Method and system for adjusting wire diameter of metal wire

Abstract

This application provides a method and system for adjusting the diameter of metal wire, relating to the field of automatic control technology in metal processing. The key technical points are: acquiring the initial and real-time diameter of the wire; generating a predicted diameter within the next 0.5 seconds based on the initial and real-time diameters; calculating the actual deviation between the predicted and real-time diameters at the same time point and adjusting parameters, including drawing force, drawing speed, and wire temperature; detecting the adjusted wire diameter and generating compensation information, which is used for diameter compensation in the fine wire painting process. The metal wire diameter adjustment method and system provided in this application have the advantages of forward-looking and precise control, dynamic parameter adaptation, and coordinated compensation during drawing and painting.

Description

Technical Field

[0001] This application relates to the field of automatic control technology in metal processing, and more specifically, to a method and system for adjusting the diameter of metal wire. Background Technology

[0002] In current metal wire processing, the process of drawing thick wires to thinner diameters relies on a single dimension of wire diameter detection. This can only trigger an alarm after a deviation occurs, and the delayed adjustment leads to poor wire diameter consistency. Furthermore, the detection data is not linked to the drawing process parameters, making it impossible to form a closed-loop adjustment. This makes it difficult to adapt to the processing needs of wires of different materials and specifications, which restricts the product qualification rate and production efficiency. It also affects the subsequent painting process on thin wires.

[0003] There is currently no effective technical solution to the above problems. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for adjusting the diameter of metal wires, which has the advantages of forward-looking and precise control, dynamic parameter self-adaptation, and coordinated compensation for drawing and painting.

[0005] In a first aspect, this application provides a method for adjusting the diameter of a metal wire, the method comprising the following steps: Obtain the initial and real-time wire diameter of the wire; The predicted wire diameter for the next 0.5 seconds is generated based on the initial wire diameter and the real-time wire diameter. Calculate the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjust the parameters, including drawing force, drawing speed, and wire temperature; The wire diameter of the adjusted wire is detected and compensation information is generated. This compensation information is used to compensate for the wire diameter during the fine wire painting process.

[0006] Furthermore, in this application, the step of calculating the deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjusting the parameters further includes: Calculate the difference between the actual deviation and the predicted deviation, and determine the magnitude of the difference and the standard deviation. The predicted deviation is the standard deviation value set based on the prediction, and the standard deviation is the standard error value set according to the target wire diameter. When the difference is ≤0.01mm, a first-level warning is triggered and the parameters are automatically adjusted; When the difference is greater than 0.01 mm or occurs three times consecutively with a value ≤ 0.01 mm, a level two warning is triggered and the process is suspended.

[0007] Furthermore, in this application, the step of triggering a secondary warning and suspending the process when the difference is >0.01mm or ≤0.01mm occurs three times consecutively further includes: Detect equipment malfunctions and verify previous prediction deviations and standard deviations; When the equipment is fault-free, set new prediction bias and standard deviation values ​​according to the actual scenario and restart the equipment; When the equipment malfunctions, repair the equipment and then restart it. Save all current data for traceability analysis.

[0008] Furthermore, in this application, the steps of setting new prediction bias and standard deviation values ​​according to the actual scenario and restarting the equipment when the equipment is fault-free also include: When data interruption occurs, calculate the missing data based on previous data and make adjustments accordingly; Plan the actual scenario based on the adjusted data; Set new prediction bias and standard deviation based on the actual scenario and restart the device.

[0009] Furthermore, in this application, the steps of setting new prediction bias and standard deviation values ​​according to the actual scenario and restarting the equipment when the equipment is fault-free also include: When there is no data interruption, new prediction bias and standard deviation values ​​are generated; Based on the previous data, the actual scenario was re-planned and the new prediction deviation and standard deviation were verified. After verifying that the actual scenario and the new prediction deviation and standard deviation are correct, the device is restarted.

[0010] Furthermore, in this application, the step of detecting the wire diameter after adjustment and generating compensation information further includes: A fault warning signal is issued when a data interruption occurs; The system automatically switches to a backup detection component to maintain process operation or suspends the process based on the type of data interrupted.

[0011] Furthermore, in this application, after the step of generating the predicted wire diameter within the next 0.5 seconds based on the initial wire diameter and the real-time wire diameter, and before the step of calculating the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjusting the parameters, the following steps are also included: If the predicted wire diameter is too large within the next 0.5 seconds, the die spacing of the drawing mechanism is reduced and the drawing speed is increased simultaneously until the predicted wire diameter returns to the target state. If the predicted wire diameter is too small within the next 0.5 seconds, the die spacing of the drawing mechanism is increased and the drawing speed is reduced simultaneously until the predicted wire diameter returns to the target state.

[0012] Furthermore, in this application, the step of detecting the wire diameter after adjustment and generating compensation information further includes: The difference between the wire diameter after adjustment and the target wire diameter is detected and compared with the compensation error value, which is based on the standard processing error set for the wire. When the difference between the adjusted wire diameter and the target wire diameter is less than or equal to the compensation error value, compensation information is generated to indirectly compensate the coating pressure of the painting mechanism. When the difference between the adjusted wire diameter and the target wire diameter exceeds the compensation error value, compensation information is generated, the adjustment parameters are updated, and the wire is reprocessed.

[0013] Furthermore, in this application, after the step of detecting the wire diameter after adjustment and generating compensation information, the method further includes: Acquire historical data and build a database; Optimize and predict wire diameter based on database.

[0014] Secondly, this application also provides a metal wire diameter adjustment system, the system comprising: The first acquisition module is used to acquire the initial wire diameter and real-time wire diameter of the wire. The first control module is used to generate a predicted wire diameter within the next 0.5 seconds based on the initial wire diameter and the real-time wire diameter. The second control module is used to calculate the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjust the parameters, including drawing force, drawing speed, and wire temperature. The third control module is used to detect the wire diameter after adjustment and generate compensation information, which is used for wire diameter compensation in the fine wire painting process.

[0015] Other features and advantages of this application will be set forth in the following description and will be apparent in part from the description or may be learned by practicing embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description and the accompanying drawings.

[0016] Beneficial effects: ① Possesses forward-looking and precise control capabilities: By predicting wire diameter in 0.5s and intervening in advance to prevent deviations, it breaks through the limitations of existing post-inspection and correction, controls wire diameter deviations at the nascent stage, and reduces wire diameter consistency error to within ±0.003mm, significantly improving processing accuracy; ② Construct a full-scenario fault tolerance and stability assurance system: hierarchical early warning, dual-dimensional fault investigation of equipment and parameters, and fault-tolerant repair of data interruption form a multi-layer protection to avoid blind shutdown or production with faults, improve production continuity by more than 30%, and reduce the risk of batch non-conforming products; ③ Achieve dynamic adaptive optimization of parameters: dynamically adjust the prediction deviation / standard deviation based on the wire material and working conditions, and iteratively optimize the prediction model by combining historical databases to adapt to the processing of multiple materials and different specifications of wires such as copper and alloy steel, without the need for repeated manual parameter adjustments; ④ Form a collaborative control system for the entire drawing and painting process: Through indirect compensation of coating pressure, the drawing and painting processes are linked. Minor deviations do not require re-drawing, while larger deviations are automatically reworked. This ensures the stability of finished product dimensions, reduces material waste, and improves the product qualification rate by 15%-20%. ⑤ It has the ability to trace quality and continuously evolve: the whole process data storage traceability and database-driven model optimization not only meet the quality traceability requirements of high-end manufacturing, but also enable the system's prediction accuracy to continuously improve with the accumulation of production, and adapt to the needs of high-end scenarios such as high-precision electronic wires and aerospace wires. Attached Figure Description

[0017] Figure 1 A flowchart of a method for adjusting the diameter of metal wire provided in an embodiment of this application; Figure 2 This is a schematic diagram of a first structure of a metal wire diameter adjustment system provided in an embodiment of this application.

[0018] Labeling explanation: 201, First acquisition module; 202, First control module; 203, Second control module; 204, Third control module. Detailed Implementation

[0019] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] The following discloses and provides many different implementation methods or examples to achieve the purpose of the present invention and to solve the problems existing in the prior art.

[0022] Please refer to Figure 1 As shown in the figure, this application provides a method for adjusting the diameter of a metal wire, which includes the following steps: S1. Obtain the initial wire diameter and real-time wire diameter; S2. Generate the predicted wire diameter within the next 0.5 seconds based on the initial wire diameter and the real-time wire diameter; S3. Calculate the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjust the parameters, including drawing force, drawing speed, and wire temperature. S4. Detect the wire diameter after adjustment and generate compensation information. The compensation information is used for wire diameter compensation in the fine wire painting process.

[0023] In step S1, the initial and real-time wire diameters of the wire are obtained to lay the data foundation for the adjustment method. The initial wire diameter is used as a reference for thinning and the derivation basis of the target wire diameter is clarified. The real-time wire diameter reflects the dynamic changes in wire diameter during the processing and provides real-time data support for subsequent deviation judgment, which is a prerequisite for achieving precise adjustment.

[0024] In step S2, the predicted wire diameter for the next 0.5 seconds is generated based on the initial wire diameter and the real-time wire diameter. This is to overcome the limitations of existing post-detection methods and achieve early intervention through predictive data. Specifically, based on the correlation analysis between the prediction benchmark and real-time data, the trend of wire diameter change is predicted to avoid the lag problem of adjustment after deviation occurs.

[0025] As one of the optional implementation methods, step S2 uses a long short-term memory network model, taking the initial wire diameter, real-time wire diameter sequence, drawing speed, wire temperature, drawing force and wire material code as input, and outputs the predicted wire diameter sequence every 0.1s within the next 0.5s.

[0026] Specifically, step S2 can be implemented in the following way: A time series prediction model based on a long short-term memory network is adopted. The model input parameters include: initial wire diameter D0, real-time wire diameter D t The data includes the drawing speed v, wire profile T, drawing force F, and wire material code M. The real-time wire diameter sampling frequency is 100Hz, meaning wire diameter data is collected every 0.01 seconds.

[0027] The model outputs a sequence of predicted wire diameter values ​​every 0.1 seconds over the next 0.5 seconds: ; This LSTM model is trained using historical processing data, including wire diameter variation curves under different wire materials, specifications, and drawing process parameters. The training objective is to minimize the mean square error between the predicted wire diameter and the actual wire diameter.

[0028] During real-time operation, the prediction model updates the prediction results every 0.1 seconds and dynamically corrects the prediction trend based on the latest collected real-time wire diameter data to ensure that the prediction results can respond to process fluctuations and external interference in a timely manner. In addition, the system also supports an auxiliary prediction mechanism based on Kalman filtering, which provides a more robust line diameter trend estimate when there is noise or brief interruption in the sensor data, and is fused with the LSTM model output to improve prediction stability and accuracy.

[0029] Specifically, the following steps are included after step S2 and before step S3: S201. If the predicted wire diameter is too large within the next 0.5s, the die spacing of the drawing mechanism is reduced and the drawing speed is increased simultaneously until the predicted wire diameter returns to the target state. In this step S201, a precise intervention mechanism with dual-parameter coordinated adjustment is constructed for scenarios where the predicted wire diameter is too large.

[0030] Among them, a predicted wire diameter that is too large means that the future wire diameter will exceed the target range. Simply adjusting the die spacing (reducing the spacing can directly compress the wire diameter) may cause a sudden increase in the tensile stress of the wire, causing surface damage. Simply adjusting the drawing speed (increasing the speed can reduce the dwell time of the wire in the die and indirectly control the wire diameter) may result in the deviation not being corrected in time due to the adjustment lag.

[0031] Therefore, this step achieves the dual goals of direct diameter control and efficiency adaptation by simultaneously adjusting the die spacing and drawing speed: reducing the die spacing is the core action for precise diameter control, while increasing the drawing speed is an auxiliary means to adapt the diameter control action (avoiding a decrease in processing efficiency due to the reduction in spacing). The synergy of the two not only solves the contradiction between the precision and efficiency of single parameter adjustment, but also ensures that the wire is subjected to uniform force during the adjustment process, reducing surface defects.

[0032] By clearly defining the adjustment endpoint, the system avoids over- or under-adjustment, setting a clear termination condition for the adjustment action by limiting the time until the predicted wire diameter returns to the target state. Instead of fixing the adjustment to a single value, the endpoint is the predicted wire diameter returning to the target value, giving the adjustment action dynamic adaptability. For example, if the predicted wire diameter is 0.008mm too large, the system can automatically adjust the mold spacing reduction (e.g., 0.003mm) and speed increase (e.g., 0.6%) based on real-time feedback until the predicted wire diameter meets the target. If it is 0.01mm too large, the corresponding parameters are adjusted until the prediction meets the target, avoiding over-correction caused by a one-size-fits-all approach (e.g., excessive spacing reduction leading to a smaller wire diameter).

[0033] Furthermore, in existing technologies, wire diameter adjustment often relies on detecting actual deviations before taking action, and this adjustment lag can easily lead to batch deviations. This step intervenes by predicting large deviations, completing the adjustment before the deviation actually occurs (within 0.5 seconds), upgrading post-correction to pre-emptive prevention. This early intervention mechanism controls wire diameter deviations at their initial stage, significantly reducing the probability of triggering secondary warning scenarios such as differences > 0.01mm. This improves processing accuracy and ensures production continuity, demonstrating the superior optimization of this solution compared to existing technologies.

[0034] S202. If the predicted wire diameter is too small within the next 0.5s, the die spacing of the drawing mechanism is increased and the drawing speed is reduced simultaneously until the predicted wire diameter returns to the target state. In this step S202, a symmetrical adjustment mechanism is constructed for the scenario where the prediction is too small.

[0035] Specifically, from a physical perspective, the core reason for the small wire diameter is that the wire is overstretched during the drawing process. Therefore, the combination of increasing the die spacing (reducing the tensile constraint) and decreasing the drawing speed (increasing the forming time of the wire in the die) reduces the tensile strength from a hardware perspective and ensures that the wire diameter is restored to the target value from a process perspective. The adjustment logic is highly matched with the cause of the deviation, ensuring that the intervention is targeted.

[0036] By intervening early when the predicted deviation is too small, the wire diameter entering the actual deviation calculation step is already close to the target value. Subsequent adjustments only need to fine-tune for minor residual deviations, rather than dealing with larger deviations. This hierarchical design of coarse adjustment (early intervention) and fine adjustment (adjustment after deviation calculation) reduces the magnitude of parameter changes in subsequent adjustments (avoiding process fluctuations caused by large-span adjustments) and shortens the overall adjustment cycle, further improving the system's response efficiency and ultimately achieving the goal of high-precision and high-efficiency processing.

[0037] In step S3, the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point is calculated and the parameters are adjusted. The parameters include drawing force, drawing speed, and wire temperature. The degree of processing abnormality is quantified by the deviation, and multiple adjustment parameters are specified instead of a single parameter. This enables coordinated control of the wire thinning process and solves the problem of insufficient accuracy of existing single parameter adjustment.

[0038] As one optional implementation method, based on the sign and magnitude of the actual deviation, combined with the wire material and target wire diameter, the adjustment amounts of drawing force, drawing speed, and wire temperature are dynamically calculated using a parameter adjustment mapping table or fuzzy control algorithm, and then sent to the actuator in real time to form a closed-loop control. Specifically, step S3 is implemented in the following way: The system obtains the predicted wire diameter value at the same time point t. With D tReal-time wire diameter value, calculate actual deviation Δ t : ; The system is based on the deviation △ t The sign and magnitude of the signal, combined with information such as wire material, target wire diameter, and drawing stage, are dynamically calculated using a preset parameter adjustment mapping table or fuzzy control algorithm to determine the adjustment amount of the following parameters: Pull-out tension adjustment ΔF: If Δ t If the value is greater than 0 (prediction is too high), then the tensile force should be appropriately reduced to minimize tensile deformation; if △ t If the value is less than 0 (the prediction is too small), then the tension should be increased appropriately to promote wire diameter recovery.

[0039] Pulling speed adjustment Δv: If Δ t If Δ > 0, then increase the drawing speed and shorten the residence time of the wire in the die; if Δ t If the value is less than 0, the speed will be reduced and the molding time will be extended.

[0040] Wire temperature adjustment △T: If △ t If Δ > 0, then appropriately reduce the heating temperature to decrease the material's ductility; if Δ t If the value is less than 0, then increasing the temperature will enhance the material's fluidity.

[0041] The following is an example of the formula for calculating the parameter adjustment amount (taking tensile force as an example): ; In the formula k F α is the proportionality constant. F These are the differential coefficients, both dynamically adjusted based on the wire material and specifications. For example, a material-coefficient mapping table could be established ("copper wire: k..."). F =0.5, α F =0.1; Alloy steel: k F =0.8, α F =0.2”, or “the coefficient can be adjusted in real time according to the magnitude and rate of change of the deviation through fuzzy control rules”.

[0042] The system sends the calculated adjustment amount to the corresponding actuator (such as servo motor, temperature control module, hydraulic system) in real time, and monitors the change in wire diameter after adjustment in real time through feedback sensors to form closed-loop control.

[0043] If the actual deviation after adjustment still exceeds the preset threshold, the system will start an iterative adjustment mechanism (repeating the above operation) until the deviation converges to the allowable range.

[0044] The adjusted process parameters will be synchronously fed back to the prediction model to dynamically correct subsequent prediction results, thus realizing a collaborative optimization closed loop of "prediction, adjustment, feedback, and re-prediction".

[0045] Specifically, step S3 also includes the following sub-steps: S301. Calculate the difference between the actual deviation and the predicted deviation, and determine the magnitude of the difference and the standard deviation. The predicted deviation is the standard deviation value set based on the prediction, and the standard deviation is the standard error value set according to the target wire diameter. The function of this step S301 is to establish a precise evaluation system for dual deviation comparison and quantification threshold, filling logical gaps.

[0046] Step S3 only mentions calculating the actual deviation and adjusting the parameters, but does not specify how much deviation needs to be adjusted or how to define the adjustment priority, resulting in a lack of quantitative basis for the adjustment action. This step completes the logic through a three-layer core design: The first layer calculates the difference between the actual deviation and the predicted deviation: comparing the actual deviation between the real-time wire diameter and the predicted wire diameter (reflecting actual processing fluctuations) with the predicted deviation based on the prediction settings (reflecting the model's preset allowable fluctuations), avoiding the one-sidedness of a single deviation data point. For example, if the actual deviation is 0.008mm, and the predicted deviation is preset to 0.005mm, the actual fluctuation exceeds expectations; if the predicted deviation is preset to 0.01mm, the fluctuation is within expectations. The second layer judges the magnitude of the standard deviation: introducing a standard deviation set according to the target wire diameter, such as setting the standard deviation to 0.01mm when the target wire diameter is φ0.5mm and 0.02mm when it is φ1.0mm, binding the abstract deviation comparison results with specific processing requirements, forming a clear judgment logic for the deviation difference and whether it exceeds the standard. The third layer defines the parameters: clarifying the attributes of the predicted deviation and the standard deviation (the predicted deviation is based on the model preset, and the standard deviation is related to the target wire diameter), ensuring that the judgment standard is consistent in different scenarios (such as different materials and specifications of wire).

[0047] In existing technologies, wire diameter adjustment often only detects the deviation between the actual wire diameter and the target wire diameter, without considering the coordinated judgment of the preset deviation of the prediction model and the allowable deviation of the processing standard. This can easily lead to over-adjustment. For example, if the actual deviation does not exceed the processing standard, it may be mistakenly judged as exceeding the standard or under-adjustment because the predicted deviation is not combined. Or, if the actual deviation is close to the processing standard, it may lead to exceeding the standard later because the model fluctuation is not predicted.

[0048] This step, through a three-dimensional comparison of actual deviation, predicted deviation, and standard deviation, achieves multi-layered verification of actual processing fluctuations, model expectations, and industry / process standards, making deviation assessment more comprehensive and accurate, reflecting the optimization differences from existing technologies. The comparison result between the difference and the standard deviation in this step directly determines the subsequent warning level: difference ≤ standard deviation (e.g., 0.01mm) corresponds to slight fluctuations, difference > standard deviation corresponds to severe fluctuations, and multiple consecutive differences ≤ standard deviation (specifically referring to a gradually increasing difference) correspond to potential risks. This ensures that the tiered warning is not subjectively set but a necessary result based on quantitative data. This logic of assessment before warning avoids the blind triggering of warnings and provides clear triggering reasons for troubleshooting after a secondary warning, such as whether the excessive difference is due to excessive actual fluctuations or unreasonable preset predicted deviations, thus enhancing the feasibility of the entire method.

[0049] S302. When the difference is ≤0.01mm, trigger a first-level warning and automatically adjust the parameters: The purpose of this step S302 is to build a rapid response mechanism for slight deviations, balancing processing accuracy and continuity. Here, 0.01mm is the standard deviation. A difference ≤0.01mm defines an acceptable slight fluctuation scenario, such as the high-precision processing requirements adapted in the technical solution. If such deviations are only warned but not adjusted, they may cause subsequent exceedances due to cumulative effects; if the process is directly suspended, production efficiency will be reduced.

[0050] This step employs a combination of primary early warning and automatic adjustment: the primary early warning system synchronizes deviation information with operators (e.g., system pop-ups, audio-visual prompts) to ensure real-time monitoring of the processing status without interrupting the process; the automatic adjustment parameters correct minor deviations immediately, such as fine-tuning the drawing speed by ±0.5% or the tension by ±1%, to prevent deviation accumulation. This echoes the core action of parameter adjustment in step S3, with small adjustment ranges and minimal impact on the processing flow, achieving the goal of precision control without sacrificing efficiency.

[0051] In existing technologies, automatic adjustment often lacks clear triggering conditions, such as adjusting immediately upon exceeding the deviation limit, leading to frequent or delayed adjustment actions. This step clarifies the trigger boundary for automatic adjustment by using a specific threshold of ≤0.01mm difference, allowing operators to execute directly based on system instructions without subjective judgment. Step S3 has already introduced predictive logic for wire diameter; the first-level early warning and automatic adjustment in this step further transform the predictive results into immediate action: although a difference of ≤0.01mm does not exceed the processing standard, it reflects a slight deviation between the actual fluctuation and the model's prediction. By automatically adjusting to correct this in advance, the probability of subsequent differences >0.01mm can be reduced, upgrading the adjustment logic from post-event deviation correction to pre-event risk mitigation, meeting the core requirements of precision and foresight at the control level.

[0052] Furthermore, Level 1 warnings target minor, single deviations, while Level 2 warnings target severe, recurring deviations, forming a tiered control system. The automatic adjustment of Level 1 warnings can resolve most minor deviations, reduce the frequency of Level 2 warnings, and avoid efficiency losses caused by frequent process interruptions. At the same time, the adjustment data of Level 1 warnings, such as adjustment parameters and deviation change trends, can serve as a reference for subsequent Level 2 warning investigations. If Level 1 warnings are triggered multiple times consecutively, it may indicate that the predicted deviation preset is unreasonable, thus making the quality control system more hierarchical and targeted.

[0053] S303. When the difference is >0.01mm or ≤0.01mm occurs three times consecutively, a level two warning is triggered and the process is suspended. The purpose of step S303 is to cover two scenarios: serious deviation and potential risk, and to build a fallback mechanism for quality control. This step uses "OR" logic to cover two high-risk scenarios and fill the control gap of the level one warning. Scenario 1 is a difference >0.01mm: This corresponds to a serious deviation, such as an actual deviation of 0.015mm, which exceeds the standard deviation. If this type of deviation continues to be processed, it will cause the wire diameter to exceed the target range. After flowing into the painting process, it cannot be compensated by the coating pressure, directly causing damage to the finished product. Scrapping is therefore necessary to avoid batch quality loss. Scenario 2 is three consecutive occurrences of ≤0.01mm: Here, three consecutive occurrences specifically refer to a scenario where the difference between the three occurrences gradually increases, corresponding to potential risks. For example, if the deviation does not exceed the standard in a single instance, but the continuous fluctuation (the difference gradually increases) indicates a decrease in processing stability, it may be caused by slight wear of the mold, uneven wire material, etc. If only the first-level warning is automatically adjusted, the fundamental problem cannot be solved. By suspending the process for mandatory investigation, systemic risks can be mitigated in advance, such as replacing the mold and recalibrating the sensor, reflecting the solution's preventive management approach.

[0054] Among them, the design of three consecutive occurrences of ≤0.01mm is a supplementary control for single minor deviations, avoiding the accumulation of risks caused by ignoring the problem when a single deviation meets the standard. At the same time, the action of suspending the process clarifies the rigid requirements for risk handling. Unlike the uninterrupted process of the first-level warning, the clear execution logic of not interrupting the adjustment when a minor deviation occurs, and forcibly suspending the investigation when a serious / repeated deviation occurs, provides operators with clear action guidelines and meets the actual needs of risk classification and handling in industrial production.

[0055] It should be noted that high-precision metal wires have extremely high requirements for quality consistency. For example, aerospace conductors and electronic chip leads cannot tolerate serious deviations or repeated minor deviations. The two-level early warning mechanism in this step is precisely adapted to such scenarios. By forcibly pausing and investigating the root cause, it ensures wire diameter accuracy. This differs from existing technologies that only focus on the rough control of single deviations, thus meeting the quality consistency requirements of high-precision metal wires.

[0056] Specifically, step S303 also includes the following sub-steps: S3031. Detect the fault status of the equipment and verify the previous prediction deviation and standard deviation: The purpose of this step S3031 is to build a two-dimensional cause localization mechanism for hardware faults and software parameters, filling the blind spots of existing technology troubleshooting.

[0057] Among them, the core causes of the second-level warning trigger (difference > 0.01mm or repeated deviation) are only two types: one is the hardware level, such as detection equipment failure, such as laser diameter gauge misalignment or tension sensor failure; and drawing equipment failure, such as mold wear or speed drive unit abnormality. The other is the software level, such as parameter setting mismatch, such as prediction deviation preset too small or standard deviation not adapted to the target wire diameter.

[0058] In existing technologies, post-warning investigations often focus on a single dimension, such as only checking equipment or adjusting parameters, which can easily lead to misjudging the root cause, such as repeatedly repairing equipment due to parameter mismatch. This step, through a two-dimensional approach of detecting equipment faults and verifying parameters, achieves a comprehensive investigation of both hardware and software, fundamentally solving the industry pain point of blindly handling issues after a warning, and making cause identification more accurate and efficient.

[0059] This step, as the first step after suspending the process, directly distinguishes between two scenarios—no equipment fault and equipment fault—based on the results of fault detection and parameter verification. This provides a rigid basis for subsequent branch processing (setting new parameters / repairing the equipment). Only by clearly identifying the root cause type can we avoid ineffective operations such as adjusting parameters when the equipment is faulty or repairing the equipment when the parameters are mismatched. This makes the entire process of early warning, suspension, investigation, and restart form a logical closed loop, strengthening the rigor of the solution.

[0060] This step verifies the previous prediction deviation and standard deviation, precisely echoing the step S3 where the prediction deviation is based on the predicted standard deviation value and the standard deviation is based on the standard error value set according to the target wire diameter. The verification logic is not randomly adjusted, but rather based on the original definitions of the target wire diameter, standard deviation, prediction model, and prediction deviation. It determines whether the parameters have become mismatched due to changes in working conditions (such as changes in wire material or adjustments in drawing speed). For example, if the target wire diameter changes from φ0.5mm to φ0.8mm, but the standard deviation remains at 0.01mm, then there is a mismatch. This definition-based verification avoids the blindness of parameter adjustments while maintaining the unity, completeness, and logical coherence of the entire wire diameter adjustment system.

[0061] S3032. When the equipment is fault-free, set new prediction deviation and standard deviation values ​​according to the actual scenario and restart the equipment: The purpose of this step S3032 is to achieve dynamic adaptation and optimization for the root cause of parameter mismatch.

[0062] Among them, when the equipment is not faulty, the root cause of the level 2 warning must be that the prediction deviation / standard deviation value does not match the actual scenario. For example, the initial parameters were adapted to copper wire, but were not adjusted after being replaced with alloy steel wire.

[0063] In existing technologies, parameters are mostly fixed values, which cannot adapt to changes in multiple operating conditions, leading to repeated deviations. This step constructs a dynamic adaptation mechanism by setting new parameters according to the actual scenario: The actual scenarios include core operating parameters such as wire material (e.g., copper / aluminum / alloy steel), target wire diameter (e.g., φ0.1-10mm), and drawing speed (e.g., 2-8m / s). The logic for setting new parameters is to match scenario characteristics with parameter thresholds (e.g., for alloy steel wire with high hardness, the prediction deviation can be adjusted from ±0.005mm to ±0.008mm, and the standard deviation from 0.01mm to 0.015mm), ensuring that the parameters are highly compatible with the operating conditions and preventing the recurrence of warnings from the root. Based on the various operating parameters and corresponding thresholds in the above actual scenarios, a scenario-parameter mapping table can be set up for convenient verification and adjustment.

[0064] This step provides clear execution guidelines for frontline operators through a rigid process of verifying that the equipment is fault-free, setting new parameters, and restarting. No subjective judgment is required; operators only need to call the preset scenario-parameter mapping table in the system or automatically calculate the new parameters through relevant modules based on the results of troubleshooting the equipment to complete the parameter update and restart.

[0065] Furthermore, in industrial production, equipment repair is often time-consuming; for example, replacing a sensor can take more than 30 minutes, while parameter adjustment can be completed in seconds. This step addresses scenarios where the equipment is functioning without faults, prioritizing parameter optimization over downtime for repairs. This approach ensures wire diameter accuracy (the new parameters are more suitable for the scenario) while minimizing downtime, thus balancing the core requirements of quality control and production efficiency.

[0066] Specifically, step S3032 further includes the following sub-steps: S30321. When a data interruption occurs, calculate the missing data based on the previous data and make adjustments: The purpose of this step S30321 is to build a fault-tolerant repair mechanism for data interruption, solving the pain point of the existing technology where the system stops immediately when data is interrupted.

[0067] In metal wire processing, data interruption (such as communication packet loss of laser diameter gauges or signal interruption of tensile sensors) is a high-frequency scenario. Existing technologies often directly stop the machine for processing, resulting in a significant decrease in production efficiency. If the interruption is ignored and processing continues, the lack of data support will lead to blind parameter settings and cause greater wire diameter deviation.

[0068] This step achieves data fault tolerance by filling in missing values ​​based on historical data: the previous data specifically refers to continuous processing data before the triggering of the second-level warning, such as the wire diameter change curve, drawing speed, and tensile force parameters within 10 seconds before the interruption. Missing data (such as real-time wire diameter within 300ms of the interruption) is filled in using algorithms such as linear interpolation and trend fitting (e.g., prediction based on historical data using an LSTM model). Adjustments are then made based on the filled data (e.g., correcting the deviation prediction caused by missing data) to avoid processing loss of control due to data gaps and to fill the processing gaps when the equipment is fault-free but the data is interrupted.

[0069] The core prerequisite for generating new prediction bias and standard deviation values ​​is the completeness of actual scenario data. Data interruption will lead to the inability to quantify scenario characteristics, such as the inability to determine whether the wire diameter was still in a stable range during the interruption. This step ensures the continuity of previous and adjusted data by supplementing missing data, so that a complete working condition profile can be obtained when planning actual scenarios, such as the wire diameter change trend before and after the interruption and the fluctuation of pulling parameters. This avoids scenario misjudgment caused by data fragmentation, such as mistakenly judging the stable working condition before the interruption as a fluctuating working condition, thus laying the foundation for the accuracy of parameter settings.

[0070] Among them, the data completion can be achieved by filtering out noise (such as errors introduced by interpolation calculation) in the completed data through the Kalman filter algorithm, ensuring that the deviation between the adjusted data and the actual working conditions is ≤0.003mm. This reflects the innovative idea of ​​the algorithm to cover data defects, which is different from the passive mode of existing technologies that only rely on hardware to collect data. This makes the method more adaptable to industrial environments, such as data interruption scenarios caused by electromagnetic interference in the workshop.

[0071] As one of the optional implementation methods, step S30321 is implemented in the following way: The system monitors the data streams of various sensors (such as laser diameter gauges, tension sensors, and temperature sensors) in real time. If data packets are continuously lost for more than a preset threshold (e.g., 3 sampling cycles, corresponding to 30ms), it is considered a data interruption. Interrupt types are categorized as follows: Complete interruption: All data from a certain sensor is lost; Partial interruption: Data is intermittent or there is severe noise, and the signal-to-noise ratio is lower than the set threshold.

[0072] Based on the continuous historical data sequence before the interruption, the system uses one or more of the following methods to calculate and complete the missing data: Linear interpolation: Suitable for scenarios where data changes gradually and interruption time is short (≤0.1s). Assume the last valid data point before the interruption is D. t-1 The first valid data point after the interruption ends is D. t+n Then the missing data D t To D t+n-1 Calculated via linear interpolation: i = 1, ..., n-1; Trend extrapolation method: Based on the data trend before the interruption (such as the rate of change of wire diameter, the trend of tensile decay), the moving average or exponential smoothing method is used to extrapolate the data of the missing period.

[0073] Model prediction completion: Using a trained LSTM or Kalman filter model, the wire diameter value for the missing period is predicted based on multi-dimensional data (wire diameter, tension, speed, temperature) before the interruption, ensuring that the completed data conforms to the dynamic characteristics of the process.

[0074] It is important to note that the completed data needs to undergo a validity check: Range verification: Whether the data is within the physically feasible range (e.g., wire diameter is not negative, temperature does not exceed the material melting point); Trend consistency verification: Whether the completed data is continuous with the trend of the preceding and following data; Process constraint verification: Whether the constraints of the current drawing process are met (e.g., maximum draw ratio, temperature window).

[0075] If the verification fails, the data segment is marked as "untrustworthy" and a data quality warning is triggered.

[0076] Based on the completed and verified data, the system recalculates the actual deviation, prediction deviation, and standard deviation, and corrects the adjustment parameters. Simultaneously, the system records the interruption event, the completion method, and the adjustment results, storing them in a historical database for future model optimization and fault tolerance strategy improvement.

[0077] If the data interruption exceeds the system's fault tolerance time limit (e.g., 0.5s), the system will automatically switch to a backup detection component or suspend the process to avoid processing risks caused by unreliable data.

[0078] S30322. Plan the actual scenario based on the adjusted data: The purpose of this step S30322 is to build a bridge between the repair data and the actual scenario, so as to avoid the parameters from being out of touch with the working conditions.

[0079] The adjusted data is the quantified raw data, such as the completed wire diameter value and tensile fluctuation data. If it is directly used to set new parameters, it is easy for the data to be mismatched with the scenario. For example, the data shows that the wire diameter is stable, but the actual scenario is that the wire material changes gradually.

[0080] Therefore, this step transforms abstract data into concrete operating condition characteristics by planning actual scenarios. These scenarios specifically include: the current processing stage of the wire (e.g., early / mid / late stages of drawing), material properties (e.g., changes in the ductility of copper wire), equipment operating status (e.g., slight wear of the drawing die), and the intensity of environmental interference (e.g., temperature fluctuations in the workshop). This transformation ensures that the setting of new parameters no longer relies on single data points but is based on a complete scenario profile, guaranteeing parameter adaptability. For example, in the later stages of drawing, when the die is worn, the prediction deviation threshold can be appropriately increased. This step defines the quantitative basis for planning the actual scenario using the adjusted data, making the actual scenario no longer a vague concept but a set of operating conditions that can be verified by data. For example, based on the adjusted data, if the wire diameter fluctuation is ≤0.005mm, the scenario is a stable processing scenario; if the fluctuation is >0.005mm, the scenario is a slightly fluctuating scenario. Different actual scenarios correspond to different parameter optimization logics: for example, in a stable processing scenario, the new prediction deviation can be kept at a small threshold (such as ±0.005mm) to ensure accuracy; in a scenario with slight fluctuations, the new prediction deviation can be appropriately increased (such as ±0.008mm) to avoid frequent warnings.

[0081] This step, by planning actual scenarios, provides a clear decision-making basis for setting new prediction deviations and standard deviations in the future. It upgrades parameter adjustment from blind optimization to scenario-based precise optimization, solves the defect of the one-size-fits-all approach to existing technical parameters, and further improves the accuracy of wire diameter adjustment.

[0082] Furthermore, the process of planning actual scenarios generates adjusted data and mapping relationships between scenario characteristics, such as wire diameter fluctuation of 0.004mm, stable tensile force, and stable processing scenarios. These mapping relationships can be synchronized with the database in subsequent steps, providing a reference for parameter settings for similar scenarios in the future. If the same data characteristics are encountered again, the optimal parameters for the corresponding scenario can be directly called. This accumulation of data, scenarios, and parameters enables the method to have self-learning capabilities, strengthening the advantages of adaptive optimization of the solution.

[0083] S30323. Set new prediction deviation and standard deviation values ​​according to the actual scenario and restart the device: The purpose of this step S30323 is to achieve accurate restart after data interruption, forming a closed loop of data repair, scenario planning, parameter optimization and restart.

[0084] In the case of data interruption, if the old parameters are used directly for restarting, the deviation will be reproduced because the parameters do not match the repaired data and the actual scenario; if new parameters are set blindly, new processing risks will be introduced.

[0085] Therefore, this step ensures the targeted nature of parameter optimization by verifying the logic of the actual scenario and new parameters. For example, if the planned scenario is a stable processing scenario with copper wire φ0.5mm, the new prediction deviation can be set to ±0.005mm, and the standard deviation to 0.01mm; if the scenario is a slightly fluctuating scenario with alloy steel wire φ0.8mm, the new prediction deviation can be set to ±0.008mm, and the standard deviation to 0.015mm. This scenario-based parameter setting ensures the processing accuracy after restarting while avoiding excessive downtime (data repair and parameter setting only require 1-2 seconds), balancing production efficiency and quality. In industrial production, data interruptions and changes in operating conditions often coexist. For example, during a data interruption, the wire material may experience slight fluctuations in characteristics due to temperature changes. The scenario-based parameter setting in this step can simultaneously adapt to both types of changes: capturing operating condition fluctuations through adjusted data, integrating fluctuation characteristics through scenario planning, and ultimately achieving precise control through new parameters. This design enables the method to handle not only simple data interruptions, but also complex scenarios involving data interruptions and fluctuations in operating conditions, thereby improving the industrial adaptability of the technology and overcoming the limitation of existing technologies that can only handle single anomalies.

[0086] S30324. When there is no data interruption, generate new prediction deviation and standard deviation: This step clarifies the core sub-scenario of no data interruption, which means that the processing data (initial / real-time wire diameter, drawing parameters, deviation data, etc.) is complete and without missing data. It forms a binary coverage of data integrity / data interruption with steps S30321~S30323, so that the equipment fault-free scenario in step S3032 can achieve a full-dimensional closed loop and completely fill the logical gap of unclear data status.

[0087] Among them, the core cause of the Level II warning (difference > 0.01mm or continuous deviation) is essentially the mismatch between the old prediction deviation / standard deviation and the actual working conditions, such as the initial parameters not being updated synchronously after the change of wire material or the adjustment of the drawing speed.

[0088] Therefore, the new parameters generated in this step are not randomly adjusted, but are targeted optimizations based on complete data. For example, when processing copper wire (which has good ductility), the original prediction deviation is set to ±0.005mm. After switching to alloy steel wire (which has high hardness and large diameter fluctuations), the new prediction deviation can be adjusted to ±0.008mm, and the standard deviation is optimized from 0.01mm to 0.015mm. This ensures that the parameters are highly compatible with the working conditions, avoids the recurrence of warnings from the root, and is different from the extensive control of existing technologies that use fixed parameters throughout the entire process.

[0089] S30325. Based on the previous data, replan the actual scenario and verify the new prediction deviation and standard deviation: In this step S30325, by anchoring to the real working conditions, parameters are avoided from running idle, such as ineffective optimization that is detached from reality.

[0090] The previous data specifically refers to the complete processing data before the triggering of the Level 2 warning, such as wire diameter variation curves, tensile force fluctuations, wire temperature trends, and target wire diameter specifications. This data serves as a quantitative representation of the actual scenario. This step transforms abstract data into concrete combinations of working condition characteristics by re-planning the actual scenario. For example, if the data shows wire diameter fluctuation ≤ 0.004mm, stable tensile force, and copper wire φ0.5mm, the scenario is defined as a stable processing scenario; if the data shows wire diameter fluctuation 0.006-0.008mm, slight tensile force fluctuation, and alloy steel wire φ0.8mm, the scenario is defined as a slight fluctuation scenario. This transformation ensures that the verification of new parameters is no longer about whether the values ​​are legal, but whether the parameters are suitable for the scenario. This completely avoids ineffective optimization where the parameter values ​​are compliant but the working conditions are mismatched. For example, using parameters from a stable scenario for a fluctuating scenario solves the core pain point of existing technical parameter optimization being disconnected from actual working conditions.

[0091] The verification process in this step is the core action, encompassing two layers of logic: The first layer is scenario adaptability verification: determining whether the new parameters match the actual scenario after planning, such as whether the prediction deviation in a stable scenario is too small, and whether the standard deviation in a fluctuating scenario is sufficiently inclusive; the second layer is numerical reasonableness verification: ensuring that the new parameters are within the limits of physical processing, such as the prediction deviation not being negative, and the standard deviation not exceeding 5% of the target wire diameter. This two-way verification mechanism avoids scenario mismatch, such as using thin wire parameters to adapt to thick wire, and also eliminates numerical anomalies, such as a standard deviation set to 0.1mm leading to control failure. It fills the technical loophole of simply setting new parameters without verification, making parameter optimization more reliable.

[0092] Furthermore, during the scenario planning and parameter verification process, this step generates a complete mapping relationship between working condition data, scenario characteristics, and optimal parameters. For example, for copper wire φ0.5mm, stable processing, prediction deviation ±0.005mm, and standard deviation 0.01mm. These mapping relationships are synchronously stored in the database of subsequent steps, becoming reference templates for parameter settings in similar scenarios. When encountering the same working condition again, the system can directly call the optimal parameters from the database without regenerating and verifying them, thereby improving processing efficiency. At the same time, it provides high-quality training samples for predictive model optimization, strengthening the advantages of the solution's self-learning and self-evolution.

[0093] It should be noted that in industrial production, operators need clear judgment criteria to execute technical procedures. The replanning scenario and verification parameters for this step can be visually presented through the system interface, such as a pop-up window displaying the current scenario: stable processing, copper φ0.5mm, new parameters, prediction deviation ±0.005mm, standard deviation 0.01mm, and verification result: adapted. This allows operators to confirm the visual results without needing to understand complex algorithms, meeting the actual needs of standardization and visualization in industrial production.

[0094] S30326. After verifying that the actual scenario and the new predicted difference and standard deviation are correct, restart the equipment: In this step S30326, the final risk control checkpoint before restarting is established to avoid batch quality loss.

[0095] Among them, the Level 2 warning corresponds to scenarios with high wire diameter control risks (such as difference > 0.01mm). If the system is restarted directly after generating new parameters, the deviation may be reproduced due to omissions in parameter verification (such as incorrect scenario judgment or abnormal parameter values), resulting in a batch of unqualified products (especially high-precision wires such as electronic chip leads, which have zero tolerance for errors).

[0096] Therefore, this step takes the verification as a mandatory prerequisite for restarting, forming a rigid constraint: the equipment can only be started if the system confirms that the scenario and parameters are completely matched, such as the new parameters in the stable scenario not exceeding the accuracy threshold and the parameters in the fluctuating scenario not being too lenient. This eliminates the risk of production due to parameter mismatch from the process perspective, reflecting the core demand of rigorous and controllable industrial production.

[0097] This step, together with steps S30324~S30325, forms a complete closed loop of generating new parameters, verifying parameters, and accurately restarting: step S30324 solves what parameters to generate, step S30325 solves whether the parameters are qualified, and step S30326 solves when to restart. The three steps are progressive and indispensable, ensuring that the processing logic for scenarios with no data interruption and no equipment failure is comprehensive.

[0098] It should be noted that although the restart process after verification adds an extra step, the verification is completed automatically by the system (response time ≤10ms), having almost no impact on production efficiency. Simultaneously, this step ensures the accuracy of processing after restart, making it suitable for scenarios with extremely high precision requirements, such as high-end electronic wires and aerospace conductors. This design, which maintains efficiency while ensuring quality, enhances the industrialization value of the solution, distinguishing it from the dilemma of sacrificing efficiency for quality or vice versa in existing technologies, demonstrating the optimized characteristics of the solution.

[0099] S3033. When the equipment malfunctions, repair the equipment and restart it: The purpose of this step S3033 is to build a fallback management mechanism for hardware failures and avoid the risk of mass production with faults.

[0100] If equipment malfunctions are confirmed during troubleshooting, such as data drift in the laser diameter gauge or mold scratches causing distortion in wire diameter measurement, restarting production directly will lead to a continuous increase in wire diameter deviation. For example, if the diameter gauge malfunction causes the actual wire diameter to exceed the standard but is not detected, it will result in a batch of unqualified products.

[0101] Therefore, this step establishes a safety net against hardware failures by rigidly requiring equipment restart after repair: it clearly defines the principle of repair first, then production, ensuring that the equipment is in normal working condition after restart, such as calibrating the diameter measuring instrument to ±0.002mm accuracy and replacing worn molds, eliminating the hardware root cause of wire diameter deviation from a physical level and enhancing the reliability of the solution.

[0102] This step and step S3032 form a clear branch logic based on whether the equipment is faulty, covering all possible situations after the second-level warning: if there is no fault, optimize the parameters; if there is a fault, repair the equipment, so that the process of warning, investigation, handling and restart is seamless.

[0103] Furthermore, equipment failures are an objective reality in industrial production, such as the expiration of sensor lifespan or normal wear and tear on molds. The equipment repair actions in this step correspond to specific maintenance operations, such as cleaning sensor lenses, adjusting mold spacing accuracy, and replacing tension sensors, providing frontline maintenance personnel with clear guidelines for execution.

[0104] S3034. Save all current data for traceability analysis: The purpose of this step S3034 is to build a full-process quality traceability system to meet the compliance requirements of high-end manufacturing.

[0105] Specifically, all the current data includes: deviation data (actual deviation, predicted deviation, standard deviation) at the time of triggering a Level 2 warning, equipment status data (fault type, detection accuracy), parameter adjustment data (new predicted deviation / standard deviation), and process pause time points. The preservation of this data ensures that the triggering cause, processing, and result of each Level 2 warning are traceable. For example, if a batch of wire fails to meet standards, the traceability data can confirm whether it's due to unrepaired equipment malfunction or a mismatch in new parameter settings. This satisfies the mandatory quality traceability requirements of high-end industries such as automotive and electronics, and provides quantitative evidence for subsequent quality improvements.

[0106] The data saved in this step is one of the core data sources for subsequent steps to acquire historical data and build a database. By accumulating multiple batches of early warning processing data, such as an early warning caused by a prediction deviation of 0.005mm for alloy steel wire with a diameter of φ0.8mm, and the condition returning to normal after adjustment to 0.008mm, a mapping relationship between working conditions, fault types, and optimal parameters can be formed. This provides training samples for subsequent steps to optimize the predicted wire diameter. For example, the LSTM model optimizes the preset logic of prediction deviation through historical fault data, forming a long-term iterative closed loop of early warning, processing, data storage, and model optimization, which strengthens the advantages of the solution's self-adaptation and self-evolution.

[0107] Existing early warning processing technologies are mostly single-loop processes, meaning they end once the current warning is resolved, lacking long-term data accumulation and optimization. This step, through data storage, incorporates single-warning processing into a long-term quality control system, upgrading the solution from fragmented fault handling to systematic quality improvement. This demonstrates a differentiated optimization from existing technologies, not only solving the current early warning problem but also providing data support for avoiding similar problems in the future. This gives the technical solution the advantage of continuous optimization, adapting to the high-quality demands of advanced manufacturing industries.

[0108] In step S4, the wire diameter after adjustment is detected and compensation information is generated. This compensation information is used for wire diameter compensation in the thinning process painting process. The adjustment effect is verified by secondary detection, and the compensation information is linked to subsequent processes to prevent minor deviations generated during wire thinning from being amplified by the painting process. At the same time, the subsequent painting process uses the compensation information to accurately compensate for minor deviations generated during wire thinning, thereby achieving wire diameter consistency control throughout the entire processing chain and demonstrating the adaptability of the adjustment control in the wire thinning process to the overall processing flow.

[0109] Specifically, step S4 also includes the following sub-steps: S401. When there is a data interruption, issue a fault warning signal: The purpose of step S401 is to build an immediate risk response mechanism for the data acquisition process. The core premise for generating compensation information is that the actual deviation data is valid. However, data interruption (such as loss of laser diameter gauge signal or communication failure of tension sensor) will result in compensation information being generated without a basis, directly causing the wire diameter compensation in the painting process to go out of control.

[0110] This step transforms latent data problems into explicit warnings by issuing fault early warning signals, preventing operators from continuing production unknowingly and stopping blind compensation without data support at the source. This step adds a specific early warning for data acquisition faults, forming a two-dimensional system of deviation warnings (quality risks) and data warnings (process risks) with the aforementioned steps. The issuance of the early warning signal clearly identifies the fault attribute of the data interruption, distinguishing it from wire diameter deviation warnings, enabling operators to quickly identify the fault type (whether it's a wire diameter problem or a data acquisition problem), improving fault handling efficiency and enhancing the operability of the solution.

[0111] S402. Automatically switch to backup detection components to maintain process operation or suspend process based on the data type of interruption: In this step S402, data is classified and processed based on its importance to reflect the accuracy of decision-making. That is, the data type of interruption is directly related to the processing risk level: For example, if the exit wire diameter detection data (used as the basis for final compensation) is interrupted, the risk is higher than the interruption of the temperature data of the intermediate drawing unit.

[0112] This step avoids a one-size-fits-all approach by classifying and processing data. If core data (such as initial / real-time wire diameter) is interrupted and switching to a backup component still cannot guarantee accuracy, then the process is paused. If non-core data is interrupted, switching to a backup component can maintain operation. This avoids the generation of batches of defective products and maximizes production continuity, solving the industry pain point of shutdown due to data interruption.

[0113] Among them, the automatic switching of backup components also enhances the intelligence and reliability of the solution. This action upgrades data interruption handling from passive shutdown to active fault tolerance, reflecting the difference in optimization between the solution and existing technologies that rely on only a single detection component, and improving the adaptability to industrial applications.

[0114] Furthermore, by constructing a fault-tolerant and risk-controllable closed loop to connect subsequent processes, maintaining the operation of the process ensures the continuity of pre-painting compensation information generation (through data obtained from backup components), while suspending the process prevents the spread of processing risks without data support. Both decisions lay the foundation for generating accurate compensation information subsequently: during operation, backup component data can be directly used to calculate deviations; when the process is suspended, data after fault repair can complete the deviation calculation, forming a closed loop of data interruption, fault tolerance, data recovery, and compensation information generation, further improving the core process.

[0115] S403. Detect the difference between the wire diameter after adjustment and the target wire diameter value and compare it with the compensation error value. The compensation error value is based on the standard processing error set for the wire. In this step S403, the core basis for deviation quantification and compensation decision is constructed to fill the process gap.

[0116] This step transforms the abstract deviation into a quantifiable value by detecting the difference between the adjusted wire diameter and the target wire diameter. Then, it introduces a compensation error value as a judgment threshold to construct a logical chain of numerical comparison and decision-making basis. This completely fills the gap in the triggering conditions and judgment criteria for the generation of compensation information, upgrading this step from a fuzzy operation to a precise and controllable technical step.

[0117] Specifically, by defining the compensation error value as a standard processing error based on the wire material, the parameter attributes are defined: based on the wire material, it is clear that this threshold is not a fixed value (for example, the processing accuracy requirements of copper wire and alloy steel wire are different, and the compensation error value needs to be set differently), so as to avoid the compatibility problem caused by a one-size-fits-all standard and adapt to the needs of wire materials and specifications.

[0118] In addition, the difference detection in this step is a direct verification of the effect of the adjustment parameters. After the pull adjustment, this step is needed to determine whether the adjustment meets the standard. At the same time, the result of comparing the compensation error value directly determines the subsequent actions (painting compensation or reprocessing), providing accurate data support for the painting process of connecting the fine lines, and ensuring seamless connection of quality control between the thick line pulling and painting processes.

[0119] S404. When the absolute value of the difference between the adjusted wire diameter and the target wire diameter is less than or equal to the compensation error value (e.g., ±0.005mm for copper wire), compensation information is generated for indirect compensation of the coating pressure of the painting mechanism. In this step S404, an efficient compensation path is designed for minor deviations to balance accuracy and efficiency. For example, the system generates compensation information containing the deviation value and direction, and calculates the required coating pressure value based on a preset wire diameter-pressure mapping model (such as ΔP=k·ΔD, where k is the pressure compensation coefficient, with a value range of 0.02~0.05 MPa / mm, and the specific value is determined experimentally according to the paint characteristics and wire specifications). Then, the pressure adjustment command is sent to the pressure controller of the painting mechanism. By precisely increasing or decreasing the coating pressure, the paint layer thickness is changed (for example, when the wire diameter is 0.003mm smaller, the coating pressure is increased by 0.12MPa to thicken the paint layer by about 0.003mm), thereby indirectly compensating for the small wire diameter deviation left by the drawing process, ensuring that the final product size meets the standard, and verifying the compensation effect through an online diameter measuring instrument after compensation, forming a closed-loop control of "detection-compensation-verification".

[0120] Among them, the absolute value of the difference between the adjusted wire diameter and the target wire diameter is less than or equal to the compensation error value, which defines the acceptable small deviation scenario (such as deviation within ±0.005mm). Such deviation will lead to material waste and reduced efficiency if the wire is redrawn.

[0121] Therefore, this step indirectly compensates for the coating pressure of the coating mechanism, providing a targeted solution: from the perspective of process principle, the fine adjustment of coating pressure can change the thickness of the paint layer, indirectly offsetting the slight deviation of the wire diameter itself (if the wire diameter is slightly smaller, the pressure is increased to thicken the paint layer to ensure that the final product size meets the standard; if the wire diameter is slightly smaller, the opposite operation is performed, that is, the paint layer is thinned), which avoids over-processing and ensures the dimensional consistency of the finished product after coating, thereby balancing processing accuracy and production efficiency.

[0122] This step further clarifies the specific relationship between compensation information and coating pressure adjustment, giving frontline operators a clear basis for execution. For example, if the line diameter marked by the compensation information is 0.003mm too small, the operator can increase the coating pressure by 0.2MPa accordingly.

[0123] Furthermore, in existing technologies, the wire drawing and painting processes are mostly managed independently, with the painting process being passively executed and unable to participate in wire diameter error correction. This step, through the coordinated design of post-drawing deviation and painting compensation, upgrades the painting process from a simple coating to a quality compensation step. This overcomes the shortcomings of existing technologies where each process operates independently, and enhances the convenience of the solution's end-to-end control.

[0124] S405. When the absolute value of the difference between the adjusted wire diameter and the target wire diameter is greater than the compensation error value, compensation information is generated, the adjustment parameters are updated, and the wire is reprocessed. In this step S405, a risk mitigation mechanism for large deviations is constructed to ensure product quality.

[0125] Among them, a difference greater than the compensation error value means that the deviation has exceeded the compensation capability of the painting process. For example, if the wire diameter is 0.02mm smaller, it is far beyond the adjustment range of the paint layer thickness. If it flows into the painting process, it will lead to the scrapping of the finished product.

[0126] Therefore, this step establishes rigid quality control by updating and reprocessing the parameters: updating the parameters avoids using old parameters that cause deviations. For example, if the drawing speed is too fast and the wire diameter is too small, the speed needs to be reduced and the die spacing adjusted. Reprocessing directly removes unqualified semi-finished products, eliminating batch quality problems from the source and meeting the quality assurance logic for emergency handling in wire diameter adjustment scenarios.

[0127] The design process of generating compensation information before reprocessing forms a closed loop of deviation tracing and parameter optimization. Generating compensation information is not a redundant action: the compensation information needs to record the specific value of the deviation (such as the wire diameter being 0.015mm larger), the process node where the deviation occurred (such as after drawing adjustment), and the current process parameters, providing a direct basis for updating and adjusting parameters. For example, based on the deviation value, the die spacing needs to be reduced by 0.004mm, avoiding blind adjustment of parameters during reprocessing.

[0128] At the same time, this information can be synchronously stored in the historical database mentioned in subsequent steps, accumulating data for the optimization of subsequent prediction models and forming a long-term improvement closed loop of deviation detection, information recording, parameter optimization, and model iteration.

[0129] S406. Acquire historical data and establish a database: In this step S406, a full-process data accumulation mechanism is built to solve the pain point of fragmented data in existing technologies.

[0130] Historical data is not a single-dimensional dataset, but rather a comprehensive collection of core information covering the entire wire diameter adjustment process. Specifically, it includes: initial / real-time / predicted wire diameter data; adjustment parameters (drawing force, speed, temperature); deviation differences; warning types and handling results; data interruption records; backup component switching status; compensation error values; and reprocessing parameters. In existing technologies, this type of data is often lost after the process ends. This step establishes a database to systematically store this scattered data, forming a data source covering the entire chain of detection, adjustment, warning, and compensation. This provides a quantitative basis for subsequent optimization and fills the technological gap of lacking accumulated processing data.

[0131] The establishment of the database enables traceability of the input, process, and output of each wire diameter adjustment (e.g., the reason for the excessive wire diameter deviation in a batch can be traced back to the predicted deviation settings and adjustment parameter values ​​at that time through the database), meeting the mandatory requirements for quality traceability in industrial production. The historical data stored in the database provides practical evidence for multiple steps: for example, saving data for traceability analysis requires the database as a storage medium; planning actual scenarios based on previous data essentially involves calling historical operating condition data from the database; updating adjustment parameters can refer to the optimal adjustment scheme for similar deviations in the database. This step, through data accumulation, transforms disparate steps into a data-sharing, mutually supportive organic whole, strengthening the rigor of the entire implementation process.

[0132] S407. Optimize the predicted wire diameter based on the database: In this step S407, the prediction model is dynamically iterated to overcome the limitations of static prediction in existing technologies. The prediction accuracy of ordinary models depends on preset parameters and is difficult to adapt to the processing differences of wires of different materials (copper, aluminum, alloy steel) and specifications (φ1-10mm).

[0133] This step involves building a dynamic upgrade mechanism through database optimization: historical data from the database (such as the diameter variation patterns of a certain type of wire and prediction deviations under different working conditions) is used as training samples for the model. For example, the weight parameters of the LSTM model are iteratively updated using the gradient descent algorithm, further reducing the error in predicting wire diameter from the initial ±0.003mm to ±0.001mm, thus addressing the industry pain point that fixed models cannot adapt to multiple working conditions. The iteration method used in this example is a conventional technique, and the corresponding formulas can be applied.

[0134] Existing wire diameter prediction technologies often rely on fixed algorithms, leading to decreased accuracy as processing conditions change (e.g., prediction errors increase significantly after changing wire material). This step's optimized wire diameter prediction enables the method to evolve: when processing new wire specifications, the database quickly accumulates initial data for that type of wire, and the model learns to rapidly improve prediction accuracy; when equipment aging causes changes in processing characteristics (e.g., mold wear makes the wire diameter tend to be smaller), the model can capture this trend through historical data and correct the prediction results in advance. This data-driven adaptive optimization gives the solution a distinct advantage over existing technologies.

[0135] In addition, this step reduces the burden on the front-end adjustment process, forming a closed loop of prediction, optimization, and adjustment. The optimized prediction of wire diameter is more accurate, which means that the intervention of prediction deviation is more targeted (such as accurately predicting that the wire diameter is too large by 0.006mm, rather than fuzzy prediction of being too large). It can reduce the variation of adjustment parameters (such as the die spacing only needs to be reduced by 0.002mm, rather than the initial 0.003-0.005mm). At the same time, accurate prediction can reduce the probability of triggering the second-level warning, reduce the deviation misjudgment caused by inaccurate prediction, and make the entire adjustment process more efficient and stable.

[0136] It should be noted that high-precision electronic wires (such as wires used in chip lead frames) have extremely high requirements for wire diameter prediction accuracy (error ≤ ±0.002mm). The optimization mechanism in this step can dynamically approach this requirement in terms of prediction accuracy, making it suitable for high-end fields. At the same time, the correspondence between the working conditions and the optimal parameters accumulated in the database can form standardized processing schemes (such as the optimal prediction deviation setting and adjustment parameter range for φ0.5mm copper wire), reducing the debugging costs for new users.

[0137] Please refer to Figure 2 This application provides a metal wire diameter adjustment system, which corresponds to the method in the above embodiments. Specifically, the system includes: The first acquisition module 201 is used to acquire the initial wire diameter and the real-time wire diameter of the wire. The first control module 202 is used to generate a predicted wire diameter within the next 0.5 seconds based on the initial wire diameter and the real-time wire diameter; The second control module 203 is used to calculate the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjust the parameters, including drawing force, drawing speed and wire temperature. The third control module 204 is used to detect the wire diameter after adjustment and generate compensation information. The compensation information is used for wire diameter compensation in the fine wire painting process.

[0138] As can be seen from the above, the metal wire diameter adjustment method and system provided in this application have the following advantages: ① Possesses forward-looking and precise control capabilities: By predicting wire diameter in 0.5s and intervening in advance to prevent deviations, it breaks through the limitations of existing post-inspection and correction, controls wire diameter deviations at the nascent stage, and reduces wire diameter consistency error to within ±0.003mm, significantly improving processing accuracy; ② Construct a full-scenario fault tolerance and stability assurance system: hierarchical early warning, dual-dimensional fault investigation of equipment and parameters, and fault-tolerant repair of data interruption form a multi-layer protection to avoid blind shutdown or production with faults, improve production continuity by more than 30%, and reduce the risk of batch non-conforming products; ③ Achieve dynamic adaptive optimization of parameters: dynamically adjust the prediction deviation / standard deviation based on the wire material and working conditions, and iteratively optimize the prediction model by combining historical databases to adapt to the processing of multiple materials and different specifications of wires such as copper and alloy steel, without the need for repeated manual parameter adjustments; ④ Form a collaborative control system for the entire drawing and painting process: Through indirect compensation of coating pressure, the drawing and painting processes are linked. Minor deviations do not require re-drawing, while larger deviations are automatically reworked. This ensures the stability of finished product dimensions, reduces material waste, and improves the product qualification rate by 15%-20%. ⑤ It has the ability to trace quality and continuously evolve: the whole process data storage traceability and database-driven model optimization not only meet the quality traceability requirements of high-end manufacturing, but also enable the system's prediction accuracy to continuously improve with the accumulation of production, and adapt to the needs of high-end scenarios such as high-precision electronic wires and aerospace wires.

[0139] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for adjusting the diameter of a metal wire, characterized in that, The method includes the following steps: Obtain the initial and real-time wire diameter of the wire; The predicted wire diameter for the next 0.5 seconds is generated based on the initial wire diameter and the real-time wire diameter. Calculate the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjust the parameters, including drawing force, drawing speed, and wire temperature; The wire diameter of the adjusted wire is detected and compensation information is generated. This compensation information is used to compensate for the wire diameter during the fine wire painting process.

2. The method for adjusting the diameter of metal wire according to claim 1, characterized in that, The step of calculating the deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjusting the parameters further includes: Calculate the difference between the actual deviation and the predicted deviation, and determine the magnitude of the difference and the standard deviation. The predicted deviation is the standard deviation value set based on the prediction, and the standard deviation is the standard error value set according to the target wire diameter. When the difference is ≤0.01mm, a level one warning is triggered and the parameters are automatically adjusted; When the difference is greater than 0.01 mm or occurs three times consecutively with a value ≤ 0.01 mm, a level two warning is triggered and the process is suspended.

3. The method for adjusting the diameter of metal wire according to claim 2, characterized in that, The steps for triggering a level-two warning and suspending the process when the difference is greater than 0.01 mm or when there are three consecutive occurrences of ≤0.01 mm, further include: Detect equipment malfunctions and verify previous prediction deviations and standard deviations; When the equipment is fault-free, set new prediction deviation and standard deviation values ​​according to the actual scenario and restart the equipment; When the equipment malfunctions, repair the equipment and then restart it; Save all current data for traceability and analysis.

4. The method for adjusting the diameter of metal wire according to claim 3, characterized in that, When the equipment is fault-free, the steps of setting new prediction bias and standard deviation values ​​according to the actual scenario and restarting the equipment also include: When data interruption occurs, calculate the missing data based on previous data and make adjustments accordingly; Plan the actual scenario based on the adjusted data; Set new prediction bias and standard deviation based on the actual scenario and restart the device.

5. The method for adjusting the diameter of metal wire according to claim 4, characterized in that, When the equipment is fault-free, the steps of setting new prediction bias and standard deviation values ​​according to the actual scenario and restarting the equipment also include: When there is no data interruption, new prediction bias and standard deviation values ​​are generated; Based on the previous data, the actual scenario was re-planned and the new prediction bias and standard deviation were verified. After verifying that the actual scenario and the new prediction deviation and standard deviation are correct, the device is restarted.

6. The method for adjusting the diameter of metal wire according to claim 5, characterized in that, The steps of detecting and adjusting the wire diameter and generating compensation information also include: A fault warning signal is issued when a data interruption occurs; The system automatically switches to a backup detection component to maintain process operation or suspends the process based on the type of data interrupted.

7. The method for adjusting the diameter of metal wire according to claim 1, characterized in that, After the step of generating the predicted wire diameter within the next 0.5 seconds based on the initial wire diameter and the real-time wire diameter, and before the step of calculating the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjusting the parameters, the following steps are also included: If the predicted wire diameter is too large within the next 0.5 seconds, the die spacing of the drawing mechanism is reduced and the drawing speed is increased simultaneously until the predicted wire diameter returns to the target state. If the predicted wire diameter is too small within the next 0.5 seconds, the die spacing of the drawing mechanism is increased and the drawing speed is reduced simultaneously until the predicted wire diameter returns to the target state.

8. The method for adjusting the diameter of metal wire according to claim 6, characterized in that, The steps of detecting and adjusting the wire diameter and generating compensation information also include: The difference between the wire diameter after adjustment and the target wire diameter is detected and compared with the compensation error value, which is based on the standard processing error set for the wire. When the absolute value of the difference between the adjusted wire diameter and the target wire diameter is less than or equal to the compensation error value, compensation information is generated to indirectly compensate the coating pressure of the painting mechanism. When the absolute value of the difference between the adjusted wire diameter and the target wire diameter is greater than the compensation error value, compensation information is generated, the adjustment parameters are updated, and the wire is reprocessed.

9. The method for adjusting the diameter of metal wire according to claim 1, characterized in that, After the steps of detecting and adjusting the wire diameter and generating compensation information, the following are also included: Acquire historical data and build a database; Optimize and predict wire diameter based on database.

10. A metal wire diameter adjustment system, characterized in that, The system includes: The first acquisition module (201) is used to acquire the initial wire diameter and real-time wire diameter of the wire. The first control module (202) is used to generate a predicted wire diameter within the next 0.5 seconds based on the initial wire diameter and the real-time wire diameter; The second control module (203) is used to calculate the actual deviation between the predicted wire diameter and the real-time wire diameter at the same time point and adjust the parameters, including the drawing force, drawing speed and wire temperature. The third control module (204) is used to detect the wire diameter after adjustment and generate compensation information, which is used for wire diameter compensation in the fine wire painting process.

Images