas, and thus generates a laser cutting path adjustment command.

[0031] By analyzing a digital map of coating defects constructed from the target roll material, corresponding linked early warning information is automatically generated based on the marked microbubble defect types, defect levels, and defect density distribution. This linked early warning information indicates potential quality risks the target roll material may face during subsequent processing. Warning content includes alerts for high-risk overlapping areas, warnings of reduced local structural strength, and potential coating peeling risks. After the linked early warning information is generated, the system further correlates the warning area with the air escape paths on the target roll material surface, identifying dangerous path segments that are prone to coating lifting, abnormal laser ablation, or thermal stress accumulation. Based on the resistance gradient and defect sensitivity of the dangerous path segments, the system automatically generates laser cutting path adjustment instructions, locally offsetting, speed compensating, or adjusting the energy of the original laser cutting trajectory to optimize the laser cutting path by avoiding high-risk defect areas and following air escape paths. Ultimately, by executing the laser cutting path adjustment instructions, the system achieves proactive avoidance of quality risks and stable and controllable cutting results during the laser cutting process of the target roll material.

[0032] The above technical solution firstly ensures that the target roll material is in the correct initial state and position before entering the processing flow by performing feeding and unwinding preparation steps. This lays the foundation for subsequent leveling, laminating, and cutting processes, effectively avoiding problems caused by improper initial positioning or preparation. When the target roll material passes through the leveling machine after the preparation step, acquiring key real-time data on the equipment and materials during the leveling process helps to promptly identify and correct potential problems in the leveling stage, improving production stability. By synchronously correcting the laminating position and feeding path of the target roll material, the accurate position of the roll material during the laminating process can be ensured, avoiding uneven laminating or laminating failures caused by positional deviations, improving laminating quality and accuracy, and reducing production costs. By acquiring information on the surface condition of the sheet material, defects on the roll material surface can be detected in a timely manner. Early identification of surface defects and corresponding measures can effectively reduce the generation of defective products caused by surface defects, improving the product qualification rate. By constructing a defect distribution map, the distribution of surface defects of the roll material can be intuitively understood, avoiding laminating roll material with serious defects, saving material and time costs, and improving laminating quality. When the target roll material enters the single-sided laminator for lamination, and electrical response signals are collected, changes during the lamination process can be monitored in real time, allowing for timely detection of problems. Based on the electrical response signals, microbubble defects can be identified, accurately detecting hidden defects such as microbubbles within the lamination area. This information is then integrated with bare board defect information to form a comprehensive digital lamination defect map, improving lamination quality. By constructing air escape paths, laser cutting paths can be rationally planned, reducing factors during the cutting process and improving cutting accuracy. Laser cutting path adjustment commands are generated by linking early warning information and air escape paths to perform laser cutting on the target roll material. This achieves linkage between lamination defect information and downstream laser cutting, dynamically optimizing the cutting path, reducing cutting defects and material waste, and improving cutting accuracy and production line stability.

[0033] In one embodiment of this invention, the method further includes: S210. After laser cutting is completed, the robot is controlled to grab the cut target roll for stacking and the waste material of the cut target roll is transported to the waste collection box through the belt conveyor. S220. After completing the stacking process of the target roll material, start the cycle control program of the laser leveling and blanking line to realize the continuous processing process of the new roll of target roll material.

[0034] In this embodiment, after the target roll material completes the laser cutting process, the system controls an industrial robot to perform a gripping action. The robot is equipped with a multi-suction cup end effector, which can identify the finished product area of ​​the target roll material based on the cutting completion signal. Following a preset gripping path and stacking rules, it grips each piece of the cut target roll material and neatly stacks it onto a designated stacking platform or pallet for subsequent handling, packaging, or warehousing. Simultaneously, the system controls a belt conveyor to automatically transport the scraps, skeleton residue, and non-finished parts generated during the cutting process to a waste collection bin at the rear, achieving centralized collection and unified disposal of waste. The robot gripping and belt conveying processes can be performed synchronously, thereby achieving integrated automated operation of cutting, finished product handling, and waste recycling, improving production efficiency and reducing the risks of manual operation.

[0035] Once the target roll material has completed the laser cutting process and finished product stacking, the system automatically detects that the current processing flow has ended and triggers a loop control program to achieve continuous operation of the entire line. The loop control program includes: controlling the robot to complete the gripping and stacking of the last cut finished product, confirming that the stacking platform status meets the termination conditions; simultaneously, controlling the belt conveyor to transport non-finished parts such as edge scraps and skeleton residue generated during the cutting process to the waste collection box at the tail end for centralized waste processing. After the finished product and waste processing actions are completed, the system will start the next processing cycle, controlling the loading trolley to transport a new roll of material to the uncoiler, and completing preparatory actions such as roll installation, pressing, unwinding, traction, and shearing, so that the new roll enters the leveling section and re-collects operating data and roll status data.

[0036] By performing the above steps, manual waiting and repetitive operations can be effectively avoided, the roll change cycle time can be shortened, the overall line operating efficiency can be improved, and labor costs can be reduced.

[0037] In one embodiment of this invention, real-time operating data includes feed roller speed and film tension data, and roll material status data includes actual roll material speed, roll material edge trajectory data, and roll material stress distribution data. The real-time operating data and roll material status data are combined to synchronously correct the coating position and feeding path of the target roll material, including: S310. Perform polynomial fitting based on the roll material edge trajectory data to obtain the roll material edge fitting result; S320. Determine the lateral alignment deviation of the target roll material based on the edge fitting results; S330. Calculate the centerline of the target roll material based on the roll material edge trajectory data, and calculate the deviation distance and center deviation fluctuation frequency between the roll material centerline and the reference centerline. S340. Calculate the membrane tension fluctuation frequency by combining the membrane tension data, and determine the cause of the deviation by the center deviation fluctuation frequency and the membrane tension fluctuation frequency. S350. Simultaneously correct the lamination position of the target roll material by combining the deviation distance and the cause of the deviation; S360: The feeding path is corrected by combining the stress distribution data of the coil material and the lateral alignment deviation.

[0038] In this embodiment, real-time operating data refers to the dynamic operating parameters collected in real time by the equipment control system during the operation of the laser leveling and blanking line. This data reflects the current operating status and control behavior of the process equipment. Real-time operating data includes, but is not limited to, feed roller speed data and film tension data. Feed roller speed data refers to the rotational speed of the feed roller per unit time, used to determine the sheet material feeding speed and feeding stability. Film tension data refers to the tension value of the film material during unwinding and lamination in the laminating unit. Roll material status data refers to the physical state parameters related to the current processing target roll material. Roll material status data includes, but is not limited to, actual roll material speed, roll material edge trajectory data, and roll material stress distribution data. Actual roll material speed refers to the actual linear speed of the roll material in the production line. Roll material edge trajectory data refers to the trajectory data of the roll material's edge position changing over time during operation. Roll material stress distribution data refers to the spatial distribution of residual stress on the surface of the roll material before or during processing.

[0039] First, polynomial fitting is performed on the edge trajectory data of the rolled material to obtain the edge fitting result. By collecting the edge trajectory data of the target rolled material during processing, polynomial curve fitting is performed on the actual position changes of the left and right edges of the rolled material along the length direction. Polynomial fitting can smooth the edge trajectory of the rolled material to obtain a continuous and stable edge fitting result, which is used to characterize the edge shape and offset trend of the rolled material over the entire processing length. For example, the system collects the following edge trajectory points as longitudinal positions: 0, 100, 200, 300; and lateral positions: 100.2, 100.5, 101.0, 101.8. The system will use these points to perform polynomial fitting, such as X(Y) = aY 3 +bY 2 +cY+d, finally we get a=1.67×10 -8 b = 5 × 10 -6 c = 2.3 × 10 -3 d=100.2.

[0040] Secondly, the lateral alignment deviation of the target roll material is determined based on the edge fitting results. The system compares the fitted left and right edge curves with the lateral reference line of the processing equipment. By calculating the deviation between the roll material edge and the reference position, the lateral offset information of the target roll material during processing is obtained. The lateral reference line of the processing equipment can be determined based on the geometric center of the equipment, the center position of the standard roll material width, the guide rail of the feeding guide mechanism, and the initial roll material alignment calibration results. Specifically, several sampling points along the length of the roll material are selected, and the deviation between the roll material edge and the lateral reference line of the processing equipment at each sampling point is calculated using the fitted left and right edge curves, thus obtaining the lateral deviation value of the roll material center. Statistical analysis of the center deviations of all sampling points, such as calculating the average value, yields the lateral alignment deviation of the target roll material along the entire processing length. The lateral alignment deviation can be used as the control basis for the correction device, enabling real-time lateral position adjustment of the roll material on the processing equipment. It can also serve as a positioning reference for lamination, laser cutting, etc., thereby ensuring the stability of the lateral alignment accuracy during the roll material processing process.

[0041] The centerline of the roll material is calculated based on the left and right edge trajectory data during processing. This involves averaging the left and right edge coordinates at each longitudinal position to obtain a continuous center position curve along the roll's length, i.e., the centerline. Subsequently, the centerline is compared with a reference centerline set by the processing equipment or process to calculate the lateral deviation distance, characterizing the overall lateral offset of the roll material during processing. Simultaneously, frequency analysis is performed on the deviation of the centerline as it changes longitudinally to obtain the fluctuation frequency of the center deviation. Specifically, the lateral deviation of the target roll material at each longitudinal sampling point is calculated based on the centerline and the reference centerline of the processing equipment, forming a deviation curve. This deviation curve is input as a one-dimensional signal into a Fast Fourier Transform (FFT) to analyze the main periodic changes and corresponding frequencies in the deviation signal, thus obtaining the center deviation fluctuation frequency. This frequency reflects the lateral movement stability of the roll material during operation. The lateral deviation distance and the deviation fluctuation frequency can serve as the control basis for an automatic correction system, guiding the real-time lateral adjustment of the roll material on the processing equipment.

[0042] The system calculates the membrane tension fluctuation frequency by combining membrane tension data and determines the cause of deviation by using the center deviation fluctuation frequency and membrane tension fluctuation frequency. Specifically, based on real-time acquired membrane tension data, the system uses the Fast Fourier Transform (FFT) method to extract its frequency domain characteristics and calculate the membrane tension fluctuation frequency. Simultaneously, based on the roll edge trajectory fitting results, the system calculates the time-varying curve of the roll centerline offset relative to the equipment reference centerline and further extracts the center deviation fluctuation frequency. The system compares and analyzes these two types of frequencies. If the two frequencies are consistent or significantly correlated, the deviation is determined to be mainly caused by membrane tension disturbance; if the two frequencies are inconsistent or have a low correlation, the deviation is determined to be mainly caused by the active offset behavior of the roll body, such as residual stress release or geometric inhomogeneity of the sheet material. Through this frequency comparison analysis method, the system can quickly identify the cause of coating deviation, providing a decision-making basis for subsequent coating tension adjustment or feeding path correction.

[0043] Subsequently, the coating position of the target roll material is synchronously corrected based on the deviation distance and its cause. Specifically, when the system detects a misalignment between the roll material's centerline or edge line and the processing equipment's reference centerline, it automatically adjusts the position of the roll material or coating machine based on the deviation analysis results, achieving real-time alignment between the roll material's center and the film material's center. Specifically, when the deviation is mainly caused by film tension fluctuations, the film tension or conveyor speed can be adjusted appropriately to reduce the deviation; when the deviation is mainly caused by uneven roll material edges or equipment vibration, the roll material's position can be corrected using a roll material correction device or mechanical adjustment. This synchronous correction process ensures coating uniformity and coverage accuracy, reduces the risk of bubbles and wrinkles, and improves processing accuracy and product quality stability.

[0044] By analyzing the stress distribution data of the target roll material during processing, the system identifies stress concentration areas along its length and width. Simultaneously, it determines the actual deviation of the roll material during feeding by considering the lateral deviation of the roll material's centerline or edge line relative to the processing equipment's reference centerline. Based on the analysis results, the system adjusts the position of the roll material feeding trajectory, ensuring the target roll material runs along an ideal path. This reduces lateral offset and stress concentration, preventing wrinkles, uneven stretching, or localized damage. Feeding path correction guides lamination, laser cutting, and other subsequent processing steps, ensuring lateral alignment accuracy and product quality stability during roll material processing.

[0045] By comprehensively analyzing real-time operational data and roll material status data, and synchronously correcting the lamination position of the target roll material based on the overall deviation distance and causes of the deviation, the precise positioning of the lamination was ensured. Furthermore, the feeding path was corrected by combining roll material stress distribution data and lateral alignment deviation, optimizing the feeding path and improving lamination quality, thereby effectively enhancing the accuracy of the entire lamination process.

[0046] In one embodiment of this invention, the feeding path is corrected by combining the coil stress distribution data and the lateral alignment deviation, including: S410. Determine the lateral offset trajectory of the target roll material based on the lateral alignment deviation; S420. By superimposing the stress distribution data of the target roll material with the lateral offset trajectory, it is determined whether there is an offset aggravation segment corresponding to the high stress area. S430. If there is a high-stress region corresponding to an aggravated offset segment, it is determined to be an active offset for residual stress release. S440. Use the preset reverse compensation amount to perform reverse compensation on the offset aggravation section, so as to adjust the feeding path and correct the target roll material.

[0047] Determining the lateral offset trajectory of the target coil material based on lateral alignment deviation involves using the lateral deviation values ​​of the coil's centerline or edge line at various longitudinal sampling points relative to the processing equipment's reference centerline as lateral position coordinates. These coordinates are then used to form a continuous curve along the coil's length, resulting in the lateral offset trajectory. This trajectory visually reflects the lateral movement and offset trend of the coil material during feeding or processing. Furthermore, it can be used to analyze the causes of lateral deviation, including factors such as uneven coil tension, uneven edges, or equipment vibration, thereby guiding tension control optimization.

[0048] By superimposing the stress distribution data of the target coil material with the lateral offset trajectory in the longitudinal direction, the offset characteristics of the coil material during processing can be determined. Specifically, the stress values ​​of each longitudinal sampling point of the coil material are matched with the lateral offset at the same location. When a section of the coil material has a high stress value and the corresponding lateral offset is significantly increased, this section can be identified as a high-stress area corresponding to an intensified offset section. The analysis results can be used to identify key locations where the coil material is prone to problems such as deviation, wrinkles, or tension concentration during feeding or processing.

[0049] By overlaying and analyzing the stress distribution data and lateral offset trajectory of the target coil material, when a section of the coil material exhibits both high stress values ​​and a corresponding significant increase in lateral offset, it can be determined that the offset in this section is an active offset caused by the release of residual stress. Active offset is due to the gradual release of internal stress remaining during the manufacturing, winding, or preceding processing of the coil material in subsequent processing, rather than being caused by external mechanical interference or feeding errors. By identifying the active offset section caused by the release of residual stress, appropriate tension pre-adjustment, gradual correction, or dynamic feeding strategies can be adopted for this section of the coil material to reduce the impact of lateral movement of the coil material on lamination, laser cutting, and other subsequent processing steps, thereby improving the lateral stability of the coil material processing and the reliability of product quality.

[0050] Subsequently, when the system determines that there is a high-stress area corresponding to an aggravated offset segment in the target roll material based on the superposition analysis of the roll material stress distribution data and the lateral offset trajectory, the control system calculates the corresponding reverse compensation amount based on the offset amplitude, offset gradient, and duration of the aggravated offset segment. The reverse compensation amount is used to indicate the amount of active adjustment that the feeding path needs to make in the opposite direction of the offset to counteract the lateral offset trend of the roll material caused by the release of residual stress. Specifically, the system extracts the maximum offset amount Δx_max and the offset growth rate k of the aggravated offset segment (e.g., by calculating the first derivative of the lateral offset curve), and combines them with a preset compensation coefficient α. The preset compensation coefficient can be determined comprehensively based on factors such as the historical offset response data of the target roll material. The final compensation value is obtained by calculating the reverse compensation amount C = α·Δx_max. Then, the control system outputs a compensation control command to the feeding mechanism (such as an active straightening roller, a servo-side guide mechanism, or a lateral positioning slide) to move the feeding path in the opposite direction of the offset by the compensation amount C. During the compensation process, the system continuously monitors the trend of the centerline change of the compensated roll material; when the centerline offset returns to the preset stable threshold range, the compensation amount is gradually reduced to zero to achieve a smooth return to its original position.

[0051] By adjusting the feeding path and correcting the target roll material, the problem of roll material deviation caused by residual stress can be effectively solved, the stability of the roll material during the feeding process can be improved, quality defects caused by deviation can be avoided, thereby improving production efficiency and product quality, and ensuring the accuracy and reliability of roll material processing.

[0052] In one embodiment of this invention, microbubble defect information within the coated area is identified based on electrical response signals, and a digital coated defect map is generated by combining the microbubble defect information with a bare board surface defect distribution map, including: S501. Normalize the electrical response signal; S502. Construct a two-dimensional electrical characteristic phase space, and use the normalized eddy current response signal as the horizontal axis of the electrical characteristic phase space and the normalized capacitance response signal as the vertical axis of the electrical characteristic phase space. S503. Map the electrical response signal to the phase space of electrical characteristics. Within the period window of any composite harmonic signal, draw the electrical response signals collected under the same time reference into a closed trajectory loop to obtain the electrical signal trajectory. S504. Perform trajectory distortion analysis on the electrical signal trajectory to obtain distortion characteristic parameters; S505. Calculate the similarity index based on the distortion feature parameters and the baseline trajectory; S506. When the similarity index is lower than the preset defect threshold, it is determined that there is a microbubble defect in the corresponding coating area, and the microbubble defect is classified according to the type based on the distortion feature parameters to obtain microbubble defect information. S507. Determine the spatial location of microbubble defects in the coated area through microbubble defect information, and match the spatial location of microbubble defects in the coated area with the defect distribution map on the bare plate surface to form the same spatial reference system. S508. Under the same spatial reference system, superimpose the microbubble defect information onto the defect distribution map on the bare plate surface, and perform hierarchical rendering according to the type of microbubble defect. S509. When the defects of microbubbles in the coating overlap with the defects on the surface of the bare board in space, they are marked as high-risk overlapping areas, and the composite risk index is calculated. S510, combined with composite risk indicators, to obtain a digital coating defect map.

[0053] Figure 2 A schematic diagram of a digital coating defect map provided in this application embodiment, as shown below. Figure 2 As shown, this includes composite risk indicators corresponding to different microbubble defect information.

[0054] First, the acquired raw electrical response signals are normalized. These signals include, but are not limited to, capacitance, voltage, current, and eddy current response signals, and their original amplitude ranges vary significantly due to differences in sensor type, material properties, or manufacturing processes. The system employs a linear normalization algorithm or an interval mapping algorithm to uniformly transform various electrical response signals into a preset numerical range, thereby eliminating the influence of dimensional differences and amplitude shifts on subsequent analysis.

[0055] Secondly, a two-dimensional electrical characteristic phase space is constructed, with the normalized eddy current response signal as the horizontal axis and the normalized capacitance response signal as the vertical axis. This involves acquiring the eddy current response signal from the eddy current sensor located after the coating section and the capacitance response signal from the capacitance sensor, respectively; then, the two types of electrical response signals are normalized. Further, the normalized eddy current response signal is set as the horizontal axis of the electrical characteristic phase space, and the normalized capacitance response signal is set as the vertical axis. Under the same time reference, each pair of eddy current-capacitance response data points is mapped to a two-dimensional coordinate position in the phase space. As time-series data is continuously input, all mapped points form a continuous trajectory curve in the phase space. When the electrical response signal has a significant periodic component, the system segments the mapped points using a preset composite harmonic period window, causing the trajectory within each period window to automatically close, thus obtaining a closed electrical characteristic trajectory that characterizes the electrical coupling relationship of the coating area.

[0056] The electrical response signal is mapped onto the electrical characteristic phase space. Within the periodic window of any composite harmonic signal, the electrical response signals acquired under the same time reference are plotted as a closed trajectory loop to obtain the electrical signal trajectory. In the two-dimensional electrical characteristic phase space, using the period of the composite harmonic excitation signal as the time reference window, the eddy current response signal and the capacitance response signal acquired within the same periodic window with a unified time reference are mapped onto the two-dimensional phase space, and the mapping points are connected sequentially according to the sampling time sequence to obtain the closed trajectory loop formed within that period. The electrical signal trajectory refers to the closed curve formed by mapping the normalized eddy current response signal and the capacitance response signal to the two-dimensional electrical characteristic phase space according to a unified time reference within the preset composite harmonic excitation periodic window, and connecting them sequentially according to the sampling order. This curve characterizes the coupled response relationship between the conductivity and dielectric properties of the target region within that period.

[0057] Subsequently, trajectory distortion analysis was performed on the electrical signal trajectory to obtain distortion characteristic parameters. Using the closed electrical signal trajectory obtained in the electrical characteristic phase space, parameters such as the envelope area, major-minor axis ratio, trajectory closure, trajectory centroid offset, and number of local abrupt changes were calculated. This allows for the quantitative characterization of the geometric changes and spatial offset of the trajectory relative to a preset standard reference trajectory. Numerical calculation software (such as MATLAB) was used for calculation and extraction, comparing the closed electrical signal trajectory formed in the two-dimensional electrical characteristic phase space with the preset standard reference trajectory. The preset standard reference trajectory can be set according to scenario requirements and constraints. Next, the geometric shape and dynamic characteristics of the trajectory were quantitatively analyzed. Based on the quantification results, parameters describing trajectory distortion, i.e., distortion characteristic parameters, were extracted. Among them, the envelope area describes the overall variation amplitude of the electrical response intensity, the major-minor axis ratio reflects the stretching of the trajectory in different directions, the closure characterizes the periodic consistency of the composite signal, the centroid offset describes the overall drift trend of the signal, and the number of local abrupt changes and the degree of multimodal splitting are used to identify local abnormal responses. Based on the distortion characteristic parameters of the electrical signal trajectory and a pre-established standard reference trajectory, a similarity index is calculated to quantify the degree of anomaly in the electrical response of the target area. The distortion characteristic parameters of the actual trajectory are compared one by one with the corresponding characteristic parameters of the reference trajectory, and the similarity index is calculated using methods such as Euclidean distance. The similarity index characterizes the morphological deviation of the actual electrical signal trajectory from the reference trajectory. A low similarity index value indicates significant trajectory distortion, suggesting the possible presence of microbubbles or other local defects in the coated area. The pre-established standard reference trajectory is a reference trajectory obtained by acquiring and analyzing electrical signals from the target roll material coated area under normal conditions.

[0058] When the similarity index is lower than a preset defect threshold, the system can determine that a microbubble defect exists in the corresponding coated area. Further, based on the distortion characteristic parameters of the electrical signal trajectory, the system extracts amplitude distortion parameters and phase distortion parameters. The amplitude distortion parameters include eddy current axis compressibility and capacitance axis compressibility, while the phase distortion parameters include the phase rotation angle. By judging the distortion parameters according to preset thresholds, the system can classify the microbubble defect type: when the capacitance axis compressibility exceeds a first capacitance compression threshold and the eddy current axis compressibility is less than a first eddy current compression threshold, the microbubble defect type is determined to be an air-type microbubble; when the capacitance axis compressibility exceeds a second capacitance compression threshold and the phase rotation angle exceeds a phase rotation threshold, the microbubble defect type is determined to be a water vapor-containing microbubble; when the eddy current axis compressibility exceeds a third eddy current compression threshold and the capacitance axis compressibility is less than a second capacitance compression threshold, the microbubble defect type is determined to be a microbubble induced by under-film metal microparticles. The preset defect thresholds and threshold parameters used for microbubble classification, such as the first capacitance compression threshold, the second capacitance compression threshold, the first eddy current compression threshold, and the phase rotation threshold, are not arbitrarily set, but are obtained based on experimental calibration results of the coating structure characteristics and electrical response laws. Specifically, standardized coated samples containing different types of typical defects (air bubbles, water vapor bubbles, under-film metal microparticles, etc.) are constructed. Multiple scans of the samples are performed using a multi-array eddy current-capacitance composite sensing device to obtain the electrical response trajectories of different defect categories in the amplitude and phase directions. After normalizing the acquired electrical waveforms, the system statistically analyzes the distribution ranges of the eddy current axis compression ratio, capacitance axis compression ratio, and phase rotation angle corresponding to different defect categories. Box plot analysis is used to extract the feature intervals with the highest discriminative power. The corresponding classification thresholds are determined using the minimum false positive rate criterion and the maximum inter-class variance criterion. The thresholds obtained in this way have clear physical meaning and statistical basis, ensuring the separability of different types of microbubble defects in the electrical feature space. To determine the preset defect threshold, this application employs a normal distribution fitting or percentile thresholding method, using the upper limit of the 95% or 99% confidence interval as the defect judgment threshold. This ensures that when the similarity index is below this threshold, the difference between the detected area and the standard defect-free template has reached a significant level, thus reliably identifying it as a microbubble defect. Through the above method, quantitative classification information of microbubble defects within the target coating area can be obtained.

[0059] Based on microbubble defect information, the spatial location of microbubbles within the target roll material coating area is determined, including the specific coordinates of the microbubbles in the longitudinal roll material length direction and the transverse roll material width direction. The spatial location is determined by mapping the sampling points of the electrical signal trajectory to the actual roll material coordinates. Subsequently, the system matches the spatial location of the microbubble defects in the coating area with a pre-constructed bare board surface defect distribution map. Through coordinate normalization, translation, and rotation correction, the microbubble defect information and the bare board surface defect information form a unified spatial reference system. Under this unified spatial reference system, the microbubble defect information can be superimposed on the bare board surface defect distribution map, and hierarchical rendering can be performed according to the type of microbubble defect, thereby clarifying the distribution characteristics of various microbubble defects in the coating area and their spatial relationship with bare board defects.

[0060] The microbubble defect information is overlaid in a unified spatial reference system with the bare plate surface defect distribution map. This involves mapping the vertical and horizontal spatial coordinates of the microbubble defects onto the bare plate defect distribution map, thus displaying the spatial correspondence between microbubble defects and bare plate defects in the same coordinate system. Furthermore, the system performs hierarchical rendering based on the type of microbubble defect. For example, it uses different colors, labels, or brightness levels to label air-type microbubbles, water-vapor-containing microbubbles, and microbubbles induced by under-film metal microparticles, thereby clearly displaying the distribution characteristics of different types of microbubble defects on the bare plate surface defect distribution map.

[0061] When the system detects spatial overlap between microbubble defects and bare board surface defects, it marks this area as a high-risk overlap region. Further, the system calculates a composite risk index based on the bare board defect level, microbubble defect type and quantity, and spatial overlap degree to quantify the defect severity of the high-risk overlap region. This composite risk index can be used to generate a digital coating defect map. Specifically, the system first calculates the spatial overlap So between the microbubble defect region and the bare board surface defect region, and extracts the severity weights Wb and Ws of the microbubble defects and bare board defects, respectively. Simultaneously, it sets a location sensitivity factor Lf based on whether the region is located in a critical functional area. Finally, the system calculates the composite risk index using the following formula: CRI = So·(Wb+Ws)·Lf CRI represents the composite risk index; Wb is calculated based on the area, trajectory distortion amplitude, and defect type of microbubbles, reflecting their impact on the integrity of the coating; Ws is comprehensively evaluated based on parameters such as the type, depth, area, and image contrast of bare board defects, reflecting their potential impact on substrate performance; Lf is set according to whether the defect location falls within a critical functional area (such as the welding area, conductive area, or positioning reference area), and a higher sensitivity weight is assigned if it falls within a critical area. The spatial overlap So is the normalized overlap area between the microbubble defect area and the bare board defect area in two-dimensional space, So=A. b ∩A s / A b ∪A s A b A represents the projection area of ​​the microbubble defect in the image. s The projection area of ​​the bare board defect in the image is represented by the value So, which ranges from 0 to 1. If there is complete overlap, So=1; if there is no overlap, So=0. It is neither a distance function nor an isolated parameter, but a geometric ratio that can be directly calculated from the image. The bare board defect severity weight Ws is calculated based on the normalized weighted result of the defect area and depth, using the formula Ws=wA·SA+wD·SD, where SA and SD are the normalized scores of area and depth, respectively, and wA and wD are weight coefficients, satisfying wA+wD=1. The location sensitivity factor Lf is used to characterize the importance of the defect area. If the defect is located in a critical functional area such as the welding area, conductive area, or positioning reference area, Lf=1.0 is set; otherwise, Lf=0.5 is set. Finally, the system uses the formula CRI=So×(Wb+Ws)×Lf to calculate the composite risk index, which is used to quantify the collaborative risk level of spatially overlapping areas. It should be noted that CRI focuses on risk assessment of overlapping areas; if So=0, then CRI=0. However, the system also defines single-defect risk indicators Rb=Wb×Lf and Rs=Ws×Lf for risk quantification in non-overlapping areas, thus ensuring the completeness and rationality of the overall risk assessment. In this embodiment, multiplication is used to characterize synergistic risk; that is, risk is amplified only when spatial overlap, defect severity, and location sensitivity coexist. This structure ensures that the risk indicators reflect the interactive effects of defects, rather than simply being additive.

[0062] Finally, the composite risk indicators are bound to their corresponding spatial coordinates to generate a digital coating defect map. Based on a two-dimensional coordinate system of the target roll material surface, the digital defect map labels the spatial location, defect type, severity weight, and composite risk indicators of each defect point in structured data format. It is then visualized using color gradients, risk level labels, or graphic symbols, and can be output in JSON, CSV, or image heatmap formats for use by subsequent linkage early warning modules, laser cutting path adjustment modules, or quality traceability systems. By combining composite risk indicators to generate a digital defect map, the system achieves accurate representation of the spatial distribution and risk level of coating defects, improving the intelligence level of defect identification and the response efficiency of process control.

[0063] The above steps can provide precise guidance for quality control and defect repair in the production process, effectively improving product quality and production efficiency, and reducing the defect rate.

[0064] In one embodiment of this invention, microbubble defect information is obtained by classifying microbubble defects based on distortion feature parameters, including: S610. Determine the amplitude distortion parameter and phase distortion parameter through distortion characteristic parameters, wherein the amplitude distortion parameter includes the compressibility in the eddy current axis direction and the compressibility in the capacitance axis direction, and the phase distortion parameter includes the phase rotation angle. S620. When the compressibility in the capacitor axis direction exceeds the first capacitor compression threshold and the compressibility in the eddy current axis direction is less than the first eddy current compression threshold, the microbubble defect type is determined to be an air-type microbubble. S630. When the compressibility of the capacitor axis exceeds the second capacitor compression threshold and the phase rotation angle exceeds the phase rotation threshold, the microbubble defect type is determined to be a water vapor microbubble. S640. When the eddy current axial compression ratio exceeds the third eddy current compression threshold and the capacitance axial compression ratio is less than the second capacitance compression threshold, the microbubble defect type is determined to be microbubble induced by under-film metal microparticles.

[0065] The amplitude distortion parameter and phase distortion parameter are determined by distortion characteristic parameters, that is, based on the distortion characteristic parameters of the electrical signal trajectory, the amplitude distortion parameter and phase distortion parameter are further determined. In this embodiment, the amplitude distortion parameter includes the eddy current axis compression ratio and the capacitance axis compression ratio, and the phase distortion parameter includes the phase rotation angle. The distortion characteristic parameters are obtained by analyzing the electrical signal trajectory (a closed trajectory composed of eddy current and capacitance response signals) and are used to characterize the local or overall deviation of the trajectory shape. The amplitude distortion parameter characterizes the degree of compression or stretching of the trajectory in the coordinate axis direction, that is, the deformation amplitude of the trajectory in the eddy current axis direction (X-axis) and the capacitance axis direction (Y-axis). The eddy current axis compression ratio is the proportion of contraction or stretching of the trajectory in the eddy current response direction, used to reflect the influence of the metal substrate on the electrical response; the capacitance axis compression ratio is the proportion of contraction or stretching of the trajectory in the capacitance response direction, used to reflect the influence of the coating layer or air gap on the signal. Phase distortion parameters characterize the overall rotation or phase shift of the trajectory; the phase rotation angle is the rotation angle of the trajectory relative to the standard reference trajectory, used to reflect the disturbance of the signal phase by microbubbles or water vapor defects.

[0066] When the capacitance-axis compression ratio exceeds a preset first capacitance compression threshold and the eddy current-axis compression ratio is less than a preset first eddy current compression threshold in the amplitude distortion parameters of the electrical signal trajectory in the coated area, the corresponding microbubble defect type is determined to be an air-type microbubble. Air-type microbubbles are mainly caused by residual air under the coating, resulting in a significant abrupt change in dielectric constant and a relatively weak change in conductivity. In the electrical characteristic phase space, they exhibit significant compression in the longitudinal direction while maintaining a stable morphology in the transverse direction. Since air-type microbubbles mainly affect the coating layer and capacitance response, and have a relatively small impact on the eddy current signal, this specific trajectory distortion pattern is formed. The first eddy current compression threshold and the first capacitance compression threshold can be determined through experimental calibration, historical experience, or a classification model based on machine learning.

[0067] When the compressibility along the capacitor axis exceeds the second capacitor compression threshold and the phase rotation angle exceeds the phase rotation threshold, the system determines that the microbubble defect type is a water vapor microbubble. Water vapor microbubbles are typically formed by residual moisture or water vapor media during the coating process, leading to a drastic change in the dielectric constant and a significant shift in the phase coupling relationship between the capacitance and eddy current signal. In the electrical characteristic phase space, this manifests as significant compression along the longitudinal axis and rotational distortion of the overall trajectory. Using this characteristic, the system can quantitatively identify and classify water vapor microbubbles within the coating area, providing a basis for subsequent defect analysis and coating quality assessment. In this embodiment, the second capacitor compression threshold and the phase rotation threshold can be determined through experimental calibration by collecting and analyzing a large number of samples under experimental conditions.

[0068] When the eddy current axis compressibility exceeds the third eddy current compression threshold and the capacitance axis compressibility is lower than the second capacitance compression threshold, the system determines that the microbubble defect type is a microbubble induced by sub-film metal microparticles. Microbubbles are usually caused by residual metal particles during the coating process, resulting in significant compression distortion of the eddy current response signal in phase space, while the capacitance response signal shows little change. In the electrical characteristic phase space, it exhibits significant compression in the transverse direction while maintaining a stable morphology in the longitudinal direction. That is, when the eddy current axis compressibility exceeds the third eddy current compression threshold and the capacitance axis compressibility is lower than the second capacitance compression threshold, it indicates that the trajectory is significantly compressed in the eddy current direction, but the capacitance direction is almost unaffected. Therefore, the microbubble defect type can be determined to be a microbubble induced by sub-film metal microparticles.

[0069] The above classification method can accurately distinguish different types of microbubble defects, providing a clear basis for subsequent defect handling and quality control, thereby improving the accuracy and efficiency of defect detection in the production process and ensuring the stability of product quality.

[0070] In one embodiment of this invention, the multi-type sensor array includes eddy current sensors and capacitive sensors, and before identifying microbubble defect information in the coated area based on electrical response signals, it includes: S710, apply composite harmonic signals to the eddy current probe coil in the eddy current sensor and the measurement circuit in the capacitance sensor respectively; S720: Obtains the time reference pulse provided by the main encoder of the laser leveling and blanking line; S730: When the time reference pulse is triggered, the eddy current response signal of the eddy current sensor and the capacitance response signal of the capacitance sensor are collected. S740, the eddy current response signal and the capacitance response signal are used as electrical response signals under the same time reference.

[0071] The system applies composite harmonic signals to the eddy current probe coil in the eddy current sensor and the measurement circuit in the capacitance sensor to excite the sensors and acquire the electrical response characteristics of the target material. The composite harmonic signal consists of multiple frequency components and is periodic. The eddy current probe coil, excited by the composite harmonic current, generates an alternating magnetic field to detect disturbances in the conductivity of the metal. The capacitance measurement circuit, excited by the composite harmonic voltage, senses changes in the dielectric constant to identify bubbles, water vapor, or interface anomalies under the membrane. By applying a composite harmonic signal with a uniform structure to the dual-channel sensors, synchronous acquisition of the electrical response signal and construction of the phase space trajectory can be achieved.

[0072] The system acquires the time reference pulse signal provided by the main encoder, which serves as a unified time reference for all control modules and data acquisition modules. The main encoder is mounted on the main drive shaft or key feeding mechanism of the blanking line. Its output time reference pulse possesses high stability and high resolution characteristics, and can be used to calibrate the position and time points of the target roll material during processing. The system receives this pulse signal through a hardware interface and uses it as a trigger to perform operations such as electrical response signal acquisition, image recognition, cutting control, and defect marking, thereby achieving high-precision synchronous control and data consistency assurance during the processing.

[0073] Triggered by a time reference pulse, the system simultaneously acquires eddy current response signals from both the eddy current sensor and the capacitance response signals from the capacitance sensor. The time reference pulse synchronizes the acquisition from both types of sensors, ensuring that the eddy current and capacitance response signals correspond to the same location on the target roll material at the same time point. This synchronous acquisition method allows the system to obtain multi-sensor electrical response data with strong temporal consistency and accurate spatial correspondence, providing a reliable data foundation for subsequent construction of the electrical characteristic phase space, trajectory closure analysis, and microbubble defect identification.

[0074] The eddy current response signal acquired by the eddy current sensor and the capacitance response signal acquired by the capacitance sensor are paired on the same time reference to form a joint electrical response signal. This joint electrical response signal can simultaneously reflect the state information of both the metal layer and the coating layer of the target roll material, ensuring that the signals from different sensors correspond to the same location in time. Through this synchronous pairing, the system can use the joint electrical response signal as input for the phase space construction and trajectory analysis of electrical characteristics.

[0075] By applying a composite harmonic signal and using the time reference pulse provided by the main encoder of the laser leveling and blanking line to synchronously acquire the electrical response signal, accurate identification of microbubble defect information in the coating area is achieved, improving the accuracy and reliability of the signal.

[0076] In one embodiment of this invention, constructing an air escape path on the surface of the target roll material based on a digital coating defect map includes: S810: Map each microbubble defect type in the digital coating defect map to an air resistance value, and integrate the air resistance values ​​into an air resistance matrix for the target roll surface; S820. Based on the air resistance matrix, perform connectivity analysis on the surface of the target roll material, and use the region growing algorithm to divide the surface of the target roll material into air escape regions. S830. When the distance between any two air escape regions is less than a preset connected component threshold, the air connected component is determined. S840. For any air-connected domain, the defect boundary points of the air-connected domain are used as candidate nodes, and the adjacency relationship between the candidate nodes is established using a graph structure. S850. Based on the adjacency relationship, the shortest path algorithm is used to screen the path with the least resistance from the air connectivity domain to obtain local air escape candidate paths; S860, Constructing a global air resistance map of the target roll material based on local candidate paths; S870. Combine the global air resistance map with the path stitching algorithm of the resistance gradient to determine the path with minimum air resistance; S880 uses the path of least air resistance as the air escape path.

[0077] First, for each type of microbubble defect in the digital coating defect map, a corresponding air resistance value is pre-set to quantify the degree to which microbubble defects hinder air escape. Based on the microbubble defect type, such as air-type microbubbles, water-vapor-containing microbubbles, and microbubbles induced by under-film metal microparticles, the corresponding air resistance value is determined by calling a pre-set resistance value table. According to the spatial coordinates of the defects on the roll surface, each resistance value is filled into the corresponding position to construct a two-dimensional air resistance matrix. This matrix is ​​used to characterize the spatial distribution of air escape capability on the target roll surface during the coating process, providing basic data support for subsequent air escape path mapping, microbubble risk warning, and bonding process optimization. The pre-set resistance value table is obtained through a combination of experimental measurements, flow field simulation, and empirical data. Airflow experiments are used to measure the air resistance per unit area for each defect type, and computational fluid dynamics simulations are used to calculate the resistance coefficient for different defect types.

[0078] Secondly, connectivity analysis is performed on the air resistance matrix to identify the connectivity between different regions on the roll surface along the air escape path. Connectivity analysis is used to determine whether regions with low resistance are spatially connected, thereby identifying channels through which air can flow smoothly. Subsequently, using a region growing algorithm, starting from a point where the resistance is below a preset threshold, the algorithm gradually expands to neighboring grid points with resistance also below the threshold, dividing the entire connected region into air escape regions.

[0079] When the distance between any two air-escapeable areas on the surface of the target roll material is less than a preset connectivity threshold, the system classifies these two air-escapeable areas as an air connectivity region. The preset connectivity threshold is used to determine whether adjacent air-escapeable areas are close enough to ensure smooth airflow between them, and can be determined based on the airflow capacity in the coated area. When this condition is met, adjacent air-escapeable areas are merged to form a larger air connectivity region, providing basic data for subsequent screening of local air escape paths and global minimum air resistance path planning.

[0080] For any identified air connectivity region, the boundary points of microbubble defects within that region are used as candidate nodes, and a graph structure is used to establish adjacency relationships between these candidate nodes. The air connectivity region is the escapeable air area identified in the air drag matrix using a region growing algorithm. Within this region, the system extracts the edge pixels of all microbubble defects as candidate nodes for constructing escape paths. The system further determines whether any two nodes are adjacent based on the spatial distance between candidate nodes, drag gradient changes, and connectivity conditions. If a preset distance threshold and drag difference tolerance are met, an edge is established in the graph structure to connect the corresponding nodes.

[0081] Based on the established candidate node adjacency graph structure, a shortest path algorithm is used to select the path with the least air resistance within the air connectivity domain, and this path is selected as a local air escape candidate path. The adjacency graph structure consists of candidate nodes formed by the boundary points of microbubble defects within the air connectivity domain and the connecting edges between them, with the edge weights corresponding to the air resistance between nodes. This is achieved by calling a shortest path algorithm (such as Dijkstra's algorithm or A*). The algorithm searches a graph structure for the path that minimizes the total air resistance from the starting node to the target node, thus optimizing the selection of local air escape channels. Local air escape candidate paths characterize the optimal air escape trajectory within that connected region.

[0082] After screening candidate air escape paths within multiple air connectivity domains, a global air resistance map of the target roll material is constructed based on these local candidate paths. The node coordinates and corresponding air resistance values ​​in each local path are mapped to a unified roll material surface coordinate system. An interpolation algorithm is then used to integrate the resistance values ​​in path intersections or overlaps, forming a two-dimensional air resistance distribution map covering the entire roll material surface. This global air resistance map characterizes the overall spatial distribution of the target roll material's air escape capability during the lamination process, providing a basis for pressure distribution optimization and laser cutting path adjustment.

[0083] By combining the global air resistance map of the target roll material surface, a path stitching algorithm based on the resistance gradient is used to determine the path with minimum air resistance. The global air resistance map is composed of multiple local air escape candidate paths, representing the spatial distribution of air escape capability across the entire roll material surface. By calculating the resistance gradient, the direction and magnitude of air resistance value changes in space are identified. Guided by this, a path stitching algorithm (such as the RRT-Connect stitching algorithm) is used to connect multiple local paths into a continuous channel. During the path stitching process, connected nodes in the direction of decreasing resistance gradient are prioritized, and the total resistance of all feasible paths is calculated. Finally, the path with the minimum air resistance is selected as the optimal air escape path for the target roll material surface, i.e., the path with minimum air resistance. The path with minimum air resistance is used to guide defect linkage early warning, bonding pressure distribution adjustment, and laser cutting path optimization, significantly improving the intelligence level and response time of coating quality control.

[0084] By analyzing the global air resistance map of the target roll surface, the path with the minimum total resistance from the roll's starting point to its exit point is selected and designated as the air escape path. The air escape path represents the optimal channel for airflow across the roll surface, effectively reducing the risk of air stagnation and coating defects.

[0085] This application also provides a machine-readable storage medium storing instructions that cause the machine to execute the above-described intelligent monitoring and early warning method for the entire process of a laser leveling and blanking line.

[0086] This application also provides a full-process intelligent monitoring and early warning system for laser leveling and blanking lines, including: The memory is configured to store instructions; and The processor is configured to retrieve instructions from memory and, when executing instructions, to implement the aforementioned intelligent monitoring and early warning method for the entire process of the laser leveling and blanking line.

[0087] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0091] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0092] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0093] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0094] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0095] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A fully intelligent monitoring and early warning method for a laser leveling and blanking line, characterized in that, An application is made in a laser leveling and blanking monitoring system, wherein the laser leveling and blanking monitoring system is used in a laser leveling and blanking line, and the laser leveling and blanking line includes a leveling machine, comprising: The steps for feeding and unwinding the target roll material are as follows; When the target roll material passes through the leveling machine after the preparation step, real-time operating data of the laser leveling and blanking line and the roll material status data of the target roll material are collected. By combining real-time operation data and roll material status data, the coating position and feeding path of the target roll material are simultaneously corrected. After the target roll material has undergone the correction process, obtain the surface condition information of the corrected target roll material. Based on the surface condition information of the board, construct a bare board surface defect distribution map of the target roll material, and determine whether the target roll material meets the coating conditions based on the bare board surface defect distribution map; If the conditions are met, the target roll material is controlled to enter the single-sided laminator for lamination, and the electrical response signal of the target roll material is collected by multiple types of sensors set after the lamination section. Based on the electrical response signal, microbubble defect information in the coating area is identified, and a digital coating defect map is generated by combining the microbubble defect information and the bare board surface defect distribution map. Construct air escape paths on the surface of the target roll material based on a digital coating defect map; A digital coating defect map is used to generate a linkage early warning information, and a laser cutting path adjustment command is generated based on the linkage early warning information and the air escape path to perform laser cutting on the target roll material.

2. The method according to claim 1, characterized in that, The method further includes: After laser cutting is completed, the robot is controlled to grab the cut target roll for stacking and the waste material of the cut target roll is transported to the waste collection box via a belt conveyor. After the target roll material is completed in the stacking process, the cyclic control program of the laser leveling and blanking line is started to realize the continuous processing process of the new roll of target material.

3. The method according to claim 1, characterized in that, The real-time operating data includes film tension data, and the roll state data includes roll edge trajectory data and roll stress distribution data. The synchronous correction of the coating position and feeding path of the target roll by combining the real-time operating data and roll state data includes: The edge fitting result of the roll material is obtained by performing polynomial fitting based on the roll material edge trajectory data; The lateral alignment deviation of the target roll material is determined by the edge fitting results; The centerline of the target roll is calculated based on the roll edge trajectory data, and the deviation distance and center deviation fluctuation frequency are calculated by combining the roll centerline and the reference centerline. The membrane tension fluctuation frequency is calculated by combining the membrane tension data, and the cause of the deviation is determined by the center deviation fluctuation frequency and the membrane tension fluctuation frequency. The lamination position of the target roll material is simultaneously corrected based on the deviation distance and the cause of the deviation; The feeding path is corrected by combining the stress distribution data of the rolled material and the lateral alignment deviation.

4. The method according to claim 3, characterized in that, The method of correcting the feeding path by combining the coil stress distribution data and lateral alignment deviation includes: The lateral offset trajectory of the target roll material is determined based on the lateral alignment deviation; By superimposing the stress distribution data of the target roll material with the lateral offset trajectory, it can be determined whether there is an offset aggravation segment corresponding to a high stress area. If there is a high-stress region corresponding to an aggravated offset segment, it is determined to be an active offset for residual stress release; The offset aggravation section is compensated by a preset reverse compensation amount to adjust the feeding path and correct the target roll material.

5. The method according to claim 1, characterized in that, The method of identifying microbubble defect information within the coated area based on electrical response signals, and generating a digital coated defect map by combining the microbubble defect information with the bare board surface defect distribution map, includes: The electrical response signal is normalized. A two-dimensional phase space of electrical characteristics is constructed, and the normalized eddy current response signal is used as the horizontal axis of the phase space of electrical characteristics and the normalized capacitance response signal is used as the vertical axis of the phase space of electrical characteristics. The electrical response signal is mapped onto the phase space of electrical characteristics. Within the period window of any composite harmonic signal, the electrical response signals collected under the same time reference are plotted as a closed trajectory loop to obtain the electrical signal trajectory. Trajectory distortion analysis is performed on the electrical signal trajectory to obtain distortion characteristic parameters; The similarity index is calculated based on the distortion feature parameters and the baseline trajectory; When the similarity index is lower than the preset defect threshold, it is determined that there is a microbubble defect in the corresponding coating area, and the microbubble defect is classified according to the type based on the distortion feature parameters to obtain microbubble defect information. The spatial location of microbubble defects in the coated area is determined by microbubble defect information, and the spatial location of microbubble defects in the coated area is matched with the defect distribution map on the bare plate surface to form the same spatial reference system; Under the same spatial reference frame, the microbubble defect information is superimposed on the defect distribution map of the bare plate surface, and the rendering is performed in a graded manner according to the type of microbubble defect. When the defects of the coated microbubble overlap with the defects of the bare plate surface in spatial location, it is marked as a high-risk overlapping area, and the composite risk index is calculated. A digital coating defect map is obtained by combining composite risk indicators.

6. The method according to claim 5, characterized in that, The method of classifying microbubble defects based on distortion feature parameters to obtain microbubble defect information includes: The amplitude distortion parameter and the phase distortion parameter are determined by the distortion characteristic parameters. The amplitude distortion parameter includes the compressibility in the eddy current axis direction and the compressibility in the capacitance axis direction, and the phase distortion parameter includes the phase rotation angle. When the compressibility in the capacitor axis direction exceeds the first capacitor compression threshold and the compressibility in the eddy current axis direction is less than the first eddy current compression threshold, the microbubble defect type is determined to be an air-type microbubble. When the compressibility of the capacitor axis exceeds the second capacitor compression threshold and the phase rotation angle exceeds the phase rotation threshold, the microbubble defect type is determined to be a water vapor microbubble. When the eddy current axial compressibility exceeds the third eddy current compressibility threshold and the capacitance axial compressibility is less than the second capacitance compressibility threshold, the microbubble defect type is determined to be microbubble induced by under-film metal microparticles.

7. The method according to claim 5, characterized in that, The multi-type sensor array includes eddy current sensors and capacitive sensors, and before identifying microbubble defect information in the coated area based on electrical response signals, it also includes: Composite harmonic signals are applied to the eddy current probe coil in the eddy current sensor and the measurement circuit in the capacitance sensor, respectively. Obtain the time reference pulse provided by the main encoder of the laser leveling and blanking line; When the time reference pulse is triggered, the eddy current response signal of the eddy current sensor and the capacitance response signal of the capacitance sensor are acquired. The eddy current response signal and the capacitance response signal are used as electrical response signals under the same time reference.

8. The method according to claim 1, characterized in that, The step of constructing the air escape path on the surface of the target roll material based on the digital coating defect map includes: Each microbubble defect type in the digital coating defect map is mapped to an air resistance value, and the air resistance values ​​are integrated into an air resistance matrix for the target roll surface; The connectivity of the target roll material surface is analyzed based on the air resistance matrix, and the target roll material surface is divided into air escape regions using the region growing algorithm. An air connectivity region is determined when the distance between any two air escape regions is less than a preset connectivity threshold. For any air-connected domain, the defective boundary points of the air-connected domain are used as candidate nodes, and the adjacency relationship between the candidate nodes is established using a graph structure. Based on the adjacency relationship, the shortest path algorithm is used to screen the path with the least resistance from the air connectivity domain to obtain local air escape candidate paths; Construct a global air resistance map of the target roll material based on local candidate paths; The path with minimum air resistance is determined by combining the global air resistance map with the path stitching algorithm of the resistance gradient. The path of least air resistance is taken as the air escape path.

9. A machine-readable storage medium, characterized in that, The machine-readable storage medium stores instructions for causing the machine to execute the intelligent monitoring and early warning method for the entire process of the laser leveling and blanking line according to any one of claims 1 to 8.

10. A fully intelligent monitoring and early warning system for a laser leveling and blanking line, characterized in that, include: The memory is configured to store instructions; as well as The processor is configured to retrieve the instructions from the memory and, when executing the instructions, to implement the intelligent monitoring and early warning method for the entire process of a laser leveling and blanking line according to any one of claims 1 to 8.

Images