Humanoid robot use corona-resistant enamel copper flat wire curing degree control method and system

Abstract

This invention relates to the field of enameled wire curing process control technology, and discloses a method and system for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots. The method includes dynamically identifying uneven curing regions and lag stages by real-time acquisition of multi-source data such as film thickness, furnace temperature field, and solvent evaporation rate. Combined with a process knowledge base, a dynamic temperature compensation and time-enhanced control scheme is generated, and the heater power and transmission speed are collaboratively redistributed and executed accordingly. Feedback data is simultaneously acquired, compared with the expected trajectory to generate a deviation map, and the control parameters are iteratively corrected until the deviation converges, outputting a stable process parameter package. This method achieves online diagnosis and adaptive closed-loop optimization of the curing process, improving the uniformity and consistency of film curing, and reducing quality fluctuations and human intervention.

Description

Technical Field

[0001] This invention relates to the field of enameled wire curing process control technology, specifically to a method and system for controlling the curing degree of corona-resistant enameled copper flat wires used in humanoid robots. Background Technology

[0002] In the manufacturing of corona-resistant enameled copper flat wires for humanoid robots, the curing process directly affects the insulation reliability of the product. Existing technologies typically rely on preset curing temperature and time curves, temperature monitoring at a few fixed points within the curing oven, and a constant transmission speed. The setting and adjustment of process parameters are largely based on historical experience or sampling inspection of the final product, representing a static, open-loop control mode.

[0003] The drawback of this model lies in the lack of multi-dimensional real-time perception of the curing process. The actual temperature field distribution within the curing oven, the spatial variation of the paint film thickness, and the dynamic process of solvent evaporation rate all jointly determine the uniformity of curing, but current technologies cannot simultaneously acquire and correlate these key data. This results in defects such as localized under-curing or over-curing occurring during production not being detected in a timely manner, and can only be identified through sampling inspections afterward. Furthermore, fixed process parameters are difficult to adapt to real-time disturbances such as fluctuations in paint properties and changes in equipment status, making control measures lagging and passive, and failing to guarantee the consistency and stability of product performance.

[0004] A technical solution is needed to identify the spatiotemporal characteristics of curing defects online and in real time. This solution needs to be based on multi-source real-time monitoring data to dynamically diagnose uneven regions and lag stages in the curing process. Furthermore, a method is needed to dynamically coordinate and control actuators such as heating and transmission based on real-time diagnostic results, in order to achieve adaptive optimization of process parameters and thus output stable and reliable curing process conditions. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for controlling the curing degree of corona-resistant enameled copper flat wires for humanoid robots, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a method for controlling the curing degree of corona-resistant enameled copper flat wires for humanoid robots, the method comprising:

[0007] Real-time monitoring data of corona-resistant enameled copper flat wire during the production process is obtained, including coating thickness distribution, temperature field inside the curing oven, and solvent evaporation rate.

[0008] By analyzing the real-time monitoring data, the uneven development of curing degree and the lagging stage in the current production batch are identified, and a preliminary set of characteristics of curing defects is obtained.

[0009] By integrating the preliminary feature set of the curing defects with the preset process knowledge base, a dynamic temperature compensation path for the current uneven region and a time enhancement strategy for the lag stage are constructed to form a preliminary control scheme.

[0010] Based on the preliminary control scheme, the power output curve of the heater in the curing oven and the feed rate of the transmission chain are coordinated and redistributed to generate a detailed set of execution instructions containing time and space dimensions.

[0011] The detailed execution instruction set is sent to the production equipment for execution, and the process parameters are collected synchronously during the execution cycle. The feedback process parameters are compared with the expected parameter trajectory to obtain the execution deviation map.

[0012] Based on the execution deviation map, the key parameters in the dynamic temperature compensation path and time enhancement strategy are iteratively corrected until the execution deviation map converges within the allowable threshold range, and a stable curing process parameter package is output.

[0013] Preferably, the step of acquiring the real-time monitoring data specifically includes:

[0014] At the entrance stage of the curing oven, a thickness detection unit is deployed to collect the coating thickness of the entire line by scanning and record it as coating thickness distribution data.

[0015] In multiple temperature zones of the curing oven, thermocouple arrays are uniformly set up to continuously monitor the temperature values ​​at various locations inside the oven. The temperature values ​​are then organized according to spatial coordinates and time series to form temperature field data inside the curing oven.

[0016] A gas composition sensor is installed at the exhaust vent of the curing oven to measure the evaporation rate of a specific solvent and obtain solvent evaporation rate data.

[0017] The real-time monitoring data is formed by aligning the paint film thickness distribution data, the temperature field data inside the curing oven, and the solvent evaporation rate data with timestamps and fusing the data.

[0018] Preferably, the step of obtaining the preliminary feature set of the curing defects specifically includes:

[0019] The coating thickness distribution data is divided into regions, and the average thickness and variance of each region are calculated. Regions with variances exceeding a preset thickness fluctuation limit are marked as thickness unevenness regions.

[0020] In the temperature field data inside the curing oven, locate the temperature zone corresponding to the spatial coordinates of the thickness uneven area, analyze the temperature change curve of the temperature zone over time, identify the time period when the slope of the temperature curve is lower than the preset heating rate threshold, and mark the time period as the temperature lag stage.

[0021] By correlating the solvent evaporation rate data with the time point of the temperature lag phase, the solvent evaporation rate at the time point is calculated. If the solvent evaporation rate is lower than a preset evaporation rate threshold, it is confirmed that the temperature lag phase is accompanied by solvent evaporation lag.

[0022] By summarizing all the information on the thickness unevenness zone, the temperature hysteresis stage, and the accompanying solvent evaporation delay, a preliminary feature set of the curing defects is obtained.

[0023] Preferably, the steps for forming the preliminary control plan are as follows:

[0024] The process knowledge base is queried, which stores the temperature compensation and time adjustment amounts corresponding to different thickness regions in historical successful processes.

[0025] For each thickness unevenness region in the preliminary feature set of the curing defects, a suggested initial temperature compensation amount is matched from the process knowledge base;

[0026] For each of the temperature lag stages, a suggested initial time extension is matched from the process knowledge base;

[0027] Based on the initial temperature compensation amount and the initial time extension amount, and combined with the heat transfer model of the curing oven and the wire motion model, the heater power change curve and the transmission chain speed change curve that need to be adjusted for implementing compensation are calculated. The heater power change curve and the transmission chain speed change curve together constitute the dynamic temperature compensation path and time enhancement strategy, and are packaged into the preliminary control scheme.

[0028] Preferably, the detailed execution instruction set generation step specifically includes:

[0029] The preliminary control scheme is analyzed, and all heater numbers that need to be adjusted and their corresponding power change curves are extracted to form a subset of heater power control instructions.

[0030] The preliminary control scheme is analyzed to extract the transmission chain segment numbers that need speed adjustment and their corresponding speed change curves, forming a subset of transmission chain speed control instructions.

[0031] Establish a unified timeline and assign precise effective time points and durations to each instruction in the subset of heater power control instructions and the subset of transmission chain speed control instructions to ensure temporal and spatial coordination of actions;

[0032] All timestamped instructions are sorted and encapsulated to generate the detailed execution instruction set.

[0033] Preferably, the step of obtaining the execution deviation map specifically includes:

[0034] During the execution of the detailed execution instruction set, the actual power output value of the heater and the actual feed rate value of the transmission chain are collected in real time as actual execution data;

[0035] On the same time axis, the expected power output value and expected feed rate value at the corresponding time point are read from the detailed execution instruction set as the expected parameter trajectory;

[0036] Calculate the difference between the actual execution data and the expected parameter trajectory at each sampling point to obtain the power deviation sequence and the rate deviation sequence;

[0037] Using time as the horizontal axis and spatial location as the vertical axis, the power deviation sequence and the rate deviation sequence are mapped onto a two-dimensional coordinate system to form a visual deviation map, thus obtaining the execution deviation map.

[0038] Preferably, the iterative correction process specifically includes:

[0039] Analyze the execution deviation map to identify regions and time periods where the deviation amplitude is consistently positive or negative;

[0040] For regions and time periods where the deviation remains positive, it is determined that the compensation or enhancement amount in the preliminary control scheme is excessive, and the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path is reduced accordingly, or the extension amount of the corresponding stage in the time enhancement strategy is reduced.

[0041] For regions and periods where the deviation is consistently negative, it is determined that the compensation or enhancement amount in the preliminary control scheme is insufficient, and the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path is increased accordingly, or the extension amount of the corresponding stage in the time enhancement strategy is increased.

[0042] Using the corrected temperature compensation and time extension, a new subset of heater power control instructions and a new subset of drive train speed control instructions are recalculated and generated to form a new detailed execution instruction set.

[0043] The new detailed execution instruction set is executed, and data is collected again to generate a new execution deviation map. This process is repeated until all deviation values ​​in the new execution deviation map are within the preset allowable threshold range.

[0044] Preferably, the output steps of the stable curing process parameter package are as follows:

[0045] When the execution deviation map meets the convergence condition, record all parameters of the final version of the dynamic temperature compensation path and time enhancement strategy used at this time.

[0046] The power change curves of all heaters in the final version of the dynamic temperature compensation path are parameterized, and the power values ​​and time points of key control points are extracted.

[0047] The speed change curves of all transmission chains in the final version of the time enhancement strategy are parameterized, and the speed values ​​and time points of key control points are extracted.

[0048] The power change curve, rate change curve and their corresponding curing oven temperature zone number and transmission chain segment number are bound and packaged to form the stable curing process parameter package.

[0049] Preferably, the method further includes:

[0050] The stable curing process parameter package is associated with and stored with the corresponding real-time monitoring data, and updated to the process knowledge base as a new successful process case for the construction of control schemes for subsequent production batches.

[0051] Preferably, the present invention also includes a curing degree control system for corona-resistant enameled copper flat wire for humanoid robots. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the curing degree control method for corona-resistant enameled copper flat wire for humanoid robots as described above.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] This method collects real-time data on paint film thickness distribution, temperature field inside the curing oven, and solvent evaporation rate, and analyzes their coupling relationship to dynamically identify spatially uneven areas and temporally lagging stages of curing degree. Unlike conventional techniques that rely solely on temperature monitoring or post-production sampling, this method synchronously correlates paint film morphology, thermal environment, and chemical reaction processes, enabling precise online location of curing defects. Defect detection is shifted from post-production to the curing process, instantly reflecting over-cured or under-cured areas, improving the real-time nature and accuracy of defect detection, and providing direct evidence for immediate intervention.

[0054] Based on defect identification results, a dynamic temperature compensation path and time enhancement strategy are constructed to collaboratively redistribute the heater power output curve and drive train feed rate, generating a detailed set of execution instructions in the spatiotemporal dimensions. After execution, process parameters are synchronously collected and compared with the expected trajectory to generate an execution deviation map, and key parameters in the temperature compensation path and time enhancement strategy are iteratively corrected. Unlike conventional single-variable control or open-loop regulation, this method achieves coordinated action of multiple actuators in time and space, and continuously optimizes through closed-loop feedback. Process parameters can adapt to real-time production conditions, improving curing uniformity and batch-to-batch consistency, and reducing manual intervention and downtime. Attached Figure Description

[0055] Figure 1 This is a schematic diagram illustrating the working principle of the curing degree control method for corona-resistant enameled copper flat wires used in humanoid robots according to the present invention.

[0056] Figure 2 A flowchart for real-time monitoring data acquisition;

[0057] Figure 3 A flowchart generated for detailed execution of the instruction set;

[0058] Figure 4 Box plots showing heater power deviation under different sampling periods;

[0059] Figure 5 This is a thermal map showing the spatiotemporal distribution of power deviation in enameled copper flat wire. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0061] Please see Figure 1This invention provides a method for controlling the curing degree of corona-resistant enameled copper flat wires for humanoid robots. The method includes: deploying a sensor network on the production line to acquire real-time monitoring data including coating thickness distribution, temperature field inside the curing oven, and solvent evaporation rate; based on this data, using algorithmic analysis to identify regions with uneven curing degree development and lagging stages in the current production batch, thereby extracting a preliminary feature set of curing defects. This feature set is integrated with a preset process knowledge base, which stores experience data from historically successful processes. Through matching and calculation, a dynamic temperature compensation path is constructed for the identified uneven regions, and a time enhancement strategy is developed for the lagging stages, forming a preliminary control scheme. The preliminary control scheme is converted into executable equipment instructions, namely, the collaborative design and reallocation of the power output curve of a specific heater in the curing oven and the feed rate of a specific section of the drive train, generating a detailed set of execution instructions with precise targeting in both time and space dimensions. The instruction set is issued to the production equipment for execution, and the actual process parameters are collected and compared with the expected parameter trajectory specified in the instruction set during execution. This generates an execution deviation map reflecting the difference between the actual control effect and the expected target. Based on this deviation map, key parameters in the dynamic temperature compensation path and time enhancement strategy are iteratively corrected. Through multiple "execution-feedback-correction" cycles, the execution deviation is gradually reduced and eventually converges within the allowable threshold range. At this point, a set of verified and stable solidified process parameters is output to guide stable production under the same or similar conditions in the future.

[0062] Example 1: See Figure 2At the entrance stage where the enameled copper flat wire enters the curing oven, a thickness detection unit based on laser or ultrasonic principles is deployed. This unit continuously collects coating thickness data along the length of the copper flat wire using a scanning method, recording it as coating thickness distribution data. Within multiple temperature zones inside the curing oven, thermocouple arrays are uniformly arranged. These thermocouples continuously monitor temperature values ​​at different spatial locations within the oven. The collected temperature values ​​are organized according to their spatial coordinates and time series, forming internal temperature field data reflecting the spatiotemporal changes in temperature within the curing oven. At the exhaust vent of the curing oven, a gas composition sensor is placed to measure the concentration changes of specific solvents related to coating curing, thereby calculating the solvent evaporation rate data. After acquiring the above data, the coating thickness distribution data, internal temperature field data, and solvent evaporation rate data are aligned and fused in timestamps using data synchronization technology to form complete real-time monitoring data. The coating thickness distribution data is divided into regions, and the average thickness and variance of each region are calculated. Regions with variances exceeding a preset thickness fluctuation limit are marked as thickness unevenness areas. In the temperature field data inside the curing oven, the temperature zone corresponding to the spatial coordinates of the thickness unevenness area is located. The temperature change curve of this zone over time is analyzed, and the period when the slope of the temperature curve is lower than the preset heating rate threshold is identified and marked as the temperature lag stage. The solvent evaporation rate data is correlated with the time point of the temperature lag stage, and the solvent evaporation rate at that time point is calculated. If the rate is lower than the preset evaporation rate threshold, it is confirmed that the temperature lag stage is accompanied by a solvent evaporation delay. All identified thickness unevenness areas, temperature lag stages, and associated solvent evaporation delay information are summarized to obtain a preliminary feature set of curing defects.

[0063] In practical implementation, at the entrance stage where the enameled copper flat wire enters the curing oven, a thickness detection unit based on the laser triangulation principle is deployed. This unit performs a lateral scan of the moving copper flat wire at a fixed frequency, collecting and recording the entire copper film thickness data as film thickness distribution data. This distribution data is a sequence containing position coordinates and thickness values. Within the ten independent temperature zones of the curing oven, each zone is uniformly equipped with a thermocouple array consisting of five thermocouples. These thermocouples continuously monitor the temperature values ​​at various locations within the oven. The control system organizes the temperature values ​​according to the spatial coordinates of the thermocouples and the acquisition time sequence, forming internal temperature field data for the curing oven. This internal temperature field data is a three-dimensional dataset. At the main exhaust vent of the curing oven, a Fourier transform infrared spectroscopy gas composition sensor is deployed. This sensor continuously measures the concentration changes of specific solvent components in the exhaust gas, calculating the solvent evaporation rate data by analyzing the concentration changes and exhaust flow rate. The data acquisition system timestamps the film thickness distribution data, the internal temperature field data of the curing oven, and the solvent evaporation rate data using a unified time base and integrates them into the same database table, forming complete real-time monitoring data.

[0064] In some embodiments, the step of dividing the coating thickness distribution data into regions involves dividing the entire length of the copper flat wire into intervals of one meter each, and calculating the average thickness and variance of each one-meter interval. The formula for calculating the thickness variance is:

[0065] ;

[0066] in: This represents the thickness variance of a partitioned region. This indicates the number of sampling points in the area. This represents the thickness value at a single sampling point. This represents the average thickness of the region. The calculated thickness variance is compared with a preset thickness fluctuation limit, which is set at 0.05 square millimeters. Regions with a thickness variance exceeding 0.05 square millimeters are marked as thickness unevenness areas.

[0067] It is understandable that after locating the temperature zone corresponding to the spatial coordinates of the thickness unevenness area in the temperature field data inside the curing oven, the temperature change curve of the corresponding temperature zone over time is analyzed, and the instantaneous slope of the temperature curve at each time point is calculated. Periods where the instantaneous slope of the temperature curve is continuously lower than a preset heating rate threshold (set to 5 degrees Celsius per second) are identified and marked as temperature lag stages. The solvent evaporation rate data is correlated with the start and end times of the temperature lag stage, and the average solvent evaporation rate within that period is calculated. If the calculated average solvent evaporation rate is lower than a preset evaporation rate threshold (set to 0.2 grams per second), then this temperature lag stage is confirmed to be accompanied by solvent evaporation lag. The data processor summarizes the geographical location information of all marked thickness unevenness areas, the time range information of all identified temperature lag stages, and the determination information of accompanying solvent evaporation lag, outputting a structured dataset. This dataset is the preliminary feature set of curing defects.

[0068] Optionally, the scanning direction of the thickness detection unit is perpendicular to the movement direction of the copper flat wire to ensure the acquisition of a complete transverse thickness profile. The thermocouple array is arranged within the temperature zone according to the principle of spatial uniformity, with five thermocouples located at the four corners and the center of the temperature zone, respectively. The sampling frequency of the gas composition sensor is synchronized with the acquisition frequency of the temperature and thickness data, both triggered by a central clock source.

[0069] In some embodiments, the data fusion process employs a timestamp-based interpolation alignment method to unify the film thickness distribution data, curing oven internal temperature field data, and solvent evaporation rate data from different sampling frequencies onto the same time axis. The labeling information for thickness unevenness zones includes the interval start position, interval end position, and the calculated thickness variance value. The labeling information for temperature lag stages includes the temperature zone number, stage start time, stage end time, and the average heating rate during that period. The preliminary feature set of curing defects is stored in list form, where each record corresponds to an identified defect feature and all its associated attributes.

[0070] Example 2: A pre-built process knowledge base is queried. This knowledge base stores effective temperature compensation data for different film thickness regions in historical successful process cases, as well as time adjustment data for different curing lag stages. For each thickness imbalance region in the preliminary feature set of curing defects, a suggested initial temperature compensation amount matching the thickness characteristics of that region is matched from the process knowledge base. For each temperature lag stage, a suggested initial time extension amount matching the lag characteristics of that stage is matched from the process knowledge base. Based on the matched initial temperature compensation amount and initial time extension amount, combined with the heat transfer model inside the curing oven and the motion model of the wire inside the oven, simulation calculations are performed to calculate the power change curves of each heater and the speed change curves of each segment of the transmission chain that require specific adjustments to implement these compensations and enhancements. These calculated power change curves and speed change curves together constitute the dynamic temperature compensation path and time enhancement strategy for the current production batch, and are packaged into a preliminary control scheme.

[0071] In practical implementation, after receiving the preliminary feature set of curing defects generated by the steps of Example 1, the control system automatically queries the process knowledge base stored in the database. The process knowledge base stores historical successful process cases in a relational data table structure. Each case record includes the characteristics of the paint film thickness region in the historical production batch, the corresponding temperature compensation amount, the characteristics of the curing lag stage, and the corresponding time adjustment amount. For each thickness uneven area identified in the preliminary feature set of curing defects, the system matches historical cases from the process knowledge base. The matching logic is to find the record in the historical cases where the average paint film thickness is closest to the average thickness of the current thickness uneven area, and the historical thickness variance is also within a similar range. The system then extracts the suggested initial temperature compensation amount from the matched historical successful process case record. For each temperature lag stage identified in the preliminary feature set of curing defects, the system similarly matches historical cases from the process knowledge base. The matching logic is to find the record in the historical cases where the duration and average heating rate of the temperature lag stage are closest to the characteristics of the current temperature lag stage. The system then extracts the suggested initial time extension amount from the matched historical successful process case record.

[0072] In some embodiments, the matching process for initial temperature compensation values ​​stored in the process knowledge base can be represented by a relational formula. When matching for a thickness imbalance region, the system calculates the average thickness of the current thickness imbalance region. The average thickness compared to a historical case record in the process knowledge base absolute difference And calculate the thickness variance of the current thickness uneven region. thickness variance recorded in this historical case absolute difference The system sets a threshold for the difference in average thickness. With a thickness variance difference threshold ,when and When the historical case record is determined to match the characteristics of the current thickness imbalance region, the stored information in the historical case record is extracted. As a suggested initial temperature compensation amount, where: This indicates the initial temperature compensation amount. This represents the average thickness of the current uneven thickness region. This represents the average thickness of historical case records. This represents the thickness variance of the current uneven thickness region. This represents the thickness variance of the historical case record. This represents the threshold value for the difference in average thickness. This represents the threshold value for the thickness variance difference.

[0073] Understandably, after obtaining the initial temperature compensation and initial time extension for all defect characteristics, the system uses these initial parameters as input to perform joint simulation calculations using the curing oven's heat transfer model and the wire motion model. The curing oven's heat transfer model is a physical-mathematical model describing the relationship between heater power, oven temperature field distribution, and wire surface temperature. The wire motion model is a mathematical model describing the relationship between the drive train feed rate and the wire's residence time in the curing oven. The goal of the simulation calculation is to determine how the heater power output needs to change over time to achieve the matched initial temperature compensation, forming a heater power change curve. Simultaneously, it determines how the drive train feed rate needs to be adjusted to achieve the matched initial time extension, forming a drive train rate change curve. The calculated set of heater power change curves constitutes the dynamic temperature compensation path for the current unbalanced region, and the calculated set of drive train rate change curves constitutes the time enhancement strategy for the current lag stage. The dynamic temperature compensation path and time enhancement strategy are encapsulated in a data structure, which is called the preliminary control scheme.

[0074] Optionally, the historical successful process case records in the process knowledge base also include contextual parameters such as the curing oven model, heater number, and ambient humidity. During the matching process, while comparing thickness and hysteresis characteristics, records with contextual parameters consistent with current production conditions are also selected to improve matching accuracy. The heater power variation curve is represented as a discrete time-point sequence, with each time point corresponding to a power setpoint. Similarly, the drive train speed variation curve is also represented as a discrete time-point sequence, with each time point corresponding to a feed rate setpoint.

[0075] In some embodiments, when no historical case can be found in the process knowledge base that perfectly matches the characteristics of the current thickness imbalance zone or temperature lag stage, the system uses interpolation to calculate the initial temperature compensation or initial time extension from multiple historical case records with similar characteristics. The joint simulation calculation of the heat transfer model and the wire motion model is an iterative optimization process. By adjusting the shapes of the power change curve and the rate change curve, the model predicts that the final temperature of the wire in the thickness imbalance zone reaches the target compensation temperature, and simultaneously, the model predicts that the dwell time of the wire in the temperature lag stage reaches the target extension duration. The data structure of the preliminary control scheme includes the scheme number, generation timestamp, preliminary feature set identifiers of the associated curing defects, and all parameter data of the dynamic temperature compensation path and time enhancement strategy.

[0076] Example 3: See Figure 3The initial control scheme is analyzed, extracting the heater numbers that require adjustment and their corresponding pre-calculated power change curves. These are then categorized by equipment address to form a subset of heater power control commands. Similarly, the initial control scheme is analyzed to extract the drive train segment numbers that require speed adjustment and their corresponding pre-calculated speed change curves, forming a subset of drive train speed control commands. A unified production process timeline is established, assigning precise effective times and durations to each command in both the heater power control command subset and the drive train speed control command subset. This ensures that heater power adjustments and drive train speed adjustments can coordinate in time and space, avoiding mutual interference or control conflicts. All heater control commands and drive train control commands with precise timestamps are sorted and encapsulated to generate a detailed set of execution commands that can be directly issued to the production equipment control system.

[0077] In practice, the process of parsing the preliminary control scheme is executed by the central processing unit. The preliminary control scheme is loaded into memory as input data. The data structure of the preliminary control scheme explicitly contains a list of heater numbers that need to be adjusted and the power change curve corresponding to each heater number. The central processing unit traverses the list of heater numbers. For each heater number in the list, it reads the associated power change curve data. The power change curve data consists of a series of time-power value coordinate points. The central processing unit encodes this data according to the prescribed device communication protocol format to generate independent device control commands. All control commands for the heaters are categorized and collected together to form a subset of heater power control instructions. The same process of analyzing the preliminary control scheme is also performed on the drivetrain adjustment section. The central processing unit extracts the list of drivetrain segment numbers that need speed adjustment and the corresponding speed change curve for each segment number from the data structure of the preliminary control scheme. For each drivetrain segment number in the list, it reads the associated speed change curve data. The speed change curve data is also composed of a series of time-speed value coordinate points. The central processing unit encodes these data into control commands for the drivetrain driver. All control commands for the drivetrain are categorized and collected together to form a subset of drivetrain speed control instructions.

[0078] In some embodiments, establishing a unified timeline requires a reference clock. The central processing unit uses the absolute time of the start of the production batch as the time zero point to calculate the precise effective time point relative to the time zero point for each instruction in the heater power control instruction subset and the drive train speed control instruction subset. The effective time point of an instruction... Its time coordinate in the original change curve Add a fixed amount to compensate for the delay in command issuance and system response. The decision, the relationship is:

[0079] ;

[0080] in: Indicates the absolute effective time of the instruction. This represents the relative time coordinates of data points in a power change curve or rate change curve. This represents the system delay compensation constant. The central processing unit assigns a duration parameter to each instruction. The duration parameter is directly taken from the time interval between adjacent data points in the power change curve or the rate change curve, ensuring that the power adjustment action of the heater and the rate adjustment action of the transmission chain are completely synchronized in time. Spatially, the central processing unit avoids generating conflicting temperature and speed setting instructions for the same physical area by checking the physical correspondence between the heater number and the transmission chain segment number.

[0081] It is understandable that sorting and encapsulating all timestamped instructions is a crucial step in generating the final executable instructions. The central processing unit merges all instructions from the heater power control instruction subset and the drive train speed control instruction subset into a list, based on the effective time of each instruction. The instructions are sorted in ascending order to generate a sequence of instructions arranged chronologically. The sorted instruction sequence is then further encapsulated by adding a header containing a batch identifier, instruction set version number, and data checksum. The encapsulated complete data packet is the detailed execution instruction set, which is then distributed to the curing oven heater controller and drive train servo drive via an industrial Ethernet or fieldbus network.

[0082] Optional, system delay compensation constant The value is determined by performing multiple no-load command response tests on the control system, and the average delay time obtained from the tests is taken. The command sorting algorithm uses a stable merge sort to ensure that the original relative order of commands with the same effective time point is maintained. The data checksum is generated using a cyclic redundancy check algorithm to ensure the integrity of the detailed execution command set during transmission.

[0083] In some embodiments, a conflict verification check is performed before instruction encapsulation. The logic of the conflict verification check is to scan the sorted instruction sequence and check whether there are setpoint instructions with different values ​​assigned to the same device at the same time. If such a conflict is detected, the system will mark the error and re-examine the relevant parameter settings in the preliminary control scheme. The file format of the detailed execution instruction set can be binary format or text-based markup language format, depending on the interface protocol requirements of the downstream device controller. After successfully sending the detailed execution instruction set, the central processing unit waits for and receives confirmation acknowledgments from the heater controller and the drive chain driver. The confirmation acknowledgments indicate that the detailed execution instruction set has been received and is ready for execution.

[0084] Example 4: During the execution of the detailed execution instruction set, the actual power output value of each heater and the actual feed rate value of each transmission chain segment are collected in real time by the equipment control system as actual execution data. Under the same production process time axis, the expected power output value and expected feed rate value at the corresponding time point are read from the detailed execution instruction set as expected parameter trajectory. The difference between the actual execution data and the expected parameter trajectory at each sampling time point is calculated to obtain a series of power deviation sequences and rate deviation sequences arranged by time. With time as the horizontal axis and the spatial location number of the equipment or temperature zone as the vertical axis, the power deviation sequence and rate deviation sequence are mapped to a two-dimensional coordinate system to form a visualized deviation map, i.e., the execution deviation map. The execution deviation map is analyzed to identify the regions and time periods in which the deviation amplitude is continuously positive or continuously negative. For regions and time periods in which the deviation is continuously positive, it is determined that the compensation or enhancement amount in the preliminary control scheme is excessive, and the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path is reduced accordingly, or the extension amount of the corresponding stage in the time enhancement strategy is reduced. For regions and time periods where the deviation remains negative, it is determined that the compensation or enhancement amount in the initial control scheme is insufficient. Accordingly, the temperature compensation amount for the corresponding heater in the dynamic temperature compensation path is increased, or the extension amount of the corresponding stage in the time enhancement strategy is increased. Using the corrected temperature compensation amount and time extension amount, a new subset of heater power control commands and a subset of drive train speed control commands are recalculated and generated, forming a new detailed execution command set. The new detailed execution command set is executed, and actual execution data is collected again to generate a new execution deviation map. This cycle of analysis, judgment, correction, and execution is repeated until all deviation values ​​in the newly generated execution deviation map are within the preset allowable threshold range.

[0085] In practice, after the detailed execution instruction set is issued and executed, the data acquisition system reads the actual power output value fed back by the heater controller in real time at a fixed sampling period, and simultaneously reads the actual feed rate value fed back by the drive train servo drive. These read values ​​are recorded as actual execution data and timestamped. On the same time axis, the monitoring system parses the expected power output value and expected feed rate value corresponding to each sampling time point from the detailed execution instruction set. These values ​​constitute the expected parameter trajectory. For each sampling time point, the difference between the actual power output value and the expected power output value is calculated to obtain the power deviation, and the difference between the actual feed rate value and the expected feed rate value is calculated to obtain the rate deviation. All power deviations and rate deviations arranged in chronological order constitute the power deviation sequence and the rate deviation sequence. With the sampling time as the horizontal axis and the physical location number of the heater or drive train segment as the vertical axis, the power deviation value or rate deviation value corresponding to each sampling time point and each spatial location is mapped into a two-dimensional matrix. The matrix data is converted into a two-dimensional image with color mapping by the graphics rendering engine, where the color represents the magnitude and sign of the deviation. The generated image is the visualized deviation map, which is the execution deviation map. See Table 1.

[0086] Table 1: Execution Deviation Data Table

[0087]

[0088] In some embodiments, the process of analyzing the deviation map is performed by an image processing and analysis algorithm. The algorithm identifies regions in the image that continuously exhibit warm tones (representing positive deviation) or cool tones (representing negative deviation) and records the spatial location number range and time range corresponding to these regions. For regions and time periods where the identified deviation amplitude is continuously positive, the decision logic determines that the compensation or enhancement amount in the initial control scheme is excessive, and the control system correspondingly reduces the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path, for example, reducing the original temperature compensation amount by a fixed step or reducing it proportionally. For regions and time periods where the identified deviation amplitude is continuously negative, the decision logic determines that the compensation or enhancement amount in the initial control scheme is insufficient, and the control system correspondingly increases the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path, for example, increasing the original temperature compensation amount by a fixed step. The same logic applies to the adjustment of the time enhancement strategy: positive deviation corresponds to a reduction in time extension, and negative deviation corresponds to an increase in time extension.

[0089] It is understandable that after correcting the compensation and extension amounts, control commands need to be regenerated to use the corrected temperature compensation amounts. and time extension As input, the heat transfer model and wire motion model are recalculated, generating new sets of heater power variation curves and new sets of drive train speed variation curves. The new power variation curve set is encoded into a new subset of heater power control commands, and the new speed variation curve set is encoded into a new subset of drive train speed control commands. These new subsets of heater power control commands and drive train speed control commands are aligned, sorted, and encapsulated along a unified time axis to form a new detailed execution command set. The control system issues and executes this new detailed execution command set. During execution, the data acquisition system synchronously collects and re-collects the actual power output value and actual feed rate value, and generates a new execution deviation graph.

[0090] Optionally, the rule for determining whether the deviation is "persistent" is to check whether the sign of the deviation is consistent over N consecutive sampling periods, where N is a preset integer, for example, N=5. The proportional coefficient for correcting the compensation amount is then applied. The adjustment can be adaptively made based on the absolute value of the deviation, and the adjustment relationship is as follows:

[0091] ;

[0092] in: This indicates the corrected temperature compensation amount. This represents the original temperature compensation amount. This represents a correction factor related to the system response characteristics. This represents the normalized average deviation value. The iterative process sets a maximum number of iterations to prevent infinite loops. Each iteration generates a new execution deviation map, which is compared to a preset allowable threshold range—an upper limit on the absolute value of the deviation.

[0093] In some embodiments, the allowable threshold ranges are set separately for power deviation and rate deviation, for example, the power deviation threshold is set to ±0.3 kW and the rate deviation threshold is set to ±0.05 m / min. The convergence condition is defined as the deviation values ​​of all spatial locations at all sampling time points within their corresponding allowable threshold ranges in the new execution deviation map. The correction parameters, the generated new instruction set, and the corresponding new execution deviation map for each iteration are recorded in the log for process traceability and analysis. When the convergence condition is not met but the maximum number of iterations is reached, the system will trigger an alarm and prompt manual intervention.

[0094] See Figure 4This is a box plot of heater power deviation under different sampling periods, used to analyze the distribution characteristics of heater power deviation within different continuous sampling periods during the curing degree control process. It belongs to the category of charts for process stability assessment. The stable deviation direction of the two types of heaters indicates that their power output deviates from the expected value continuously. This type of chart is the core basis for "deviation persistence judgment" in curing degree control. By observing the deviation distribution in different periods, the power compensation strategy for the heater can be determined (e.g., H07 requires increased compensation, H08 requires decreased compensation), while also assisting in setting a reasonable "persistent deviation judgment period".

[0095] Example 5: When the execution deviation graph meets the convergence condition, record all parameters in the final version of the dynamic temperature compensation path and time enhancement strategy used at this time. Parametrically describe all heater power change curves in the final version of the dynamic temperature compensation path, extracting the power values ​​and corresponding time points of key control points on each curve. Parametrically describe all drive train speed change curves in the final version of the time enhancement strategy, extracting the speed values ​​and corresponding time points of key control points on each curve. Bind and package these parametrically described power change curves, speed change curves, and their corresponding curing oven temperature zone numbers and drive train segment numbers to form a complete, reusable, and stable curing process parameter package. Associate and store this stable curing process parameter package with the real-time monitoring data and final control results of the batch that generated it, and update it to the process knowledge base as a new successful process case for querying and matching in subsequent production batches when constructing control schemes.

[0096] In practical implementation, when the deviation graph meets the convergence condition (i.e., all deviation values ​​in the graph are within the preset allowable threshold range), the control system triggers the parameter package generation process. This process records all parameters in the final version of the dynamic temperature compensation path and the final version of the time-enhancing strategy currently in use. These parameters include the power change curve function expressions or discrete data points for all heaters, the rate change curve function expressions or discrete data points for all drive train segments, and the spatiotemporal ranges to which these curves apply. The recording process parameterizes all heater power change curves in the dynamic temperature compensation path. This parameterization method extracts key control points from each power change curve. Key control points include curve inflection points, extreme points, and start and end points. The power value corresponding to each key control point is recorded. and time point ,in This represents the power value of the i-th critical control point. This represents the time point of the i-th critical control point. The recording process parameterizes all the transmission chain speed change curves in the final version of the time-enhancing strategy. The parameterization method extracts critical control points from each speed change curve. These critical control points include curve inflection points, extreme points, start points, and end points. The speed value corresponding to each critical control point is recorded. and time point ,in This represents the rate value of the j-th critical control point. This represents the time point of the j-th critical control point.

[0097] In some embodiments, the parameterized power change curves, rate change curves, and their corresponding curing oven temperature zone numbers and drive train segment numbers are bound and packaged. The binding logic involves establishing a relational mapping table. One record in the mapping table contains a curing oven temperature zone number, the heater number corresponding to that temperature zone, the parameterized description data set of the heater's power change curve, and the applicable wire spatial location range for that curve. Another record in the mapping table contains a drive train segment number, the parameterized description data set of the segment's rate change curve, and the applicable time period range for that curve. The packaging process encapsulates these relational mapping tables, parameter data of all key control points, production batch identifiers, and execution deviation graph summary information at convergence into a structured file. This structured file is named the stable curing process parameter package, and the stable curing process parameter package is stored in non-volatile memory in file format.

[0098] It is understandable that the process of associating and storing stable curing process parameter packages with corresponding real-time monitoring data is a data archiving operation. The control system establishes an index association between all real-time monitoring data collected from the beginning to the end of this production batch, including film thickness distribution data, temperature field data inside the curing oven, solvent evaporation rate data, preliminary feature sets of all intermediate curing defects, preliminary control schemes, detailed execution instruction sets, and execution deviation maps, and stores them as a complete data set in the historical process database. This complete data set is marked as a new successful process case. The data structure of a successful process case includes all the aforementioned input data, process data, output parameters, and the final product quality inspection result label. The operation of updating to the process knowledge base involves converting this new successful process case according to the schema defined in the process knowledge base and adding it to the case table of the process knowledge base for querying and matching in subsequent production batches when constructing control schemes.

[0099] Optionally, the stable curing process parameter package file format adopts a highly readable markup language such as JSON or XML, facilitating manual review or cross-system exchange. When storing related data, the production batch number is used as the primary key, and the stable curing process parameter package file path, real-time monitoring data file path, and other metadata of successful process cases are recorded in the same row of the database. Before adding a new case, the case table in the process knowledge base undergoes a deduplication check. The check mechanism compares the similarity of the key process parameters of the new case with existing cases in the database. If the similarity exceeds a preset threshold, a merge update can be selected instead of a simple addition.

[0100] In some embodiments, the quality inspection result label of a successful process case originates from performance tests conducted offline in a laboratory on the batch of enameled copper flat wire samples. Test items include the withstand voltage, adhesion, and corona resistance of the coating. Only when these test results meet predetermined standards is the case marked as "successful" and available for subsequent matching. Adding a new successful process case to the process knowledge base triggers internal index reconstruction to ensure the efficiency of subsequent query operations. The query logic used for subsequent production batch matching prioritizes successful process cases most similar to the current production conditions. Factors evaluating similarity include wire specifications, coating type, ambient temperature and humidity, and curing oven equipment model.

[0101] See Figure 5 This is a heatmap showing the spatiotemporal distribution of power deviation in enameled copper flat wire. It illustrates power deviation across the dimensions of "spatial location" and "execution time" during process execution, commonly seen in industrial production process monitoring scenarios. After 40 minutes, the positive deviation (red area) increases significantly, indicating that the power is likely to exceed expectations in the later stages. The positive deviation is more concentrated in the 8-10 meter spatial location area, while the negative deviation is relatively prominent in the 0-4 meter area, reflecting differences in power control stability at different locations. It is primarily used for process deviation analysis in industrial production, helping engineers locate abnormal areas in both the "time + space" dimensions, thereby optimizing control strategies and improving the stability of the production process.

[0102] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0103] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for controlling the curing degree of corona-resistant enameled copper flat wires used in humanoid robots, characterized in that, The method includes: Real-time monitoring data of corona-resistant enameled copper flat wire during the production process is obtained, including coating thickness distribution, temperature field inside the curing oven, and solvent evaporation rate. By analyzing the real-time monitoring data, the uneven development of curing degree and the lagging stage in the current production batch are identified, and a preliminary set of characteristics of curing defects is obtained. By integrating the preliminary feature set of the curing defects with the preset process knowledge base, a dynamic temperature compensation path for the current uneven region and a time enhancement strategy for the lag stage are constructed to form a preliminary control scheme. Based on the preliminary control scheme, the power output curve of the heater in the curing oven and the feed rate of the transmission chain are coordinated and redistributed to generate a detailed set of execution instructions containing time and space dimensions. The detailed execution instruction set is sent to the production equipment for execution, and the process parameters are collected synchronously during the execution cycle. The feedback process parameters are compared with the expected parameter trajectory to obtain the execution deviation map. Based on the execution deviation map, the key parameters in the dynamic temperature compensation path and time enhancement strategy are iteratively corrected until the execution deviation map converges within the allowable threshold range, and a stable curing process parameter package is output. The steps for obtaining the preliminary feature set of the curing defects are as follows: The coating thickness distribution data is divided into regions, and the average thickness and variance of each region are calculated. Regions with variances exceeding a preset thickness fluctuation limit are marked as thickness unevenness regions. In the temperature field data inside the curing oven, locate the temperature zone corresponding to the spatial coordinates of the thickness uneven area, analyze the temperature change curve of the temperature zone over time, identify the time period when the slope of the temperature curve is lower than the preset heating rate threshold, and mark the time period as the temperature lag stage. By correlating the solvent evaporation rate data with the time point of the temperature lag phase, the solvent evaporation rate at the time point is calculated. If the solvent evaporation rate is lower than a preset evaporation rate threshold, it is confirmed that the temperature lag phase is accompanied by solvent evaporation lag. By summarizing all the information on the thickness unevenness zone, the temperature hysteresis stage, and the accompanying solvent evaporation delay, a preliminary feature set of the curing defects is obtained.

2. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 1, characterized in that, The specific steps for acquiring the real-time monitoring data are as follows: At the entrance stage of the curing oven, a thickness detection unit is deployed to collect the coating thickness of the entire line by scanning and record it as coating thickness distribution data. In multiple temperature zones of the curing oven, thermocouple arrays are uniformly set up to continuously monitor the temperature values ​​at various locations inside the oven. The temperature values ​​are then organized according to spatial coordinates and time series to form temperature field data inside the curing oven. A gas composition sensor is installed at the exhaust vent of the curing oven to measure the evaporation rate of a specific solvent and obtain solvent evaporation rate data. The real-time monitoring data is formed by aligning the paint film thickness distribution data, the temperature field data inside the curing oven, and the solvent evaporation rate data with timestamps and fusing the data.

3. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 2, characterized in that, The specific steps for forming the preliminary control plan are as follows: The process knowledge base is queried, and it stores the temperature compensation and time adjustment amounts corresponding to different thickness regions in historical successful processes. For each thickness unevenness region in the preliminary feature set of the curing defects, a suggested initial temperature compensation amount is matched from the process knowledge base; For each of the aforementioned temperature hysteresis stages, a suggested initial time extension is matched from the process knowledge base; Based on the initial temperature compensation amount and the initial time extension amount, and combined with the heat transfer model of the curing oven and the wire motion model, the heater power change curve and the transmission chain speed change curve that need to be adjusted for implementing compensation are calculated. The heater power change curve and the transmission chain speed change curve together constitute the dynamic temperature compensation path and time enhancement strategy, and are packaged into the preliminary control scheme.

4. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 3, characterized in that, The specific steps for generating the detailed execution instruction set are as follows: The preliminary control scheme is analyzed, and all heater numbers that need to be adjusted and their corresponding power change curves are extracted to form a subset of heater power control instructions. The preliminary control scheme is analyzed to extract the transmission chain segment numbers that need speed adjustment and their corresponding speed change curves, forming a subset of transmission chain speed control instructions. Establish a unified timeline and assign precise effective time points and durations to each instruction in the subset of heater power control instructions and the subset of transmission chain speed control instructions to ensure temporal and spatial coordination of actions; All timestamped instructions are sorted and encapsulated to generate the detailed execution instruction set.

5. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 4, characterized in that, The specific steps for obtaining the execution deviation map are as follows: During the execution of the detailed execution instruction set, the actual power output value of the heater and the actual feed rate value of the transmission chain are collected in real time as actual execution data; On the same time axis, the expected power output value and expected feed rate value at the corresponding time point are read from the detailed execution instruction set as the expected parameter trajectory; Calculate the difference between the actual execution data and the expected parameter trajectory at each sampling point to obtain the power deviation sequence and the rate deviation sequence; Using time as the horizontal axis and spatial location as the vertical axis, the power deviation sequence and the rate deviation sequence are mapped onto a two-dimensional coordinate system to form a visual deviation map, thus obtaining the execution deviation map.

6. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 5, characterized in that, The iterative correction process is specifically as follows: Analyze the execution deviation map to identify regions and time periods where the deviation amplitude is consistently positive or negative; For regions and time periods where the deviation remains positive, it is determined that the compensation or enhancement amount in the preliminary control scheme is excessive, and the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path is reduced accordingly, or the extension amount of the corresponding stage in the time enhancement strategy is reduced. For regions and periods where the deviation is consistently negative, it is determined that the compensation or enhancement amount in the preliminary control scheme is insufficient, and the temperature compensation amount of the corresponding heater in the dynamic temperature compensation path is increased accordingly, or the extension amount of the corresponding stage in the time enhancement strategy is increased. Using the corrected temperature compensation and time extension, a new subset of heater power control instructions and a new subset of drive train speed control instructions are recalculated and generated to form a new detailed execution instruction set. The new detailed execution instruction set is executed, and data is collected again to generate a new execution deviation map. This process is repeated until all deviation values ​​in the new execution deviation map are within the preset allowable threshold range.

7. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 6, characterized in that, The specific steps for outputting the stable curing process parameter package are as follows: When the execution deviation map meets the convergence condition, record all parameters of the final version of the dynamic temperature compensation path and time enhancement strategy used at this time. The power change curves of all heaters in the final version of the dynamic temperature compensation path are parameterized, and the power values ​​and time points of key control points are extracted. The speed change curves of all transmission chains in the final version of the time enhancement strategy are parameterized, and the speed values ​​and time points of key control points are extracted. The power change curve, rate change curve and their corresponding curing oven temperature zone number and transmission chain segment number are bound and packaged to form the stable curing process parameter package.

8. The method for controlling the curing degree of corona-resistant enameled copper flat wire for humanoid robots according to claim 7, characterized in that, The method further includes: The stable curing process parameter package is associated with and stored with the corresponding real-time monitoring data, and updated to the process knowledge base as a new successful process case for the construction of control schemes for subsequent production batches.

9. A curing degree control system for corona-resistant enameled copper flat wire for humanoid robots, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the curing degree control method for corona-resistant enameled copper flat wire for humanoid robots as described in any one of claims 1 to 8.

Images