A high-voltage liquid-cooled cable in a vehicle and its process

Abstract

The present invention relates to the technical field of cable manufacturing, and specifically discloses an in-vehicle high-pressure liquid-cooled cable and a process thereof, and proposes a surface defect detection and control method integrating image acquisition, segmentation, defect recognition, quality judgment and dynamic pressure adjustment. A camera is installed on a production line to collect cable surface images in real time, and image segmentation technology is used to highlight the cable surface; computer vision technology is then used to identify surface defects, calculate defect areas and contour features, form a comprehensive score and compare it with a preset threshold to generate a quality signal; at the same time, extruder pressure data is collected, and the correlation between defects and pressure fluctuations is analyzed; a prediction model is constructed based on the obtained data, the extruder pressure setting is dynamically adjusted, and the production process is optimized until the defects are effectively controlled. This system improves the efficiency of cable surface quality detection and effectively deals with quality problems caused by pressure fluctuations in production.

Description

Technical Field

[0001] The invention relates to the technical field of cable manufacturing, and in particular to an in-vehicle high-voltage liquid-cooled cable and a process thereof. Background Art

[0002] With the rapid development of the new energy vehicle industry, the demand for in-vehicle high-voltage liquid-cooled cables is increasing. However, during the cable production process, various defects may appear on the cable surface, such as scratches, pits and bumps, which not only affect the appearance quality of the cable, but also its electrical performance and service life. Therefore, it is very important to establish an effective cable surface defect detection and control system.

[0003] At present, the commonly used cable surface defect detection methods mainly include manual visual inspection, traditional image processing technology and machine vision technology. Manual visual inspection is inefficient and easily affected by subjective factors; although traditional image processing technology can identify defects to a certain extent, its robustness and accuracy in complex environments are limited; although machine vision technology has the advantages of high precision and high speed, it still faces problems such as difficulty in defect feature extraction and inaccurate classification in practical applications.

[0004] The present invention aims to provide an efficient and accurate cable surface defect detection and control system. By introducing advanced computer vision technology and data analysis methods, it can detect cable surface defects in real time on the production line, and automatically adjust production equipment parameters by analyzing the relationship between defects and production parameters, thereby effectively improving the quality of cable products. Summary of the invention

[0005] The purpose of the present invention is to provide an in-vehicle high-pressure liquid-cooled cable and a process thereof. The technical solution of the present invention includes image acquisition and processing, defect identification and quantification, quality judgment, and pressure data collection and analysis and dynamic adjustment. Specifically, a camera is installed at a key position of the cable production line to collect the cable surface image in real time, and the cable surface area is separated from the background by image segmentation technology, and each pixel point in the segmented cable surface area image is mapped to a plane coordinate system; computer vision technology is used to automatically detect cable surface defects, and the area of ​​each defect area is calculated by an image processing algorithm, and the outline of the defect area is extracted by an edge detection algorithm and its aspect ratio is calculated, and then according to the defect, the cable surface area is automatically detected. The comprehensive defect score is calculated based on factors such as the area, length, position and shape of the cable; the comprehensive defect score is compared with the preset threshold to determine whether the cable surface quality is qualified, and a surface defect qualified or unqualified signal is generated based on the comparison result; the pressure data of the extruder is regularly recorded, the mean and standard deviation are calculated, and the relationship between the cable surface defects and the extruder pressure fluctuations is analyzed using statistical methods; based on the correlation analysis results, a linear regression model is used to predict the effect of the extruder pressure on the cable surface defects, the extruder pressure setting is dynamically adjusted according to the model prediction results, the production process is optimized, and the adjusted cable surface defects are monitored in real time. If necessary, the above steps are repeated until the defects are effectively controlled.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A process for in-vehicle high-voltage liquid-cooled cables, comprising the following steps:

[0008] Collect cable surface images in real time to obtain defect areas in the cable surface area images;

[0009] Process the cable surface area image to identify and obtain the cable surface defect information;

[0010] Among them, the cable surface defect information includes the area, length, location and shape of each defect area;

[0011] Based on the cable surface defect signal, a comprehensive defect score is obtained to generate a cable surface defect status signal;

[0012] The cable surface defect status signal includes a surface defect qualified signal and a surface defect unqualified signal;

[0013] Based on the unqualified signal of cable surface defects, the relationship between cable surface defects and extruder pressure fluctuations is analyzed;

[0014] Based on the relationship between cable surface defects and extruder pressure fluctuations, a prediction model is constructed to obtain the optimal extruder pressure value that minimizes cable surface defects.

[0015] Dynamically adjust the extruder pressure setting based on the optimal extruder pressure value.

[0016] As a further solution of the present invention: the identification process of the defective area in the cable surface area image is specifically as follows:

[0017] During the cable production process, collect cable surface images;

[0018] Segment the cable surface image and extract the cable surface area image;

[0019] Based on the cable surface area image, obtaining position information of each pixel in the cable surface area image;

[0020] Map each pixel in the cable surface area image as a point into the plane coordinate system to obtain the coordinate position of each pixel, which is recorded as a pixel point;

[0021] The cable surface area image is processed for recognition, and the defect area in the cable surface area image is identified.

[0022] As a further solution of the present invention: the process of obtaining the cable surface defect information is specifically as follows:

[0023] Based on the identified defective area in the cable surface area image, all pixel points in each defective area in the cable surface area image are obtained;

[0024] Process the pixels in each defect area to obtain the area of ​​the defect area;

[0025] Process all pixels in the defect area to obtain the complete outline of the defect area;

[0026] Based on the complete outline of the defect area, the coordinates of all pixel points on the edge outline of the defect area are obtained;

[0027] The distance between every two points on the edge contour of the defect area is calculated, and the distance between every two points is processed to obtain the distance between the farthest two points on the edge contour of the defect area, which is recorded as the length of the defect area.

[0028] As a further solution of the present invention: based on the area and length of the defect area, all the pixels in the defect area are processed to calculate the coordinates of the center point of each defect area;

[0029] Based on the coordinates of the center point of each defect area, the position of each defect is marked in the cable surface area image, and a defect position distribution map is drawn;

[0030] Convert the defect location distribution map into a heat map;

[0031] Analyze the thermal map of defects to obtain the distribution of defects on the entire cable surface;

[0032] Based on the distribution of defects on the entire cable surface, the number of defects per unit area is obtained.

[0033] As a further solution of the present invention: based on the complete outline of the defect area, the product of 4 times π and the area of ​​the defect area is ratio-calculated with the square of the perimeter of the defect area to obtain the roundness value of the defect area;

[0034] If the roundness value of the defect area is greater than the roundness value threshold, it means that the contour of the defect area is close to a circle, and vice versa;

[0035] The width and length of the contour of the defect area are compared to obtain the ellipticity value. If the ellipticity value is greater than the ellipticity value threshold, it means that the contour of the defect area is close to an ellipse, and vice versa.

[0036] Based on the roundness and ellipticity values, the defect areas are divided into circular defects, elliptical defects and irregular defects according to their shapes.

[0037] As a further solution of the present invention: the formula for the comprehensive defect score is ZH=a1×M+a2×C+a3×W+a4×XZ;

[0038] Among them, M is the total area of ​​cable surface defects, C is the total length of cable surface defects, and W is the total position score of all defects. x z and z are the ordinate and abscissa of the center point of the defect area, j is the number of defect areas on the cable surface, j = 1, 2, ..., n, XZ is the total shape score of all defect areas on the cable surface,

[0039] As a further solution of the present invention: comparing the comprehensive defect score of the cable surface with a defect score threshold;

[0040] If the defect comprehensive score is greater than or equal to the defect score threshold, a surface defect unqualified signal is generated;

[0041] If the defect comprehensive score is less than the defect score threshold, a surface defect qualified signal is generated.

[0042] As a further solution of the present invention: the process of analyzing the relationship between the cable surface defects and the extruder pressure fluctuations is:

[0043] Divide the monitoring period into a number of time sub-units of equal length, obtain the pressure data of the extruder when the time sub-unit is in the middle moment, and record it as the unit pressure value;

[0044] Processing all unit pressure values ​​within the monitoring period to obtain a pressure data group;

[0045] The pressure data group is processed according to the mean value calculation formula to obtain the pressure mean value within the monitoring period;

[0046] The pressure data group is processed according to the standard deviation calculation formula to obtain the pressure standard deviation within the monitoring period;

[0047] Calculate the difference between each unit pressure value and the pressure mean value to obtain the deviation value between the unit pressure and the mean value;

[0048] The deviation value of the unit pressure from the mean value is calculated by ratio with the pressure standard deviation to obtain the standard value of the unit pressure value;

[0049] Obtain the comprehensive defect score of the cable surface when the time subunit is in the middle of the monitoring period;

[0050] The relationship between cable surface defects and extruder pressure fluctuations was analyzed using the Spearman rank correlation coefficient;

[0051] The analysis results were tested by the t-value test method, and the t-value of the correlation between the cable surface defects and the extruder pressure fluctuation was obtained;

[0052] The absolute value of t is compared with the significance threshold. If the absolute value of t is greater than the critical value threshold, a defect pressure impact signal is generated.

[0053] If the absolute value of t is less than or equal to the critical value threshold, a defect pressure non-influence signal is generated.

[0054] As a further solution of the present invention: based on the defect pressure influence signal, a prediction model is constructed by linear regression;

[0055] The cable surface defects under different pressure settings are predicted through the prediction model;

[0056] Based on the model prediction results, the optimal extruder pressure value is obtained;

[0057] Based on the optimal extruder pressure value, the pressure of the extruder is regulated.

[0058] As a further solution of the present invention: real-time monitoring of cable surface defects after adjusting the extruder pressure value; if the comprehensive score of cable surface defects after adjustment is less than the comprehensive score of cable surface defects before adjustment, but still greater than the defect score threshold, the extruder pressure is readjusted according to the latest score.

[0059] Beneficial effects of the present invention:

[0060] (1) The present invention achieves high-precision recognition of cable surface defects by adopting advanced image acquisition and processing technology. First, a high-definition camera is used to obtain the cable surface image, and the cable surface area is accurately extracted using image segmentation technology to ensure the accuracy of subsequent analysis. Secondly, the specific location and distribution of the defect are accurately located by mapping the pixel position to a plane coordinate system, and then the defect area is automatically marked and the area is calculated with the help of a computer vision algorithm. Compared with the traditional manual visual inspection method, the detection efficiency and consistency are greatly improved. The present invention also introduces edge detection and contour tracking technology to obtain detailed morphological information of defects, so that even defects as small as micron level can be detected without omission. The combined effect of these improvement measures not only greatly reduces the false detection rate and missed detection rate, but also provides reliable data support for subsequent quality judgment, ensuring that each cable can meet high quality standards.

[0061] (2) In response to the common surface defect problems in the cable production process, the present invention proposes a method for adjusting the extruder working parameters based on the defect analysis results. After accumulating a large amount of data on the type, location and severity of cable surface defects in the early detection stage, the data are deeply mined using statistical principles, and the correlation between the extruder pressure fluctuations and the cable surface defects is analyzed. When the extruder pressure is in an unstable state, the cable surface is more likely to have unevenness or other forms of defects. The present invention establishes a mathematical model between the extrusion pressure and the cable surface quality, and uses a linear regression algorithm to predict the expected defect probability under different pressure settings, thereby guiding factory technicians to reasonably set the operating parameters of the extruder. This measure not only helps to suppress the occurrence of surface defects from the root, but also can significantly improve the appearance flatness of the cable products, ultimately achieving the goal of improving both production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] The present invention will be further described below in conjunction with the accompanying drawings.

[0063] Figure 1 It is a flowchart of the specific steps of an in-vehicle high-voltage liquid-cooled cable and its process of the present invention;

[0064] Figure 2 is a flow chart of obtaining cable surface defect information in the present invention;

[0065] Figure 3 It is a flow chart for analyzing the relationship between cable surface defects and extruder pressure fluctuations in the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Example 1

[0068] See also Figure 1 As shown, the present invention is a process for in-vehicle high-voltage liquid-cooled cables, comprising the following steps:

[0069] Collect cable surface images in real time to obtain cable surface area images;

[0070] Process the cable surface area image to identify and obtain the cable surface defect information;

[0071] Among them, the cable surface defect information includes the area, length, location and shape of each defect area;

[0072] Based on the cable surface defect signal, a comprehensive defect score is obtained to generate a cable surface defect status signal;

[0073] The cable surface defect status signal includes a surface defect qualified signal and a surface defect unqualified signal;

[0074] Based on the unqualified signal of cable surface defects, the relationship between cable surface defects and extruder pressure fluctuations is analyzed;

[0075] Based on the relationship between cable surface defects and extruder pressure fluctuations, a prediction model is constructed to obtain the optimal extruder pressure value that minimizes cable surface defects.

[0076] Dynamically adjust the extruder pressure setting based on the optimal extruder pressure value.

[0077] Example 2

[0078] See also Figure 2 As shown, during the cable production process, the surface image of the produced cable is collected;

[0079] Segment the cable surface image and extract the cable surface area image;

[0080] Based on the cable surface area image, obtaining position information of each pixel in the cable surface area image;

[0081] Map each pixel in the cable surface area image as a point into the plane coordinate system to obtain the coordinate position of each pixel, which is recorded as a pixel point;

[0082] Performing recognition processing on the cable surface area image to identify defect areas in the cable surface area image;

[0083] It should be noted that the processing process of the cable surface area is the existing computer vision technology, which automatically analyzes the image through computer vision technology to identify scratches, pits and protrusions on the cable surface;

[0084] Acquire all pixel points in each defect area in the cable surface area image;

[0085] Process the pixels in each defect area to obtain the area of ​​the defect area;

[0086] It should be noted that the area of ​​the defective region can be obtained by processing the defective region area image through an image processing algorithm;

[0087] Process all pixels in the defect area to obtain the complete outline of the defect area;

[0088] It should be noted that the image of the defect area can be processed by the edge detection algorithm to identify the boundary of the defect. Once the boundary is found, the complete outline of the defect can be depicted by the contour tracking technology. The complete outline of the defect can help us determine the specific shape and location of the defect.

[0089] Based on the complete outline of the defect area, the coordinates of all pixel points on the edge outline of the defect area are obtained;

[0090] Calculate the distance between every two points on the edge contour of the defect area, process the distance between every two points, and obtain the distance between the farthest two points on the edge contour of the defect area, which is recorded as the length of the defect area;

[0091] comparing the length of each defect region with a defect length threshold;

[0092] If the length of the defect area is greater than the threshold, the defect is marked as a slender defect;

[0093] It should be noted that the edge contour measurement of the defect area and the length calculation of the defect are achieved through the edge detection algorithm built into the computer. The edge detection algorithm is an existing and mature image processing technology. The slender defects may be scratches or cracks. The surface area of ​​the defect may be small, but the severity is not small, so the length of the defect area needs to be calculated separately;

[0094] Process all the pixels in the defect area and calculate the coordinates of the center point of each defect area;

[0095] The formula for calculating the center point of the defect area is:

[0096] Among them, (x z ,y z ) is the coordinate of the center point of the defect area, x i and i is the coordinate of each pixel in the defect area, N is the number of pixels in the defect area, i = 1, 2, ..., N;

[0097] Based on the coordinates of the center point of each defect area, the position of each defect is marked in the cable surface area image, and a defect position distribution map is drawn;

[0098] Convert the defect location distribution map into a heat map;

[0099] Analyze the thermal map of defects to obtain the distribution of defects on the entire cable surface;

[0100] Based on the distribution of defects on the entire cable surface, the number of defects per unit area is obtained;

[0101] Based on the complete outline of the defect area, the actual shape of the defect area is differentiated from the ideal circle or ellipse;

[0102] Among them, the roundness formula for differential analysis of the actual shape of the defect area and the ideal circle is: MA is the area of ​​the defect area, ZP is the perimeter of the defect area, and YD is the roundness value. If the roundness value is greater than the roundness value threshold, it means that the outline of the defect area is close to a circle, and vice versa;

[0103] The difference analysis between the actual shape of the defect area and the ellipse is to calculate the ratio of the width to the length of the outline of the defect area to obtain the ellipse value. If the ellipse value is greater than the ellipse value threshold, it means that the outline of the defect area is close to an ellipse, and vice versa.

[0104] The aspect ratio of the defect area is obtained by calculating the ratio of the length to the width of the contour of the defect area. If the aspect ratio is greater than the aspect ratio threshold, it means that the contour of the defect area deviates from a circle or an ellipse.

[0105] Comprehensively analyze the roundness value, ellipticity value and aspect ratio of the actual shape of the above defect area;

[0106] If the roundness value and ellipticity value are greater than the threshold, and the aspect ratio is less than or equal to the threshold, the actual shape of the defect area is considered regular;

[0107] Otherwise, the actual shape of the defect area is considered irregular;

[0108] It should be noted that by analyzing the regularity or irregularity of the shape of the defect area, the types of cable surface defects can be obtained. Different defect types often have different shape characteristics. For example, scratches usually appear as thin and long lines, while pits appear as circles or ovals. By analyzing the shape characteristics, defects can be classified into different types, so as to better understand the nature of the defects. Defects of different shapes may represent different degrees of severity. For example, a large-area irregular defect may be more serious than a small-area circular defect.

[0109] Combining shape features with other indicators (such as area, position, etc.) can more comprehensively evaluate the severity of defects. By analyzing the shape of the contour, the location of the defect can be more accurately located, improving the accuracy of defect detection. In the continuous production process, by tracking the changes in the contour, the development of defects can be monitored in real time and timely measures can be taken;

[0110] Process the above indicators, specifically:

[0111] The areas of all defective regions in the cable surface area image are processed to obtain the total defective area of ​​the cable surface;

[0112] The lengths of all defective areas in the cable surface area image are processed to obtain the total length of cable surface defects;

[0113] Based on the roundness value and ellipticity value, the defect area is divided into circular defects, elliptical defects and irregular defects according to shape;

[0114] Among them, the shape of the defect area is divided into 3 levels according to the level, recorded as Z k , k=1,2,...,n, n is the total number of defective areas on the cable surface. If the kth defective area is a circular defect, then Z k =1, if the kth defect area is an elliptical defect, then Z k =2, if the kth defect area is an irregular defect, then Z k =3;

[0115] The comprehensive score of cable surface defects is calculated by the formula ZH = a1×M+a2×C+a3×W+a4×XZ;

[0116] Among them, M is the total area of ​​cable surface defects, C is the total length of cable surface defects, and W is the total position score of all defects. x z and z are the ordinate and abscissa of the center point of the defect area, j is the number of defect areas on the cable surface, j = 1, 2, ..., n, XZ is the total shape score of all defect areas on the cable surface,

[0117] comparing the composite defect score of the cable surface with a defect score threshold;

[0118] If the comprehensive defect score is greater than or equal to the defect score threshold, it means that the surface defect of the cable section is too large, affecting the cable quality, and a surface defect unqualified signal is generated;

[0119] If the comprehensive defect score is less than the defect score threshold, it means that the surface defect of the cable section is small and does not affect the cable quality, and a surface defect qualified signal is generated;

[0120] It should be noted that comprehensive quantitative evaluation of cable surface defects has multiple benefits, significance and functions. By detecting minor defects at an early stage and accurately controlling production parameters, product quality can be significantly improved, scrap rate can be reduced, and resource allocation can be optimized, thereby reducing production costs and improving production efficiency. Automated detection and real-time monitoring not only reduce the errors and workload of manual detection, but also achieve rapid response, reducing downtime and maintenance costs.

[0121] Example 3

[0122] See also Figure 3 As shown, based on the surface defect unqualified signal generated above, due to the different extruder pressures in the production process of the high-pressure liquid-cooled cable, the cable surface defects will be unqualified. It is analyzed whether the cable surface defects are related to the pressure of the extruder. If there is a relationship of the influence of the extruder pressure, it is convenient to adjust the extruder pressure in time in the subsequent production process, reduce the influence of the extruder pressure on the cable surface in the production process, effectively maintain the quality of the cable in the production process, and then the yield rate of the high-pressure liquid-cooled cable production process, specifically:

[0123] In a specific embodiment:

[0124] Divide the monitoring period into a number of time sub-units of equal length, obtain the pressure data of the extruder when the time sub-unit is in the middle moment, and record it as the unit pressure value;

[0125] Processing all unit pressure values ​​within the monitoring period to obtain a pressure data group;

[0126] The pressure data group is processed according to the mean value calculation formula to obtain the pressure mean value within the monitoring period;

[0127] The pressure data group is processed according to the standard deviation calculation formula to obtain the pressure standard deviation within the monitoring period;

[0128] Calculate the difference between each unit pressure value and the pressure mean value to obtain the deviation value between the unit pressure and the mean value;

[0129] The deviation value of the unit pressure from the mean value is calculated by ratio with the pressure standard deviation to obtain the standard value of the unit pressure value;

[0130] Obtain the comprehensive defect score of the cable surface when the time subunit is in the middle of the monitoring period;

[0131] Arrange the defect comprehensive scores of the cable surface and the standard values ​​of the extruder pressure values ​​in chronological order, and align the timestamps of the defect comprehensive scores and the standard values ​​of the pressure values;

[0132] The relationship between cable surface defects and extruder pressure fluctuations was analyzed using the Spearman rank correlation coefficient;

[0133] The specific steps of Spearman's rank correlation coefficient are:

[0134] The cable surface defect comprehensive score data ZH and the extruder pressure value YL are ranked;

[0135] Among them, the comprehensive defect score data ZH of the cable surface during the monitoring period are ZH1, ZH2, ..., ZH g , the extruder pressure values ​​YL are YL1, YL2, ..., WL g ;

[0136] Sort ZH and YL by rank, and get the ranks rank(ZH1), rank(ZH2), ..., rank(ZH g ) and rank(YL1),rank(YL2),...,rank(YL g );

[0137] The formula for the Spearman rank correlation coefficient is: Among them, d a is the level difference of the a-th time subunit, i.e., d a =rank(ZH a )-rank(YL a ), g is the number of samples;

[0138] By formula Examining the connection between cable surface defects and extruder pressure fluctuations;

[0139] The absolute value of t is compared with the significance threshold. If the absolute value of t is greater than the critical value threshold, it means that there is a significant correlation between the cable surface defect and the extruder pressure fluctuation, and a defect pressure influence signal is generated.

[0140] If the absolute value of t is less than or equal to the critical value threshold, it means that there is no significant correlation between the cable surface defect and the extruder pressure fluctuation, and a defect pressure non-influence signal is generated;

[0141] It should be noted that through the above steps, we can effectively analyze the connection between cable surface defects and extruder pressure fluctuations and find the correlation between the two, which helps us take corresponding measures in the production process to optimize the extruder pressure setting and thus improve the quality of the cable.

[0142] Example 4

[0143] Based on the defect pressure influence signal, a prediction model is constructed through linear regression to predict cable surface defects under different pressure settings;

[0144] The linear regression model formula is ZH = β0 + β1 × YL, where β0 and β1 are model parameters;

[0145] The model parameters β0 and β1 are estimated using the least squares method, that is, minimizing the residual sum of squares (RSS), and the formula is: Among them, ZH a is the comprehensive score of cable surface defects in the ath time subunit, YL a is the extruder pressure at the ath time subunit;

[0146] Based on the model prediction results, the optimal extruder pressure value is obtained;

[0147] Based on the optimal extruder pressure value, the extruder pressure is regulated to minimize the cable surface defects;

[0148] After the adjustment, the cable surface defects are monitored in real time. If the comprehensive score of the cable surface defects after the adjustment is less than the comprehensive score of the cable surface defects before the adjustment, but still greater than the defect score threshold, the above steps are repeated, and the extruder pressure is readjusted according to the latest score;

[0149] It should be noted that through the above steps, we can effectively use the linear regression model to predict and optimize the pressure setting of the extruder, thereby reducing the defects on the cable surface and improving the quality of the cable.

[0150] Example 5

[0151] The in-vehicle high-voltage liquid-cooled cable described in this embodiment is prepared by the above-mentioned in-vehicle high-voltage liquid-cooled cable process.

[0152] Working principle of the invention: The purpose of the invention is to provide an in-vehicle high-pressure liquid-cooled cable and its process. The technical solution of the invention includes image acquisition and processing, defect identification and quantification, quality judgment, and pressure data collection and analysis and dynamic adjustment. Specifically, cameras are installed at key positions of the cable production line to collect cable surface images in real time, and the cable surface area is separated from the background by image segmentation technology, and each pixel point in the segmented cable surface area image is mapped to a plane coordinate system; computer vision technology is used to automatically detect cable surface defects, and the area of ​​each defect area is calculated by an image processing algorithm. The outline of the defect area is extracted by an edge detection algorithm and its aspect ratio is calculated, and then The comprehensive defect score is calculated based on factors such as the area, length, location and shape of the defect; the comprehensive defect score is compared with the preset threshold to determine whether the cable surface quality is qualified, and a surface defect qualified or unqualified signal is generated based on the comparison result; the pressure data of the extruder is regularly recorded, its mean and standard deviation are calculated, and the relationship between the cable surface defects and the extruder pressure fluctuations is analyzed using statistical methods; based on the correlation analysis results, a linear regression model is used to predict the impact of the extruder pressure on the cable surface defects, and the extruder pressure setting is dynamically adjusted according to the model prediction results to optimize the production process, and the adjusted cable surface defects are monitored in real time. If necessary, the above steps are repeated until the defects are effectively controlled.

[0153] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A process for producing high-voltage liquid-cooled cables in vehicles, characterized in that: The following steps are involved: Collect cable surface images in real time to obtain defect areas in the cable surface area images; Process the cable surface area image to identify and obtain the cable surface defect information; Among them, the cable surface defect information includes the area, length, location and shape of each defect area; Based on the cable surface defect signal, a comprehensive defect score is obtained to generate a cable surface defect status signal; The cable surface defect status signal includes a surface defect qualified signal and a surface defect unqualified signal; Based on the unqualified signal of cable surface defects, the relationship between cable surface defects and extruder pressure fluctuations is analyzed; Based on the relationship between cable surface defects and extruder pressure fluctuations, a prediction model is constructed to obtain the optimal extruder pressure value that minimizes cable surface defects. Dynamically adjust the extruder pressure setting based on the optimal extruder pressure value.

2. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 1, characterized in that: The identification process of defective areas in the cable surface area image is as follows: During the cable production process, collect cable surface images; Segment the cable surface image and extract the cable surface area image; Based on the cable surface area image, obtaining position information of each pixel in the cable surface area image; Map each pixel in the cable surface area image as a point into the plane coordinate system to obtain the coordinate position of each pixel, which is recorded as a pixel point; The cable surface area image is processed for recognition, and the defect area in the cable surface area image is identified.

3. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 2, characterized in that: The specific process of obtaining cable surface defect information is as follows: Based on the identified defective area in the cable surface area image, all pixel points in each defective area in the cable surface area image are obtained; Process the pixels in each defect area to obtain the area of ​​the defect area; Process all pixels in the defect area to obtain the complete outline of the defect area; Based on the complete outline of the defect area, the coordinates of all pixel points on the edge outline of the defect area are obtained; The distance between every two points on the edge contour of the defect area is calculated, and the distance between every two points is processed to obtain the distance between the farthest two points on the edge contour of the defect area, which is recorded as the length of the defect area.

4. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 3, characterized in that: Based on the area and length of the defect area, all pixels in the defect area are processed to calculate the coordinates of the center point of each defect area; Based on the coordinates of the center point of each defect area, the position of each defect is marked in the cable surface area image, and a defect position distribution map is drawn; Convert the defect location distribution map into a heat map; Analyze the thermal map of defects to obtain the distribution of defects on the entire cable surface; Based on the distribution of defects on the entire cable surface, the number of defects per unit area is obtained.

5. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 3, characterized in that: Based on the complete outline of the defect area, the roundness value of the defect area is obtained by calculating the ratio of the product of 4 times π and the area of ​​the defect area to the square of the perimeter of the defect area; If the roundness value of the defect area is greater than the roundness value threshold, the contour of the defect area is close to a circle, and vice versa; The width and length of the contour of the defect area are compared to obtain the ellipticity value. If the ellipticity value is greater than the ellipticity value threshold, the contour of the defect area is close to an ellipse, and vice versa. Based on the roundness and ellipticity values, the defect areas are divided into circular defects, elliptical defects and irregular defects according to their shapes.

6. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 5, characterized in that: The formula for the comprehensive defect score is ZH = a1×M+a2×C+a3×W+a4×XZ; Among them, M is the total area of ​​cable surface defects, C is the total length of cable surface defects, and W is the total position score of all defects. x z and z are the ordinate and abscissa of the center point of the defect area, j is the number of defect areas on the cable surface, j = 1, 2, ..., n, XZ is the total shape score of all defect areas on the cable surface, 7. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 6, characterized in that: comparing the composite defect score of the cable surface with a defect score threshold; If the defect comprehensive score is greater than or equal to the defect score threshold, a surface defect unqualified signal is generated; If the defect comprehensive score is less than the defect score threshold, a surface defect qualified signal is generated.

8. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 7, characterized in that: The process of analyzing the relationship between cable surface defects and extruder pressure fluctuations is: Divide the monitoring period into a number of time sub-units of equal length, obtain the pressure data of the extruder when the time sub-unit is in the middle moment, and record it as the unit pressure value; Processing all unit pressure values ​​within the monitoring period to obtain a pressure data group; The pressure data group is processed according to the mean value calculation formula to obtain the pressure mean value within the monitoring period; The pressure data group is processed according to the standard deviation calculation formula to obtain the pressure standard deviation within the monitoring period; Calculate the difference between each unit pressure value and the pressure mean value to obtain the deviation value between the unit pressure and the mean value; The deviation value of the unit pressure from the mean value is calculated by ratio with the pressure standard deviation to obtain the standard value of the unit pressure value; Obtain the comprehensive defect score of the cable surface when the time subunit is in the middle of the monitoring period; The relationship between cable surface defects and extruder pressure fluctuations was analyzed using the Spearman rank correlation coefficient; The analysis results were tested by the t-value test method, and the t-value of the correlation between the cable surface defects and the extruder pressure fluctuation was obtained; The absolute value of t is compared with the significance threshold. If the absolute value of t is greater than the critical value threshold, a defect pressure impact signal is generated. If the absolute value of t is less than or equal to the critical value threshold, a defect pressure non-influence signal is generated.

9. The process for producing an in-vehicle high-voltage liquid-cooled cable according to claim 8, characterized in that: Based on the defect pressure impact signal, a prediction model is constructed through linear regression; The cable surface defects under different pressure settings are predicted through the prediction model; Based on the model prediction results, the optimal extruder pressure value is obtained; Based on the optimal extruder pressure value, the pressure of the extruder is regulated; Real-time monitoring is performed on the defects on the cable surface after the extruder pressure value is adjusted. If the comprehensive defect score of the cable surface after adjustment is less than the comprehensive defect score of the cable surface before adjustment, but still greater than the defect score threshold, the extruder pressure is readjusted according to the latest score.

10. A high-voltage liquid-cooled cable in a vehicle, characterized in that: The product is prepared by the process described in any one of claims 1 to 9.

Images