Method for integrating glue stripping and thread trimming by additionally arranging yarn cutting mechanism on glue dissolving and stripping machine

Abstract

The invention relates to the technical field of cable processing equipment, and discloses a method for integrating glue stripping and wire cutting by additionally arranging a yarn cutting mechanism on a glue dissolving and stripping machine. The method comprises the steps of capturing a cable surface image through a visual acquisition unit, processing the cable surface image to obtain a clear feature map, and separating a rubber area and a clutter line distribution feature according to the clear feature map. And dynamically judging the operation stage by analyzing the change trend of the integrity of the rubber region. In the rubber stripping completion period, a dynamic yarn cutting path is generated by combining pre-stored parameters and real-time clutter line characteristics, and a cutter displacement vector is calculated to drive a mechanism. And synchronously monitoring and calibrating the temperature of the rubber stripping heating unit, so that the rubber stripping heating unit and the yarn cutting action are alternately and parallelly executed on the time axis according to the cooperative control logic. And the characteristic database is updated in real time, so that the self-adaptive matching of the stripping and trimming parameters is realized. The problems that in the prior art, stage judgment is not accurate, the yarn cutting path is fixed, and the working procedure collaboration is poor are solved, and efficient and accurate integrated automatic operation of the rubber stripping and thread trimming working procedures is achieved.

Description

Technical Field

[0001] This invention relates to the field of cable processing equipment technology, specifically a method for integrating adhesive stripping and wire cutting by adding a yarn cutting mechanism to a melt-stripping machine. Background Technology

[0002] In existing cable handling equipment, debonding and cutting loose wires are often performed as two separate processes. Conventional equipment typically uses simple optical sensors or mechanical limit switches to detect the state of the adhesive, triggering a debonding completion signal based on a single physical threshold. This static determination mechanism is difficult to adapt to differences in cable specifications or the initial state of the adhesive, and is prone to misjudgment due to surface scratches or unexpected damage. This can cause the equipment to start cutting the wire prematurely before the adhesive is completely debonded, or to delay the action even after the adhesive has been completely removed.

[0003] In the yarn cutting process, existing technologies largely rely on preset fixed trajectories to control the cutter movement. This method assumes that the distribution and shape of loose yarns after stripping are fixed or predictable. However, in actual production, loose yarns are randomly scattered due to the uncertainty of the stripping process. Fixed cutting paths cannot completely cover all loose yarns, often resulting in missed cuts or improper cutting depths, posing a risk of damaging the internal core yarns. Furthermore, stripping and yarn cutting are sequential, independent cycles without coordination. The equipment must wait for the previous process to complete before starting the next, resulting in unnecessary idle waiting time in the overall work cycle, limiting further improvements in production efficiency. The equipment also struggles to intelligently respond to real-time changes in the processing status. Summary of the Invention

[0004] The purpose of this invention is to provide a method for integrating glue stripping and yarn cutting by adding a yarn cutting mechanism to a melt-stripping machine, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a method for integrating glue stripping and yarn cutting by adding a yarn cutting mechanism to a melt-stripping machine, the method comprising: The original image information of the cable surface is captured by the visual acquisition unit, and the original image information is subjected to multiple noise filtering processes, followed by image enhancement, contour sharpening and grayscale normalization operations to obtain a clear cable surface feature map. Based on a clear image of the cable surface features, the image features of the rubber area and the image features of the miscellaneous wire distribution are separated, and a feature map of the rubber and miscellaneous wires is established. By analyzing the trend of the integrity change of the image features of the rubber area, it is determined whether the current operation stage is the beginning stage of rubber stripping or the completion stage of rubber stripping. When it is determined that the stripping process has been completed, the pre-stored yarn cutting trajectory parameters are called, and combined with the real-time miscellaneous wire distribution image features, a dynamic yarn cutting path plan adapted to the current cable status is generated. Based on the coordinate difference between the dynamic yarn cutting path planning and the actual position of the yarn cutting mechanism, the displacement vector and motion speed of the yarn cutting tool are calculated, and a drive command sequence is generated. The temperature reading of the peeling heating unit is monitored synchronously, and the temperature reading is matched with the preset heating temperature curve. If a deviation occurs, the temperature compensation mechanism is activated to recalibrate the power output of the heating unit. By integrating the timing of yarn cutting and the temperature control cycle of degumming, a collaborative control logic is constructed to enable the yarn cutting action and the degumming operation to run in parallel on the time axis. Based on the real-time changes in the cable surface feature map, the image feature database of the rubber area and the image feature database of miscellaneous wire distribution are dynamically updated to achieve adaptive matching of the parameters for stripping rubber and cutting wires.

[0006] Preferably, the multi-noise filtering process on the original image information includes: The raw image information captured by the visual acquisition unit is input into the preprocessing channel, where illumination equalization is first performed to eliminate shadows and reflection interference. The image after illumination equalization is decomposed in the frequency domain to separate high-frequency detail components from low-frequency background components. An adaptive threshold noise reduction method is used for high-frequency detail components, and a Gaussian smoothing filter method is used for low-frequency background components. The processed high-frequency detail components and low-frequency background components are reconstructed to obtain a preliminary denoised image; Edge detection is performed on the preliminarily denoised image to extract cable contour features, and geometric correction is performed on the image based on the contour features to eliminate view distortion. The corrected image is converted into a standardized grayscale image to complete the multi-stage noise filtering process.

[0007] Preferably, the separation of the rubber region image features and the stray line distribution image features includes: Extract the pixel intensity distribution matrix from the normalized grayscale image after noise filtering; A region growing algorithm is used to gradually expand the connected region using a preset rubber pixel intensity as the seed point to generate a rubber region mask. Apply the rubber area mask to the original grayscale image to extract the rubber area image feature data set. Perform noise feature enhancement operations on the grayscale image and identify linear texture features through directional gradient histogram analysis; By performing a difference operation between the linear texture features and the rubber region mask, a binary map of the stray line distribution is obtained, and then the image feature data set of the stray line distribution is extracted.

[0008] Preferably, determining whether the current operation stage is the initial stage of peeling or the completion stage of peeling includes: Establish a time series model of the rubber area image feature data set, and calculate the rate of change of the rubber area within consecutive sampling periods; When the rate of change of the rubber area changes from negative to positive and continues to exceed the set threshold, it is marked as the beginning of the peeling period; When the rate of change of the rubber area changes from a positive value to a negative value and the area of ​​the rubber area reaches the preset integrity threshold, it is marked as the rubber peeling completion period. The density changes of the image feature dataset of stray line distribution are analyzed synchronously. When the stray line density reaches its peak during the degumming completion period, a cutting ready signal is triggered.

[0009] Preferably, the generation of a dynamic yarn cutting path plan adapted to the current cable state includes: Receive the set of data on the marker signal for the completion of degumming and the image characteristics of the distribution of stray lines; Call the reference cutting path template from the pre-stored yarn cutting trajectory parameter library; Input the image feature dataset of clutter distribution into the path optimization engine to calculate the clutter clustering heatmap; Adjust the key path points in the baseline cutting path template based on the heat map of yarn aggregation to generate a dynamic yarn cutting path plan that includes path point coordinates and cutting sequence. The dynamic yarn cutting path planning is converted into a sequence of coordinate instructions that the yarn cutting mechanism can recognize.

[0010] Preferably, the calculation of the displacement vector and movement speed of the yarn cutting tool includes: Real-time acquisition of the actual position coordinates fed back by the photoelectric encoder of the yarn cutting mechanism; Read the coordinates of the target path point in the dynamic yarn cutting path planning; Calculate the Euclidean distance and orientation angle between the current position and the target path point, and generate a displacement vector; Based on the displacement vector modulus and the acceleration constraints of the yarn cutting mechanism, calculate the uniform speed segment speed and acceleration / deceleration curve of the cutter. The feed rate of the tool is dynamically adjusted by combining the path curvature radius, generating a drive command sequence that includes position, rate, and acceleration parameters.

[0011] Preferably, the step of activating the temperature compensation mechanism if a deviation occurs includes: Continuously collect thermocouple temperature readings from the adhesive stripping heating unit to construct a temperature time series. The temperature time series is compared with the preset heating temperature curve through a sliding window, and the average deviation value is calculated. When the average deviation exceeds the tolerance range, the proportional-integral-derivative controller is activated to calculate the power compensation amount. A pulse width modulation signal duty cycle adjustment command is generated based on the power compensation amount; The adjustment command is sent to the thyristor power control module of the heating unit to change the effective power of the heating element.

[0012] Preferably, the construction of the collaborative control logic includes: Establish a timestamp list for the yarn cutting path execution sequence and a timestamp list for the stripping temperature control cycle; Analyze the phase relationship between the two timestamp lists and calculate the optimal interleaving delay time; After the rising edge of the power of the stripping heating unit, a set staggered delay time is inserted before the yarn cutting mechanism drive command is triggered. After the yarn cutting mechanism completes each path point action, the temperature stability index of the stripping heating unit is checked. If the temperature stability index does not meet the standard, the triggering time of the next yarn cutting action is dynamically adjusted.

[0013] Preferably, the time series model for establishing the rubber area image feature data set includes: Image information of the cable surface is captured at fixed sampling intervals, and the area value of the rubber area at each sampling time is extracted; The area values ​​of the rubber region at multiple consecutive sampling times are arranged in chronological order to form a time series dataset. Linear interpolation is then performed on the time series dataset to fill in the missing data points. The moving average algorithm is applied to smooth the time series dataset, generating a smoothed time series. Calculate the area difference between adjacent sampling periods in the smoothed time series, divide it by the sampling interval to obtain the instantaneous rate of change, sum the instantaneous rates of change, and divide them by the number of sampling periods to obtain the average rate of change.

[0014] Preferably, the continuous acquisition of thermocouple temperature readings from the adhesive stripping heating unit to construct a temperature time series includes: Thermocouple temperature readings are acquired at a constant frequency, and a timestamp is appended to each reading; Temperature readings are stored in chronological order as a circular buffer structure to form a temperature time series. The step of comparing the temperature time series with a preset heating temperature curve using a sliding window includes: Set the sliding window size to a fixed time span, and extract the temperature data points within the current sliding window from the temperature time series; The system obtains reference temperature data points with the same time span from the preset heating temperature curve, calculates the absolute deviation between each temperature data point in the current window and the corresponding reference data point, calculates the arithmetic mean of all absolute deviation values ​​to obtain the average deviation value, and triggers the temperature compensation mechanism when the average deviation value exceeds the tolerance range.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This method determines the operational stage by analyzing the integrity changes of image features in the rubber area, replacing the method that relies on a single static threshold. This dynamic trend analysis can distinguish between normal rubber peeling and existing surface defects during the peeling process, accurately capturing the entire process from start to finish. The system can more reliably identify the true completion time of peeling, avoiding premature or delayed actions due to misjudgment, and improving the accuracy and adaptability of stage transition judgment. This allows the equipment to handle more diverse cable types and more complex initial surface conditions, reducing operational failures caused by sensor erroneous signals.

[0016] A dynamic yarn-cutting path is generated by combining pre-stored trajectory parameters with real-time acquired images of miscellaneous wire distribution. This path planning method responds to the uncertainty of the actual miscellaneous wire distribution after each stripping operation, allowing the tool movement trajectory to adapt to the specific state of the current cable in real time. The generated path accurately covers the actual location of the miscellaneous wires, improving the thoroughness and accuracy of the cut and avoiding missed cuts and potential damage to the internal core wires. The yarn-cutting path execution sequence and the stripping temperature control cycle are integrated to construct a collaborative control logic, allowing the two processes to run in parallel on the time axis. This collaborative mechanism allows the yarn-cutting action to intervene when the stripping reaches the appropriate stage, rather than passively waiting for the entire stripping process to be completed. Similarly, temperature control can be fine-tuned according to the overall progress. This achieves partial overlap between the stripping and yarn-cutting sub-processes in the time dimension, shortening the overall processing cycle of a single cable and improving the production efficiency and resource utilization efficiency of the equipment. Attached Figure Description

[0017] Figure 1 This is a schematic diagram illustrating the working principle of the method for integrating glue stripping and thread cutting by adding a yarn cutting mechanism to the melt-stripping machine described in this invention. Figure 2 A flowchart for multiple noise filtering processes; Figure 3 A flowchart for generating dynamic yarn cutting path planning; Figure 4 Heat map of miscellaneous line distribution and cutting path planning diagram; Figure 5 This is a performance analysis diagram for a temperature monitoring and control system. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 This invention provides a method for integrating glue stripping and thread cutting into a melt-and-strip machine by adding a yarn cutting mechanism. The method includes: achieving coordinated glue stripping and thread cutting operations through integrated vision processing, path planning, and temperature control technologies. The overall implementation scheme is as follows: The method initiates the cable processing flow, captures the original image information of the cable surface through a vision acquisition unit, performs multiple noise filtering processes on the original image information, and sequentially performs image enhancement, contour sharpening, and grayscale normalization operations to obtain a clear cable surface feature map; based on the clear cable surface feature map, the image features of the rubber area and the image features of the miscellaneous thread distribution are separated, and a feature map of the rubber and miscellaneous thread is established. By analyzing the completeness change trend of the rubber area image features, the current operation stage is determined to be either the glue stripping start stage or the glue stripping completion stage; when the glue stripping completion stage is determined, the pre-stored yarn cutting trajectory parameters are called, combined with the real-time miscellaneous thread distribution image features, to generate a model suitable for the current cable state. Dynamic yarn cutting path planning; based on the coordinate difference between the dynamic yarn cutting path planning and the actual position of the yarn cutting mechanism, the displacement vector and movement speed of the yarn cutting tool are calculated, and a drive command sequence is generated; the temperature reading of the stripping heating unit is monitored synchronously, and the temperature reading is matched with the preset heating temperature curve. If a deviation occurs, the temperature compensation mechanism is activated to recalibrate the power output of the heating unit; the yarn cutting path execution sequence and the stripping temperature control cycle are integrated to construct a collaborative control logic, so that the yarn cutting action and the stripping operation are interleaved and parallel on the time axis; according to the real-time changes of the cable surface feature map, the rubber area image feature database and the miscellaneous wire distribution image feature database are dynamically updated to achieve adaptive matching of stripping and cutting parameters.

[0020] Example 1: See Figure 2 In specific implementation, the vision acquisition unit uses an industrial camera fixed above the operating station of the melt-and-strip machine. Under trigger signal control, the industrial camera captures raw image information including the cables. The raw image information is transmitted in RGB format to the preprocessing channel of the image processing module. In the preprocessing channel, illumination equalization processing is first performed. This processing uses an algorithm based on Retinex theory to correct the image pixel values ​​to eliminate interference caused by shadows due to uneven ambient lighting and reflections from metal surfaces. In some embodiments, a Fast Fourier Transform (FFT) is performed on the image after illumination equalization processing to convert the image from the spatial domain to the frequency domain, thereby separating the high-frequency components representing detail edges and the low-frequency components representing smooth background regions. An adaptive threshold denoising method is used for the high-frequency components, with the threshold dynamically determined based on the variance of pixels within a local window. For the low-frequency components, a Gaussian smoothing filter with a standard deviation of 1.5 pixels is used for smoothing. The high-frequency and low-frequency components, after separate denoising processing, are reconstructed using an inverse Fourier transform to obtain a preliminary denoised image.

[0021] In the specific implementation, edge detection is performed on the preliminary denoised image. The edge detection operation uses the Canny operator to extract the contour features of the cable. Based on the extracted contour features, the orientation of the cable in the image is estimated by calculating the minimum bounding rectangle of the contour. Then, affine transformation is applied to geometrically correct the image to eliminate perspective distortion caused by the non-perpendicular view of the industrial camera. The corrected image is converted into a standardized grayscale image with pixel values ​​between 0 and 255, thus completing the multi-noise filtering process.

[0022] In the specific implementation, a region growing algorithm is used for rubber region segmentation. The algorithm uses a preset rubber pixel intensity value of 85 as a seed point, typically located in the rubber portion of the image's center. Based on the 8-neighborhood connectivity criterion, the algorithm progressively merges pixels whose grayscale value differs from the seed point by less than a threshold of 10, generating a complete rubber region mask. This generated mask is applied to the original grayscale image, and through pixel-to-pixel multiplication, an image feature data set containing only the rubber region is extracted. This can be understood as performing a clutter feature enhancement operation on the original grayscale image. This operation analyzes the gradient magnitude and direction of pixels by calculating the image's directional gradient histogram, thereby identifying linear texture features with significant directionality. These linear texture features typically correspond to clutter. The binarized image of the identified linear texture features is then subjected to a difference operation with the rubber region mask—that is, the portion covered by the rubber region mask is subtracted from the linear texture feature image—to obtain a binary image accurately representing the clutter location. From this, a clutter distribution image feature data set is extracted. This dataset contains information on the number, location coordinates, and orientation angles of the stray lines. In some embodiments, the stopping condition for the region growing algorithm is determined by the following formula:

[0023] in: Represents the pixel currently to be judged. grayscale value, This represents the average gray value of the currently grown area. This represents the standard deviation of the currently grown area. This represents a preset threshold constant; when the inequality holds, the pixel... It is not included in the growth area.

[0024] Example 2: In specific implementation, a time series model of the rubber area image feature data set is established. The construction of the time series model is based on a fixed sampling interval of 50 milliseconds. At each sampling moment, the visual acquisition unit is triggered to capture image information of the cable surface, and an image processing algorithm is called to extract the area value of the rubber area in the current image. The area values ​​of the rubber area at 100 consecutive sampling moments are arranged in chronological order to form a time series dataset of length 100. For individual missing data points due to data transmission delays, linear interpolation is used to fill in the missing data points to ensure the continuity of the time series. It can be understood that the moving average algorithm is applied to smooth the completed time series dataset to suppress random fluctuations. The moving average algorithm uses a sliding window with a window width of 5, and calculates the arithmetic mean of the area values ​​within the window as the smoothed value at the center of the window, thereby generating the smoothed time series.

[0025] In practice, the area difference between adjacent sampling periods in the smoothed time series is calculated. Adjacent sampling periods refer to the nth sampling time and the (n-1)th sampling time. The area difference is divided by the sampling interval of 50 milliseconds to obtain the instantaneous rate of change. The instantaneous rates of change from multiple consecutive sampling periods are accumulated and summed, and then divided by the number of instantaneous rates of change involved in the summation to obtain the average rate of change. The operation stage is determined by monitoring the sign and value of the rubber area change rate. When the rubber area change rate changes from negative to positive, and its absolute value exceeds a set threshold of 5 square millimeters per second for three consecutive sampling periods, the system marks the current operation stage as the peeling start period. In other words, when the rubber area change rate changes from positive to negative, and simultaneously the rubber area reaches a preset integrity threshold of 95%, the system marks the current operation stage as the peeling completion period.

[0026] In specific implementation, the density change of the stray yarn distribution image feature data set is analyzed synchronously. The stray yarn distribution density is calculated by counting the number of white pixels in the binary image of stray yarn distribution per unit area. When the system marks the entry into the stripping completion period, the stray yarn density value is monitored in real time. If the stray yarn density value reaches more than 90% of its maximum peak value during the entire processing during the stripping completion period, a cutting ready signal is triggered. This signal is used to start the subsequent yarn cutting path planning process. In some embodiments, the average rate of change... The calculation uses the following formula:

[0027] in: Representative from the first From the sampling time to the... This sampling time Average rate of change over a period of time Representing the The area value of the rubber region at each sampling time. Represents the sampling interval time. This represents the number of sampling periods used to calculate the average rate of change. The value is 10. Optionally, the decision logic for the operation phase is written as a state machine program and runs in the programmable logic controller. Optionally, the integrity threshold of the rubber area can be preset and configured on the operation interface according to different cable models.

[0028] Example 3: See Figure 3 In practical implementation, the path planning module receives the stripping completion marker signal sent by the operation stage determination module, along with a set of stray wire distribution image feature data at the current moment. This set of stray wire distribution image feature data contains the two-dimensional coordinate information of the stray wires. The path planning module retrieves a reference cutting path template matching the current cable model from a pre-stored yarn cutting trajectory parameter library in non-volatile memory. The reference cutting path template contains a series of two-dimensional path point coordinates defined under standard conditions. It can be understood that the received stray wire distribution image feature data set is input into the path optimization engine. Based on the distribution of stray wire coordinate points, the path optimization engine uses a kernel density estimation algorithm to generate a stray wire aggregation heatmap. This heatmap visually reflects the density distribution of stray wires on a two-dimensional plane. The kernel density estimation algorithm calculates the density estimate of each grid cell by assigning a smooth kernel function to each stray line coordinate point and superimposing the contribution values ​​of all kernel functions on the two-dimensional grid. This generates a heatmap of stray line aggregation, which represents the degree of stray line aggregation in grayscale or color intensity. This heatmap transforms the discrete distribution of stray line coordinate points into a continuous density distribution field, so that high-density areas correspond to stray line aggregation areas and low-density areas correspond to stray line sparse areas, providing an intuitive spatial density reference for subsequent path adjustment.

[0029] In practical implementation, the path optimization engine adjusts the key path points in the baseline cutting path template based on the calculated heatmap of miscellaneous thread aggregation. The adjustment principle is to prioritize the cutting path through areas with high miscellaneous thread aggregation. Specific adjustment methods involve inserting new path points within areas where the miscellaneous thread aggregation exceeds a set threshold, or fine-tuning the coordinates of existing path points, ultimately generating a dynamic yarn cutting path plan containing optimized path point coordinates and a defined cutting sequence. The dynamic yarn cutting path plan is converted into a sequence of Cartesian coordinate instructions recognizable by the yarn cutting mechanism's motion control system via a coordinate transformation module. In some embodiments, the yarn cutting mechanism employs a two-dimensional Cartesian coordinate robot structure.

[0030] In practical implementation, the motion control unit acquires pulse signals from photoelectric encoders mounted on the X-axis and Y-axis servo motors of the yarn cutting mechanism in real time. The current actual position coordinates of the yarn cutting tool are obtained by multiplying the pulse count by the pulse equivalent. Simultaneously, the motion control unit reads the coordinates of the target path point to be reached in the dynamic yarn cutting path planning. In other words, the motion control unit calculates the difference between the current position coordinates and the target path point coordinates. and The motion control unit calculates the Euclidean distance and the required direction angle of the tool movement based on these two differences, thereby generating a displacement vector. Optionally, the motion control unit calculates the uniform speed segment and the acceleration and deceleration curves of the tool during the process of moving from the current position to the target position, based on the calculated displacement vector magnitude and the inherent maximum acceleration constraint of the servo axis of the yarn cutting mechanism. Optionally, the motion control unit anticipates the direction of subsequent path points in the dynamic yarn cutting path planning and dynamically adjusts the tool feed rate in combination with the radius of curvature of the current path segment, appropriately reducing the rate when the path curvature is large to ensure cutting stability. In some embodiments, the direction angle of the displacement vector... Calculated using the following formula:

[0031] in: This represents the angle between the displacement vector and the positive X-axis. This represents the difference between the Y-coordinate of the target path point and the Y-coordinate of the current position. This represents the difference between the X-coordinate of the target path point and the X-coordinate of the current position. The motion control unit ultimately generates a sequence of drive commands containing the target position, planned speed, and acceleration parameters, and sends it to the servo driver for execution.

[0032] See Figure 4 This diagram demonstrates the intelligent optimization process of cable surface stray wire distribution and yarn cutting path planning. The heatmap, created using color gradients, clearly shows the density distribution of stray wires on a two-dimensional plane; darker areas represent highly concentrated stray wires, while lighter areas are relatively sparse. Based on the heatmap, the system plots two critical path trajectories: one is the initially set baseline cutting path, and the other is the optimized path dynamically adjusted according to stray wire density. It is evident that the optimized path actively approaches areas of dense stray wires, inserting new path points and fine-tuning existing coordinates to ensure the yarn cutting tool accurately covers these areas. The entire diagram vividly showcases the complete intelligent decision-making process from visual recognition to path generation through multiple visual elements such as coordinate grids, path points, and color gradients, demonstrating the system's adaptability and optimization capabilities to complex spatial distributions.

[0033] Example 4: In specific implementation, the temperature monitoring system continuously acquires temperature readings of the K-type thermocouples in the peeling heating unit at a constant frequency of 100 Hz, and adds a timestamp accurate to the millisecond level to each acquired temperature reading. These timestamped temperature readings are stored sequentially in a circular buffer data structure with a capacity of 1000 data points, thus forming a continuous temperature time series. It can be understood that the design of the circular buffer structure allows new data to overwrite old data, ensuring that the temperature history within the most recent 10 seconds is always preserved. The preset heating temperature curve is stored in the read-only memory of the control system in the form of a function, which defines the expected value of the heating temperature changing with time under ideal conditions.

[0034] In practical implementation, the temperature control algorithm sets the sliding window size to a fixed time span of 2 seconds. It extracts 200 temperature data points within the current sliding window from the circular buffer of the temperature time series, and simultaneously obtains 200 reference temperature data points with the same 2-second time span from the preset heating temperature curve. It calculates the absolute deviation between each actual temperature data point within the current window and its corresponding reference data point, and calculates the arithmetic mean of all 200 absolute deviation values ​​to obtain the average deviation value. When the calculated average deviation value exceeds the preset tolerance range of ±5 degrees Celsius, the system immediately activates the temperature compensation mechanism. The core of the temperature compensation mechanism is a proportional-integral-derivative (PID) controller. The PID controller calculates the required power compensation amount based on the average deviation value, the integral term of the deviation, and the derivative term. Optionally, the parameters of the PID controller are tuned according to the thermal characteristics of the heating unit. Optionally, the average deviation value... The calculation uses the following formula:

[0035] in: This represents the average deviation value within the sliding window. This represents the number of data points within the sliding window, which is 200 in this case. Represents timestamp The actual temperature readings collected at all times. Represents timestamp The reference temperature value is obtained from the preset heating temperature curve at all times.

[0036] In specific implementation, based on the power compensation calculated by the proportional-integral-derivative (PID) regulator, the signal generation module maps it into a duty cycle adjustment command for a pulse-width modulation (PWM) signal. This duty cycle adjustment command is sent digitally to the thyristor power control module of the heating unit. The thyristor power control module adjusts the effective voltage applied across the heating element by changing the conduction angle within one AC cycle, thereby altering the effective power output of the heating element and achieving closed-loop temperature control. In some embodiments, the carrier frequency of the pulse-width modulation signal is set to 1 kHz. In some embodiments, refer to Table 1 for temperature time series data.

[0037] Table 1: Temperature Time Series Data Table

[0038] See Figure 5 This chart showcases the real-time operating status and regulation performance of the temperature monitoring and control system. Two main curves compare the actual measured temperature with the preset ideal temperature, highlighting the difference in control accuracy. The chart also displays the average deviation curve calculated using a sliding window, along with clearly marked temperature tolerance ranges and compensation trigger zones. When the actual temperature deviates from the preset range, the system activates the compensation mechanism; these periods are highlighted in the chart, visually demonstrating the dynamic response of the closed-loop control system. The entire chart, employing a dual-axis design, region filling, and threshold marking, comprehensively presents the closed-loop control process from temperature acquisition and deviation analysis to power regulation, demonstrating the system's ability to maintain temperature stability under complex operating conditions.

[0039] Example 5: In specific implementation, the construction of the collaborative control logic begins with establishing two independent timestamp lists. The timestamp list for the yarn cutting path execution sequence records the precise moment when the action command for each path point in the dynamic yarn cutting path planning is issued. The timestamp list for the stripping temperature control cycle records the moment when the heating unit power adjustment event occurs, particularly the trigger moment of the power rising edge. It can be understood that analyzing the phase relationship between the two timestamp lists is achieved by calculating the time interval between the power rising edge moment of each heating unit and the trigger moment of the nearest subsequent yarn cutting action. The system performs statistical analysis on all such time intervals and selects the time interval that maximizes the execution efficiency of both processes while avoiding resource conflicts as the optimal interleaving delay time. In some embodiments, the optimal interleaving delay time... Calculated using the following formula:

[0040] in: This represents the calculated optimal interleaving delay time. This represents the timestamp of the first yarn cutting action after the j-th heating cycle. This represents the power rising edge timestamp for the j-th heating cycle. This represents the number of heating cycles used for statistical analysis. This represents the safety offset set based on the system's response characteristics.

[0041] In practical implementation, the collaborative control logic inserts a calculated optimal staggered delay time after the rising edge event of the power of the stripping heating unit, and then sends a drive command to the yarn cutting mechanism to trigger the yarn cutting action. This timing arrangement ensures that the heating operation has sufficient time for the rubber to reach the predetermined softening state before cutting. It can be understood that after the yarn cutting mechanism completes the cutting action at each path point in the dynamic yarn cutting path planning, the system will detect the temperature stability index of the stripping heating unit. The temperature stability index is measured by calculating the standard deviation of the temperature reading within the most recent second. If the calculated temperature standard deviation exceeds the preset stability threshold, it indicates that the temperature fluctuation is large, and the collaborative control logic will dynamically adjust the triggering timing of the next yarn cutting action, for example, by appropriately increasing the delay time of the next action to wait for the temperature to stabilize. Optionally, the temperature stability threshold is set to 2 degrees Celsius. Optionally, a high-precision timer is used to manage the timestamp list to ensure the accuracy of the time interval control.

[0042] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0043] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for achieving the integrated operation of stripping and cutting by installing a cutting mechanism on a solution stripping coater, characterized in that, The method starts a cable processing flow, specifically comprising: Capture the original image information of the cable surface by the visual acquisition unit, perform multiple noise filtering on the original image information, sequentially perform image enhancement, contour sharpening and gray scale normalization, and obtain a clear cable surface feature map; Based on the clear cable surface feature map, separate the rubber area image feature and the wire distribution image feature, establish the feature map of rubber and wire, and determine whether the current operation stage is the stripping starting period or the stripping completion period by analyzing the integrity change trend of the rubber area image feature; When it is determined that the stripping completion period is entered, call the pre-stored cutting yarn trajectory parameters, combine the real-time wire distribution image feature, and generate a dynamic cutting yarn path planning suitable for the current cable state; According to the coordinate difference value between the dynamic cutting yarn path planning and the actual position of the cutting yarn mechanism, calculate the displacement vector and motion rate of the cutting yarn cutter, and generate a driving instruction sequence; Synchronously monitor the temperature reading of the stripping heating unit, match the temperature reading with the preset heating temperature curve, and if there is a deviation, start the temperature compensation mechanism to recalibrate the power output of the heating unit; Integrate the cutting yarn path execution time sequence and the stripping temperature regulation period, construct a cooperative control logic, and make the cutting yarn action and the stripping operation staggered and parallel on the time axis; According to the real-time change of the cable surface feature map, dynamically update the rubber area image feature database and the wire distribution image feature database, and realize adaptive matching of the stripping and cutting parameters.

2. The method of claim 1, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The multiple noise filtering on the original image information comprises: Input the original image information captured by the visual acquisition unit into a preprocessing channel, first perform light equalization processing to eliminate shadow and reflection interference; Perform frequency domain decomposition on the image after light equalization to separate high-frequency detail components and low-frequency background components, use an adaptive threshold noise reduction method for high-frequency detail components, and use a Gaussian smoothing filter method for low-frequency background components; Reconstruct the processed high-frequency detail components and low-frequency background components to obtain a preliminary denoising image; Perform edge detection operation on the preliminary denoising image to extract the cable contour feature, and perform geometric correction on the image based on the contour feature to eliminate perspective distortion; Convert the corrected image into a standardized gray scale image to complete the multiple noise filtering process.

3. The method of claim 2, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The separation of the rubber area image feature and the wire distribution image feature comprises: Extract the pixel intensity distribution matrix from the standardized gray scale image after noise filtering; Use a region growing algorithm to gradually expand the connected region with the preset rubber pixel intensity as the seed point to generate a rubber area mask; Apply the rubber area mask to the original gray scale image to extract a rubber area image feature data set; Perform wire feature enhancement operation on the gray scale image to identify linear texture features through histogram of oriented gradients analysis; Differential operation of the linear texture features and the rubber area mask obtains a wire distribution binary image, and further extracts a wire distribution image feature data set.

4. The method of claim 3, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The determination of whether the current operation stage is the stripping starting period or the stripping completion period comprises: Establish a time series model of the rubber area image feature data set, and calculate the change rate of the rubber area in the continuous sampling period; When the area change rate of the rubber region changes from a negative value to a positive value and continues to exceed a set threshold, the rubber stripping start period is marked; When the area change rate of the rubber region changes from a positive value to a negative value and the area of the rubber region reaches a preset completeness threshold, the rubber stripping completion period is marked; Synchronously analyze the density change of the feature data set of the wire distribution image, and when the wire density reaches a peak value at the rubber stripping completion period, a cutting readiness signal is triggered.

5. The method of claim 4, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The generation of the dynamic cutting path planning adapted to the current cable state comprises: receiving the rubber stripping completion period marking signal and the feature data set of the wire distribution image; calling a reference cutting path template from a pre-stored cutting trajectory parameter library; inputting the feature data set of the wire distribution image into a path optimization engine to calculate a wire aggregation degree heat map; adjusting key path points in the reference cutting path template according to the wire aggregation degree heat map to generate a dynamic cutting path planning containing path point coordinates and a cutting sequence; converting the dynamic cutting path planning into a coordinate instruction sequence recognizable by the cutting mechanism.

6. The method of claim 5, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The calculation of the displacement vector and the motion rate of the cutting tool comprises: real-time acquisition of actual position coordinates fed back by a photoelectric encoder of the cutting mechanism; reading target path point coordinates in the dynamic cutting path planning; calculating the Euclidean distance and the direction angle between the current position and the target path point to generate a displacement vector; calculating the uniform speed segment rate and the acceleration / deceleration curve of the tool according to the module length of the displacement vector and the acceleration constraint condition of the cutting mechanism; combining the path curvature radius to dynamically adjust the feed rate of the tool to generate a driving instruction sequence containing position, rate and acceleration parameters.

7. The method of claim 1, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the thread is stripped by the stripping and cutting machine. The temperature compensation mechanism is started when the deviation occurs, comprising: continuous acquisition of thermocouple temperature readings of the rubber stripping heating unit to construct a temperature time sequence; sliding window comparison of the temperature time sequence with a preset heating temperature curve to calculate an average deviation value; when the average deviation value exceeds a tolerance range, starting a proportional-integral-derivative regulator to calculate a power compensation amount; generating a pulse width modulation signal duty cycle adjustment instruction according to the power compensation amount; sending the adjustment instruction to a silicon-controlled power regulating module of the heating unit to change the effective power of the heating element.

8. The method of claim 7, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The construction of the cooperative control logic comprises: establishing a timestamp list of the cutting path execution timing and a timestamp list of the rubber stripping temperature regulation period; analyzing the phase relationship of the two timestamp lists to calculate an optimal staggered delay time; inserting a set staggered delay time after the power rising edge of the rubber stripping heating unit and then triggering the cutting mechanism driving instruction; after the cutting mechanism completes an action at each path point, detecting a temperature stability indicator of the rubber stripping heating unit, and if the temperature stability indicator does not meet the standard, dynamically adjusting the triggering time of the next cutting action.

9. The method of claim 4, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the stripping operation. The time sequence model of the feature data set of the rubber region image comprises: capturing image information of the cable surface at a fixed sampling interval and extracting the rubber region area value at each sampling time; arranging the rubber region area values at a plurality of continuous sampling times in time sequence to form a time sequence data set, performing linear interpolation processing on the time sequence data set to fill in missing data points; applying a moving average algorithm to smooth the time sequence data set to generate a smoothed time sequence; The area difference of adjacent sampling periods in the smoothed time series is calculated, and divided by the sampling interval time to obtain the instantaneous change rate, the instantaneous change rate is cumulatively summed, and divided by the number of sampling periods to obtain the average change rate.

10. The method of claim 7, wherein the cutting mechanism is installed in the stripping and cutting machine, and the cutting mechanism cuts the thread after the thread is stripped by the stripping and cutting machine. The continuous acquisition of the thermocouple temperature reading of the stripping and heating unit, and the construction of the temperature time series include: The thermocouple temperature reading is acquired at a constant frequency, and a time stamp is added to each reading; The temperature readings are stored in time sequence as a ring buffer structure to form a temperature time series; The temperature time series is compared with the preset heating temperature curve in a sliding window, which includes: The size of the sliding window is set to a fixed time span, and the temperature data points in the current sliding window are extracted from the temperature time series; The reference temperature data points of the same time span are obtained from the preset heating temperature curve, the absolute deviation of each temperature data point in the current window and the corresponding reference data point is calculated, and the arithmetic mean of all absolute deviation values is calculated to obtain the average deviation value, and when the average deviation value exceeds the tolerance range, the temperature compensation mechanism is triggered.

Images