Cabling quality on-line detection method and system for cable cabling machine

Abstract

The present application relates to the technical field of quality detection, in particular to a braiding quality online detection method and system for a cable braiding machine, which first analyzes the vibration difference between braiding spindles according to the proportional relationship between the main traction speed and the pay-off speed, obtains the vibration non-synchronization degree at the current time, and further analyzes the non-synchronization continuous influence degree at the current time; then obtains the recess area and recess depth of the candidate recess area, analyzes the recess characteristics of the candidate recess area in combination with the non-synchronization continuous influence degree, further analyzes the recess consistency of all candidate recess areas, and obtains the recess influence degree of each candidate recess area recognized at the current time; finally, the abnormal probability of each candidate recess area recognized at the current time is obtained, and hierarchical early warning is performed. Through the present application, the reliability and accuracy of cable braiding defect recognition can be effectively improved, so as to improve the mechanical strength, electrical performance and long-term use reliability of the cable product.

Description

Technical Field

[0001] This invention relates to the field of quality inspection technology, and specifically to an online inspection method and system for the braiding quality of cable braiding machines. Background Technology

[0002] The braided layer (usually a metal wire braided shielding layer or reinforcing layer) of cables (such as communication cables, coaxial cables, power cables, etc.) is a key structure to ensure their electrical performance, mechanical strength, and anti-interference ability. During the wire and cable production stage, the braiding machine wraps multiple fine metal wires, such as copper wire, aluminum wire, and tinned copper wire, around the outside of the core conductor or inner insulation material of the cable according to a preset braiding angle and density, thereby forming a dense mesh structure.

[0003] During the braiding process of a cable braiding machine, factors such as mechanical vibration and load variations can easily cause fluctuations in the tension of each braiding spindle during the unwinding and take-up phases. This leads to fluctuations in the fixed ratio between the unwinding speed and the main traction speed of the braided wire, resulting in asynchronous movement of the braided wire, micro-stress concentration, and other problems, posing a risk of wire breakage. This manifests as localized depressions on the surface of the braided layer. Various factors (such as material quality and vibration) can cause these surface depressions. Traditional methods typically detect surface depressions and uneven textures using visual systems, but it's difficult to accurately correlate these surface defects with parameter fluctuations (such as abnormal unwinding speed or unstable main traction speed). This makes it impossible to accurately guide subsequent process adjustments, potentially allowing cables with potential breakage to enter later applications, affecting the mechanical strength, electrical performance, and long-term reliability of the cable products. Summary of the Invention

[0004] To address the above-mentioned technical problems, the present invention aims to provide an online detection method and system for the braiding quality of a cable braiding machine.

[0005] According to a first aspect of the present invention, an online detection method for braiding quality of a cable braiding machine is provided, the specific technical solution of which is as follows: The main traction speed and unwinding speed of the braided wire, as well as the circumferential contour and surface image of the cable, are collected, and candidate concave regions are identified based on the circumferential contour and the surface image. Based on the ratio between the main traction speed and the wire feeding speed, the vibration difference between the braiding spindles is analyzed to obtain the degree of vibration asynchrony at the current moment. Set a sliding window, and analyze the stability of the overall asynchronous change of the braided spindle within the sliding window according to the degree of vibration asynchrony, so as to obtain the degree of continuous impact of asynchrony at the current moment. The depression area and depression depth of the candidate depression regions are obtained. Combined with the degree of asynchronous continuous influence, the depression characteristics of each candidate depression region identified at the current time are analyzed. Then, the depression consistency of all candidate depression regions identified at the current time is analyzed to obtain the degree of depression influence of each candidate depression region identified at the current time. Based on the degree of impact of the depression, the abnormal probability of each candidate depression region identified at the current time is obtained, and a graded early warning is performed.

[0006] In an embodiment of the present invention, based on the proportional relationship between the main traction speed and the unwinding speed, the vibration difference between the braiding spindles is analyzed to obtain the degree of vibration asynchrony at the current moment, including: Calculate the ratio of the pay-off speed to the main traction speed, and combine it with the theoretical proportionality coefficient between the pay-off speed and the main traction speed to obtain the degree of proportional deviation of each braiding spindle at the current moment; For the current moment, obtain the minimum proportional deviation of all braiding spindles, analyze the difference between the proportional deviation of each braiding spindle and the minimum proportional deviation, and obtain the degree of vibration asynchrony at the current moment.

[0007] In an embodiment of the present invention, a sliding window is set, and based on the degree of vibration asynchrony, the overall stability of the asynchronous change of the braiding spindle within the sliding window is analyzed to obtain the degree of continuous asynchrony influence at the current moment, including: Set up a sliding window to slide across the time series; By analyzing the differences in the degree of vibration asynchrony between adjacent moments within the sliding window, the overall stability of the asynchronous change of the braided spindle within the sliding window can be obtained. The average degree of vibration asynchrony within the sliding window is obtained, and combined with the degree of change stability, the degree of continuous impact of asynchrony at the current moment is obtained.

[0008] In an embodiment of the present invention, the depression area and depression depth of the candidate depression regions are obtained, and the depression features of each candidate depression region identified at the current moment are analyzed in conjunction with the degree of asynchronous continuous influence, including: The depression area of ​​the candidate depression region is obtained based on the pixel coordinates of the candidate depression region. The depth of the candidate concave region is obtained by combining the depth values ​​of the pixel coordinates of all contour points in the candidate concave region with the reference depth values ​​of the corresponding positions on the reference surface. Based on the depression area and the depression depth, and combined with the degree of asynchronous continuous influence, the depression feature value of each candidate depression region identified at the current time is obtained.

[0009] In an embodiment of the present invention, analyzing the concavity consistency of all candidate concavity regions identified at the current time to obtain the concavity influence degree of each candidate concavity region identified at the current time includes: For the candidate depression regions identified at the current time, calculate the average depression feature value of all candidate depression regions; Analyze the difference between the depression feature value and the average depression feature value of each candidate depression region to obtain the degree of depression consistency of all candidate depression regions identified at the current time.

[0010] In an embodiment of the present invention, obtaining the degree of indentation influence of each candidate indentation region identified at the current time includes: Based on the depression feature value and the degree of depression consistency, the degree of depression influence of each candidate depression region identified at the current time is obtained.

[0011] In an embodiment of the present invention, tiered early warning is performed, including: Preset a first warning threshold, a second warning threshold, and a third warning threshold; If the anomaly probability is less than the first warning threshold, then the weaving quality of the candidate depression region identified at the current time is determined to be qualified. If the abnormal probability is greater than or equal to the first warning threshold and less than the second warning threshold, then the candidate depression region identified at the current time will be continuously monitored. If the abnormal probability is greater than or equal to the second warning threshold and less than the third warning threshold, then the candidate depression region identified at the current time is manually investigated. If the anomaly probability is greater than or equal to the third warning threshold, then the candidate depression region identified at the current time will be stopped for inspection.

[0012] In an embodiment of the present invention, identifying candidate concave regions based on the circumferential contour and the surface image includes: Each pixel in the surface image is matched with the three-dimensional coordinate points in the circumferential contour data to obtain a matching image; Texture analysis is performed on the matched image to obtain suspected concave regions; Connectivity analysis is performed on the suspected depression region to obtain candidate depression regions, and the pixel coordinates of the candidate depression regions are recorded.

[0013] According to a second aspect of the present invention, an online inspection system for braiding quality of a cable braiding machine is provided, comprising: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read program code stored in the memory and execute the method described in the first aspect of the present invention.

[0014] In some embodiments of the present invention, the processor includes: The data acquisition and candidate depression region acquisition module is used to acquire the main traction speed, unwinding speed, circumferential contour and surface image of the braided wire, and identify candidate depression regions based on the circumferential contour and the surface image. The braided yarn motion asynchrony analysis module is used to analyze the vibration differences between braided spindles based on the ratio of the main traction speed and the unwinding speed, and to obtain the degree of vibration asynchrony at the current moment; and to set a sliding window to analyze the overall stability of the asynchrony of the braided spindles within the sliding window based on the degree of vibration asynchrony, and to obtain the degree of continuous asynchrony impact at the current moment. The indentation impact analysis module is used to obtain the indentation area and indentation depth of the candidate indentation regions, and, in combination with the degree of asynchronous continuous impact, analyze the indentation characteristics of each candidate indentation region identified at the current time, and further analyze the indentation consistency of all candidate indentation regions identified at the current time, so as to obtain the degree of indentation impact of each candidate indentation region identified at the current time. The graded early warning module is used to obtain the abnormal probability of each candidate depression region identified at the current time according to the degree of influence of the depression, and to perform graded early warning.

[0015] Compared with existing technologies, the online detection method and system for braiding quality of cable braiding machines provided by this invention have the following advantages: This invention first acquires the main traction speed and unwinding speed of the braided wire, as well as the circumferential contour and surface image of the wire, and identifies candidate concave regions based on the circumferential contour and surface image. Then, based on the ratio of the main traction speed to the unwinding speed, it analyzes the vibration differences between the braided spindles to obtain the degree of vibration asynchrony at the current moment. Next, it sets a sliding window and analyzes the stability of the overall asynchrony changes of the braided spindles within the sliding window based on the degree of vibration asynchrony to obtain the degree of sustained asynchrony impact at the current moment. Then, it acquires the concave area and depth of the candidate concave regions, and combines this with the degree of sustained asynchrony impact to analyze the concave characteristics of each candidate concave region identified at the current moment. Furthermore, it analyzes the concave consistency of all candidate concave regions identified at the current moment to obtain the degree of concave impact of each candidate concave region identified at the current moment. Finally, based on the degree of concave impact, it obtains the anomaly probability of each candidate concave region identified at the current moment and performs graded early warning. This invention acquires multi-source data during the cable braiding process of a cable braiding machine, analyzes the micro-stress concentration caused by tension fluctuations during wire feeding and take-up due to equipment vibration, quantifies the asynchronous movement of the braided wire, and combines image analysis to identify the dynamic characteristics of abnormal areas during the braiding process, quantifying the probability of abnormal areas. This enables online detection of the braiding quality of the cable braiding machine, improves the reliability and accuracy of cable braiding defect identification, and prevents cables with hidden defects from entering subsequent application stages, thereby improving the mechanical strength, electrical performance, and long-term reliability of cable products. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the basic process of an online detection method for braiding quality of a cable braiding machine according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the basic components of an online inspection system for the braiding quality of a cable braiding machine, provided as an embodiment of the present invention. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of the online detection method and system for cable braiding machines proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device comprising a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of additional identical elements in the article or device that includes the element.

[0020] The following describes in detail, with reference to the accompanying drawings, a specific scheme for an online detection method for braiding quality of a cable braiding machine provided by the present invention.

[0021] Please see Figure 1 This illustrates the basic flow of an online detection method for braiding quality of a cable braiding machine provided by an embodiment of the present invention.

[0022] like Figure 1 As shown, an embodiment of the present invention provides an online detection method for braiding quality of a cable braiding machine, which specifically includes: S100: Collects the main traction speed and unwinding speed of the braided wire, as well as the circumferential contour and surface image of the cable, and identifies candidate recessed areas based on the circumferential contour and surface image.

[0023] To achieve online inspection of the braiding quality of a cable braiding machine, it is first necessary to collect data on the main traction speed, unwinding speed, circumferential profile, and surface image of the braided wire (the wire used for surface braiding of cables). The specific implementation method is as follows: The main traction speed of the braided wire is monitored in real time using a high-precision encoder. A video stream is acquired by a high-speed industrial camera positioned above the braiding area. The video stream is then processed using image processing algorithms (such as optical flow, feature point tracking, or inter-frame difference) to track the motion trajectory of each braiding spindle, calculate its instantaneous angular velocity, and combine it with the diameter of the braiding spindle to convert the instantaneous angular velocity into linear velocity, which is used as the unwinding speed of the braided wire.

[0024] A high-resolution industrial camera or high-speed video system is installed at the exit of the braiding machine to acquire surface images of the cable braid layer in real time at a fixed frequency. Additionally, a laser diameter gauge and a laser profile scanner are installed along the cable's travel path to measure the overall outer diameter of the cable in real time and perform high-precision profile scanning in the circumferential direction to obtain the cable's circumferential profile. The acquisition frequency is synchronized with the camera's acquisition frequency.

[0025] Then, candidate depression regions are identified based on the circumferential contour and surface image. The specific implementation method is as follows: The acquired circumferential contour and surface images are preprocessed, including filtering and grayscale conversion. Since the cable moves continuously, and the camera and laser contour scanner are typically distributed longitudinally along the production line with installation gaps, it is necessary to compensate for missing motion in the circumferential contour and surface images, i.e., axial alignment. This axial alignment is achieved through feature matching and motion estimation methods (existing technology, not detailed here). Additionally, since the cable is cylindrical, slight twisting or multi-camera splicing requires a clear mapping reference between the horizontal coordinate of the 2D surface image and the angle of the 3D circumferential contour. Therefore, circumferential angle alignment is needed between the circumferential contour and surface images, specifically achieved through quasi-synchronous sampling methods (existing technology, not detailed here). After achieving these two alignments, pre-calibrated system parameters (including camera intrinsic and extrinsic parameters and spatial transformation relationships with the contour scanner) are used to precisely match each pixel in the surface image with the three-dimensional coordinates in the circumferential contour data, ensuring that both describe the same physical location on the cable surface, resulting in a matched image.

[0026] On the matched image, texture analysis-based methods (such as local binary mode variance analysis) or deep learning-trained surface defect segmentation models are applied to identify regions with uneven texture, abnormal brightness, or obvious boundary abrupt changes, which are then marked as suspected depression regions. Connectivity analysis is then performed on these suspected depression regions to obtain candidate depression regions, and the pixel coordinates of these candidate depression regions are recorded.

[0027] During the braiding process of a cable braiding machine, equipment vibration causes fluctuations in the tension of the braided wire during both the unwinding and winding stages. This results in uneven tension in the braided wire, specifically manifested as fluctuations in the ratio of the unwinding speed to the main traction speed. The greater the difference between the unwinding and main traction speeds and their theoretical values, the greater the impact of vibration. Furthermore, the more asynchronous the movements of the braiding spindles, the greater the difference in tension fluctuations, and the more significant the impact of overall vibration on braid uniformity. Moreover, this asynchronous nature is not instantaneous but typically has a certain duration. When some braided wires are continuously unwinded below the main traction speed—meaning the duration of asynchronous influence is greater—micro-stress concentration can lead to localized depressions on the surface of the braided layer. The greater the degree of depression, the greater the stress concentration, the higher the potential for wire breakage in that area, and the more severe the impact on braid quality, requiring special attention. In the embodiments of the present invention, through steps S200 to S500, the effects of vibration asynchrony and indentation are analyzed, thereby realizing online detection of the braiding quality of the cable braiding machine.

[0028] S200: Based on the ratio of the main traction speed to the unwinding speed, analyze the vibration differences between the braiding spindles to obtain the degree of vibration asynchrony at the current moment.

[0029] During the braiding process of a cable braiding machine, equipment vibration causes fluctuations in the tension of the braided wire during the weaving process. This manifests as fluctuations in the ratio of the wire release speed to the main traction speed. The greater the difference between the wire release speed and the main traction speed and the theoretical values, the greater the influence of vibration and the greater the degree of micro-stress concentration in the braided wire caused by uneven tension. Furthermore, the more asynchronous the movement of each braiding spindle, the greater the difference in the degree of tension fluctuation in the braided wire, indicating that the overall vibration has a more significant impact on the uniformity of the braiding.

[0030] Based on the above analysis, in an embodiment of the present invention, the vibration difference between the braiding spindles is analyzed according to the ratio of the main traction speed to the unwinding speed to obtain the degree of vibration asynchrony at the current moment. Further, this includes: First, the ratio of the pay-off speed to the main traction speed is calculated. Combined with the theoretical proportionality coefficient between the pay-off speed and the main traction speed, the proportional deviation of each braiding spindle at the current moment is obtained. The specific implementation method is as follows: For each braiding spindle, the pay-off speed of each braiding spindle is collected in real time using a high-precision encoder. The collected pay-off speeds are arranged in chronological order to generate time series data of the pay-off speed of each braiding spindle, where the... The weaving spindle at the current time The collected wire laying speed value is denoted as Braided yarn at the current time The main traction speed is denoted as Under normal circumstances, the unwinding speed of each braiding spindle and the main traction speed of the braided yarn maintain a fixed ratio to ensure uniform braiding pitch and structural stability. This fixed ratio is determined by the braiding pitch required by the braiding process, the number of braiding spindles, and the braiding angle (the specific calculation method is existing technology and will not be elaborated here). The theoretical ratio is denoted as... Because the vibration of the cable braiding machine is transmitted to key components such as the bearings and transmission gears of the braiding spindle, the actual rotational speed of the braiding spindle changes, causing fluctuations in the ratio of the unwinding speed to the main traction speed during the cable braiding process. Therefore, in this embodiment of the invention, the degree of proportional deviation of each braiding spindle at the current moment is obtained by calculating the absolute value of the difference between the ratio of the unwinding speed to the main traction speed and the theoretical proportionality coefficient. Furthermore, the first... The weaving spindle at the current moment The formula for calculating the degree of proportional deviation is:

[0031] In the formula, Indicates the first The weaving spindle at the current moment The degree of proportional deviation; Indicates the first The weaving spindle at the current moment The wire feeding speed; Indicates the current time of the braided yarn. Main traction speed; This represents the theoretical proportionality coefficient between the current weaving process parameters (weaving pitch, number of weaving spindles, weaving angle) and the main traction speed. This indicates taking the absolute value.

[0032] It should be noted that when the cable braiding machine starts, stops, is being debugged, or malfunctions and causes traction to pause, there is a possibility of… The probability of the main traction speed being 0 is zero. Therefore, a preset main traction speed threshold is set (the value can be 80% of the main traction speed during normal operation). When the main traction speed is detected to be greater than or equal to the preset main traction speed threshold, the proportional deviation is calculated.

[0033] It reflects the severity of the proportional deviation between the wire laying speed and the main traction speed. The larger the value, the greater the difference between the proportional coefficient between the wire laying speed and the main traction speed and the theoretical value, and the greater the influence of vibration.

[0034] Whether the vibration states of all braiding spindles are synchronized at the same moment, i.e., whether the overall vibration of the equipment causes coordinated speed fluctuations in multiple braiding spindles, reflects the degree of influence of the overall vibration on braiding uniformity. The more asynchronous the movements of the braiding spindles, the more significant the impact of the overall vibration on braiding uniformity. Therefore, for the current moment, the minimum proportional deviation of all braiding spindles is obtained, and the difference between the proportional deviation of each braiding spindle and the minimum proportional deviation is analyzed to obtain the degree of vibration asynchrony at the current moment. Specifically, based on the current moment... The formula for calculating the degree of vibration asynchrony of all braiding spindles is:

[0035] In the formula, Indicates the current moment The degree of asynchronous vibration among all the braiding spindles; Indicates the first The weaving spindle at the current moment The degree of proportional deviation; This indicates that all weaving spindles are at the current moment. The minimum proportional deviation; This indicates the total number of spinning spindles.

[0036] The larger the value, the greater the vibration difference between the weaving spindles. The larger the value, the worse the consistency of movement of each weaving spindle, and the more significant the impact of vibration on weaving uniformity.

[0037] S300: Set a sliding window and analyze the stability of the overall asynchronous change of the braided spindle within the sliding window based on the degree of vibration asynchrony, so as to obtain the degree of continuous impact of asynchrony at the current moment.

[0038] Various transient disturbances exist during the weaving process (such as short-term mechanical vibration, signal noise, and short-term tension fluctuations). These fluctuations may temporarily affect the weaving speed and main traction speed, and the degree of vibration asynchrony fluctuates drastically within a short period. However, weaving layer defects (such as runaway yarns and dents) are usually not formed instantaneously, but gradually appear after a period of time due to asynchrony and uneven tension, with the corresponding degree of vibration asynchrony remaining relatively large over a long period. Slight fluctuations in the weaving speed within a short period may be absorbed by the elasticity of the weaving structure and will not immediately form visible defects; however, if a weaving spindle or a group of weaving spindles is continuously in an asynchronous state, it will cause the local weaving yarn to be in a state of abnormal tension or stress concentration for a long time, thereby inducing the accumulation of microscopic damage.

[0039] Based on the above analysis, in an embodiment of the present invention, by setting a sliding window, the stability of the overall asynchronous change of the braiding spindle within the sliding window is analyzed according to the degree of vibration asynchrony, thereby obtaining the degree of continuous asynchrony influence at the current moment. Further details include: First, set up a sliding window that slides across the time series. Specifically, define a window with a length of... Sliding window, The value can correspond to the number of time sampling points corresponding to 1 to 2 braiding pitches of the cable. It slides sequentially on the time series, sliding 1 time series each time, which is the length of the time difference between two sampling moments.

[0040] Then, the differences in the degree of vibration asynchrony between adjacent moments within the sliding window are analyzed to obtain the overall stability of the asynchronous changes of the braided spindle within the sliding window. Specifically, the absolute value of the difference between the degree of vibration asynchrony between adjacent moments within the sliding window is calculated. This process is repeated for all adjacent moments, and the absolute values ​​of the differences are summed. The inverse proportional mapping result of the sum of the absolute values ​​of the differences is then used as the overall stability of the asynchronous changes of the braided spindle within the sliding window.

[0041] Finally, the average degree of vibration asynchrony within the sliding window is obtained, and combined with the degree of change stability, the degree of sustained impact of asynchrony at the current moment is obtained. Specifically, the current moment is constructed. The formula for calculating the degree of asynchronous persistent impact is:

[0042] In the formula, Indicates the current time The degree of asynchronous and persistent impact; This represents the average degree of vibration asynchrony across all moments within the sliding window. Indicates the first [number]th ... The degree of vibration asynchrony at each moment; Indicates the first [number]th ... The degree of vibration asynchrony at each moment; Indicates the duration of the sliding window; Indicates taking the absolute value; This represents a linear normalization function, such as a max-min normalization function, which normalizes the degree of persistent asynchrony at all times. The normalization mapping ranges from 1 to 2. .

[0043] This represents the sum of the absolute values ​​of the differences in the degree of asynchrony between adjacent moments within the sliding window, used to measure the severity of fluctuations in the degree of asynchrony. The smaller the value, the smaller the difference in the degree of vibration asynchrony between adjacent moments, and the more stable the change within the sliding window; This indicates the average severity of vibration asynchrony; the degree of sustained impact of asynchrony. The larger the value, the greater and more persistent the degree of vibration asynchrony within the current window period, and the greater the risk of knitting quality problems caused by asynchronous motion within the current window period.

[0044] S400: Obtain the depression area and depression depth of the candidate depression regions, combine the degree of asynchronous continuous influence, analyze the depression characteristics of each candidate depression region identified at the current time, and then analyze the depression consistency of all candidate depression regions identified at the current time to obtain the degree of depression influence of each candidate depression region identified at the current time.

[0045] Steps S200 and S300 analyze the fluctuations in the pay-off and take-up tensions during the braiding process caused by factors such as equipment vibration, resulting in a difference between the pay-off speed and the main traction speed of the braided yarn, thus quantifying the degree of persistent impact of asynchrony. Further analysis is needed to determine when the pay-off speed of some braided yarns is consistently lower than the main traction speed, i.e., when the degree of persistent impact of asynchrony is greater, micro-stress concentration will lead to localized depressions on the surface of the braided layer. The greater the degree of depression, the greater the stress concentration, the greater the possibility of yarn breakage in that area, and the more severe the impact on braiding quality.

[0046] Based on the above analysis, in the embodiments of the present invention, by obtaining the depression area and depression depth of the candidate depression regions, and combining the degree of asynchronous continuous influence, the depression characteristics of each candidate depression region identified at the current moment are analyzed, and then the depression consistency of all candidate depression regions identified at the current moment is analyzed to obtain the degree of depression influence of each candidate depression region identified at the current moment. Wherein: Obtain the depression area and depth of the candidate depression regions, and analyze the depression characteristics of each candidate depression region identified at the current time, taking into account the degree of asynchronous persistence. Further analysis includes: First, the depression area of ​​the candidate depression region is obtained based on the pixel coordinates of the candidate depression region. Specifically, for step S100 at the current time... For each identified candidate depression region, based on the recorded pixel coordinates of the candidate depression region, the 3D coordinates of all points at the corresponding spatial location are extracted from the spatiotemporally aligned 3D circular contour point cloud data to obtain the current time. The first identification The depression area of ​​each candidate depression region is denoted as . The specific calculation method for the depression area uses existing technology and will not be elaborated here.

[0047] Then, based on the depth values ​​of the pixel coordinates of all contour points in the candidate concave region, combined with the reference depth values ​​of the corresponding positions on the reference surface, the concave depth of the candidate concave region is obtained. Specifically, before the cable braiding machine officially starts working, a sample cable is braided in advance, and the normal braiding area is manually determined. The surface fitted by the circumferential contour data of the normal braiding area is used as the reference surface. When the cable braiding machine officially starts working, the current time is calculated. The first identification The deviation of the pixel coordinate depth value (z value) of all contour points in a candidate concave region from the reference depth value at the corresponding position on the reference plane is used as the average of the deviation values ​​corresponding to all contour points at the current moment. The first identification The depth of each candidate depression region is denoted as . This reflects the average degree of depression in the candidate depression area.

[0048] Finally, based on the depression area and depth, and considering the degree of persistent asynchronous influence, the depression feature value of each candidate depression region identified at the current moment is obtained. Specifically, when part of the braided wire is continuously lower than the main traction speed due to the unwinding speed, that is, when the degree of persistent asynchronous influence is greater, local depressions will form on the surface of the braided layer due to micro-stress concentration. The greater the depression degree, the greater the stress concentration of the braided wire. In addition, the depression area and depth of the candidate depression region also indicate that the greater the depression degree, the greater the stress concentration of the braided wire. Therefore, based on the depression area and depth, and considering the degree of persistent asynchronous influence, the depression feature value of each candidate depression region at the current moment is constructed. The first identification The formula for calculating the depression feature value of each candidate depression region is as follows:

[0049] In the formula, Indicates the current time The first identification Characteristic values ​​of the degree of depression in each candidate depression region; Indicates the current time The first identification The depression area of ​​each candidate depression region Indicates the current time The first identification The depth of the depression in each candidate depression region; Indicates the current time The degree of asynchronous and persistent impact; This represents a linear normalization function, such as a max-min normalization function, used to normalize the area and depth of depressions in all candidate regions identified at all time points. The normalized values ​​range from [value missing]. .

[0050] The value represents the degree of indentation. The larger the value, the more severe the damage to the indented area under the current process conditions, and the higher the risk of structural weakening in the area.

[0051] Analyze the concavity consistency of all candidate concavity regions identified at the current time to obtain the concavity influence degree of each candidate concavity region identified at the current time, further including: First, for the candidate depression regions identified at the current time, calculate the average depression feature value of all candidate depression regions. Specifically, for the current time... The identified candidate depression regions are calculated, and the average depression feature value of all candidate depression regions is denoted as . .

[0052] Then, the difference between the depression characteristic value and the average depression characteristic value of each candidate depression region is analyzed to determine the consistency of depressions among all candidate depression regions identified at the current time. Specifically, if multiple candidate depression regions appearing at the same time have similar depression characteristic values, it indicates a high degree of similarity among all candidate depression regions, suggesting that they may be caused by the same systematic factor (such as an anomaly in a certain weaving spindle group, leading to periodic depressions). The higher the consistency of the depression characteristic values ​​of candidate depression regions, the higher the probability that the depressions are caused by the same systematic factor. Therefore, the calculation of the current time... The first identification The depression feature values ​​of each candidate depression region With average concave characteristic value The absolute value of the difference, iterating through the current time. Sum the absolute values ​​of the differences among all identified candidate depression regions to obtain the current time step. The degree of concavity uniformity among all identified candidate concavity regions is:

[0053] In the formula, Indicates the current time The degree of concavity consistency among all identified candidate concavity regions; Indicates the current time The first identification Characteristic values ​​of the degree of depression in each candidate depression region; Indicates the current time The average indentation degree feature value of all identified candidate indentation regions; Indicates the current time The number of all candidate depression regions identified; This indicates taking the absolute value.

[0054] This indicates the deviation between the characteristic value of the indentation degree and the average value. If the characteristic values ​​of each candidate indentation region are close, it means that the candidate indentation regions have a high degree of similarity. The higher the degree of consistency of the indentation, the more likely it is to originate from overall process fluctuations.

[0055] It should be noted that when there is only one isolated defect ( )hour, If the value is 0, then Reaching a maximum value of 1 (indicating extremely high consistency) can lead to a greater degree of impact, potentially amplifying the severity of an isolated defect and misjudging it as a "systemic high-risk failure," which contradicts the original intention that "consistency reflects systemic factors." Therefore, when Force setting Or smaller, to reduce the systematic weight of isolated defects, when Then, the formula for calculating the degree of indentation uniformity is applied again.

[0056] The degree of indentation impact for each candidate indentation region identified at the current moment is obtained, further including: based on the indentation feature value and the indentation consistency degree, the degree of indentation impact for each candidate indentation region identified at the current moment is obtained. Specifically, the degree of indentation impact of the candidate indentation region is calculated by combining the indentation degree feature value and the indentation consistency degree. The larger the value, the greater the degree of micro-stress concentration caused by uneven tension during weaving, the greater the degree of indentation, the greater the possibility of yarn breakage in this area, and the more serious the impact on weaving quality. Therefore, based on the current moment... The first identification Feature value of the degree of depression in each candidate depression region and the current moment Concavity consistency of all identified candidate concavity regions Construct the current moment The first identification The formula for calculating the degree of depression influence of each candidate depression region is:

[0057] In the formula, Indicates the current time The first identification The degree of depression influence of each candidate depression region; Indicates the current time The degree of concavity consistency among all identified candidate concavity regions; Indicates the current time The first identification The feature value of the degree of depression in each candidate depression region.

[0058] Degree of impact of dent The larger the value, the greater the influence of the candidate depression area, the greater the possibility of stress concentration causing yarn breakage, and the more serious the impact on weaving quality.

[0059] S500: Based on the degree of impact of the depression, obtain the abnormal probability of each candidate depression region identified at the current time, and perform graded early warning.

[0060] Based on the degree of impact of the depression, the anomaly probability of each candidate depression region identified at the current moment is obtained, and a graded early warning is performed.

[0061] Specifically, firstly, based on the current moment The first identification Degree of impact of depression in each candidate depression region Construct the current moment The first identification The formula for calculating the anomaly probability of each candidate depression region is:

[0062] In the formula, Indicates the current time The first identification The probability of anomalies in each candidate depression region; Indicates the current time The first identification The degree of depression influence of each candidate depression region.

[0063] abnormal probability This reflects the degree to which factors such as vibration affect the weaving quality of the braided layer, providing a basis for quality inspection. The anomaly probability of candidate depression areas. The higher the value, the greater the possibility of stress concentration causing yarn breakage, and the more serious the impact on weaving quality.

[0064] Then, based on the anomaly probability of each candidate depression region A quality judgment threshold is set to achieve real-time, automated online detection and graded early warning of knitting quality. Specifically, based on the statistical distribution of anomalies in historical qualified products, a first, second, and third early warning threshold are preset, denoted as... In practical applications, the values ​​of the three thresholds should be dynamically set based on the statistical distribution of historical data (such as quantiles). For example, they can be set to... , , If the probability of an anomaly is less than the first warning threshold, i.e. Then determine the current time. The first identification If the weaving quality of a candidate concave region is acceptable, the warning measure is to send a "region weaving quality acceptable" warning; if the anomaly probability is greater than or equal to the first warning threshold but less than the second warning threshold, i.e. Then determine the current time. The first identification If a candidate indentation region exhibits slight anomalies, the early warning measure is to issue an alert stating "Slight stress concentration exists, continuous monitoring required," and then conduct continuous monitoring. If the anomaly probability is greater than or equal to the second early warning threshold but less than the third early warning threshold, i.e. Then determine the current time. The first identification If a candidate depression region exhibits moderate anomalies, the warning measure is to send an alert stating "Significant stress concentration, high risk of wire breakage, manual inspection required," and mark the abnormal area for manual inspection by staff. If the anomaly probability is greater than or equal to the third warning threshold, i.e. Then determine the current time. The first identification If a candidate depression area is severely abnormal, the warning measure is to send a message saying "Severe structural weakening, high risk of wire breakage, machine needs to be stopped for inspection", and mark the abnormal area and indicate that a red indicator light is constantly on, and then stop the machine for inspection.

[0065] The system automatically generates a braiding quality inspection report periodically (e.g., after each roll of cable is completed), including the pass rate, the number and distribution of defects (the number of independent, unconnected recessed areas), and related fluctuations in process parameters (e.g., abnormal pay-off speed, unstable traction speed). Early warning measures and parameter calculations are integrated into the online braiding quality inspection system of the cable braiding machine, thereby achieving online inspection of the braiding quality.

[0066] Based on the same inventive concept as the above method, this embodiment also provides an online detection system for braiding quality of cable braiding machines.

[0067] Please see Figure 2 This illustrates the basic components of an online inspection system for the braiding quality of a cable braiding machine, provided by an embodiment of the present invention.

[0068] like Figure 2 As shown, an online inspection system for braiding quality of a cable braiding machine includes a memory 10 and a processor 20, wherein: Memory 10 is used to store program code; The processor 20 is used to read the program code stored in the memory 10 and execute it to acquire the main traction speed and unwinding speed of the braided wire, as well as the circumferential contour and surface image of the wire. Based on the circumferential contour and surface image, it identifies candidate concave regions. Based on the ratio of the main traction speed to the unwinding speed, it analyzes the vibration differences between the braided spindles to obtain the degree of vibration asynchrony at the current moment. It sets a sliding window and analyzes the stability of the overall asynchrony changes of the braided spindles within the sliding window based on the degree of vibration asynchrony to obtain the degree of continuous asynchrony impact at the current moment. It acquires the concave area and depth of the candidate concave regions, and combined with the degree of continuous asynchrony impact, analyzes the concave characteristics of each candidate concave region identified at the current moment. Furthermore, it analyzes the concave consistency of all candidate concave regions identified at the current moment to obtain the degree of concave impact of each candidate concave region identified at the current moment. Based on the degree of concave impact, it obtains the abnormal probability of each candidate concave region identified at the current moment and performs graded early warning.

[0069] Furthermore, the processor 20 includes a data acquisition and candidate depression region acquisition module 21, a braided wire movement asynchrony analysis module 22, a depression impact analysis module 23, and a graded early warning module 24, wherein: The data acquisition and candidate depression region acquisition module 21 is used to acquire the main traction speed, unwinding speed, circumferential contour and surface image of the braided wire, and identify candidate depression regions based on the circumferential contour and surface image. The braided wire motion asynchrony analysis module 22 is used to analyze the vibration difference between braided spindles based on the ratio of the main traction speed and the wire release speed, and to obtain the degree of vibration asynchrony at the current moment; and to set a sliding window to analyze the overall asynchrony change stability of the braided spindles within the sliding window based on the degree of vibration asynchrony, and to obtain the degree of continuous asynchrony influence at the current moment. The depression impact analysis module 23 is used to obtain the depression area and depression depth of the candidate depression region, and combined with the degree of asynchronous continuous impact, analyze the depression characteristics of each candidate depression region identified at the current time, and then analyze the depression consistency of all candidate depression regions identified at the current time to obtain the depression impact degree of each candidate depression region identified at the current time. The graded early warning module 24 is used to obtain the abnormal probability of each candidate depression region identified at the current time according to the degree of depression impact, and to perform graded early warning.

[0070] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0071] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for online detection of braiding quality in a cable braiding machine, characterized in that, The method includes: The main traction speed and unwinding speed of the braided wire, as well as the circumferential contour and surface image of the cable, are collected, and candidate concave regions are identified based on the circumferential contour and the surface image. Based on the ratio between the main traction speed and the wire feeding speed, the vibration difference between the braiding spindles is analyzed to obtain the degree of vibration asynchrony at the current moment. Set a sliding window, and analyze the stability of the overall asynchronous change of the braided spindle within the sliding window according to the degree of vibration asynchrony, so as to obtain the degree of continuous impact of asynchrony at the current moment. The depression area and depression depth of the candidate depression regions are obtained. Combined with the degree of asynchronous continuous influence, the depression characteristics of each candidate depression region identified at the current time are analyzed. Then, the depression consistency of all candidate depression regions identified at the current time is analyzed to obtain the degree of depression influence of each candidate depression region identified at the current time. Based on the degree of impact of the depression, the abnormal probability of each candidate depression region identified at the current time is obtained, and a graded early warning is performed.

2. The online detection method for braiding quality of a cable braiding machine according to claim 1, characterized in that, Based on the ratio between the main traction speed and the unwinding speed, the vibration differences between the braiding spindles are analyzed to obtain the degree of vibration asynchrony at the current moment, including: Calculate the ratio of the pay-off speed to the main traction speed, and combine it with the theoretical proportionality coefficient between the pay-off speed and the main traction speed to obtain the degree of proportional deviation of each braiding spindle at the current moment; For the current moment, obtain the minimum proportional deviation of all braiding spindles, analyze the difference between the proportional deviation of each braiding spindle and the minimum proportional deviation, and obtain the degree of vibration asynchrony at the current moment.

3. The online detection method for braiding quality of a cable braiding machine according to claim 2, characterized in that, A sliding window is set up, and based on the degree of vibration asynchrony, the overall stability of the asynchronous changes of the braiding spindles within the sliding window is analyzed to obtain the degree of persistent asynchrony influence at the current moment, including: Set up a sliding window to slide across the time series; By analyzing the differences in the degree of vibration asynchrony between adjacent moments within the sliding window, the overall stability of the asynchronous change of the braided spindle within the sliding window can be obtained. The average degree of vibration asynchrony within the sliding window is obtained, and combined with the degree of change stability, the degree of continuous impact of asynchrony at the current moment is obtained.

4. The online detection method for braiding quality of a cable braiding machine according to claim 1, characterized in that, Obtain the depression area and depression depth of the candidate depression regions, and analyze the depression features of each candidate depression region identified at the current time, in conjunction with the degree of asynchronous persistent influence, including: The depression area of ​​the candidate depression region is obtained based on the pixel coordinates of the candidate depression region. The depth of the candidate concave region is obtained by combining the depth values ​​of the pixel coordinates of all contour points in the candidate concave region with the reference depth values ​​of the corresponding positions on the reference surface. Based on the depression area and the depression depth, and combined with the degree of asynchronous continuous influence, the depression feature value of each candidate depression region identified at the current time is obtained.

5. The online detection method for braiding quality of a cable braiding machine according to claim 4, characterized in that, Analyze the concavity consistency of all candidate concavity regions identified at the current time to obtain the concavity influence degree of each candidate concavity region identified at the current time, including: For the candidate depression regions identified at the current time, calculate the average depression feature value of all candidate depression regions; Analyze the difference between the depression feature value and the average depression feature value of each candidate depression region to obtain the degree of depression consistency of all candidate depression regions identified at the current time.

6. The online detection method for braiding quality of a cable braiding machine according to claim 5, characterized in that, Obtain the degree of indentation influence for each of the candidate indentation regions identified at the current time, including: Based on the depression feature value and the degree of depression consistency, the degree of depression influence of each candidate depression region identified at the current time is obtained.

7. The online detection method for braiding quality of a cable braiding machine according to claim 1, characterized in that, Implement tiered early warning systems, including: Preset a first warning threshold, a second warning threshold, and a third warning threshold; If the anomaly probability is less than the first warning threshold, then the weaving quality of the candidate depression region identified at the current time is determined to be qualified. If the abnormal probability is greater than or equal to the first warning threshold and less than the second warning threshold, then the candidate depression region identified at the current time will be continuously monitored. If the abnormal probability is greater than or equal to the second warning threshold and less than the third warning threshold, then the candidate depression region identified at the current time is manually investigated. If the anomaly probability is greater than or equal to the third warning threshold, then the candidate depression region identified at the current time will be stopped for inspection.

8. The online detection method for braiding quality of a cable braiding machine according to claim 1, characterized in that, Identifying candidate depression regions based on the circumferential contour and the surface image includes: Each pixel in the surface image is matched with the three-dimensional coordinate points in the circumferential contour data to obtain a matching image; Texture analysis is performed on the matched image to obtain suspected concave regions; Connectivity analysis is performed on the suspected depression region to obtain candidate depression regions, and the pixel coordinates of the candidate depression regions are recorded.

9. An online inspection system for braiding quality of a cable braiding machine, characterized in that, The system includes: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read program code stored in the memory and execute the method as described in any one of claims 1 to 8.

10. The online braiding quality detection system for a cable braiding machine according to claim 9, characterized in that, The processor includes: The data acquisition and candidate depression region acquisition module is used to acquire the main traction speed, unwinding speed, circumferential contour and surface image of the braided wire, and identify candidate depression regions based on the circumferential contour and the surface image. The braided yarn motion asynchrony analysis module is used to analyze the vibration differences between braided spindles based on the ratio of the main traction speed and the unwinding speed, and to obtain the degree of vibration asynchrony at the current moment; and to set a sliding window to analyze the overall stability of the asynchrony of the braided spindles within the sliding window based on the degree of vibration asynchrony, and to obtain the degree of continuous asynchrony impact at the current moment. The indentation impact analysis module is used to obtain the indentation area and indentation depth of the candidate indentation regions, and, in combination with the degree of asynchronous continuous impact, analyze the indentation characteristics of each candidate indentation region identified at the current time, and further analyze the indentation consistency of all candidate indentation regions identified at the current time, so as to obtain the degree of indentation impact of each candidate indentation region identified at the current time. The graded early warning module is used to obtain the abnormal probability of each candidate depression region identified at the current time based on the degree of influence of the depression, and to perform graded early warning.

Images