A Multi-Sensor Fusion-Based Method for Detecting Lacquer Glaze in Enameled Wire

Abstract

This invention relates to the field of enameled wire inspection technology, and more particularly to a method for detecting enamel nodules in enameled wire based on multi-sensor fusion. The invention divides the bare wire into several inspection intervals according to a preset length to determine defect risk intervals; determines the corresponding coating inspection strategy; performs pre-inspection on at least one inspection interval at the preheated end of the enameled wire to obtain the temperature rise characteristic value of the inspection interval and determine the interval current period; calculates the average temperature rise slope and holding time ratio in each temperature change curve to generate a rate adjustment coefficient and determine the inspection duration of a single inspection interval; determines whether to perform microscopic assisted inspection on the defect risk interval to obtain the false negative rate; determines whether the enameled wire is qualified, generates a corrected defect degree based on the false negative rate, and classifies qualified enameled wires based on the corrected defect degree; and increases the floating threshold in response to a low false negative rate. This invention improves the accuracy of enamel nodule detection in enameled wire based on multi-sensor fusion.

Description

Technical Field

[0001] The present invention relates to the technical field of enameled wire detection, and particularly to an enameled wire lacquer tumor detection method based on multi-sensor fusion. Background Technique

[0002] The integrity of the insulating layer on the surface of enameled wire directly determines the insulation performance, operation stability and service life of the equipment. Lacquer tumors are typical surface defects caused by factors such as fluctuations in painting process parameters, residual impurities in the paint solution, and insufficient curing during the production process of enameled wire. They not only cause the enameled wire diameter to exceed the tolerance and the insulating layer to be damaged during winding, but may also cause safety hazards such as partial discharge and short circuit. Therefore, accurate and efficient detection of lacquer tumors on the surface of enameled wire has important industrial significance. Traditional lacquer tumor detection methods mainly rely on manual visual inspection, which depends on the experience judgment of operators and has problems such as low detection efficiency, high missed detection rate, and strong subjectivity, making it difficult to meet the high-speed continuous production requirements of modern production lines; some single machine vision detection schemes only identify defects through gray threshold segmentation or simple edge detection, and are easily affected by interference such as surface reflection, color difference, and slight scratches of enameled wire, resulting in misjudgment between lacquer tumors and normal surface textures, and unable to accurately quantify key parameters such as the height and area of lacquer tumors.

[0003] For example, the Chinese patent application publication number: CN113916940A, this invention discloses a lacquer tumor detection method and system for enameled wire. It discloses a lacquer tumor detection method and system for enameled wire, which detects the current signal of the enameled wire in real time through an on-line detection device; sequentially captures the instantaneous large current signals in the current signal to construct a large current sequence; identifies the large current sequence and determines whether there is a continuous lacquer tumor; when there is a continuous lacquer tumor, determines that the enameled wire is a non-conforming product, otherwise it is a conforming product; it can avoid problems such as false alarms caused by minor defects of enameled wire, ignore some small flaws that do not affect the quality of enameled wire, and has a fast identification speed, improving the identification speed and reducing the system overhead, thereby improving the transmission quality and stability of enameled wire and the product quality of enameled wire. [[ID= twelve]]

[0004] The following problems still exist in the prior art:

[0005] Only detecting and analyzing the current signal after the enameled wire is powered on cannot accurately capture the lacquer tumor and cannot conduct targeted re-inspection while improving the detection efficiency. Summary of the Invention

[0006] Therefore, the present invention provides an enameled wire lacquer tumor detection method based on multi-sensor fusion to overcome the problem in the prior art that only detecting and analyzing the current signal after the enameled wire is powered on cannot accurately capture the lacquer tumor and cannot conduct targeted re-inspection while improving the detection efficiency.

[0007] To achieve the above objectives, the present invention provides a method for detecting enamel nodules on enameled wires based on multi-sensor fusion, comprising:

[0008] The bare wire is divided into several detection intervals according to a preset length, several bare wire images of each detection interval are obtained, and the bare wire characterization coefficient of each detection interval is determined to determine the defect risk interval.

[0009] The quality fluctuation value is calculated based on the characterization coefficient of each bare wire, and the paint detection strategy of the bare wire is determined by combining the preset fluctuation threshold.

[0010] In response to the detection of intermittent current flowing through the enameled wire, at least one detection zone at the front end of the preheated enameled wire is pre-detected to obtain a temperature rise characteristic value, and the intermittent current period is adjusted based on the temperature rise characteristic value.

[0011] In response to the interval current cycle, detection is performed to obtain the temperature change curve of each detection interval, and the average temperature rise slope and the proportion of holding time in each temperature change curve are calculated to generate a rate adjustment coefficient. The detection duration of a single detection interval is determined based on the rate adjustment coefficient.

[0012] A detection deviation value is generated based on the detection results of adjacent defect risk intervals. Based on the detection deviation value, it is determined whether to perform microscopic assisted detection on the defect risk interval. In response to performing microscopic assisted detection, the false negative rate is obtained.

[0013] The defect degree is calculated by obtaining the microscopic auxiliary inspection results for the non-defect risk range and the defect risk range respectively, and the qualification of the enameled wire is determined.

[0014] In response to the judgment result of a high missed detection rate, a corrected defect rate is generated based on the defect rate and the missed detection rate, and qualified enameled wires are classified based on the corrected defect rate.

[0015] In response to the determination that the false negative rate is low, the preset length is increased and the floating threshold is raised.

[0016] Furthermore, the process of determining the bare wire characterization coefficients includes,

[0017] Obtain the bare wire diameter at several detection points within the detection interval, and calculate the absolute value of the difference ratio between each bare wire diameter and the standard wire diameter as the wire diameter deviation factor.

[0018] The standard deviation of the wire diameter deviation factor is determined as the bare wire characterization coefficient.

[0019] Furthermore, the process of determining the defect risk range includes,

[0020] If the bare wire characterization coefficient of the detection interval exceeds a preset bare wire characterization coefficient threshold, then the detection interval is determined to be a defect risk interval.

[0021] Furthermore, the process of determining the paint detection strategy includes,

[0022] Calculate the difference between the maximum and minimum values ​​of each bare wire characterization coefficient, and the ratio of the difference to the average value of each bare wire characterization coefficient to obtain the quality fluctuation value of the bare wire.

[0023] If the quality fluctuation value is less than a preset fluctuation threshold, the paint coating detection strategy is determined to be to perform detection by passing a constant current through the enameled wire.

[0024] If the quality fluctuation value is greater than or equal to a preset fluctuation threshold, the paint detection strategy is determined to be to perform detection by passing an interval current through the enameled wire.

[0025] Furthermore, the process of adjusting the interval current period based on the aforementioned temperature rise characteristic value includes,

[0026] The ratio of the heating rate to the standard heating rate is determined as the heating effect factor;

[0027] The ratio of the temperature difference value to the standard temperature difference value is determined as the temperature difference influence factor.

[0028] The weighted sum of the temperature rise influence factor and the temperature difference influence factor is determined to be the temperature rise characteristic value;

[0029] In response to the temperature rise characteristic value being less than the minimum value of the temperature rise characteristic range, the interval current period is increased;

[0030] In response to the temperature rise characteristic value being greater than the maximum value of the temperature rise characteristic range, the interval current period is reduced.

[0031] Furthermore, the process of determining whether to increase the detection time of a single detection interval based on the rate adjustment coefficient includes,

[0032] The average temperature rise slope is calculated by obtaining several temperature rise slopes from the temperature change curve.

[0033] The ratio of the average holding time of the highest temperature in the temperature change curve to the interval current period is determined as the holding time percentage.

[0034] The ratio of the average heating slope to the holding time is determined as the rate adjustment coefficient;

[0035] If the rate adjustment coefficient is less than or equal to the rate adjustment coefficient threshold, it is determined that the detection time of a single detection interval will be increased.

[0036] Furthermore, the process of determining whether to perform microscopic-assisted inspection on the defect risk range based on the detection deviation value includes,

[0037] Calculate the ratio of the difference in the number of lacquer lumps between the previous defect risk interval and the ratio of the difference in the projected area of ​​the lacquer lumps between the next defect risk interval;

[0038] The weighted sum of the ratio of the difference in the number of lacquer lumps and the ratio of the difference in the projected area of ​​the lacquer lumps is determined as the detection deviation value;

[0039] If the detection deviation value is greater than the detection deviation threshold, it is determined that the defect risk range should be subjected to microscopic assisted detection.

[0040] The detection results include the number of lacquer lumps per unit length and the projected area of ​​the lacquer lumps.

[0041] Furthermore, the process of obtaining the false negative rate in response to microscopic-assisted detection includes,

[0042] Obtain the auxiliary detection results of microscopic auxiliary detection for defect risk range and compare and analyze them with the corresponding detection results;

[0043] Calculate the ratio of the number of lacquer ducts detected by auxiliary detection to the difference between the number of lacquer ducts detected by auxiliary detection;

[0044] Calculate the ratio of the projected area of ​​the auxiliary detection lacquer cyst to the difference between the projected area of ​​the auxiliary detection lacquer cyst and the projected area of ​​the auxiliary detection lacquer cyst.

[0045] The weighted sum of the difference ratio of the number of auxiliary detected lacquer lumps and the difference ratio of the projected area of ​​the auxiliary detected lacquer lumps is determined as the false negative rate;

[0046] The auxiliary detection results include the number of auxiliary lacquer lumps per unit length and the projected area of ​​the auxiliary lacquer lumps.

[0047] Furthermore, the process of determining whether enameled wire is qualified includes,

[0048] Based on the detection results of the non-defect risk range and the defect risk range, the total number of lacquer nodules per unit length of enameled wire and the total projected area of ​​lacquer nodules are calculated and generated.

[0049] The ratio of the total number of lacquer cysts to the standard total number of lacquer cysts is determined as the total number of lacquer cysts factor;

[0050] The ratio of the total projected area of ​​lacquer smears to the standard total projected area of ​​lacquer smears is determined as the total projected area factor of lacquer smears.

[0051] The weighted sum of the total number of lacquer ducts factor and the total projected area factor of lacquer ducts is determined as the defect degree.

[0052] When the defect degree is less than or equal to a preset defect degree threshold, the enameled wire is deemed qualified.

[0053] If the defect degree is greater than a preset defect degree threshold, the enameled wire is determined to be unqualified.

[0054] Furthermore, the process of classifying qualified enameled wires based on the corrected defect degree includes,

[0055] In response to a high false negative rate, the product of the false negative rate and 1 and the defect degree of the qualified enameled wire is determined as the corresponding corrected defect degree.

[0056] In response to the fact that the correction defect degree is greater than the correction defect degree threshold, the qualified enameled wires corresponding to the correction defect degree are classified as enameled wires to be re-inspected.

[0057] Compared with existing technologies, the beneficial effects of this invention are as follows: By acquiring the bare wire diameter at several detection points within the detection interval and calculating the deviation between the bare wire diameter and the standard wire diameter, the smoothness of the bare wire is characterized. This allows for the prediction of lacquer nodules caused by protrusions and depressions on the surface of the bare wire during the subsequent coating process. Furthermore, the invention analyzes the differences in surface conditions at different locations of the same bare wire. Based on the analysis results, a coating detection strategy using a constant current through the enameled wire is adopted for areas with small surface fluctuations, while a strategy using intermittent currents is adopted for areas with large surface fluctuations. The constant current coating detection strategy is more efficient than the intermittent current strategy, allowing for targeted coating detection and improved efficiency. By addressing the root cause of lacquer nodule formation and accurately analyzing the differences in protrusions, depressions, and fluctuations on the bare wire surface, the risk of lacquer nodule defects during the coating process can be predicted in advance, improving the accuracy of lacquer nodule detection for enameled wires based on multi-sensor fusion.

[0058] Furthermore, this invention performs pre-detection on at least one detection section at the front end of the preheated enameled wire by responding to the flow of an intermittent current within the wire. This allows for the acquisition of the heating rate and temperature difference characteristics of the defect-risk section. Based on these characteristics, the intermittent current period is determined. Simultaneously, detection is performed within this determined intermittent current period, acquiring the average heating slope and holding time percentage in the temperature change curve. The ratio of the average heating slope to the holding time percentage characterizes the efficiency of the enameled wire's temperature change, achieving a quantitative assessment of the enameled wire's temperature response characteristics. For enameled wires with slower temperature changes, the detection time is increased to obtain more accurate results; for enameled wires with faster temperature changes, the detection time is reduced to improve detection efficiency. Through feature-driven period adaptation, the intermittent current detection can accurately focus on high-risk defect points, improving the detection coverage of risk areas and avoiding redundant detection in risk-free areas, further enhancing the accuracy of enameled wire nodule detection based on multi-sensor fusion.

[0059] Furthermore, this invention evaluates the detection results by analyzing the degree of detection deviation in the number of lacquer nodules and the projected area of ​​lacquer nodules in the detection results between adjacent defect risk intervals. If the deviation in the detection results is too large, microscopic-assisted detection needs to be introduced, and the microscopic-assisted detection results are verified with the detection results of the corresponding defect risk intervals to determine the false negative rate. The distribution and characteristics of lacquer nodules in adjacent defect risk intervals are correlated. By analyzing the degree of detection deviation in the number of lacquer nodules and the projected area of ​​lacquer nodules between intervals, abnormal detection results in the multi-sensor fusion detection process can be quickly identified. By verifying the microscopic-assisted detection results with the detection results of the corresponding defect risk intervals, the false negative rate is quantitatively evaluated, effectively reducing the false positive rate and false negative rate of multi-sensor fusion detection, and further improving the accuracy of lacquer nodule detection of enameled wire based on multi-sensor fusion.

[0060] Furthermore, this invention determines the qualification of enameled wire by analyzing the detection results of the overall enamel nodule. When the false negative rate is too high, qualified enameled wires are classified according to the false negative rate to determine whether re-inspection is necessary. Differentiated classification and re-inspection management of qualified wires based on the false negative rate achieves precise allocation of testing resources and controllable management of quality risks. When the false negative rate is too high, it indicates a significant accuracy deficiency in the current testing system. Classifying qualified enameled wires and determining whether re-inspection is necessary effectively avoids the quality risk of "false acceptance" and prevents subsequent electrical equipment malfunctions due to undetected enamel nodule defects. When the false negative rate is low, by increasing the floating threshold and using constant current detection, the detection efficiency is improved, further enhancing the accuracy of enamel nodule detection based on multi-sensor fusion. Attached Figure Description

[0061] Figure 1 This is a flowchart illustrating the steps of the multi-sensor fusion-based method for detecting enamel nodules on enameled wires according to an embodiment of the present invention.

[0062] Figure 2 A logic flowchart for determining the paint detection strategy for bare wires in an embodiment of the present invention;

[0063] Figure 3 A logic flowchart for determining the interval current period in an embodiment of the present invention;

[0064] Figure 4 This is a flowchart illustrating the logic of determining whether to perform microscopic-assisted detection on the defect risk range in an embodiment of the present invention. Detailed Implementation

[0065] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0066] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0067] It should be noted that in the description of this invention, the terms "upper," "lower," "inner," "outer," etc., which indicate the direction or positional relationship, are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0068] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0069] Please see Figure 1 The diagram shows the steps of a multi-sensor fusion-based method for detecting lacquer lumps in enameled wire according to an embodiment of the present invention. The method includes:

[0070] Step S1: Divide the bare wire into several detection intervals according to a preset length, acquire several bare wire images of each detection interval, and determine the bare wire characterization coefficient of each detection interval to determine the defect risk interval.

[0071] Step S2: Calculate and generate quality fluctuation values ​​based on the characterization coefficients of each bare wire, and determine the paint detection strategy for the bare wire by combining the preset fluctuation threshold.

[0072] Step S3: In response to the detection by passing an interval current through the enameled wire, pre-detect at least one detection interval at the front end of the preheated enameled wire to obtain the temperature rise characteristic value, and adjust the interval current period based on the temperature rise characteristic value.

[0073] Step S4: In response to the interval current cycle, the temperature change curve of each detection interval is obtained, the average temperature rise slope and the holding time ratio in each temperature change curve are calculated to generate the rate adjustment coefficient, and the detection time of a single detection interval is determined based on the rate adjustment coefficient.

[0074] Step S5: Generate a detection deviation value based on the detection results of adjacent defect risk intervals, and determine whether to perform microscopic assisted detection on the defect risk interval based on the detection deviation value. In response to performing microscopic assisted detection, obtain the false negative rate.

[0075] Step S6: Obtain the microscopic auxiliary inspection results for the non-defect risk range and the defect risk range respectively, calculate the defect degree, and determine whether the enameled wire is qualified.

[0076] In response to the judgment result of a high missed detection rate, a corrected defect score is generated based on the defect score and the missed detection rate, and qualified enameled wires are classified based on the corrected defect score.

[0077] Step S7: In response to the judgment result of a low false negative rate, increase the preset length and raise the floating threshold.

[0078] Specifically, several bare wire images of each detection zone can be obtained by taking bare wire images from different angles using an industrial camera, and by using a linear array light source to preprocess and reduce noise in the acquired images in order to obtain the bare wire diameter of the detection point based on the images. The method of obtaining the diameter of the target object from the images is not limited here.

[0079] Specifically, for enameled wire, the structural differences in varnish nodules lead to different heat dissipation characteristics compared to normal varnish layers, resulting in temperature anomalies. The temperature difference between nodules and normal varnish layers arises because the enameled wire heats up according to Joule's law when energized. The bare wire is the primary heat source, while the varnish layer, being an insulating material with extremely high resistance, generates almost no Joule heat. Heat is transferred to the varnish surface via thermal conduction and then dissipated through convection and radiation. The presence of nodules disrupts this heat transfer balance, creating temperature anomalies that can be detected by thermal imaging. Nodules are essentially heat transfer defects; whether thick varnish layers or air cavities, they lead to a decrease in local heat dissipation efficiency, forming temperature anomalies identifiable by thermal imaging. Therefore, by energizing the enameled wire and identifying the temperature characteristics during the energizing process, the location and number of nodules can be identified. This method is suitable for conventional insulating varnishes such as polyurethane, polyester, and polyimide, but not for varnishes with special thermal conductivity characteristics such as ceramicized or glass fiber-coated varnishes.

[0080] Specifically, when conducting interval current testing inside the enameled wire, the enameled wire is preheated to bring the surface temperature of the bare wire to 40℃~80℃, which must be 15℃~50℃ higher than the ambient temperature. If the temperature is too low, the temperature difference will not be obvious, and if it is too high, it may cause the enamel layer to soften and age. In addition, the preheating temperature is generally lower than 1 / 2 of the heat resistance grade of the enamel layer.

[0081] Specifically, the test current for the same batch of enameled wires is determined through several tests. For example, for copper enameled wires with a diameter of 1mm, a current of 5A to 10A is required, and the temperature rises to 30℃ after 40 seconds of energization.

[0082] It is understandable that infrared thermal imaging devices, such as online infrared thermal imagers or linear infrared thermal imaging systems, which can capture local temperature differences, can be used to detect lacquer lumps.

[0083] It is understandable that for those skilled in the art, the existing technology is to collect and calculate the average temperature of different detection points at different times in each detection interval using a temperature measuring device to generate a temperature change curve, and will not be elaborated here.

[0084] It is understandable that for those skilled in the art, microscopic-assisted detection refers to the use of a microscope lens in conjunction with a high-resolution microscopic imaging system such as a high-speed industrial camera to observe the microscopic morphology, size, and structural characteristics of lacquer lumps. This is existing technology and will not be elaborated upon here.

[0085] Specifically, in step S1, the process of determining the bare wire characterization coefficients includes,

[0086] Obtain the bare wire diameter at several test points within the test interval, and calculate the absolute value of the difference between each bare wire diameter and the standard wire diameter as the wire diameter deviation factor.

[0087] It is understandable that the difference between the bare wire diameter and the standard wire diameter is the difference between the bare wire diameter and the standard wire diameter divided by the standard wire diameter.

[0088] The standard deviation of the wire diameter deviation factor is determined as the bare wire characterization coefficient.

[0089] Specifically, the standard wire diameter refers to the standard bare wire diameter corresponding to different specifications of enameled wire, which can be determined according to the process specifications. The number of inspection points is generally no less than 3, and preferably 5.

[0090] Specifically, in step S1, the process of determining the defect risk range includes,

[0091] If the bare wire characterization coefficient of a single detection interval exceeds the preset bare wire characterization coefficient threshold, the detection interval is determined to be a defect risk interval.

[0092] If the bare wire characterization coefficient of a single detection interval does not exceed the preset bare wire characterization coefficient threshold, the detection interval is determined to be a non-defect risk interval.

[0093] Specifically, the preset bare wire characterization coefficient threshold is the product of the bare wire characterization coefficient reference value and the bare wire characterization factor. The bare wire characterization coefficient reference value is the average value of the bare wire characterization coefficient detected under the same working conditions in historical data. The bare wire characterization factor can be set by those skilled in the art according to the accuracy requirements of bare wire detection. The higher the accuracy requirement, the smaller the value should be. The value range can be [1.0, 1.2], preferably 1.1.

[0094] Please see Figure 2 As shown, it is a logic flowchart of determining the paint detection strategy for bare wires according to an embodiment of the present invention. In step S2, the process of determining the paint detection strategy includes,

[0095] Calculate the difference between the maximum and minimum values ​​of each bare wire characterization coefficient, and the ratio of the difference to the average value of each bare wire characterization coefficient to obtain the quality fluctuation value of the bare wire.

[0096] If the quality fluctuation value is less than the preset fluctuation threshold, the paint coating detection strategy is determined to be to perform detection by passing a constant current through the enameled wire.

[0097] If the quality fluctuation value is greater than or equal to the preset fluctuation threshold, the paint coating detection strategy is determined to be to perform detection by passing an interval current through the enameled wire.

[0098] Specifically, the purpose of setting a floating threshold is to characterize the consistency of the surface smoothness of the bare wire. The floating threshold can be in the range of [0.9, 1.1].

[0099] Specifically, the coating detection strategy of passing a constant current through the enameled wire is to keep the detection current constant during the coating detection process.

[0100] Specifically, the coating detection strategy of passing an intermittent current through the enameled wire is to switch the detection current on and off periodically during the coating detection process.

[0101] In one specific embodiment, the coating detection strategy of passing a constant current through the enameled wire is as follows: the current is set to 5A and the detection time is set to 10s.

[0102] In a specific embodiment, the coating detection strategy of passing an interval current inside the enameled wire is as follows:

[0103] The current is set to 5A, the initial detection time is set to 10s, and the initial interval current cycle is set to 2s.

[0104] Specifically, this invention acquires the bare wire diameter at several detection points within a detection range, calculates the deviation between the bare wire diameter and the standard wire diameter, and characterizes the smoothness of the bare wire. This helps predict lacquer nodules caused by protrusions and depressions on the bare wire surface during the subsequent coating process. Simultaneously, it analyzes the differences in surface conditions at different locations on the same bare wire. Based on the analysis results, a coating detection strategy using a constant current through the enameled wire is adopted for areas with small surface fluctuations, while a strategy using intermittent currents is adopted for areas with large surface fluctuations. The constant current coating detection strategy is more efficient than the intermittent current strategy, allowing for targeted coating detection and improved efficiency. By addressing the root cause of lacquer nodule formation and accurately analyzing the differences in protrusions, depressions, and fluctuations on the bare wire surface, the risk of lacquer nodule defects during the coating process can be predicted in advance, improving the accuracy of lacquer nodule detection for enameled wires based on multi-sensor fusion.

[0105] Specifically, when performing the coating detection strategy of passing a constant current through the enameled wire, for a copper enameled wire with a diameter of 1mm, the detection current is set to 5A, the detection time is usually set to 10s, and the temperature change curve of each detection interval is obtained, and the process of steps S5 to S7 is executed.

[0106] Please see Figure 3 The diagram shown is a logic flowchart for determining the interval current period according to an embodiment of the present invention. In step S3, the process of adjusting the interval current period based on the temperature rise characteristic value includes:

[0107] The ratio of the heating rate to the standard heating rate is determined as the heating effect factor;

[0108] The ratio of the temperature difference value to the standard temperature difference value is determined as the temperature difference influence factor.

[0109] The weighted sum of the temperature rise influence factor and the temperature difference influence factor is determined as the temperature rise characteristic value;

[0110] Understandably, the temperature difference is the difference between the maximum and minimum temperatures measured in the detection range of the enameled wire during the energizing process.

[0111] Specifically, in implementation, the heating rate is determined based on the temperature rise per unit time of the defect risk zone collected during the pre-inspection process, the temperature difference is determined by the difference between the highest temperature and the initial temperature of the defect risk zone collected during the pre-inspection process, the standard heating rate is the average of the heating rates of several enameled wires of the same specification that were tested without paint lumps under the same working conditions, and the standard temperature difference is the average of the temperature differences of several enameled wires of the same specification that were tested without paint lumps under the same working conditions.

[0112] Specifically, the sum of the weighting coefficients of the temperature rise influence factor and the temperature difference influence factor is 1. Preferably, the weighting coefficient of the temperature rise influence factor is 0.6 and the weighting coefficient of the temperature difference influence factor is 0.4.

[0113] In one specific embodiment, the interval current period is increased in response to the temperature rise characteristic value being less than the minimum value of the temperature rise characteristic range;

[0114] In response to the temperature rise characteristic value being greater than the maximum value of the temperature rise characteristic range, the interval current period is reduced.

[0115] Specifically, the purpose of setting the temperature rise characteristic range is to characterize the consistency of the temperature rise rate and temperature difference value when an intermittent current is passed through the enameled wire with the same temperature rise rate and temperature difference value as that of the enameled wire of the same specification under the same working conditions. The minimum value of the temperature rise characteristic range is selected within the range [0.8, 0.9], and the maximum value of the temperature rise characteristic range is selected within the range [1.1, 1.2]. Preferably, the minimum value of the temperature rise characteristic range is 0.8, and the maximum value of the temperature rise characteristic range is 1.2.

[0116] Specifically, in step S4, the process of determining whether to increase the detection time of a single detection interval based on the rate adjustment coefficient includes:

[0117] The average temperature rise slope is calculated by obtaining several temperature rise slopes from the temperature change curve.

[0118] The ratio of the average holding time of the highest temperature in the temperature change curve to the interval current period is determined as the holding time percentage.

[0119] The ratio of the average heating slope to the holding time is determined as the rate adjustment coefficient;

[0120] Specifically, the rate adjustment coefficient = average heating slope / holding time percentage.

[0121] If the rate adjustment coefficient is less than or equal to the rate adjustment coefficient threshold, it is determined that the detection time of a single detection interval should be increased.

[0122] If the rate adjustment coefficient is greater than the rate adjustment coefficient threshold, the detection time of a single detection interval is reduced.

[0123] Specifically, the rate adjustment coefficient threshold is the product of the rate adjustment coefficient reference value and the rate adjustment factor. The rate adjustment coefficient reference value is the average value of the rate adjustment coefficient under the same working conditions in historical data. The rate adjustment factor can be set by those skilled in the art according to the accuracy requirements of temperature detection. The higher the accuracy requirement, the smaller the value should be. The value range can be [1.0, 1.3], and preferably, it can be 1.1.

[0124] In one specific embodiment, when the detection time of a single detection interval is increased, the adjusted detection time is 1.3 times the original detection time; when the detection time of a single detection interval is decreased, the adjusted detection time is 0.7 times the original detection time.

[0125] Specifically, the initial value of the detection duration of a single detection interval is an integer multiple of the interval current period. Preferably, the detection duration of a single detection interval is twice the interval current period.

[0126] Specifically, this invention detects defects by passing an intermittent current through the enameled wire. It pre-detects at least one detection zone at the front end of the preheated enameled wire, acquiring the characteristics of the heating rate and temperature difference in the defect risk zone. Based on these characteristics, the intermittent current period is determined. Simultaneously, detection is performed within the determined intermittent current period, obtaining the average heating slope and the proportion of holding time in the temperature change curve. The ratio of the average heating slope to the proportion of holding time is analyzed to characterize the efficiency of the enameled wire's temperature change, achieving a quantitative evaluation of the enameled wire's temperature response characteristics. For enameled wires with slower temperature changes, the detection time is increased to obtain more accurate results; for enameled wires with faster temperature changes, the detection time is reduced to improve detection efficiency. Through feature-driven period adaptation, the intermittent current detection can accurately focus on high-risk defect points, improving the detection coverage of risk areas and avoiding redundant detection in risk-free areas, further enhancing the accuracy of enameled wire nodule detection based on multi-sensor fusion.

[0127] Please see Figure 4 The diagram shown is a flowchart illustrating the logic of determining whether to perform microscopic-assisted detection on the defect risk area according to an embodiment of the present invention. In step S5, the process of determining whether to perform microscopic-assisted detection on the defect risk area based on the detection deviation value includes...

[0128] Calculate the ratio of the difference in the number of lacquer lumps between the previous defect risk interval and the ratio of the difference in the projected area of ​​the lacquer lumps between the next defect risk interval;

[0129] The weighted sum of the ratio of the difference in the number of lacquer lumps and the ratio of the difference in the projected area of ​​the lacquer lumps is the detection deviation value.

[0130] It is understandable that the difference between the number of lacquer ducts in the defect risk interval and the next defect risk interval is the difference between the number of lacquer ducts in the defect risk interval and the number of lacquer ducts in the next defect risk interval divided by the number of lacquer ducts in the next defect risk interval.

[0131] It is understandable that the ratio of the difference between the projected area of ​​the lacquer smear in the defect risk interval and the next defect risk interval is the difference between the projected area of ​​the lacquer smear in the defect risk interval and the projected area of ​​the lacquer smear in the next defect risk interval divided by the projected area of ​​the lacquer smear in the next defect risk interval.

[0132] Specifically, the sum of the weighting coefficients of the difference ratio of the number of lacquer lumps and the difference ratio of the projected area of ​​lacquer lumps is 1. Since the difference ratio of the number of lacquer lumps and the projected area of ​​lacquer lumps have roughly the same impact on the detection results, the weighting coefficient of the difference ratio of the number of lacquer lumps and the weighting coefficient of the difference ratio of the projected area of ​​lacquer lumps are generally taken as 0.5.

[0133] In one specific embodiment, in response to a detection deviation value being greater than a detection deviation threshold, it is determined that a microscopic-assisted inspection should be performed on the defect risk zone.

[0134] If the detection deviation value is less than or equal to the detection deviation threshold, it is determined that no microscopic auxiliary detection will be performed on the defect risk range.

[0135] The test results include the number of lacquer lumps per unit length and the projected area of ​​the lacquer lumps.

[0136] Specifically, the number of detected lacquer lumps per unit length is the ratio of the total number of detected lacquer lumps within the detection interval to the length of the detection interval; the projected area of ​​detected lacquer lumps per unit length is the ratio of the total projected area of ​​detected lacquer lumps within the detection interval to the length of the detection interval.

[0137] Specifically, the purpose of setting the detection deviation threshold is to characterize the consistency of the enameled wire detection results. The detection deviation threshold can be set in the range of [0.1, 0.3].

[0138] Specifically, in step S5, the process of obtaining the false negative rate in response to performing microscopic-assisted detection includes:

[0139] Obtain the auxiliary detection results of microscopic auxiliary detection for defect risk range and compare and analyze them with the corresponding detection results;

[0140] Calculate the ratio of the number of lacquer ducts detected by auxiliary detection to the difference between the number of lacquer ducts detected by auxiliary detection;

[0141] Calculate the ratio of the projected area of ​​the auxiliary detection lacquer cyst to the difference between the projected area of ​​the auxiliary detection lacquer cyst and the projected area of ​​the auxiliary detection lacquer cyst.

[0142] Understandably, the difference ratio of the number of lacquer cysts detected by auxiliary detection is the difference between the number of lacquer cysts detected by auxiliary detection and the total number of lacquer cysts, divided by the total number of lacquer cysts.

[0143] It is understandable that the difference ratio of the projected area of ​​the auxiliary detection lacquer cyst is the difference between the projected area of ​​the auxiliary detection lacquer cyst and the projected area of ​​the lacquer cyst divided by the projected area of ​​the lacquer cyst.

[0144] The weighted sum of the difference ratio of the number of lacquer cysts detected by auxiliary detection and the difference ratio of the projected area of ​​lacquer cysts detected by auxiliary detection is determined as the false negative rate;

[0145] The auxiliary detection results include the number of auxiliary lacquer lumps per unit length and the projected area of ​​the auxiliary lacquer lumps.

[0146] Specifically, the number of auxiliary detection lacquer lumps per unit length is the ratio of the total number of lacquer lumps detected by microscopy within the detection interval to the length of the detection interval; the projected area of ​​auxiliary detection lacquer lumps per unit length is the ratio of the total projected area of ​​lacquer lumps detected by microscopy within the detection interval to the length of the detection interval.

[0147] Specifically, the sum of the weighting coefficients of the difference ratio of the number of auxiliary lacquer lumps and the difference ratio of the projected area of ​​auxiliary lacquer lumps is 1. Since the difference ratio of the number of auxiliary lacquer lumps and the difference ratio of the projected area of ​​auxiliary lacquer lumps have roughly the same impact on the detection results, the weighting coefficient of the difference ratio of the number of auxiliary lacquer lumps and the difference ratio of the projected area of ​​auxiliary lacquer lumps are generally taken as 0.5.

[0148] Specifically, this invention evaluates the detection results by analyzing the degree of detection deviation in the number of lacquer nodules and the projected area of ​​lacquer nodules in the detection results between adjacent defect risk intervals. If the deviation in the detection results is too large, microscopic-assisted detection needs to be introduced, and the microscopic-assisted detection results are verified with the detection results of the corresponding defect risk intervals to determine the false negative rate. The distribution and characteristics of lacquer nodules in adjacent defect risk intervals are correlated. By analyzing the degree of detection deviation in the number of lacquer nodules and the projected area of ​​lacquer nodules between intervals, abnormal detection results in the multi-sensor fusion detection process can be quickly identified. By verifying the microscopic-assisted detection results with the detection results of the corresponding defect risk intervals, a quantitative assessment of the false negative rate is achieved, effectively reducing the false positive rate and false negative rate of multi-sensor fusion detection, and further improving the accuracy of lacquer nodule detection of enameled wire based on multi-sensor fusion.

[0149] Specifically, in step S6, the process of determining whether the enameled wire is qualified includes,

[0150] Based on the detection results of the non-defect risk range and the defect risk range, the total number of lacquer nodules per unit length of enameled wire and the total projected area of ​​lacquer nodules are calculated and generated.

[0151] The ratio of the total number of lacquer cysts to the standard total number of lacquer cysts is determined as the total number of lacquer cysts factor;

[0152] The ratio of the total projected area of ​​lacquer smears to the standard total projected area of ​​lacquer smears is determined as the total projected area factor of lacquer smears.

[0153] The weighted sum of the total number of lacquer nodules and the total projected area of ​​lacquer nodules is determined as the defect degree.

[0154] Specifically, in implementation, the standard total number of enamel tumors is the maximum allowable number of enamel tumors for enameled wires of the same specification under the condition of meeting the design requirements, and the standard total projected area of enamel tumors is the maximum allowable total area of enamel tumors for enameled wires of the same specification under the condition of meeting the design requirements.

[0155] Specifically, the sum of the weight coefficients of the total number factor of enamel tumors and the total projected area factor of enamel tumors is 1, the weight coefficient of the total number factor of enamel tumors is 0.5, and the weight coefficient of the total projected area factor of enamel tumors is 0.5.

[0156] In response to the defect degree being less than or equal to the preset defect degree threshold, it is determined that the enameled wire is qualified;

[0157] In response to the defect degree being greater than the preset defect degree threshold, it is determined that the enameled wire is unqualified.

[0158] Specifically, the purpose of setting the preset defect degree threshold is to characterize the consistency between the detection result of the enameled wire and the maximum value of the design-allowed detection result, and the value range of the preset defect degree threshold can be [0.8, 1.0].

[0159] Specifically, in response to the missed detection rate being less than or equal to the preset missed detection rate threshold, it is determined that the missed detection rate is low;

[0160] In response to the missed detection rate being greater than the preset missed detection rate threshold, it is determined that the missed detection rate is high.

[0161] Specifically, the purpose of setting the preset missed detection rate threshold is to characterize the consistency between the auxiliary detection result of the detection interval under microscopic assistance and the detection result of the same detection interval, and the value range of the preset missed detection rate threshold can be [1.0, 1.1].

[0162] Specifically, in step S6, the process of classifying qualified enameled wires based on the corrected defect degree includes

[0163] In response to a high missed detection rate, it is determined that the product of the sum of the missed detection rate and 1 and the defect degree of the qualified enameled wire is the corresponding corrected defect degree, that is, the corrected defect degree = (missed detection rate + 1) * defect degree of the qualified enameled wire;

[0164] Specifically, in response to the corrected defect degree being greater than the corrected defect degree threshold, the qualified enameled wire corresponding to the corrected defect degree is classified as an enameled wire to be re-inspected;

[0165] In response to the corrected defect degree being less than or equal to the corrected defect degree threshold, the qualified enameled wire corresponding to the corrected defect degree is classified as an enameled wire that does not require re-inspection.

[0166] Specifically, the purpose of setting the correction defect threshold is to characterize the consistency between the predicted detection results and the maximum value of the designed allowable detection results of the enameled wire under the condition of combined false negative rate. The correction defect threshold can be in the range of [0.9, 1.1].

[0167] Specifically, this invention determines the quality of enameled wire by analyzing the detection results of the overall enamel nodule. When the false negative rate is too high, qualified enameled wires are classified based on the false negative rate to determine whether re-inspection is necessary. Differentiated classification and re-inspection management of qualified wires based on the false negative rate achieves precise allocation of testing resources and controllable management of quality risks. When the false negative rate is too high, it indicates a significant accuracy deficiency in the current testing system. Classifying qualified enameled wires and determining whether re-inspection is necessary effectively avoids the quality risk of "false acceptance" and prevents subsequent electrical equipment malfunctions due to undetected enamel nodule defects. When the false negative rate is low, increasing the floating threshold and using constant current detection improves both testing efficiency and accuracy of enameled wire enamel nodule detection based on multi-sensor fusion.

[0168] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0169] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for detecting enamel nodules in enameled wire based on multi-sensor fusion, characterized in that, include: The bare wire is divided into several detection intervals according to a preset length, several bare wire images of each detection interval are obtained, and the bare wire characterization coefficient of each detection interval is determined to determine the defect risk interval. The quality fluctuation value is calculated based on the characterization coefficient of each bare wire, and the paint detection strategy of the bare wire is determined by combining the preset fluctuation threshold. In response to the detection of intermittent current flowing through the enameled wire, at least one detection zone at the front end of the preheated enameled wire is pre-detected to obtain a temperature rise characteristic value, and the intermittent current period is adjusted based on the temperature rise characteristic value. In response to the interval current cycle, detection is performed to obtain the temperature change curve of each detection interval, and the average temperature rise slope and the proportion of holding time in each temperature change curve are calculated to generate a rate adjustment coefficient. The detection duration of a single detection interval is determined based on the rate adjustment coefficient. A detection deviation value is generated based on the detection results of adjacent defect risk intervals. Based on the detection deviation value, it is determined whether to perform microscopic assisted detection on the defect risk interval. In response to performing microscopic assisted detection, the false negative rate is obtained. The defect degree is calculated by obtaining the microscopic auxiliary inspection results for the non-defect risk range and the defect risk range respectively, and the qualification of the enameled wire is determined. In response to the judgment result of a high missed detection rate, a corrected defect rate is generated based on the defect rate and the missed detection rate, and qualified enameled wires are classified based on the corrected defect rate. In response to the determination that the false detection rate is low, the preset length is increased and the floating threshold is raised; The process of determining the paint inspection strategy includes, Calculate the difference between the maximum and minimum values ​​of each bare wire characterization coefficient, and the ratio of the difference to the average value of each bare wire characterization coefficient to obtain the quality fluctuation value of the bare wire. If the quality fluctuation value is less than a preset fluctuation threshold, the paint coating detection strategy is determined to be to perform detection by passing a constant current through the enameled wire. If the quality fluctuation value is greater than or equal to a preset fluctuation threshold, the paint coating detection strategy is determined to be to perform detection by passing an interval current through the enameled wire. The process of adjusting the interval current period based on the aforementioned temperature rise characteristic value includes: The ratio of the heating rate to the standard heating rate is determined as the heating effect factor; The ratio of the temperature difference value to the standard temperature difference value is determined as the temperature difference influence factor. The weighted sum of the temperature rise influence factor and the temperature difference influence factor is determined to be the temperature rise characteristic value; In response to the temperature rise characteristic value being less than the minimum value of the temperature rise characteristic range, the interval current period is increased; In response to the temperature rise characteristic value being greater than the maximum value of the temperature rise characteristic range, the interval current period is reduced; The process of determining whether to increase the detection time of a single detection interval based on the rate adjustment coefficient includes: The average temperature rise slope is calculated by obtaining several temperature rise slopes from the temperature change curve. The ratio of the average holding time of the highest temperature in the temperature change curve to the interval current period is determined as the holding time percentage. The ratio of the average heating slope to the holding time is determined as the rate adjustment coefficient; If the rate adjustment coefficient is less than or equal to the rate adjustment coefficient threshold, it is determined that the detection time of a single detection interval will be increased.

2. The method for detecting enamel nodules in enameled wire based on multi-sensor fusion according to claim 1, characterized in that, The process of determining the bare wire characterization coefficients includes, Obtain the bare wire diameter at several detection points within the detection interval, and calculate the absolute value of the difference ratio between each bare wire diameter and the standard wire diameter as the wire diameter deviation factor. The standard deviation of the wire diameter deviation factor is determined as the bare wire characterization coefficient.

3. The method for detecting enamel nodules in enameled wire based on multi-sensor fusion according to claim 2, characterized in that, The process of determining the defect risk range includes, If the bare wire characterization coefficient of the detection interval exceeds a preset bare wire characterization coefficient threshold, then the detection interval is determined to be a defect risk interval.

4. The method for detecting enamel nodules in enameled wire based on multi-sensor fusion according to claim 3, characterized in that, The process of determining whether to perform microscopic-assisted inspection on the defect risk range based on the detection deviation value includes: Calculate the ratio of the difference in the number of lacquer lumps between the previous defect risk interval and the ratio of the difference in the projected area of ​​the lacquer lumps between the next defect risk interval; The weighted sum of the ratio of the difference in the number of lacquer lumps and the ratio of the difference in the projected area of ​​the lacquer lumps is determined as the detection deviation value; If the detection deviation value is greater than the detection deviation threshold, it is determined that the defect risk range should be subjected to microscopic assisted detection. The detection results include the number of lacquer lumps per unit length and the projected area of ​​the lacquer lumps.

5. The method for detecting enamel nodules in enameled wire based on multi-sensor fusion according to claim 4, characterized in that, In response to performing microscopic-assisted detection, the process of obtaining the false negative rate includes, Obtain the auxiliary detection results of microscopic auxiliary detection for defect risk range and compare and analyze them with the corresponding detection results; Calculate the ratio of the number of lacquer ducts detected by auxiliary detection to the difference between the number of lacquer ducts detected by auxiliary detection; Calculate the ratio of the projected area of ​​the auxiliary detection lacquer cyst to the difference between the projected area of ​​the auxiliary detection lacquer cyst and the projected area of ​​the auxiliary detection lacquer cyst. The weighted sum of the difference ratio of the number of auxiliary detected lacquer lumps and the difference ratio of the projected area of ​​the auxiliary detected lacquer lumps is determined as the false negative rate; The auxiliary detection results include the number of auxiliary lacquer lumps per unit length and the projected area of ​​the auxiliary lacquer lumps.

6. The method for detecting enamel nodules in enameled wire based on multi-sensor fusion according to claim 5, characterized in that, The process of determining whether enameled wire is qualified includes, Based on the detection results of the non-defect risk range and the defect risk range, the total number of lacquer nodules per unit length of enameled wire and the total projected area of ​​lacquer nodules are calculated and generated. The ratio of the total number of lacquer cysts to the standard total number of lacquer cysts is determined as the total number of lacquer cysts factor; The ratio of the total projected area of ​​lacquer smears to the standard total projected area of ​​lacquer smears is determined as the total projected area factor of lacquer smears. The weighted sum of the total number of lacquer ducts factor and the total projected area factor of lacquer ducts is determined as the defect degree. When the defect degree is less than or equal to a preset defect degree threshold, the enameled wire is deemed qualified. If the defect degree is greater than a preset defect degree threshold, the enameled wire is determined to be unqualified.

7. The method for detecting enamel nodules in enameled wire based on multi-sensor fusion according to claim 6, characterized in that, The process of classifying qualified enameled wires based on the aforementioned correction defect degree includes: In response to a high false negative rate, the product of the false negative rate and 1 and the defect degree of the qualified enameled wire is determined as the corresponding corrected defect degree. In response to the fact that the correction defect degree is greater than the correction defect degree threshold, the qualified enameled wires corresponding to the correction defect degree are classified as enameled wires to be re-inspected.

Images