Detection method of galvanized automobile sheet circle shear and related equipment

Abstract

The invention discloses a galvanized automobile sheet circle shear detection method and related equipment, and relates to the technical field of industrial automatic detection, and the method comprises the steps: obtaining image data of a circle shear region; based on the image data, performing feature extraction on the image data to generate edge thread features; judging whether the edge wire state is in an abnormal state or not based on the edge wire characteristics; and when the edge wire state is in an abnormal state, the operation parameters of the circle shear are adjusted. According to the method and the device, equipment jamming caused by edge wire accumulation or strip steel scratching caused by escape are effectively avoided, the real-time performance and the accuracy of anomaly detection are improved, the manual intervention requirement is reduced, meanwhile, the equipment running state is optimized through closed-loop control, and the production continuity and the product quality are guaranteed.

Description

Technical Field

[0001] The present application relates to the field of industrial automation detection technology, and in particular to a detection method and related equipment for a galvanized automobile sheet disc shear. Background Art

[0002] In the production of galvanized automotive sheet, the disc shearing process is a critical step in ensuring the dimensional accuracy of the strip. However, the edge wires produced during the trimming process are prone to clogging or escaping, leading to waste accumulation or scratches on the strip surface, seriously affecting product quality and production continuity. Existing technologies often rely on manual inspections or contact sensors to monitor the edge wire status. These technologies suffer from problems such as detection lag and high misjudgment rates, making it difficult to meet the real-time and stability requirements of high-speed production lines. Therefore, a detection method for disc shears of galvanized automotive sheet is urgently needed to address the above-mentioned technical issues. Summary of the Invention

[0003] The Summary of the Invention introduces a series of simplified concepts that will be further described in the Detailed Description of the Invention. The Summary of the Invention of this application is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0004] In a first aspect, the present application provides a method for detecting a galvanized automobile sheet disc shear, comprising:

[0005] Obtain image data of the disc shearing area;

[0006] Based on the image data, feature extraction is performed on the image data to generate edge features;

[0007] Based on the edge wire characteristics, determine whether the edge wire status is in an abnormal state;

[0008] When the edge wire status is in an abnormal state, adjust the operating parameters of the disc shear.

[0009] In some embodiments, the image data includes drive-side image data and operating-side image data, and acquiring image data of the disc shearing area includes:

[0010] High-resolution industrial cameras are used to collect the video stream data of the drive side and the working side of the circular shear respectively;

[0011] Based on the preset frame extraction rules, the driving side video stream data and the working side video stream data are subjected to frame extraction processing to generate the driving side target image sequence and the working side target image sequence;

[0012] A preprocessing operation is performed on the driving side target image sequence and the working side target image sequence to generate driving side image data and operating side image data.

[0013] In some embodiments, based on the image data, feature extraction is performed on the image data to generate edge features, including:

[0014] Based on the image data, the morphological features of the edge wire are extracted through the segmentation parameters of the image semantic segmentation algorithm. The morphological features include the edge wire contour information and edge wire position information. The segmentation parameters include the number of network layers, learning rate and feature fusion weight.

[0015] Based on the shape features, the motion vector of the edge wire is calculated by the optical flow method to generate the motion trajectory features of the edge wire;

[0016] The trajectory prediction parameters of the Kalman filter algorithm are used to predict and correct the motion trajectory characteristics to generate the corrected edge wire motion trajectory characteristics, wherein the trajectory prediction parameters include the state transfer matrix, the observation noise covariance and the process noise covariance;

[0017] The morphological features are integrated with the edge wire motion trajectory features to generate edge wire features.

[0018] In some embodiments, the abnormal state includes a blocking abnormal state or an escaping abnormal state. Based on the edge wire characteristics, determining whether the edge wire state is in an abnormal state includes:

[0019] Based on the edge wire motion trajectory characteristics and the preset guide groove path, the target path deviation degree is determined, and when the target path deviation degree is greater than a first preset threshold, the edge wire state is determined to be an escape abnormal state; or,

[0020] Based on the morphological characteristics and preset blockage judgment conditions, the target flow deviation and the target flow duration of the target flow deviation are determined. When the target flow deviation is less than the first preset flow threshold and the target flow duration is greater than or equal to the first preset duration, the edge wire state is determined to be a blockage abnormal state.

[0021] In some embodiments, when the edge wire state is in an abnormal state, adjusting the operating parameters of the disc shear includes:

[0022] When the edge wire state is an escape abnormal state, based on the target path deviation degree, a target first operating parameter corresponding to the target path deviation degree is determined from a preset mapping relationship between the path deviation degree and the first operating parameter, the first operating parameter including the speed reduction amplitude and the downtime duration of the disc shear; or,

[0023] When the edge wire state is an abnormal blockage state, based on the target flow deviation and the target flow duration, the target second operating parameters corresponding to the target flow deviation and the target flow duration are determined through a preset compensation model. The second operating parameters include the shear pressure compensation value and the vibration frequency compensation value of the disc shear.

[0024] In some embodiments, further comprising:

[0025] Determining a path adjustment amount of a first preset threshold based on a historical frequency of occurrence of the escape abnormal state;

[0026] Based on the path adjustment amount, the first preset threshold is updated through a dynamic parameter optimization model to reduce the missed detection rate of escape anomalies;

[0027] Determining a flow adjustment amount for a first preset flow threshold and a duration adjustment amount for a first preset duration based on a historical occurrence frequency of the abnormal congestion state;

[0028] Based on the flow adjustment amount and the duration adjustment amount, the first preset flow threshold and the first preset duration are updated through a dynamic parameter optimization model to improve the detection rate of blockage anomalies.

[0029] In some embodiments, further comprising:

[0030] Based on the abnormal type of the edge wire status and the adjusted operating parameters of the disc shear, an abnormality handling record containing the abnormality type identifier, parameter adjustment value and timestamp is generated;

[0031] Based on exception processing records, segmentation parameters are optimized through machine learning models to improve the accuracy of edge wire contour and position extraction;

[0032] Based on the exception handling records, the trajectory prediction parameters are optimized through the machine learning model to reduce the motion trajectory prediction error.

[0033] In a second aspect, the present application proposes a detection device for a galvanized automobile sheet disc shear, comprising:

[0034] a disc shear image acquisition unit, for acquiring image data of the disc shear area;

[0035] The edge wire feature extraction unit extracts features from the image data based on the image data to generate edge wire features;

[0036] An abnormal state judgment unit, which judges whether the edge wire state is in an abnormal state based on the edge wire characteristics;

[0037] The disc shear parameter adjustment unit is used to adjust the operating parameters of the disc shear when the edge wire state is abnormal.

[0038] In a third aspect, an electronic device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for detecting galvanized automobile sheet disc shears according to any one of the first aspects when executing the computer program stored in the memory.

[0039] In a fourth aspect, the present application proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the detection method for the galvanized automobile sheet disc shear according to any one of the first aspects.

[0040] In summary, this application obtains image data from the circular shear area, extracts the edge wire morphology and motion trajectory characteristics based on image processing algorithms, dynamically determines whether the edge wire state is abnormal, and automatically adjusts the circular shear operating parameters according to the type of abnormality. This application effectively solves the problem of traditional detection technology relying on manual labor and poor real-time performance, improves the detection accuracy and response speed of edge wire blockage and escape, reduces the risk of product quality defects and equipment damage caused by abnormal processing delays, and optimizes production stability through closed-loop control, providing reliable guarantees for the efficient and continuous production of galvanized automotive sheet metal. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0042] Figure 1 A schematic flow chart of a detection method for a galvanized automobile sheet disc shear provided in an embodiment of the present application;

[0043] Figure 2 Schematic diagram of the detection interface of the galvanized automobile sheet disc shear provided in the embodiment of the present application;

[0044] Figure 3 A schematic diagram of the structure of a detection device for a galvanized automobile sheet disc shear provided in an embodiment of the present application;

[0045] Figure 4 Schematic diagram of the structure of the detection equipment for the galvanized automobile sheet disc shear provided in the embodiment of the present application. DETAILED DESCRIPTION

[0046] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the embodiments described are only part of the embodiments of the present application, not all of the embodiments.

[0047] See also Figure 1 , which is a flow chart of a detection method for a galvanized automobile sheet disc shear provided in an embodiment of the present application, which may specifically include:

[0048] S110, acquiring image data of the disc shearing area;

[0049] For example, during the production of galvanized automotive sheet, image data from the circular shear area is collected in real time through non-contact detection. Specifically, high-resolution industrial cameras are used to simultaneously monitor the drive side, working side, and hopper outlet of the circular shear from multiple angles. Combined with LED light source fill-light technology, this ensures clear capture of the edge wire morphology and motion during high-speed shearing conditions. The camera layout strategy is designed based on the structural characteristics of the circular shear, covering key operating areas to comprehensively capture the dynamic behavior of the edge wire, providing basic data support for subsequent feature extraction and anomaly detection.

[0050] The captured video stream data is transmitted in real time via Ethernet to a central server, where a target image sequence is generated based on pre-set frame extraction rules. After preprocessing, the image sequence forms a standardized image dataset. This step optimizes data quality, eliminates the impact of environmental interference on image analysis, ensures the accuracy and reliability of subsequent algorithm processing, and lays a data foundation for monitoring edge wire condition.

[0051] S120, based on the image data, extracting features from the image data to generate edge features;

[0052] Exemplarily, the feature extraction process is implemented through multimodal image analysis technology. First, based on the image semantic segmentation algorithm, the morphological features of the edge wire are extracted from the image data of the disc shear area, including the geometric shape of the edge wire contour, edge clarity, and spatial position information. The algorithm uses a deep learning model to accurately distinguish the foreground (edge wire) and background (equipment structure and environment) in the image. Combined with a hierarchical feature fusion strategy, it enhances the ability to capture subtle deformations and positional offsets of the edge wire, providing high-resolution static feature data for subsequent abnormal state judgment.

[0053] The dynamic behavior of the edge wire is modeled and analyzed based on the optical flow method and the Kalman filter algorithm. The optical flow method calculates the motion vectors of the edge wire in adjacent frames and generates the edge wire's trajectory features. The Kalman filter algorithm predicts and corrects the trajectory data, eliminating noise interference and optimizing trajectory continuity. Finally, the topographic features are fused with the corrected trajectory features to form a comprehensive feature set representing the edge wire's state. This process enhances the ability to characterize edge wire blockage or escape behavior through the complementary temporal and spatial features, providing multi-dimensional dynamic data support for subsequent abnormal state judgment.

[0054] S130, judging whether the edge wire state is in an abnormal state based on the edge wire feature;

[0055] Exemplarily, the judgment of abnormal state is achieved by comprehensively analyzing the dynamic trajectory and static morphology data in the edge wire characteristics. For escape abnormality, the degree of deviation from the target path is calculated based on the matching degree between the edge wire motion trajectory characteristics and the preset guide groove path. When the degree of deviation exceeds the preset threshold, it is determined that the edge wire has deviated from the normal guide groove path, and the escape abnormality signal is triggered. For blockage abnormality, the edge wire flow at the hopper outlet is counted by morphological characteristics, and combined with the flow deviation and its duration, it is determined whether the waste flow is interrupted due to accumulation. If the flow continues to be lower than the safety threshold and is not restored after the timeout, it is determined to be a blockage abnormality.

[0056] This decision logic combines preset rules with dynamic thresholds to ensure accurate identification of abnormal conditions. The results are directly linked to the adjustment strategy for the shear's operating parameters, providing a basis for subsequent closed-loop control decisions. This allows for a rapid response to an anomaly at its earliest stages, preventing it from escalating.

[0057] S140: When the edge wire is in an abnormal state, adjust the operating parameters of the disc shear.

[0058] For example, when it is detected that the edge wire is in an abnormal state of escape or blockage, the system dynamically adjusts the operating parameters of the disc shear according to the type of abnormality to quickly eliminate the impact of the abnormality and restore production stability. For escape abnormalities, the speed reduction amplitude or shutdown duration of the disc shear is automatically adjusted through the mapping relationship between the preset path deviation degree and the operating parameters to suppress the edge wire from further deviating from the guide groove path; for blockage abnormalities, the shear pressure or vibration frequency is adjusted through the compensation model based on the flow deviation amount and duration to unclog the waste channel and optimize the edge wire flow efficiency. The above adjustment strategy ensures the coordination of abnormality handling and production rhythm through real-time feedback control, avoiding equipment damage or product quality defects due to delayed parameter adjustment.

[0059] In summary, the embodiment of the present application obtains image data of the disc shear area in real time, combines high-resolution industrial cameras with multi-angle synchronous monitoring technology, comprehensively captures the dynamic behavior of the edge wire on the driving side, working side and blanking hopper, uses image semantic segmentation algorithm to accurately extract the edge wire contour and position information, and uses optical flow method and Kalman filter algorithm to model and correct the motion trajectory to form a comprehensive feature set that integrates the morphology and dynamic trajectory. Based on the preset guide groove path matching degree and edge wire flow statistics, it dynamically judges whether the edge wire state is in an escape or blockage abnormality, and realizes early identification and accurate classification of abnormal states. Through a closed-loop control mechanism, the operating parameters such as the speed reduction amplitude, downtime duration, shearing pressure or vibration frequency of the disc shear are automatically adjusted according to the abnormality type, quickly eliminating the abnormal impact and optimizing the edge wire flow efficiency. The dynamic parameter optimization model and machine learning algorithm are introduced to continuously iteratively adjust the detection threshold, segmentation parameters and trajectory prediction parameters to improve the abnormality detection rate and algorithm robustness, and generate abnormality processing records to support the traceability and optimization of the production process. The embodiments of the present application effectively solve the problems of lag and high misjudgment rate in traditional manual inspection, improve the automation level and stability of the galvanized automobile sheet production line, reduce the scrap rate and equipment damage risk caused by abnormal edge wires, and provide technical support for quality control and efficiency improvement in high-speed continuous production environments.

[0060] In some examples, the image data includes driving side image data and operating side image data, and acquiring the image data of the disc shearing area includes:

[0061] High-resolution industrial cameras are used to collect video stream data from the drive side and working side of the circular shears;

[0062] Based on the preset frame extraction rules, the driving side video stream data and the working side video stream data are subjected to frame extraction processing to generate the driving side target image sequence and the working side target image sequence;

[0063] A preprocessing operation is performed on the driving side target image sequence and the working side target image sequence to generate driving side image data and operating side image data.

[0064] For example, in the production process of galvanized automobile sheets, the video stream data of the drive side (DS) and the working side (WS) of the disc shear are collected in real time through non-contact detection. Specifically, high-resolution industrial cameras are deployed on the drive side and the working side of the disc shear to ensure that the key areas of the entire process of edge wire generation, shearing and falling are covered. The drive side camera is used to monitor the initial motion state of the edge wire after shearing, and the working side camera is used to capture the dynamic trajectory of the edge wire entering the guide groove and the drop hopper. All cameras use LED light source fill light technology to eliminate image blurring problems caused by ambient light fluctuations or metal reflections under high-speed shearing conditions, ensuring that the collected video stream data has high definition and high frame rate characteristics. The video stream data is transmitted to the central server in real time via Gigabit Ethernet, providing the original data basis for subsequent processing.

[0065] To address the issue of massive amounts of video stream data in high-speed production line scenarios, preset frame extraction rules are used to downsample the video stream data on the drive and working sides. These frame extraction rules include a dynamic frame extraction strategy and a static frame extraction strategy. The dynamic frame extraction strategy dynamically adjusts the frame extraction frequency based on the edge wire speed. When the edge wire speed exceeds a preset speed threshold, the frame extraction rate is increased to capture key frames. The static frame extraction strategy generates a target image sequence based on a fixed time interval (for example, 10 frames per second). Through the above-mentioned rule processing, the target image sequences for the drive side and the working side are generated respectively, which reduces the computational load while retaining the key frame data, thereby improving the efficiency of subsequent feature extraction and analysis.

[0066] Standardized preprocessing is performed on the extracted drive-side and working-side target image sequences. A Gaussian filter algorithm is used to eliminate random noise in the industrial environment, and histogram equalization is used to enhance image contrast, highlighting the distinction between the edge wire outline and the background. Based on camera calibration parameters, perspective transformation and radial distortion correction are performed on the images to eliminate image distortion caused by camera installation angle or lens deformation. The image resolution is uniformly adjusted to a preset standard size (e.g., 512×512 pixels), and the original video stream format is converted to an algorithm-compatible RGB format. After this preprocessing, standardized drive-side and operating-side image data are generated, eliminating the negative impact of environmental interference on subsequent algorithm processing and ensuring the accuracy and consistency of edge wire topography feature extraction and motion trajectory analysis.

[0067] In summary, the embodiments of this application achieve efficient acquisition and standardized processing of image data from both the drive and working sides of a circular shear. Multi-angle synchronous monitoring combined with fill-light technology ensures comprehensive capture of the dynamic behavior of the edge wire. Preset frame extraction rules balance data integrity and processing efficiency. Preprocessing eliminates environmental interference and improves image quality. These three elements work together to provide reliable data input for subsequent feature extraction and abnormal state determination, thus supporting the high-precision and real-time requirements of the circular shear detection method for galvanized automotive sheet metal.

[0068] In some examples, based on the image data, feature extraction is performed on the image data to generate edge features, including:

[0069] Based on the image data, the morphological features of the edge wire are extracted through the segmentation parameters of the image semantic segmentation algorithm. The morphological features include the edge wire contour information and edge wire position information. The segmentation parameters include the number of network layers, learning rate and feature fusion weight.

[0070] Based on the shape features, the motion vector of the edge wire is calculated by the optical flow method to generate the motion trajectory features of the edge wire;

[0071] The trajectory prediction parameters of the Kalman filter algorithm are used to predict and correct the motion trajectory characteristics to generate the corrected edge wire motion trajectory characteristics, wherein the trajectory prediction parameters include the state transfer matrix, the observation noise covariance and the process noise covariance;

[0072] The morphological features are integrated with the edge wire motion trajectory features to generate edge wire features.

[0073] Exemplarily, based on image data, the morphological features of the edge wire are extracted through the segmentation parameters of the image semantic segmentation algorithm. The image semantic segmentation algorithm adopts a deep learning model, and its segmentation parameters include the number of network layers, learning rate and feature fusion weight. Specifically, the number of network layers is set to a preset value (for example, 128 convolutional layers) to construct a multi-scale feature extraction network; the learning rate is dynamically adjusted by an adaptive gradient descent algorithm, and the initial value is set to 0.01 to balance the model convergence speed and stability; the feature fusion weight is allocated through a cross-layer connection mechanism, and the weight value is optimized and determined based on the distribution characteristics of the edge wire contour in the training data. The morphological features include edge wire contour information and edge wire position information, wherein the contour information is extracted through the binary mask output by the segmentation algorithm, and the position information is obtained by converting the pixel coordinate system to the physical coordinate system. The morphological features are represented in the form of a binary mask and a coordinate matrix, providing a high-precision static data basis for subsequent motion trajectory analysis.

[0074] After obtaining the morphological features, the edge wire morphology in adjacent frame images is dynamically tracked based on the optical flow method to generate the motion trajectory features of the edge wire. Specifically, the Lucas-Kanade optical flow algorithm is used to calculate the pixel-level motion vector with the edge wire contour area in the adjacent frame images as input. The optical flow method performs local motion estimation at the key points of the edge wire contour (such as edge inflection point, center point) through a preset window size (for example, 15×15 pixels), and fits the global motion trajectory based on the weighted least squares method. The motion vector includes horizontal displacement, vertical displacement and instantaneous velocity. The generated motion trajectory features record the continuous motion state of the edge wire on the drive side, working side and blanking hopper outlet of the disc shear through time series data. This step uses non-contact motion capture technology to achieve high-frequency sampling and quantitative characterization of the dynamic behavior of the edge wire, providing original dynamic data for trajectory correction and abnormality judgment.

[0075] To eliminate noise interference potentially introduced by the optical flow method and improve the robustness of trajectory prediction, the trajectory features are predicted and corrected using the trajectory prediction parameters of the Kalman filter algorithm to generate a corrected trajectory feature for the edge wire. The trajectory prediction parameters of the Kalman filter include the state transition matrix, the observation noise covariance, and the process noise covariance. The state transition matrix defines the dynamic evolution of the trajectory state based on the Newtonian kinematic model and is used to predict the position and velocity of the edge wire at the next moment. The observation noise covariance and the process noise covariance quantify the sensor error and the uncertainty of the system model, respectively. Through two stages, prediction and update, the algorithm optimally estimates the edge wire trajectory. The observed trajectory generated by the optical flow method is integrated with the model-predicted trajectory to eliminate trajectory jumps caused by environmental noise or image jitter, and output a smooth and continuous corrected trajectory feature. This step improves the reliability and stability of the trajectory through dynamic noise suppression and trajectory optimization, providing high-confidence input for subsequent path deviation analysis.

[0076] The static morphological features of the edge wire are fused with the corrected edge wire motion trajectory features to generate a comprehensive feature set that characterizes the edge wire status. Specifically, the edge wire contour information and position information in the morphological features are associated with the time series data of the motion trajectory features through a spatial alignment algorithm to form a spatiotemporal fusion matrix. The fusion matrix contains the static geometric properties and dynamic behavior properties of the edge wire. During the fusion process, a weighted superposition strategy is adopted, in which the static feature weight is set to 0.6 and the dynamic feature weight is set to 0.4. The weight values are determined based on the training and optimization of historical abnormal data. The final generated edge wire features comprehensively characterize the deformation, position offset and motion anomalies of the edge wire through multi-dimensional parameters, providing a unified data interface for the subsequent accurate determination of escape or blockage status. This step enhances the sensitivity and discrimination ability of the abnormal state of the edge wire through the complementarity of spatiotemporal features and weight optimization, ensuring the robustness of the detection method in high-noise industrial environments.

[0077] In summary, the feature extraction method of the embodiment of the present application constructs a complete edge wire state characterization system through high-precision shape capture of the image semantic segmentation algorithm, dynamic trajectory quantization of the optical flow method, noise suppression of the Kalman filter, and multi-dimensional correlation of feature fusion. The preset values of the segmentation parameters and trajectory prediction parameters are optimized based on a large amount of experimental data, taking into account both algorithm efficiency and accuracy; the dynamic adjustment mechanism of the fusion weight further improves the adaptability of anomaly detection. The embodiment of the present application effectively solves the problem of misjudgment caused by the separation of shape and motion features in traditional detection technology, and provides technical support for the closed-loop control of galvanized automobile plate disc shears.

[0078] In some examples, the abnormal state includes a blocking abnormal state or an escape abnormal state. Based on the edge wire characteristics, determining whether the edge wire state is in an abnormal state includes:

[0079] Based on the edge wire motion trajectory characteristics and the preset guide groove path, the target path deviation degree is determined, and when the target path deviation degree is greater than a first preset threshold, the edge wire state is determined to be an escape abnormal state; or,

[0080] Based on the morphological characteristics and preset blockage judgment conditions, the target flow deviation and the target flow duration of the target flow deviation are determined. When the target flow deviation is less than the first preset flow threshold and the target flow duration is greater than or equal to the first preset duration, the edge wire state is determined to be a blockage abnormal state.

[0081] Exemplarily, the degree of deviation from the target path is determined based on the matching analysis of the edge wire motion trajectory characteristics and the preset guide groove path. Specifically, the preset guide groove path is pre-defined by the structural parameters of the disc shearing equipment and the normal flow trajectory of the edge wire. It is represented by a set of continuous path points in a two-dimensional coordinate system, and is used to characterize the ideal motion trajectory of the edge wire under normal working conditions. During the detection process, the corrected edge wire motion trajectory characteristics (including the real-time position sequence of the edge wire on the driving side, the working side and the blanking hopper outlet) are compared with the preset guide groove path frame by frame, and the Euclidean distance between the actual position of the edge wire and the corresponding point of the guide groove path at each moment is calculated. The degree of deviation from the target path is obtained by statistically averaging the distances of all frames within a preset time window (for example, 5 seconds), which is used to quantify the overall offset of the edge wire motion trajectory. This step uses a spatial matching algorithm to quantify the difference between the dynamic trajectory and the standard path, providing a comparable numerical basis for escape anomaly judgment.

[0082] When the target path deviation exceeds a first preset threshold, the edge wire is determined to have escaped the normal guide groove path, triggering an escape anomaly state. The first preset threshold is set based on the path fluctuation range in historical normal production data, for example, 20% of the guide groove path width. If the target path deviation exceeds this threshold, it indicates that the edge wire has escaped the guide groove's constraints and is at risk of escape. This mechanism ensures that the threshold parameters can adapt to different production speeds, material properties, and equipment wear conditions, improving the adaptability and robustness of escape anomaly determination.

[0083] For the judgment of blockage anomalies, the edge wire flow at the hopper outlet is statistically analyzed based on the morphological features to determine the target flow deviation and the target flow duration. Specifically, the edge wire contour information in the morphological features is extracted through the binary mask output by the image semantic segmentation algorithm, and the volume of edge wire passing through the hopper outlet per unit time is calculated in real time by combining the calibration relationship between pixel area and physical size. The preset blockage judgment condition is defined as the normal flow range, for example, the preset flow lower limit is the first preset flow threshold. The target flow deviation is calculated by the difference between the current flow value and the first preset flow threshold; the target flow duration is obtained by statistically analyzing the cumulative time (first preset duration) that the flow is continuously lower than the first preset flow threshold. The first preset duration is used to distinguish between instantaneous flow fluctuations and continuous blockages to avoid false triggering due to short-term interference. The above steps quantify the degree and duration of edge wire flow interruption through flow monitoring and time series analysis, providing dynamic data support for blockage anomaly judgment.

[0084] When the target flow deviation is less than the first preset flow threshold and the target flow duration is greater than or equal to the first preset duration, the edge wire state is determined to be a blocked abnormal state. The first preset duration is set based on the process requirements. For example, the maximum tolerance duration for flow abnormality is 10 seconds. Specifically, if the flow rate continues to be lower than the threshold and is not restored after the timeout, it indicates that there is waste accumulation at the outlet of the hopper, resulting in interruption of the edge wire flow. The judgment logic is implemented through a joint judgment of flow rate and duration, and the blockage abnormality signal is triggered only when both conditions are met at the same time. Through dual condition constraints, false triggering caused by instantaneous flow fluctuations is avoided, the reliability of blockage abnormality judgment is ensured, and a precise triggering basis is provided for the subsequent compensation adjustment of shear pressure or vibration frequency.

[0085] In some instances, when the edge wire is in an abnormal state, the operating parameters of the disc shear are adjusted, including:

[0086] When the edge wire state is an escape abnormal state, based on the target path deviation degree, a target first operating parameter corresponding to the target path deviation degree is determined from a preset mapping relationship between the path deviation degree and the first operating parameter, the first operating parameter including a speed reduction amplitude and a stoppage time of the disc shear;

[0087] When the edge wire state is an abnormal blockage state, based on the target flow deviation and the target flow duration, the target second operating parameters corresponding to the target flow deviation and the target flow duration are determined through a preset compensation model. The second operating parameters include the shear pressure compensation value and the vibration frequency compensation value of the disc shear.

[0088] Exemplary, when the edge wire state is determined to be an escape abnormality, based on the target path deviation degree (i.e., the deviation between the actual edge wire trajectory and the preset guide groove path), the target first operating parameter corresponding to the target path deviation degree is determined from the preset mapping relationship between the path deviation degree and the first operating parameter. Specifically, the preset mapping relationship is established through historical production data and experimental tests, and the path deviation degree is divided into multiple levels (e.g., mild deviation, moderate deviation, severe deviation), each level corresponding to a different speed reduction amplitude and downtime length. For example, when the target path deviation degree is mild deviation (0%-30%), that is, the target path deviation degree is greater than 0% and less than or equal to 30%, the corresponding speed reduction amplitude is 15% and the downtime length is 2 seconds; when the deviation degree is moderate deviation, that is, the target path deviation degree is greater than 30% and less than or equal to 70%, the corresponding speed reduction amplitude is 30% and the downtime length is 5 seconds; when the deviation degree is severe deviation, that is, the target path deviation degree is greater than 70%, the corresponding speed reduction amplitude is 50% and the downtime length is 10 seconds. The speed reduction is achieved by adjusting the speed control command for the circular shear drive motor, and the duration of the downtime is triggered by the control system's timing module. After this adjustment, the circular shear's cutting speed is reduced or suspended to prevent the edge wire from further deviating from the guide groove path. This also provides a buffer for manual intervention or automatic deviation correction to prevent the escaped edge wire from scratching the strip or wrapping around the equipment rollers. This step dynamically correlates the degree of deviation with operating parameters to ensure that the abnormal handling measures are precisely matched to the severity of the edge wire escape, avoiding over-adjustment or under-response. This allows for rapid suppression of further edge wire deviation from the guide groove path and restores production stability.

[0089] When the edge wire state is determined to be abnormally blocked, the corresponding target second operating parameter is determined based on the target flow deviation and the target flow duration through the preset compensation model. Specifically, the preset compensation model is a multivariate regression model, which is dynamically calculated based on process parameters and real-time data. The input parameters include flow deviation and flow duration, and the output parameters are shear pressure compensation value and vibration frequency compensation value. For example, when the flow deviation is 10% and the duration is 8 seconds, the model outputs a shear pressure compensation value of 5MPa and a vibration frequency compensation value of 20Hz; when the flow deviation increases to 15% and lasts for 12 seconds, the shear pressure compensation value and the vibration frequency compensation value are increased to 8MPa and 30Hz respectively. The second operating parameter includes the shear pressure compensation value and vibration frequency compensation value of the disc shear, wherein the shear pressure compensation value is achieved by adjusting the pressure valve opening of the hydraulic system, and the vibration frequency compensation value is implemented by adjusting the driving frequency of the vibration motor. After adjustment, the shearing pressure of the disc shear is increased to cut off accumulated wire. Simultaneously, high-frequency vibration clears the waste channel at the hopper outlet, restoring normal wire flow, eliminating the risk of blockage and ensuring production continuity. This step dynamically calculates the optimal parameter combination through a compensation model to clear waste accumulation at the hopper outlet while avoiding equipment overload or secondary wire blockage caused by excessive compensation. The adjusted operating parameters are applied to the disc shear control system to ensure that wire flow returns to a safe threshold, effectively eliminating blockage anomalies and maintaining continuous production line operation.

[0090] In some instances, this also includes:

[0091] Determining a path adjustment amount of a first preset threshold based on a historical frequency of occurrence of the escape abnormal state;

[0092] Based on the path adjustment amount, the first preset threshold is updated through a dynamic parameter optimization model to reduce the missed detection rate of escape anomalies;

[0093] Determining a flow adjustment amount for a first preset flow threshold and a duration adjustment amount for a first preset duration based on a historical occurrence frequency of the abnormal congestion state;

[0094] Based on the flow adjustment amount and the duration adjustment amount, the first preset flow threshold and the first preset duration are updated through a dynamic parameter optimization model to improve the detection rate of blockage anomalies.

[0095] Exemplarily, based on the historical frequency of the escape anomaly state, statistical analysis is performed to quantify abnormal events within a preset time period and generate an escape anomaly frequency value. The historical frequency is defined as the ratio of the number of escape anomaly triggers per unit time to the total number of detections, for example, the percentage of escape anomaly triggers per hour to the total number of detections. Based on the frequency value, a path adjustment amount for a first preset threshold is determined using a preset linear adjustment strategy. Specifically, when the historical frequency exceeds a preset safety threshold (e.g., a frequency value greater than or equal to 5%), the current first preset threshold is determined to be insufficiently sensitive to the escape anomaly, and the path adjustment amount needs to be increased to lower the threshold. Conversely, when the frequency value is below a preset lower limit (e.g., a frequency value less than or equal to 1%), the threshold is determined to be too strict and may lead to false detections, and the path adjustment amount needs to be reduced to raise the threshold. The path adjustment amount is determined by a proportionality coefficient and the difference between the actual statistical escape anomaly frequency and a preset target frequency. Specifically, the path adjustment amount is equal to the proportionality coefficient (e.g., 0.2) multiplied by the difference between the actual statistical escape anomaly frequency and the preset target frequency (e.g., 3%). The actual escape anomaly frequency is defined as the percentage of escape anomaly triggers over the total number of detections within a preset time period. The preset target frequency is the expected anomaly frequency value set based on process requirements. By quantifying the deviation between the actual and target frequencies, path adjustments are dynamically generated, providing input for subsequent threshold optimization.

[0096] The path adjustment amount is input into the dynamic parameter optimization model, and the first preset threshold is updated to reduce the missed detection rate. The dynamic parameter optimization model adopts a gradient descent algorithm, takes the path adjustment amount as the gradient direction, and iteratively adjusts the first preset threshold. Specifically, the update formula is the current threshold minus the product of the learning rate and the path adjustment amount. For example, if the current threshold is 20% of the guide groove path width, the path adjustment amount is 4%, and the learning rate is 0.05, then the updated threshold is adjusted to 20%-(0.05×4%)=19.8%. The adjusted threshold takes effect in real time through the control system, making the judgment conditions for the degree of deviation of the edge wire trajectory more stringent, thereby reducing the possibility of missed detection of escape anomalies.

[0097] According to the historical frequency of the abnormal congestion state, the cumulative duration of the flow deviation below the first preset flow threshold in the preset time period is calculated to generate a congestion abnormality frequency value. Based on the congestion abnormality frequency value, the flow adjustment amount of the first preset flow threshold and the duration adjustment amount of the first preset duration are determined by a preset piecewise function strategy. For example, when the abnormal congestion frequency exceeds 5%, it is determined that the current flow threshold or duration threshold setting is unreasonable and needs to be adjusted simultaneously; based on the historical distribution of the flow deviation, the first preset flow threshold is reduced to 90% of the original value to relax the flow lower limit judgment condition and avoid missed detection due to excessively high thresholds; based on the statistical median of the target flow duration, the first preset duration is shortened to 80% of the original value to speed up the response to persistent congestion. For example, if the original flow threshold is 100 units / second, the flow adjustment amount is -10 units / second; if the original duration threshold is 10 seconds, the duration adjustment amount is -2 seconds.

[0098] The flow rate adjustment and duration adjustment are input into the dynamic parameter optimization model to update the first preset flow rate threshold and the first preset duration to improve the detection rate of congestion anomalies. The model uses a multi-objective optimization algorithm with the objective function of maximizing the congestion anomaly detection rate and minimizing the false detection rate. The constraint is that the adjusted thresholds must be within the process tolerance. For example, if the original flow rate threshold is 100 units / second, the flow rate adjustment is -10 units / second, and the weighting factor is 1.2, the updated flow rate threshold is 100 + (-10 × 1.2) = 88 units / second. If the original duration threshold is 10 seconds, the duration adjustment is -2 seconds, and the weighting factor is 0.8, the updated duration threshold is 10 + (-2 × 0.8) = 8.4 seconds. The adjusted thresholds are applied to the real-time monitoring module, making it more sensitive to flow interruptions, thereby improving the detection rate of congestion anomalies. At the same time, the duration constraint prevents false triggering caused by transient fluctuations.

[0099] In summary, through the above steps, the dynamic parameter optimization model adjusts threshold parameters in real time based on historical anomaly frequencies, creating a balance between detection sensitivity and false positive rate. Adjusting the threshold for escape anomalies reduces the risk of missed detections, while optimizing the flow rate and duration thresholds for blocking anomalies improves detection accuracy, ultimately achieving adaptive closed-loop control of the galvanized automotive sheet metal disc shear detection system.

[0100] In some instances, this also includes:

[0101] Based on the abnormal type of the edge wire status and the adjusted operating parameters of the disc shear, an abnormality handling record containing the abnormality type identifier, parameter adjustment value and timestamp is generated;

[0102] Based on exception processing records, segmentation parameters are optimized through machine learning models to improve the accuracy of edge wire contour and position extraction;

[0103] Based on the exception handling records, the trajectory prediction parameters are optimized through the machine learning model to reduce the motion trajectory prediction error.

[0104] For example, when the edge wire status is detected as an escape or jam abnormality, a structured exception handling record is automatically generated based on the abnormality type identification (for example, "escape abnormality code E01" or "jamming abnormality code B02"), the adjusted operating parameter values of the disc shear (such as a speed reduction of 15%, a shear pressure compensation value of 5MPa) and the trigger timestamp. The record is stored in a time series database, and each record contains the following fields: the abnormality type identification field is used to classify escape or jam events; the parameter adjustment value field records the actual adjustment amount of the operating parameters; the timestamp field marks the specific time of the abnormality trigger and parameter adjustment. The abnormality handling record is synchronized to the training data set of the machine learning model in real time through the preset data interface, providing a historical data basis for subsequent parameter optimization. For example, when an escape abnormality occurs, the record stores a path deviation degree of 25%, a speed reduction of 20%, a downtime of 5 seconds and the corresponding timestamp, forming an associated data chain of escape abnormality characteristics and adjustment strategies.

[0105] Based on exception handling records, a machine learning model iteratively optimizes the segmentation parameters (number of network layers, learning rate, and feature fusion weight) of the image semantic segmentation algorithm. Specifically, the model uses the edge wire contour extraction error in the exception handling records as a loss function and uses a backpropagation algorithm to update the segmentation parameters. For example, if the exception record indicates missed detection due to blurred contours, the model uses gradient descent to reduce the learning rate (for example, from 0.01 to 0.005) to reduce the parameter update step size and improve training stability. If the edge wire position offset error is high, the feature fusion weight is increased (for example, from 0.5 to 0.7) to enhance the multi-scale feature fusion capability. The optimized segmentation parameters are validated through a combination of offline training and online deployment. For example, the model is retrained weekly based on newly added exception records, and the updated parameters (for example, increasing the number of network layers from 128 to 256) are pushed to the online detection system. This improves the accuracy of edge wire contour and position extraction and reduces misjudgment of anomalies due to segmentation errors.

[0106] Based on trajectory prediction errors in anomaly handling records (such as optical flow noise or Kalman filter lag), a machine learning model dynamically adjusts trajectory prediction parameters (state transition matrix, observation noise covariance, and process noise covariance). The model uses a reinforcement learning algorithm to iteratively update parameters, optimizing the trajectory prediction residual (the mean squared error between the actual and predicted trajectories) as the optimization objective. For example, if anomaly records indicate trajectory prediction deviations due to an overly rigid state transition matrix, the model uses a policy gradient method to adjust matrix elements (for example, reducing the acceleration component from 0.9 to 0.8) to accommodate the nonlinear characteristics of the wire's variable velocity motion. If the observation noise covariance is set too high, resulting in oversmoothing of the trajectory, it is adjusted from 0.1 to 0.05 to enhance sensitivity to instantaneous position jumps. The optimized trajectory prediction parameters are implemented through a real-time parameter configuration module, for example, automatically loading the latest parameters into the Kalman filter every 24 hours. This reduces trajectory prediction errors and improves the real-time and accuracy of escape anomaly detection.

[0107] In summary, the embodiments of the present application achieve dynamic adaptation of segmentation parameters and trajectory prediction parameters through a closed-loop optimization mechanism combining exception handling records with machine learning models. Optimizing the segmentation parameters enhances the ability to capture edge wire deformation and positional offsets; optimizing the trajectory prediction parameters improves the prediction accuracy and noise suppression of motion trajectories. These two factors work together to reduce the risk of missed or false detections caused by rigid algorithmic parameters, providing adaptive technical support for the efficient and stable operation of galvanized automotive sheet circular shears.

[0108] See also Figure 2 , is a schematic diagram of the detection interface of a galvanized automobile sheet circular shear provided in an embodiment of the present application; this interface is the "Line 1 Disc Shear Detection System", which is mainly used to display the abnormal data statistics of the current shift, the current day and the cumulative number of abnormalities of the disc shear, of which the current shift disc shear has 3 abnormalities; the current day and the cumulative number of abnormalities of the disc shear are both 10. The interface also marks the time and lists the equipment such as the East, West, Shear Mouth East, and Shear Mouth West of Line 1 Disc Shear. Overall, Figure 2 It is the human-machine interaction interface of the industrial automation detection system. It can monitor the abnormal status of the disc shear in real time and provide data support for production line abnormality diagnosis and interlocking control, which helps to reduce the missed detection rate and misjudgment rate and ensure production continuity and efficiency.

[0109] See also Figure 3 , is a schematic structural diagram of a detection device for a galvanized automobile sheet circular shear provided in an embodiment of the present application, comprising:

[0110] The disc shear image acquisition unit 21 is used to acquire image data of the disc shear area;

[0111] The edge feature extraction unit 22 extracts features from the image data based on the image data to generate edge features;

[0112] The abnormal state judgment unit 23 judges whether the edge wire state is in an abnormal state based on the edge wire characteristics;

[0113] The circular shear parameter adjustment unit 24 is used to adjust the operating parameters of the circular shear when the edge wire state is abnormal.

[0114] See also Figure 4 An embodiment of the present application also provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, the steps of any method for detecting galvanized automobile sheet disc shears are implemented.

[0115] Since the electronic device introduced in this embodiment is the equipment used to implement a detection device for a galvanized automobile plate disc shear in the embodiment of this application, based on the method introduced in the embodiment of this application, technical personnel in this field can understand the specific implementation of the electronic device of this embodiment and its various variations. Therefore, how the electronic device implements the method in the embodiment of this application will not be introduced in detail here. As long as the equipment used by technical personnel in this field to implement the method in the embodiment of this application falls within the scope of protection of this application.

[0116] During the specific implementation process, when the computer program 311 is executed by the processor, any implementation method of the embodiments corresponding to the first aspect can be implemented.

[0117] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0118] Those skilled in the art will appreciate that the embodiments of the present application may provide methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-readable program code.

[0119] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0120] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0121] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0122] The present application also provides a computer program product, which includes computer software instructions. When the computer software instructions are executed on a processing device, the processing device executes Figure 1 The process of a detection method for a galvanized automobile sheet disc shear in the corresponding embodiment.

[0123] A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium can be a magnetic medium, an optical medium or a semiconductor medium, etc.

[0124] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0126] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0127] In addition, the functional units in the various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The above-mentioned integrated units may be implemented in the form of hardware and / or software functional units.

[0128] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk.

[0129] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

[0130] Although the preferred embodiments of this specification have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this specification.

[0131] Obviously, those skilled in the art may make various changes and modifications to this specification without departing from the spirit and scope of this specification. Thus, if such changes and modifications fall within the scope of the claims of this specification and their equivalents, this specification is intended to include such changes and modifications.

Claims

1. A detection method for galvanized automobile sheet disc shears, characterized in that: include: Obtain image data of the disc shearing area; Based on the image data, feature extraction is performed on the image data to generate edge features; Based on the edge wire characteristics, determining whether the edge wire state is in an abnormal state; When the edge wire state is in an abnormal state, the operating parameters of the disc shear are adjusted.

2. The method according to claim 1, characterized in that The image data includes driving side image data and operating side image data, and the image data of the disc shearing area is obtained, including: High-resolution industrial cameras are used to collect video stream data from the drive side and working side of the circular shears; Based on a preset frame extraction rule, the driving side video stream data and the working side video stream data are subjected to frame extraction processing to generate a driving side target image sequence and a working side target image sequence; A preprocessing operation is performed on the driving side target image sequence and the working side target image sequence to generate the driving side image data and the operating side image data.

3. The method according to claim 1, characterized in that The step of extracting features from the image data based on the image data to generate edge features includes: Based on the image data, extracting the morphological features of the edge wires by using the segmentation parameters of the image semantic segmentation algorithm, wherein the morphological features include edge wire contour information and edge wire position information, and the segmentation parameters include the number of network layers, the learning rate, and the feature fusion weight; Based on the morphological features, the motion vector of the edge wire is calculated by an optical flow method to generate a motion trajectory feature of the edge wire; The motion trajectory characteristics are predicted and corrected by using trajectory prediction parameters of a Kalman filter algorithm to generate corrected edge wire motion trajectory characteristics, wherein the trajectory prediction parameters include a state transfer matrix, an observation noise covariance, and a process noise covariance; The morphological features are fused with the edge wire motion trajectory features to generate the edge wire features.

4. The method according to claim 3, characterized in that The abnormal state includes a blocking abnormal state or an escaping abnormal state, and judging whether the edge wire state is in an abnormal state based on the edge wire feature includes: Determine the target path deviation degree based on the edge wire motion trajectory characteristics and the preset guide groove path, and when the target path deviation degree is greater than a first preset threshold, determine that the edge wire state is the escape abnormal state; or Based on the morphological features and the preset blockage judgment conditions, the target flow deviation and the target flow duration of the target flow deviation are determined. When the target flow deviation is less than the first preset flow threshold and the target flow duration is greater than or equal to the first preset duration, the edge wire state is determined to be the blockage abnormal state.

5. The method according to claim 4, characterized in that When the edge wire state is in an abnormal state, adjusting the operating parameters of the disc shear includes: When the edge wire state is the escape abnormal state, based on the target path deviation degree, a target first operating parameter corresponding to the target path deviation degree is determined from a preset mapping relationship between the path deviation degree and the first operating parameter, wherein the first operating parameter includes a speed reduction amplitude and a shutdown duration of the disc shear; or When the edge wire state is the abnormal blockage state, based on the target flow deviation and the target flow duration, the target second operating parameter corresponding to the target flow deviation and the target flow duration is determined through a preset compensation model, wherein the second operating parameter includes the shear pressure compensation value and the vibration frequency compensation value of the disc shear.

6. The method according to claim 4, characterized in that Also includes: Determining a path adjustment amount of the first preset threshold based on a historical occurrence frequency of the escape abnormal state; Based on the path adjustment amount, updating the first preset threshold through a dynamic parameter optimization model to reduce the missed detection rate of escape anomalies; Determining a flow adjustment amount of the first preset flow threshold and a duration adjustment amount of the first preset duration based on a historical occurrence frequency of the abnormal congestion state; Based on the flow adjustment amount and the duration adjustment amount, the first preset flow threshold and the first preset duration are updated through the dynamic parameter optimization model to improve the detection rate of blockage anomalies.

7. The method according to claim 3, characterized in that Also includes: Based on the abnormal type of the edge wire state and the adjusted operating parameters of the disc shear, an abnormality handling record including an abnormality type identifier, a parameter adjustment value and a timestamp is generated; Based on the exception processing record, the segmentation parameters are optimized by a machine learning model to improve the extraction accuracy of the edge wire contour and edge wire position; Based on the exception handling record, the trajectory prediction parameters are optimized through the machine learning model to reduce the motion trajectory prediction error.

8. A detection device for galvanized automobile sheet disc shears, characterized in that: include: a disc shear image acquisition unit, for acquiring image data of the disc shear area; A side wire feature extraction unit is configured to extract features from the image data based on the image data to generate side wire features; an abnormal state judgment unit, for judging whether the edge wire state is in an abnormal state based on the edge wire characteristics; The disc shear parameter adjustment unit is used to adjust the operating parameters of the disc shear when the edge wire state is abnormal.

9. An electronic device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor is configured to implement the steps of the method for detecting a galvanized automobile sheet circular shear as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the detection method of the galvanized automobile sheet circular shear according to any one of claims 1 to 7 is implemented.

Images