Enameled wire paint film thickness detection method and system

Abstract

The invention discloses an enamelled wire paint film thickness detection method and system, and relates to the technical field of paint film thickness detection, and the method comprises the steps: carrying out the non-contact high-speed circumferential scanning of an advancing enamelled wire at a first station on a production line, and obtaining a complete continuous data flow of the surface morphology of the enamelled wire; converting the data flow into a one-dimensional surface feature sequence along the length direction of the enameled wire, identifying and extracting position information of a suspected abnormal section from the one-dimensional surface feature sequence based on a preset uniformity criterion, and generating an ordered fixed-point detection task list arranged along the length direction, the ordered fixed-point detection task list only contains position information of the suspected abnormal section; according to the invention, the combination of high-speed full-coverage screening and fixed-point accurate detection is realized, and the contradiction between the detection precision and the detection efficiency is solved.

Description

Technical Field

[0001] This invention belongs to the field of paint film thickness detection technology, specifically a method and system for detecting the paint film thickness of enameled wire. Background Technology

[0002] As a core insulation component in electrical equipment such as motors, transformers, and relays, the uniformity and accuracy of the enamel film thickness of enameled wire directly determine the insulation performance and service life of these devices. In high-speed continuous production lines for enameled wire, traditional enamel film thickness detection methods are mainly divided into two categories: one is online high-speed detection based on laser scanning or eddy current testing, which can achieve uninterrupted scanning across the entire line, but has low detection accuracy and struggles to distinguish between enamel film layer thickness and micro-defects; the other is offline precision detection based on metallographic microscopes or ultrasonic thickness gauges, which can acquire high-precision data, but has low detection efficiency, cannot adapt to the high-speed operation of the production line, and requires manual sampling, resulting in detection lag. Some existing technologies attempt to combine high-speed inspection with precision inspection, but most adopt a full-line synchronous inspection mode. This fails to consider that the coating thickness in most sections of the enameled wire production process is within the acceptable range. Running the precision inspection device throughout the entire process will waste equipment resources and reduce inspection efficiency. At the same time, frequent start-ups and shutdowns of the precision inspection device will shorten the equipment's lifespan. In actual production, coating abnormalities in enameled wire are usually concentrated in localized sections, such as areas of wear on the coating mold or areas of tension fluctuation. Global precision inspection will generate a large amount of redundant data, increasing data processing pressure and failing to meet the real-time quality control requirements of high-speed production lines. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for detecting the thickness of the enamel film on enameled wires, so as to solve the problems mentioned in the background art.

[0004] To make the purpose, technical solution, and advantages of this application clearer, the key terms used in this application are explained below: One-dimensional surface feature sequence: refers to a sequence formed by arranging the continuous data stream of the surface morphology of enameled wire in the sampling order along the length direction. Each data point corresponds to one or more surface feature parameters at a specific length position of the enameled wire, including but not limited to the average diameter, standard deviation of diameter fluctuation, range of circumferential thickness, and average surface roughness.

[0005] Uniformity criterion: refers to the judgment rule based on the technical standard of enameled wire products, used to judge whether the coating thickness is uniform. Its parameters include the allowable deviation of coating thickness, the allowable value of circumferential thickness uniformity, and the surface roughness threshold, which are used to compare with the feature values ​​in the one-dimensional surface feature sequence.

[0006] Deviation of the same trend: refers to the situation in a one-dimensional surface feature sequence where the same feature parameter of multiple consecutive sequence windows deviates from the normal range in the same direction (such as both being above the normal range or both being below the normal range), and the deviation amplitude does not exceed a preset range (e.g., 10% of the upper limit of the normal range).

[0007] A method for detecting the coating thickness of enameled wire, used for online detection of enameled wire on a high-speed continuous production line, includes: At the first station on the production line, a non-contact high-speed circumferential scan is performed on the moving enameled wire to obtain a complete and continuous data stream of the surface morphology of the enameled wire. The data stream is converted into a one-dimensional surface feature sequence along the length of the enameled wire. Based on a preset uniformity criterion, the location information of suspected abnormal sections is identified and extracted from the one-dimensional surface feature sequence, and an ordered fixed-point detection task list arranged along the length direction is generated. The ordered fixed-point detection task list only contains the location information of the suspected abnormal sections. At the second station located downstream of the first station, a precision detection device with a detection rate lower than that of the high-speed circular scan is provided. According to the ordered fixed-point inspection task list, selective fixed-point inspection is carried out on the enameled wire: when an abnormal section of the enameled wire that is listed in the ordered fixed-point inspection task list enters the inspection window of the precision inspection device as the production line moves, the precision inspection device is triggered to perform coating film layer thickness resolution or micro-area sampling analysis on the abnormal section. For enameled wire segments not included in the ordered fixed-point inspection task list, the precision inspection device remains in standby mode during the time period when the enameled wire segment passes through the inspection window.

[0008] As a further aspect of the present invention: the step of identifying and extracting suspected abnormal segments based on a preset uniformity criterion includes: The one-dimensional surface feature sequence is divided into continuous sequence windows; Calculate the feature value of the current sequence window and determine whether the feature value deviates from the preset normal feature value range; If the feature value deviates, it is further determined whether the feature values ​​of at least two subsequent consecutive sequence windows of the current sequence window all show the same trend of deviation. If the feature values ​​of at least two subsequent consecutive sequence windows all show the same trend of deviation, then the current sequence window and the at least two subsequent consecutive sequence windows are jointly determined as a suspected abnormal segment, and the start position information and end position information of the suspected abnormal segment are entered into the ordered fixed-point detection task list.

[0009] As a further aspect of the present invention: the step of triggering the precision detection device includes: Establish a length coordinate mapping relationship between the first workstation and the second workstation; The length coordinates of the enameled wire reaching the second workstation are obtained in real time, and the real-time length coordinates are compared cyclically with the starting coordinates of the next task to be inspected in the ordered fixed-point detection task list. When the difference between the real-time length coordinate and the starting coordinate of the next task to be inspected enters the preset warning range, the trigger preparation program is started. When the difference between the real-time length coordinate and the starting coordinate of the next inspection task returns to zero, an execution trigger signal is generated to start the precision inspection device.

[0010] As a further aspect of the present invention, the method further includes: The number of tasks that have been triggered but not completed in the ordered fixed-point detection task list is monitored in real time and recorded as the backlog of tasks. If the backlog of tasks exceeds a preset warning threshold, a production line slowdown instruction is generated. According to the speed reduction command, the production line is controlled to reduce its operating speed until the backlog of tasks is reduced to below a preset safety threshold.

[0011] As a further aspect of the present invention, the method further includes: Record the point-to-point detection results of the precision detection device for each task in the ordered point-to-point detection task list; The fixed-point detection results are correlated with the one-dimensional surface feature sequences collected at the corresponding positions of the first workstation to generate labeled detection samples; After a complete production batch is completed, based on the test sample set generated in that batch, the preset parameters of the uniformity criterion are statistically analyzed and optimized. The optimized parameters will be applied to subsequent production batches.

[0012] As a further aspect of the present invention: the step of establishing the length coordinate mapping relationship includes: At least one physical position reference mark with a known fixed spacing is provided between the first workstation and the second workstation; The high-speed circumferential scanning device records the first occurrence coordinates of the physical position reference mark on the enameled wire during the scanning process; The precision detection device or associated sensor records the second occurrence coordinates when the physical position reference mark reaches the detection window; Based on the first occurrence coordinate and the second occurrence coordinate, the mapping parameters of the length coordinate mapping relationship are periodically calculated and corrected to compensate for the cumulative slip or tensile deformation of the enameled wire during transmission.

[0013] As a further aspect of the present invention, the method further includes: At the second workstation, upstream of the precision testing device, a high-speed re-inspection unit is set up; the high-speed re-inspection unit quickly scans the enameled line segment before it enters the testing window to obtain local surface features; Before a certain abnormal section in the ordered fixed-point detection task list is about to trigger the precision detection device, the current local surface features obtained by the high-speed re-inspection unit are compared in real time with the historical features corresponding to the abnormal section in the one-dimensional surface feature sequence. If the comparison result indicates that the current feature has returned to the normal range, then the task item is removed from the ordered fixed-point detection task list, and the current triggering of the precision detection device is canceled.

[0014] As a further aspect of the present invention, the method further includes: The second workstation is equipped with at least two precision testing devices with different performance characteristics, each of which is good at analyzing different types of defects. When generating the ordered fixed-point detection task list, preliminary pattern recognition is performed on the defect type of each suspected abnormal segment based on the one-dimensional surface feature sequence. Based on the results of the preliminary pattern recognition, a recommended detection device type identifier is assigned to each task item in the list; When the trigger is executed, the system assigns the task to the corresponding precision detection device based on the recommended detection device type identifier.

[0015] As a further aspect of the present invention, the method further includes: Continuously monitor the continuity and rate of change of the real-time acquired length coordinates; If an unexpected jump or pause is detected in the length coordinate, it is determined that the position tracking system has lost synchronization, and the triggering process based on the ordered fixed-point detection task list is immediately suspended. At the same time, the precision detection device is controlled to switch to a safe detection mode: within the next preset time window, continuous scanning detection is performed on the enameled wire passing through the detection window; Once the position tracking system recovers and a reliable length coordinate mapping relationship is re-established, the normal opportunistic fixed-point detection process resumes.

[0016] Secondly, this application provides a system for detecting the thickness of the enamel coating on enameled wires, comprising: The acquisition module, set at the first station on the production line, is used to perform non-contact high-speed circumferential scanning on the moving enameled wire to acquire a complete and continuous data stream of the surface morphology of the enameled wire. The generation module, connected to the acquisition module, is used to convert the data stream into a one-dimensional surface feature sequence along the length of the enameled wire, and based on a preset uniformity criterion, to identify and extract the location information of suspected abnormal sections from the one-dimensional surface feature sequence, and generate an ordered fixed-point detection task list. The detection device is located at the second station downstream of the first station; and, The trigger control module, connected to the generation module and the detection device, is used to control the detection device to perform fixed-point detection on suspected abnormal sections arriving at its detection window according to the ordered fixed-point detection task list, and to keep non-abnormal sections in standby state.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention can distinguish the differences in detection requirements between normal and abnormal sections in the production process of enameled wire. By combining high-speed scanning full-coverage screening with opportunistic fixed-point precision detection, it avoids the limitations of insufficient accuracy of single high-speed detection and solves the pain points of low efficiency and waste of resources in traditional full-line precision detection. Furthermore, it focuses on suspected abnormal sections with quality risks, and accurately locks down the abnormal range through a combination of sequence window division, multi-parameter feature value calculation and continuous trend judgment. This effectively eliminates misjudgments caused by accidental factors, meets the actual needs of high-speed production lines for key quality and rapid response, and improves the pertinence and accuracy of abnormal diagnosis. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the method framework structure of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] First aspect: Please refer to Figure 1 This application provides a method for detecting the coating thickness of enameled wire, used for online detection of enameled wire on a high-speed continuous production line. The specific implementation process is as follows: At the first station on the production line, a non-contact high-speed circumferential scan is performed on the moving enameled wire to obtain a complete and continuous data stream of the surface morphology of the enameled wire. It should be noted that at the first station, a high-speed laser triangulation scanning sensor is used to perform non-contact, high-speed circumferential scanning of the enameled wire. The scanning frequency is 2000Hz, the sampling accuracy is 0.01μm, and the scanning angle covers 360°. Before scanning, the sensor is calibrated using standard parts, and the error is controlled within ±0.05μm. During scanning, tension fluctuations are controlled within ±5N. The scanned data is denoised using a moving average filter (5 sampling points in a window) and subjected to coordinate transformation to obtain a continuous data stream along the length direction, containing information on diameter, surface roughness, and circumferential thickness uniformity.

[0021] The data stream is transformed into a one-dimensional surface feature sequence along the length of the enameled wire. Based on a preset uniformity criterion, the location information of suspected abnormal sections is identified and extracted from the one-dimensional surface feature sequence. An ordered fixed-point detection task list is generated along the length direction. The ordered fixed-point detection task list only contains the location information of suspected abnormal sections. It should be noted that the conversion of the one-dimensional surface feature sequence needs to focus on the core parameters related to the coating thickness. Specifically, the point cloud data of the circumferential scanning of each cross section obtained by high-speed circumferential scanning is fitted and statistically calculated to obtain parameters such as the average diameter, the range of circumferential thickness, and the average surface roughness of the cross section. These parameters are then arranged in the sampling order along the length of the enameled wire, thus integrating them into a one-dimensional surface feature sequence. The sampling interval of the sequence is matched with the production line running speed to ensure that each data point corresponds to the actual length position of the enameled wire.

[0022] The preset uniformity criterion is based on the product technical standards of enameled wire and covers key indicators such as allowable deviation of enamel film thickness, allowable value of circumferential thickness uniformity, and surface roughness threshold. The criterion parameters can be adjusted according to the production requirements of enameled wire of different specifications. The process of identifying and extracting suspected abnormal sections based on this uniformity criterion is as follows: The one-dimensional surface feature sequence is divided into continuous sequence windows, each containing 10 consecutive sampling points, corresponding to an enameled wire length of approximately 4-5 mm. The window sliding step is 1 sampling point, and adjacent windows overlap by 9 sampling points to ensure continuous monitoring.

[0023] Calculate the feature values ​​of the current sequence window and determine whether the feature values ​​deviate from the preset normal feature value range. Calculate the feature values ​​of the current sequence window, including the average diameter, standard deviation of diameter fluctuation, maximum range of circumferential thickness, and average surface roughness. Compare the feature values ​​with the preset normal feature value range. For example, for 0.1mm enameled wire, the normal feature value range is shown below: If any of the above conditions is exceeded, it will be considered a deviation.

[0024] If a feature value deviates, it is further determined whether the feature values ​​of at least two subsequent consecutive sequence windows of the current sequence window all exhibit the same trend of deviation. Same trend means that the deviation direction of the same feature parameter is consistent, and the fluctuation of the deviation amplitude does not exceed 10% of the upper limit of the normal range. For example, for the average diameter (whose upper limit of the normal range is 0.105 mm), if the current window measurement is higher than the upper limit by 0.008 mm, and the subsequent window measurement is higher than the upper limit by 0.007-0.009 mm, since this fluctuation range (0.002 mm) is less than 10% of the upper limit value of 0.105 mm (i.e., 0.0105 mm), it is determined to be of the same trend. If the subsequent window deviates in the opposite direction, the fluctuation range of the deviation amplitude exceeds 10% of the upper limit value, or returns to the normal range, it is determined to be of a different trend.

[0025] If the feature values ​​of at least two subsequent consecutive sequence windows show the same trend of deviation, then the current sequence window and at least two subsequent consecutive sequence windows are jointly identified as a suspected abnormal segment, and the start and end position information of the suspected abnormal segment is entered into the ordered fixed-point detection task list.

[0026] Specifically, the location information of suspected abnormal sections needs to be accurately mapped to the actual physical length of the enameled wire. The starting position corresponds to the position of the first sampling point in the current sequence window along the length of the enameled wire, and the ending position corresponds to the position of the last sampling point in the second consecutive sequence window along the length of the enameled wire.

[0027] The position information is calculated based on the production line operation data collected by the encoder, combined with the timestamp of the sampling point and the real-time speed of the production line, to ensure that the error between the start and end positions is controlled within ±2mm.

[0028] In addition to the start and end positions, the information entered into the ordered fixed-point detection task list should also include the type of characteristic parameter that triggered the anomaly, the specific value of the deviation from the normal range, the deviation trend, etc., so as to provide a clear detection direction for the precision detection of the second station.

[0029] For sequence windows where the feature values ​​do not deviate, and for sequence windows where the features deviate but subsequent consecutive sequence windows do not show the same trend of deviation, they are all judged as normal segments and are not included in the ordered fixed-point detection task list, so as to avoid invalid detection and ensure detection efficiency.

[0030] At the second station, located downstream of the first station, there is a precision inspection device with a detection rate lower than that of high-speed circular scanning. It should be noted that the second workstation is located 20m downstream of the first workstation. This distance is determined based on the production line operating speed and the response time of the precision testing device to ensure that there is sufficient time to complete the testing preparation work.

[0031] The precision testing device uses an ultrasonic pulse-echo thickness gauge, which can distinguish the thickness of paint film layers and differentiate between the thickness of the primer layer and the topcoat layer, with a detection accuracy of 0.001μm. It is also equipped with a micro-area sampling and analysis module, which can perform micro-area component analysis on suspected abnormal sections to determine the coating quality of the paint film.

[0032] The precision inspection device has a detection rate of 50 times / min, which is lower than the 2000Hz detection frequency of the high-speed scanning sensor at the first station. Its effective detection window length is 5cm, suitable for the minimum length requirement of suspected abnormal sections. The device is equipped with a trigger control module that can receive an ordered list of fixed-point inspection tasks from the first station and automatically adjust the trigger timing based on the position information in the list to ensure accurate detection of the target section.

[0033] Based on the orderly fixed-point inspection task list, selective fixed-point inspection is carried out on the enameled wire: when an abnormal section of the enameled wire that is listed in the orderly fixed-point inspection task list enters the inspection window of the precision inspection device as the production line moves, the precision inspection device is triggered to perform coating film layer thickness resolution or micro-area sampling analysis on the abnormal section. It should be noted that the core of the opportunistic fixed-point detection lies in precise position synchronization and trigger control. First, a position synchronization system between the first and second workstations is established. The encoder collects the running speed of the production line and the travel length of the enameled wire in real time. The encoder data sampling frequency is set to 1000Hz to match the sampling rhythm of the high-speed scanning sensor to ensure the accuracy of position calculation. Every 10m of enameled wire travel distance, the error is calibrated by the scanning data of the first workstation and the position feedback of the second workstation, and the synchronization error is controlled within ±2mm.

[0034] The specific steps to trigger the precision testing device are as follows: Establishing a length coordinate mapping relationship between the first workstation and the second workstation specifically includes: setting at least one physical position reference mark with a known fixed spacing between the first workstation and the second workstation along the length direction of the production line; Specifically, physical position reference marks (circular marks, 0.5 mm wide, 0.001 mm deep, with a contrast of ≥80%) are set on the surface of the enameled wire, with a spacing of 5 m between adjacent marks.

[0035] During the scanning process, when the high-speed circumferential scanning device detects the physical position reference mark on the surface of the enameled wire, it records the first appearance position of the mark in the first station coordinate system (based on the real-time data of the first station encoder, with an error of ±0.1mm). Meanwhile, the associated vision sensor of the precision inspection device (resolution 1024×768, frame rate 500fps) monitors the inspection window area in real time. When the physical position reference mark reaches the center of the inspection window, it records the second appearance position in the second station coordinate system (based on the second station encoder data, with an error of ±0.1mm).

[0036] Based on the known fixed spacing of these two locations and reference marks, the relevant parameters of the length coordinate mapping relationship are periodically calculated and corrected to compensate for the cumulative slippage caused by the slippage of the traction wheel during the transmission of the enameled wire, or the tensile deformation caused by temperature changes and tension fluctuations (usually deformation error ≤0.5%).

[0037] The specific correction method is as follows: calculate the spacing correction value and the slip compensation coefficient (correction value = second occurrence position - first occurrence position - fixed spacing; compensation coefficient = correction value / fixed spacing + 1), and correct immediately after each mark passes to ensure that the mapping error does not exceed ±1mm.

[0038] Taking the first reference mark between the first and second workstations as an example, its known fixed distance is 5m. The first occurrence position recorded by the high-speed scanning device is 100.000m, and the second occurrence position recorded by the associated sensor of the precision detection device is 125.002m. The calculated distance correction value is 0.002m, and the slip compensation coefficient is 1.0004. In the subsequent mapping relationship, these two corrected values ​​are used to calibrate the coordinates of the suspected abnormal section to avoid triggering deviations caused by slip and deformation.

[0039] After the coordinate mapping relationship is established and periodically corrected, the starting point of enameled wire production is taken as the unified origin. The starting position of the suspected abnormal section identified by the first station is added to the fixed distance of 20m between the first and second stations, multiplied by the slip compensation coefficient, and finally the distance correction value is added to obtain the target position corresponding to the second station, thus ensuring the accuracy of coordinate transformation.

[0040] Subsequently, the length and position of the enameled wire reaching the second station are acquired in real time. The encoder at the second station collects position data every 1ms, and the real-time length and position are obtained after filtering. The data update frequency matches the response speed of the precision detection device. At the same time, the real-time length and position are compared cyclically with the starting position of the next task to be inspected in the ordered fixed-point detection task list. The comparison period is set to 1ms to ensure no position comparison delay.

[0041] When the difference between the real-time length position and the starting position of the next task to be inspected enters the preset warning range, the trigger preparation procedure is started.

[0042] Specifically, the warning range is set based on the maximum operating speed of the production line and the start-up response time of the precision detection device, with a value of 0.5-1.0m (for example, when the production line speed is 500m / min, the start-up response time of the precision detection device is 0.06s, corresponding to a travel distance of 0.5m). After the preparation program is triggered, the precision detection device switches from a low-power standby state to a ready state. Core detection components such as the ultrasonic emission module and the laser sampling module are preheated and started. The detection parameters are preset according to the abnormal characteristics in the ordered fixed-point detection task list (e.g., for diameter deviation abnormalities, the ultrasonic emission frequency is preset to 100MHz, focusing on the diameter measurement dimension).

[0043] When the real-time length position perfectly matches the starting position of the next inspection task, the trigger control module immediately generates an execution trigger signal to start the precision detection device. The trigger signal uses a level signal trigger with a response delay of no more than 10μs, ensuring that the starting end of the abnormal section accurately enters the center of the detection window. After the device starts, it performs detection according to preset parameters. The detection time is set according to the length of the abnormal section and is 1.2 times the time it takes for the abnormal section to pass through the detection window (for example, if the length of the abnormal section is 0.0125m and the production line speed is 500m / min, the passage time is 0.0015s, and the detection time is set to 0.0018s), ensuring complete coverage of the abnormal section.

[0044] Taking the abnormal section starting at 120.0m as an example, after coordinate mapping, its target position at the second station is 140.058m. When the real-time length position at the second station reaches 139.558m, the precision detection device starts the preparation program; when the real-time length position reaches 140.058m, a trigger signal is generated, and the device immediately starts detection to accurately capture the complete data of the abnormal section.

[0045] In the process of distinguishing the thickness of paint film layers, an ultrasonic pulse reflection thickness gauge emits an ultrasonic signal of a specific frequency. This signal generates reflected signals at the interface between the primer layer and the conductor, the interface between the primer layer and the topcoat layer, and the interface between the topcoat layer and the air. By analyzing the time difference and amplitude of the reflected signals, the thickness of the primer layer and the topcoat layer, as well as the total paint film thickness, are calculated.

[0046] The micro-area sampling and analysis module utilizes laser micro-area sampling technology. The laser wavelength is set to 1064nm, the spot diameter is 50μm, and the sampling depth is controlled to 0.002mm to avoid damaging the overall insulation performance of the enameled wire. A trace sample is collected from the surface of the enamel film in the abnormal section, and the composition of the sample is analyzed using an infrared spectrometer. The detection band is set to 780-2500nm, and the spectral resolution is 1cm⁻¹. By comparing the infrared spectrum of the standard enamel film sample (such as characteristic peaks of the -CH2 group at 2940cm⁻¹, N-H bending vibration at 1530cm⁻¹, and C-O stretching vibration at 1250cm⁻¹), the compositional differences in the abnormal section are analyzed (such as the intensity of impurity characteristic peaks and the position of resin crosslinking characteristic peaks) to determine whether there are problems such as uneven coating or impurity contamination.

[0047] Taking the 120-125mm thickness deviation abnormality section as an example, when this section enters the detection window, the precision detection device is triggered and started. The detection results show that the thickness of the primer layer is 0.008mm, the thickness of the topcoat layer is 0.009mm, and the total paint film thickness exceeds the allowable deviation by 0.003mm. Micro-area composition analysis shows that the intensity of the -OH characteristic peak at 3360cm⁻¹ in this section of the paint film is significantly higher than that of the standard sample, confirming the presence of paint process defects caused by free water residue.

[0048] For enameled wire segments not included in the list of ordered fixed-point inspection tasks, the precision inspection device remains in standby mode during the time period when the enameled wire segment passes through the inspection window. It should be noted that the standby state of the precision detection device is a low-power operation mode, in which only the position synchronization system and the trigger control module are kept working, while the other detection modules are turned off, thereby reducing equipment energy consumption and unnecessary mechanical wear.

[0049] The system compares the current position of the enameled wire with the segment positions in the ordered fixed-point detection task list in real time to determine whether the enameled wire segment passing through the detection window is a suspected abnormal segment. When a non-abnormal segment passes through, the trigger control module does not send a start command, and the precision detection device remains in standby mode; the detection program is only triggered when a suspected abnormal segment arrives.

[0050] Compared to the full-line precision testing mode, this opportunistic fixed-point testing mode can reduce the working time of precision testing equipment, improve testing efficiency, reduce equipment operating costs, and avoid the generation of a large amount of redundant data.

[0051] This invention can distinguish the differences in detection requirements between normal and abnormal sections in the production process of enameled wire. By combining high-speed scanning full-coverage screening with opportunistic fixed-point precision detection, it avoids the limitations of insufficient precision of single high-speed detection and solves the pain points of low efficiency and waste of resources in traditional full-line precision detection. Furthermore, it focuses on suspected abnormal sections with quality risks, and accurately locks down the abnormal range through a combination of sequence window division, multi-parameter feature value calculation and continuous trend judgment, effectively eliminating misjudgments caused by accidental factors, which meets the actual needs of high-speed production lines for key quality and rapid response, and improves the pertinence and accuracy of abnormal diagnosis. This invention not only achieves precise alignment between abnormal sections and detection windows, but also compensates for slippage and deformation during transmission through periodic correction, solving the problem of inaccurate timing control in high-speed production lines and further ensuring the integrity and reliability of detection data. The combined application of paint film layer thickness resolution and micro-area component analysis not only achieves precise quantification of thickness parameters, but also allows for in-depth tracing of the causes of anomalies, thus preserving the high efficiency of online detection.

[0052] In some embodiments of this application, the method further includes: Real-time monitoring of the number of tasks that have been triggered but not completed in the orderly fixed-point detection task list, which is recorded as the backlog of tasks. It should be noted that the core of monitoring the backlog of tasks is to track the number of suspected abnormal segments that have triggered detection but have not output detection results. The system has built an independent task status monitoring module that is linked in real time with the execution status of the precision detection device.

[0053] Each time a precision inspection task is triggered, the monitoring module marks it as pending completion. Once the precision inspection device outputs complete inspection data (including layer thickness data and micro-area composition analysis results), it automatically marks it as completed. The backlog of tasks is updated in real-time to reflect the total number of pending tasks, with the update frequency matching the task triggering frequency. This high-frequency, real-time update design ensures the accuracy of the monitoring data, enabling the immediate capture of task backlog trends and preventing missed detections of abnormal sections or data distortion due to monitoring delays, thus guaranteeing the continuity and integrity of the inspection process. If the backlog of tasks exceeds the preset warning threshold, a production line slowdown instruction will be generated. In this embodiment, the warning threshold is set based on the maximum concurrent processing capacity and detection efficiency of the precision testing device. Combining the previously set detection rate (50 times / min), single detection duration (maximum 0.0018s), and the maximum operating speed of the production line (500m / min), the warning threshold is calculated to be 8. This value both reserves equipment processing redundancy, effectively handling instantaneous task peaks, and provides early warning of the risk of overload operation, preventing the precision testing device from becoming saturated due to task accumulation.

[0054] When the monitoring module detects that the backlog of tasks has reached 9 (exceeding the warning threshold of 8), it immediately generates a speed-reduction command through the instruction output interface of the production line control system. The command includes information such as the current backlog of tasks, the warning threshold, and the suggested speed-reduction amount, facilitating precise adjustments by the control system. This early warning mechanism solves the problem of decreased detection accuracy caused by equipment overload in traditional testing, ensuring that the testing device always operates within a highly efficient and stable working range, thus improving the overall reliability of the testing system. Based on the speed reduction command, control the production line to reduce its operating speed until the backlog of tasks is reduced to below the preset safety threshold; In this embodiment, the safety threshold is set to 3, which is far lower than the maximum concurrent processing capacity of the precision detection device, thus reserving sufficient buffer space for the device and ensuring that subsequent tasks can be processed in a timely manner.

[0055] Furthermore, the speed reduction is adjusted in steps. The initial speed reduction is 20% of the current production line operating speed. If the backlog of tasks does not decrease within 3 seconds after the speed reduction, the speed will continue to be reduced in steps of 10%, until the speed is reduced to the minimum stable operating speed of the production line.

[0056] The stepped speed reduction design avoids tension fluctuations or surface damage to the enameled wire caused by sudden speed changes, and can dynamically adjust according to the backlog, minimizing the impact on production efficiency while ensuring testing effectiveness. After receiving the speed reduction command, the production line control system adjusts the frequency converter of the traction motor to achieve a smooth speed reduction. When the monitoring module detects that the backlog of tasks has dropped to 3 or less, it generates a speed recovery command, and the production line gradually restores to the original operating speed in increments of 10% per second.

[0057] This dynamic speed adjustment and recovery mechanism achieves a dynamic balance between detection accuracy and production efficiency. It not only solves the core contradiction between high-speed production and accurate detection, but also copes with scenarios where abnormal sections occur in concentrated areas due to sudden wear of painting molds and fluctuations in raw material composition, thereby improving the production adaptability and anti-interference capability of the detection system. In some embodiments of this application, the method further includes: At the second workstation, upstream of the precision inspection device, a high-speed re-inspection unit is set up; the high-speed re-inspection unit quickly scans the enameled wire segment before it enters the inspection window to obtain local surface features; It should be noted that the high-speed re-inspection unit is installed 1m upstream of the detection window of the precision testing device. This distance ensures that the re-inspection unit has sufficient time to complete scanning and comparison, and also allows it to get as close as possible to the detection window, ensuring that the acquired local surface features are consistent with the state of the section to be inspected, and avoiding secondary changes in the surface state of the enameled wire due to excessive distance.

[0058] The high-speed re-inspection unit uses the same type of laser triangulation scanning sensor as the first station to ensure that the scanning accuracy is consistent with that of the first station and to guarantee the accuracy of feature comparison. Simultaneously, non-contact scanning is employed, with a scanning angle covering the entire 360° circumference to avoid damaging the enameled wire surface coating. This unit only performs local scanning on the enameled wire segment about to enter the detection window, with the scanning range being the length of a single suspected abnormal segment (maximum 0.0125m). The scanning data is transmitted to the feature comparison module in real time, without storing redundant data, thus reducing the system processing load.

[0059] The core benefit of setting up a high-speed re-inspection unit is that it can perform a second confirmation of abnormal sections before the precision detection device is triggered, avoiding misjudgments caused by instantaneous fluctuations in the enameled wire (such as brief tension changes or slight surface impurities falling off), reducing invalid detection actions, and improving the accuracy and resource utilization of the detection system. When an abnormal section in the ordered fixed-point detection task list is about to trigger the precision detection device, the current local surface features obtained by the high-speed re-inspection unit are compared in real time with the historical features corresponding to the abnormal section in the one-dimensional surface feature sequence. It should be noted that the timing for triggering the comparison is set when the abnormal section reaches the scanning range of the high-speed re-inspection unit (i.e., 0.12 seconds before the trigger time of the precision inspection device, calculated based on the maximum operating speed of the production line of 500m / min). This time difference ensures the complete execution of the comparison process while avoiding delays in triggering the precision inspection device. The characteristic parameters for comparison are consistent with the core parameters used when identifying anomalies at the first station, including four key indicators: local diameter average, diameter fluctuation standard deviation, circumferential thickness range, and surface roughness average, ensuring the consistency of the comparison dimensions. The comparison process employs a feature similarity matching algorithm, setting the allowable range for feature deviation to 5% of the upper limit of the normal feature value range. This means that a comparison is considered valid when the deviations of all parameters of the current local surface feature from historical feature parameters do not exceed this range. Historical feature data is directly retrieved from an ordered fixed-point detection task list, which already stores historical feature parameters, start and end positions, etc., for each abnormal segment. This eliminates the need to retrieve data a second time from the first station, improving comparison efficiency. This invention effectively filters transient anomalies through dual-station feature cross-validation, ensuring that the precision testing device only tests sections with continuous anomalies, reducing the number of invalid equipment starts, extending the service life of the precision testing device, reducing detection data redundancy, and improving the efficiency of subsequent data analysis. If the comparison result indicates that the current feature has returned to the normal range, then remove the task item from the ordered fixed-point detection task list and cancel the current triggering of the precision detection device; In this embodiment, the criterion for determining whether the current feature has returned to the normal range is: all four core feature parameters acquired by the high-speed re-inspection unit fall within the preset normal feature value range (consistent with the first station's criterion), and the feature similarity matching degree is ≥98%. When this condition is met, the system automatically performs a task removal operation, deleting the task item through the dynamic update interface of the ordered fixed-point detection task list, and simultaneously sending a cancel trigger signal to the trigger control module to prevent the precision detection device from starting this detection. If the comparison result indicates that the current feature is still in the abnormal range, or the feature deviation exceeds the allowable range, the task item remains unchanged, and the control module is triggered to start the precision detection device according to the original logic; if the comparison result is ambiguous (such as some parameters returning to normal and some still being abnormal), it is treated as an abnormality, the task item is retained and the detection is triggered to avoid missing potential defects. The core beneficial effect of this invention is that it realizes the dynamic error correction capability of the detection system, which can accurately distinguish between continuous anomalies and transient fluctuations, avoids invalid detection from consuming equipment resources and detection time, improves detection accuracy, and further optimizes the collaborative efficiency of production and detection, making the detection system more in line with the dynamic changes in the surface state of enameled wire in actual production. In some embodiments of this application, the method further includes: The second workstation is equipped with at least two precision testing devices with different performance characteristics, each of which is good at analyzing different types of defects. It should be noted that the two heterogeneous precision testing devices configured in the second workstation both adopt the non-contact testing principle, and the testing accuracy reaches 0.001μm, ensuring consistency with the testing standards mentioned above.

[0060] The first type of detection device is an ultrasonic pulse-echo thickness gauge, which is good at analyzing thickness-related defects, including localized excessive thickness, excessive thinness, delamination, pinholes and other anomalies related to the thickness dimension of the paint film. The second type of detection device is the laser Raman spectroscopy analyzer, which is good at analyzing component defects, including anomalies related to material composition such as uneven composition of paint film raw materials, impurity mixing, oxidation and deterioration, and free water residue. The detection windows of the two devices are arranged side by side along the production line with a spacing of 0.2m. Both are located downstream of the high-speed re-inspection unit (0.8m from the re-inspection unit) to ensure that abnormal sections confirmed by re-inspection can enter the detection range of the corresponding detection devices in sequence, and that the devices do not interfere with each other.

[0061] This step enables specialized testing for different types of defects in the enameled wire coating, avoiding the limitations of single devices in analyzing cross-category defects, improving the comprehensiveness and accuracy of defect identification, and enhancing overall testing efficiency through division of labor and collaboration. When generating an ordered list of fixed-point detection tasks, preliminary pattern recognition is performed on the defect type of each suspected abnormal segment based on a one-dimensional surface feature sequence. It should be noted that the preliminary pattern recognition is executed synchronously in the feature analysis module of the first station, and is processed in parallel with the identification process of suspected abnormal sections, without adding additional system latency. The identification is based on the variation patterns of four core parameters (local diameter average, diameter fluctuation standard deviation, circumferential thickness range, and surface roughness average) in the one-dimensional surface feature sequence. Different defect types correspond to specific parameter variation characteristics. For example, in the preset defect pattern library, rules can be defined: when the circumferential thickness range > 0.006 mm and the surface roughness average < 0.02 μm, it is initially identified as a thickness unevenness defect; when the surface roughness average > 0.02 μm and the diameter fluctuation standard deviation > 0.001 mm, it is initially identified as a compositional abnormality defect. Thickness defects are usually characterized by a significant deviation of the diameter average and circumferential thickness range from the normal range, while the surface roughness average is not obviously abnormal; compositional defects are characterized by large fluctuations in the surface roughness average and the diameter fluctuation standard deviation exceeding the threshold, while the thickness parameter is relatively stable. The identification process uses a pre-set defect pattern library for matching. The pattern library stores parameter change thresholds and association rules corresponding to common defect types (such as excessively thick or thin paint films). For example, when the circumferential thickness range is >0.006μm and the average surface roughness is <0.02μm, it is matched as a thickness defect; when the average surface roughness is >0.02μm and the standard deviation of diameter fluctuation is >0.001μm, it is matched as a group defect.

[0062] If the parameter change pattern simultaneously matches the characteristics of multiple defect types, it is marked as a composite defect type and subsequently assigned to two detection devices for sequential detection. By classifying defect types in advance, a precise basis is provided for subsequent device assignment, avoiding waste of detection resources and improving the targeting of detection. Based on the preliminary pattern recognition results, a recommended detection device type identifier is assigned to each task item in the list; In this embodiment, device type identifiers are divided into three categories: thickness detection identifiers, composition detection identifiers, and composite detection identifiers, corresponding to heterogeneous devices and composite defect scenarios in the second workstation. The annotation process is automatically executed by the task list generation module, directly associating identifiers based on the results of preliminary pattern recognition. Tasks identified as thickness defects are marked with a thickness detection label and correspond to the first type of detection device; tasks identified as component defects are marked with a component detection label and correspond to the second type of detection device; tasks identified as composite defects are marked with a composite detection label and correspond to the combined detection of two detection devices. In addition to the existing information (starting position, ending position, historical feature parameters, etc.), a new device identifier field has been added to the task list, facilitating the subsequent trigger control module to quickly read the assignment basis. This ensures that each detection task has a clear execution subject, laying the foundation for accurate subsequent assignment, while also ensuring that complex defects are fully detected and avoiding the omission of critical defect information. When the trigger is executed, the system assigns the task to the corresponding precision detection device based on the recommended detection device type identifier; In this embodiment, after receiving the detection task, the trigger control module first reads the device type identifier of the task item, and then sends a trigger signal to the corresponding precision detection device according to the identifier: The task of labeling thickness detection marks only triggers Type I ultrasonic thickness gauges; the task of labeling composition detection marks only triggers Type II Raman spectrometers. The task of marking composite detection marks triggers two devices in sequence. First, the ultrasonic thickness gauge is triggered to complete the thickness detection. When the section enters the detection window of the second type of device, the Raman spectrometer is triggered to perform the composition detection. The time interval between the two triggers is calculated based on the device spacing and the production line running speed. If a detection device is busy (e.g., processing a previous task), the control module is triggered to add the current task to the task queue of the corresponding device and execute it in sequence. At the same time, the production line speed is dynamically adjusted in conjunction with the task backlog control mechanism mentioned above to avoid task backlog.

[0063] This invention achieves precise matching between inspection tasks and heterogeneous devices, which can improve the depth and accuracy of defect detection; at the same time, through queue management and speed coordination, it ensures the orderliness and stability of the inspection process and avoids device conflicts and missed inspections.

[0064] In some embodiments of this application, the method further includes: Continuously monitor the continuity and rate of change of the length coordinates acquired in real time; It should be noted that the monitoring of the length coordinates is performed independently by the status monitoring module of the position tracking system. The monitored objects include the real-time length coordinates collected by the encoders of the first and second stations, as well as the target coordinates after mapping and calibration. The monitoring frequency is consistent with the coordinate acquisition frequency (1ms / time) to ensure that coordinate anomalies are captured as soon as possible.

[0065] The core of continuous monitoring is to determine whether the difference between two adjacent measurements is within a reasonable range: the theoretical coordinate change is calculated based on the production line running speed, and the allowable deviation is ±5% of the theoretical value. If it exceeds this range, it is determined that the coordinates are discontinuous. The core of rate of change monitoring is to determine the matching degree between the coordinate change rate and the actual running speed of the production line. The actual rate is obtained by calculating the coordinate change per unit time and compared with the running speed fed back by the production line control system. If the deviation exceeds ±10%, it is determined that the rate is abnormal.

[0066] By proactively identifying potential faults in the location tracking system, errors in task assignment or missed detection of abnormal sections due to coordinate jumps or stagnation can be avoided, providing early warning for emergency response and enhancing the system's resilience. If an unexpected jump or pause in the length coordinate is detected, it is determined that the position tracking system has lost synchronization, and the triggering process based on the ordered fixed-point detection task list is immediately suspended. It should be noted that the criteria for determining an unexpected jump are: the amount of a single coordinate change exceeds three times or more of the reasonable range, and there are no preset coordinate correction instructions (such as the spacing correction and slip compensation mentioned above). The criteria for determining unexpected shutdowns are: no change in coordinates for three consecutive monitoring cycles (3ms) and the production line speed not dropping to 0 (excluding normal shutdown scenarios). When any of the above conditions are met, the status monitoring module immediately determines that the position tracking system has lost synchronization and sends a pause command to the trigger control module. After receiving the command, the trigger control module stops executing the triggering operation of the ordered fixed-point detection task list within 10μs, freezes the current task queue, and avoids task assignment based on abnormal coordinates. The innovation of this invention lies in quickly cutting off the abnormal triggering path when a fault occurs, preventing the mis-triggering of precision detection devices or the omission of key sections, minimizing the impact of the fault on the detection quality, and reserving buffer time for fault recovery. At the same time, the precision detection device is switched to a safe detection mode: within the next preset time window, continuous scanning detection is performed on the enameled wire passing through the detection window; In this embodiment, the preset time window length is determined based on the historical recovery time statistics of the position tracking system and is set to 10 seconds. This duration can cover the recovery cycle of most out-of-step faults and avoid resource waste caused by long-term continuous detection.

[0067] In the safety inspection mode, the operating parameters of the precision inspection device are adaptively adjusted: the scanning frequency of the ultrasonic thickness gauge remains unchanged at 2000Hz, and the spectral acquisition frequency of the laser Raman spectrometer is increased to 100Hz (50Hz in normal mode) to ensure continuous scanning of the enameled wire passing through the inspection window without dead angles. The detection data is stored in association with the actual location according to the timestamp. Defect judgment is not performed at this time. After the system is restored to normal, the data can be backtracked and analyzed in conjunction with the coordinate correction results. If the second workstation is equipped with heterogeneous precision inspection devices (such as the thickness and composition inspection devices mentioned above), then in the safety inspection mode, the two devices simultaneously perform continuous scanning, respectively covering thickness-related and composition-related defects, ensuring comprehensive inspection in emergency situations. The beneficial effect of this safety inspection mode is that, during periods of system malfunction, continuous scanning replaces opportunistic fixed-point inspection, avoiding missed inspections due to unreliable coordinates, ensuring inspection coverage of the enameled wire during this time period, and without affecting the normal workflow after subsequent fault recovery. Once the position tracking system recovers and a reliable length coordinate mapping relationship is re-established, the normal opportunistic fixed-point detection process will resume. It should be noted that the criteria for determining the recovery of the position tracking system are: within 10 consecutive monitoring cycles (10ms), the continuity and rate of change of the length coordinates meet the normal standards, and the encoder signal strength is stable. The core of re-establishing a reliable coordinate mapping relationship is to repeat the physical position reference mark calibration process described above: using the next passed physical position reference mark, recalculate the spacing correction value and slip compensation coefficient to ensure that the mapping error is ≤±1mm. At the same time, perform coordinate correction on the continuous scan data stored during the out-of-step period to complete the position information. When the normal process is restored, the trigger control module first unfreezes the task queue, recalibrates the detection tasks that were not executed before the step loss (updates the target position based on the newly established coordinate mapping relationship), and then resumes the opportunistic fixed-point detection according to the original logic. At the same time, it stops the continuous scanning mode of the precision detection device and switches back to the normal working parameters.

[0068] The innovation of this invention lies in its ability to quickly reconnect to the normal testing process after a fault is recovered, ensuring the continuity of testing. At the same time, it uses coordinate correction to supplement the testing data during the emergency period, avoiding data gaps and improving the completeness of the testing results. Secondly, the present invention also proposes a system for detecting the thickness of the enameled wire coating, comprising: The acquisition module, set at the first station on the production line, is used to perform non-contact high-speed circumferential scanning on the moving enameled wire to acquire a complete and continuous data stream of the surface morphology of the enameled wire. The generation module, connected to the acquisition module, is used to convert the data stream into a one-dimensional surface feature sequence along the length of the enameled wire, and based on a preset uniformity criterion, to identify and extract the location information of suspected abnormal sections from the one-dimensional surface feature sequence, and generate an ordered list of fixed-point detection tasks. The detection device is located at the second station downstream of the first station; and, The trigger control module, connected to the generation module and the detection device, is used to control the detection device to perform fixed-point detection on suspected abnormal sections that arrive at its detection window according to the ordered fixed-point detection task list, and to keep non-abnormal sections in standby state.

[0069] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for detecting the coating thickness of enameled wire, used for online detection of enameled wire on a high-speed continuous production line, characterized in that, include: At the first station on the production line, a non-contact high-speed circumferential scan is performed on the moving enameled wire to obtain a complete and continuous data stream of the surface morphology of the enameled wire. The data stream is converted into a one-dimensional surface feature sequence along the length of the enameled wire. Based on a preset uniformity criterion, the location information of suspected abnormal sections is identified and extracted from the one-dimensional surface feature sequence. An ordered fixed-point detection task list is generated along the length direction. The ordered fixed-point detection task list only contains the location information of the suspected abnormal sections. At the second station located downstream of the first station, a precision detection device with a detection rate lower than that of the high-speed circular scan is provided. According to the ordered fixed-point inspection task list, selective fixed-point inspection is carried out on the enameled wire: when an abnormal section of the enameled wire that is listed in the ordered fixed-point inspection task list enters the inspection window of the precision inspection device as the production line moves, the precision inspection device is triggered to perform coating film layer thickness resolution or micro-area sampling analysis on the abnormal section. For enameled wire segments not included in the ordered fixed-point inspection task list, the precision inspection device remains in standby mode during the time period when the enameled wire segment passes through the inspection window.

2. The method for detecting the coating thickness of enameled wire according to claim 1, characterized in that, The steps for identifying and extracting suspected abnormal segments based on a preset uniformity criterion include: The one-dimensional surface feature sequence is divided into continuous sequence windows; Calculate the feature value of the current sequence window and determine whether the feature value deviates from the preset normal feature value range; If the feature value deviates, it is further determined whether the feature values ​​of at least two subsequent consecutive sequence windows of the current sequence window all show the same trend of deviation. If the feature values ​​of the subsequent at least two consecutive sequence windows all show the same trend of deviation, then the current sequence window and the subsequent at least two consecutive sequence windows are jointly determined as a suspected abnormal segment, and the start position information and end position information of the suspected abnormal segment are entered into the ordered fixed-point detection task list.

3. The method for detecting the coating thickness of enameled wire according to claim 1, characterized in that, The steps for triggering the precision detection device to perform include: Establish a length coordinate mapping relationship between the first workstation and the second workstation; The length coordinates of the enameled wire reaching the second workstation are obtained in real time, and the real-time length coordinates are compared cyclically with the starting coordinates of the next task to be inspected in the ordered fixed-point detection task list. When the difference between the real-time length coordinate and the starting coordinate of the next task to be inspected enters the preset warning range, the trigger preparation program is started. When the difference between the real-time length coordinate and the starting coordinate of the next inspection task returns to zero, an execution trigger signal is generated to start the precision inspection device.

4. The method for detecting the coating thickness of enameled wire according to claim 3, characterized in that, The method further includes: The number of tasks that have been triggered but not completed in the ordered fixed-point detection task list is monitored in real time and recorded as the backlog of tasks. If the backlog of tasks exceeds a preset warning threshold, a production line slowdown instruction is generated. According to the speed reduction command, the production line is controlled to reduce its operating speed until the backlog of tasks is reduced to below a preset safety threshold.

5. The method for detecting the coating thickness of enameled wire according to claim 1, characterized in that, The method further includes: Record the point-to-point detection results of the precision detection device for each task in the ordered point-to-point detection task list; The fixed-point detection results are correlated with the one-dimensional surface feature sequences collected at the corresponding positions of the first workstation to generate labeled detection samples; After a complete production batch is completed, based on the test sample set generated in that batch, the preset parameters of the uniformity criterion are statistically analyzed and optimized. The optimized parameters will be applied to subsequent production batches.

6. The method for detecting the coating thickness of enameled wire according to claim 3, characterized in that: The step of establishing the length coordinate mapping relationship includes: At least one physical position reference mark with a known fixed spacing is provided between the first workstation and the second workstation; The high-speed circumferential scanning device records the first occurrence coordinates of the physical position reference mark on the enameled wire during the scanning process; The precision detection device or associated sensor records the second occurrence coordinates when the physical position reference mark reaches the detection window; Based on the first occurrence coordinate and the second occurrence coordinate, the mapping parameters of the length coordinate mapping relationship are periodically calculated and corrected to compensate for the cumulative slip or tensile deformation of the enameled wire during transmission.

7. The method for detecting the thickness of the enamel film on enameled wire according to claim 1, characterized in that, The method further includes: At the second workstation, upstream of the precision testing device, a high-speed re-inspection unit is set up; the high-speed re-inspection unit quickly scans the enameled line segment before it enters the testing window to obtain local surface features; Before a certain abnormal section in the ordered fixed-point detection task list is about to trigger the precision detection device, the current local surface features obtained by the high-speed re-inspection unit are compared in real time with the historical features corresponding to the abnormal section in the one-dimensional surface feature sequence. If the comparison result indicates that the current feature has returned to the normal range, then the task item is removed from the ordered fixed-point detection task list, and the current triggering of the precision detection device is canceled.

8. The method for detecting the coating thickness of enameled wire according to claim 1, characterized in that, The method further includes: The second workstation is equipped with at least two precision testing devices with different performance characteristics, each of which is good at analyzing different types of defects. When generating the ordered fixed-point detection task list, preliminary pattern recognition is performed on the defect type of each suspected abnormal segment based on the one-dimensional surface feature sequence. Based on the results of the preliminary pattern recognition, a recommended detection device type identifier is assigned to each task item in the list; When the trigger is executed, the system assigns the task to the corresponding precision detection device based on the recommended detection device type identifier.

9. The method for detecting the coating thickness of enameled wire according to claim 3, characterized in that, The method further includes: Continuously monitor the continuity and rate of change of the real-time acquired length coordinates; If an unexpected jump or pause is detected in the length coordinate, it is determined that the position tracking system has lost synchronization, and the triggering process based on the ordered fixed-point detection task list is immediately suspended. At the same time, the precision detection device is controlled to switch to a safe detection mode: within the next preset time window, continuous scanning detection is performed on the enameled wire passing through the detection window; Once the position tracking system recovers and a reliable length coordinate mapping relationship is re-established, the normal opportunistic fixed-point detection process resumes.

10. A system for detecting the thickness of enameled wire coating, comprising: The acquisition module, set at the first station on the production line, is used to perform non-contact high-speed circumferential scanning on the moving enameled wire to acquire a complete and continuous data stream of the surface morphology of the enameled wire. The generation module, connected to the acquisition module, is used to convert the data stream into a one-dimensional surface feature sequence along the length of the enameled wire, and based on a preset uniformity criterion, to identify and extract the location information of suspected abnormal sections from the one-dimensional surface feature sequence, and generate an ordered fixed-point detection task list. The detection device is located at the second station downstream of the first station; and, The trigger control module, connected to the generation module and the detection device, is used to control the detection device to perform fixed-point detection on suspected abnormal sections arriving at its detection window according to the ordered fixed-point detection task list, and to keep non-abnormal sections in standby state.

Images