An eccentric wire package head tension automatic regulation method and system

Abstract

The present application belongs to the technical field of eccentric wire package head tension control, and discloses an automatic regulation and control method and system for the tension of an eccentric wire package head; the method comprises the following steps: acquiring the tension of the eccentric side and the non-eccentric side of the wrapping material, the speed of the wire core, the load and rotating speed of the servo motor and the variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder in real time; inputting the power supply voltage, the load and the rotating speed of the variable frequency motor in a unit time into a pre-trained rotating speed prediction model to output the rotating speed deviation estimation value of the variable frequency motor at the next moment, and combining the historical wire core speed and the transmission ratio in the set time window to calculate the wire core speed estimation value; the stable regulation and control of the tension in the eccentric wire wrapping process is realized, the stress balance of the wrapping material is ensured, the breakage, relaxation or accumulation of the wrapping material is avoided, and the product quality and production efficiency are improved.

Description

Technical Field

[0001] This invention relates to the field of eccentric yarn wrapping tension control technology, and more specifically, to an automatic tension control method and system for eccentric yarn wrapping. Background Technology

[0002] In the eccentric wire wrapping process of electromagnetic wire production, tension control is crucial to product quality. Its core lies in using tension sensors to monitor the stress difference between the wrapping material on the eccentric and non-eccentric sides in real time. A control system drives a hysteresis brake motor and tension levers to dynamically adjust the material supply on both sides, balancing the uneven stress caused by the eccentric structure and ensuring a tight wrapping layer on the eccentric side while preventing loosening on the non-eccentric side. Compared to ordinary concentric wrapping, eccentric wire wrapping involves a greater thickness of wrapping material in the eccentric direction, and the supply and stress state of the material on both sides naturally differ. This makes the tension control system more sensitive to various interference factors, and the stability of the wire core's movement is a key issue that can easily lead to control failure.

[0003] Specifically, the stable operation of the tension control system relies on the dynamic matching of the wrapping material supply speed and the core wire speed. The core wire speed is driven by two types of motors: the main drive system uses a variable frequency motor, which adjusts the power supply frequency via a frequency converter to achieve continuous speed adjustment over a wide range, adapting to the production needs of different specifications of electromagnetic wire; the tension adjustment component of the wrapping mechanism is equipped with a servo motor, which uses an encoder to provide real-time position feedback, achieving millisecond-level fine speed correction to ensure precise synchronization between the wrapping material supply and the core wire movement. However, the inherent defects of these two types of motors can cause abnormal fluctuations in the core wire speed: the servo motor, due to encoder feedback delay, is prone to speed jumps during sudden load changes; the variable frequency motor is significantly affected by power supply voltage fluctuations, and unstable power supply can cause periodic speed jitters. Both of these fluctuations will cause the actual core wire speed to deviate from the expected value, directly disrupting the balance of tension control.

[0004] At this point, the lag in the response of the tension control system is amplified: when the core speed suddenly increases, the eccentric side experiences a greater shortage of wrapping material, resulting in a more pronounced tension surge and potentially causing excessive stretching of the wrapping material; conversely, when the core speed suddenly decreases, more excess wrapping material accumulates on the eccentric side, leading to a greater tension drop, while the non-eccentric side may also experience slack due to oversupply. This tension imbalance caused by abnormal core speed repeatedly plunges the already complex eccentric wire tension control into regulatory chaos, making it difficult to maintain the stability of the wrapping layers on both sides. The harm of this tension control failure to the eccentric wire wrapping is particularly pronounced: a sudden increase in tension on the eccentric side can cause the wrapping materials such as glass fibers and films to break, or cause the insulation layer on that side to become excessively thin, affecting the voltage withstand performance of the electromagnetic wire; a sudden decrease in tension on the non-eccentric side will result in loosening and wrinkling of the wrapping material, creating localized weak points in the insulation. Meanwhile, the drastic fluctuations in tension will react on the wire core, causing it to undergo irregular elastic deformation, further exacerbating speed instability and creating a vicious cycle of tension imbalance and speed fluctuations. Ultimately, this leads to problems such as uneven insulation layer thickness and eccentricity deviation in the eccentrically wound electromagnetic wire, significantly reducing the product qualification rate and severely restricting production efficiency.

[0005] In the production of enameled wire, this failure in tension control is particularly detrimental to eccentric wire wrapping: a sudden increase in tension on the eccentric side can cause the wrapping materials, such as glass fibers and films, to break, or cause the insulation layer on that side to become excessively thin, affecting the voltage withstand performance of the enameled wire; a sudden drop in tension on the non-eccentric side will cause the wrapping material to loosen and wrinkle, forming local weak points in the insulation. At the same time, the drastic fluctuations in tension will have a reaction effect on the wire core, causing it to produce irregular elastic deformation, further aggravating speed instability, forming a vicious cycle of tension imbalance and speed fluctuation, ultimately leading to problems such as uneven insulation layer thickness and eccentricity deviation in eccentric wire-wrapped enameled wire, significantly reducing the product qualification rate and severely restricting production efficiency.

[0006] In view of this, the present invention proposes an automatic tension control method and system for eccentric yarn wrapping to solve the above problems. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art and achieve the above objectives, the present invention provides the following technical solution: a method for automatic tension control of eccentric yarn wrapping, comprising:

[0008] The system acquires in real time the tension of the wrapping material on both the eccentric and non-eccentric sides, the core speed, the load and speed of the servo motor and the variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder; the servo motor is used to adjust the tension of the wrapping material; the variable frequency motor is used to adjust the core speed.

[0009] The power supply voltage, load, and speed of the variable frequency motor per unit time are input into a pre-trained speed prediction model, and the estimated value of the speed deviation of the variable frequency motor at the next moment is output. Combined with the historical core speed and transmission ratio within a set time window, the estimated value of the core speed is calculated. The transmission ratio is the conversion coefficient between the speed of the variable frequency motor and the core speed.

[0010] Based on the estimated core speed, the current core speed, and the current speed of the variable frequency motor, adjust the power supply frequency of the variable frequency motor's inverter; based on the feedback delay time of the servo motor encoder and the current speed of the servo motor, adjust the speed of the servo motor at the next moment.

[0011] Based on the estimated core speed, the current core speed, and the tension of the wrapping material on the eccentric and non-eccentric sides, adjust the speed of the servo motor and the variable frequency motor at the next moment.

[0012] Furthermore, based on the estimated core speed and the corresponding core speed at that moment, as well as the rotational speed of the variable frequency motor, the method for adjusting the power supply frequency of the inverter for the variable frequency motor includes:

[0013] The difference between the estimated core speed and the current core speed is calculated to obtain the core speed deviation;

[0014] The difference between the current speed of the variable frequency motor and the set speed of the variable frequency motor is calculated to obtain the speed fluctuation of the variable frequency motor.

[0015] Calculate the difference between the current servo motor speed and the set servo motor speed to obtain the servo motor speed fluctuation.

[0016] The compensation frequency is obtained by calculating the weighted sum of the absolute value of the core speed deviation and the absolute value of the speed fluctuation of the variable frequency motor.

[0017] When the absolute value of the core speed deviation is greater than the set speed deviation threshold, the inverter outputs the corresponding power supply frequency to the variable frequency motor based on the positive and negative values ​​of the core speed deviation and the speed fluctuation of the variable frequency motor.

[0018] Furthermore, based on the positive and negative values ​​of the wire core speed deviation and the speed fluctuation of the variable frequency motor, the methods for outputting the corresponding power supply frequency from the frequency converter to the variable frequency motor include:

[0019] If the core speed deviation is positive and the variable frequency motor speed fluctuation is negative, then the variable frequency drive outputs the sum of the current power supply frequency and the compensation frequency to the variable frequency motor.

[0020] If the core speed deviation is negative and the variable frequency motor speed fluctuation is positive, then the variable frequency drive outputs the difference between the current power supply frequency and the compensation frequency to the variable frequency motor.

[0021] Furthermore, the method for adjusting the servo motor speed at the next moment based on the feedback delay time of the servo motor encoder and the current servo motor speed includes:

[0022] Based on the current servo motor speed, the previous speed, and the feedback delay time of the servo motor encoder, the speed lag is obtained.

[0023] The corrected speed is obtained by adding the speed lag to the current speed of the servo motor.

[0024] If the speed lag is not zero, a corrected speed is output to the servo motor.

[0025] Furthermore, based on the estimated core speed, the current core speed, and the tension of the wrapping material on both the eccentric and non-eccentric sides, the methods for adjusting the speeds of the servo motor and the variable frequency motor at the next moment include:

[0026] The tension difference is obtained by calculating the tension difference between the wrapping materials on the eccentric side and the non-eccentric side;

[0027] Based on the material of the wrapping material, a safe range for tension difference is pre-set;

[0028] If the tension difference exceeds the safe range, the speed of the servo motor and the variable frequency motor will be adjusted at the next moment based on the deviation of the wire core speed and the sign of the tension difference.

[0029] Furthermore, based on the sign of the wire core speed deviation and tension difference, the methods for adjusting the speeds of the servo motor and the frequency converter at the next moment include:

[0030] Calculate the speed correction amount based on the core speed deviation, the current tension difference, and the current load of the servo motor and the variable frequency motor;

[0031] If the core speed deviation is positive and the tension difference is positive, then the speed of the eccentric side servo motor will be corrected to the sum of the current speed and the speed correction amount. If the speed of the variable frequency motor is less than the set speed of the variable frequency motor, then the speed of the variable frequency motor will be corrected to the sum of the current speed and the speed correction amount.

[0032] If the core speed deviation is positive and the tension difference is negative, the speed of the non-eccentric servo motor will be corrected to the difference between the current speed and the speed correction amount. If the load of the variable frequency motor is greater than the rated load of the preset ratio, the speed of the variable frequency motor will be corrected to the difference between the current speed and the speed correction amount.

[0033] If the core speed deviation is negative and the tension difference is negative, the speed of the eccentric servo motor will be corrected to the difference between the current speed and the speed correction amount. If the speed of the variable frequency motor is less than the speed of the variable frequency motor set by the preset ratio, the speed of the variable frequency motor will be corrected to the sum of the current speed and the speed correction amount.

[0034] If the core speed deviation is negative and the tension difference is positive, the speed of the non-eccentric servo motor will be corrected to the difference between the current speed and the speed correction amount. If the load of the variable frequency motor is less than the preset ratio of the rated load, the speed of the variable frequency motor will be corrected to the difference between the current speed and the speed correction amount.

[0035] Furthermore, based on the core speed deviation, the current tension difference, and the current load of the servo motor and the variable frequency motor, the method for calculating the speed correction includes:

[0036] By fitting the wire core speed deviation, tension difference, and load from historical data, along with the corresponding rotational speed correction, the speed influence coefficient, tension influence coefficient, and load influence coefficient are obtained.

[0037] The speed influence coefficient is the speed correction amount corresponding to each unit of wire core speed deviation; the tension influence coefficient is the speed correction amount corresponding to each unit of tension difference; the load influence coefficient is the speed correction amount corresponding to each unit of load.

[0038] Rotational speed correction = speed influence coefficient × absolute value of core speed deviation + tension influence coefficient × absolute value of current tension difference + load influence coefficient × current load.

[0039] Furthermore, the training methods for the speed prediction model include:

[0040] Collect historical power supply voltage, historical load, and historical speed of the variable frequency motor to form a training set;

[0041] The training set is preprocessed by removing outliers. The power supply voltage, load and speed of the variable frequency motor at multiple consecutive time points are used as input samples according to the time series, and the speed deviation estimate at the corresponding time point is used as the output label to obtain the preprocessed data.

[0042] A temporal network architecture is built based on LSTM, and the corresponding hidden layers, neurons and activation functions are set to obtain the initial model;

[0043] The preprocessed data is divided into training set, test set and validation set according to the proportion. The initial model is trained by backpropagation algorithm and the error of the validation set is monitored in real time. Training is stopped when the error no longer decreases.

[0044] The initial model performance is evaluated using a test set. If the prediction error exceeds the set range, the initial model parameters are adjusted and the model is retrained until the preset accuracy requirements are met, resulting in a well-trained speed prediction model.

[0045] Furthermore, based on the estimated rotational speed deviation, the method for calculating the estimated core speed by setting the historical core speed and transmission ratio within a time window includes:

[0046] The pre-trained speed prediction model outputs the speed deviation estimate of the variable frequency motor, which is the difference between the predicted speed and the set speed.

[0047] Calculate the average historical core velocity within a set time window, and use it as a historical reference value for core velocity;

[0048] Multiply the estimated rotational speed deviation by the transmission ratio to obtain the estimated core speed deviation.

[0049] The estimated core speed is obtained by adding the historical reference value of the core speed to the estimated core speed deviation.

[0050] The data acquisition module is used to acquire in real time the tension of the wrapping material on the eccentric and non-eccentric sides, the core speed, the load and speed of the servo motor and the variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder.

[0051] The speed prediction module is used to input the power supply voltage, load and speed of the variable frequency motor per unit time into the pre-trained speed prediction model, output the speed deviation estimate of the variable frequency motor at the next moment, and calculate the core speed estimate by combining the historical core speed and transmission ratio within the set time window.

[0052] The motor adjustment module adjusts the power supply frequency of the inverter of the variable frequency motor based on the estimated core speed, the current core speed, and the current speed of the variable frequency motor; and adjusts the speed of the servo motor at the next moment based on the feedback delay time of the servo motor encoder and the current speed of the servo motor.

[0053] The speed adjustment module adjusts the speed of the servo motor and the frequency converter motor at the next moment based on the estimated core speed, the current core speed, and the tension of the wrapping material on the eccentric and non-eccentric sides.

[0054] Compared with existing technologies, the technical effects and advantages of the automatic tension control method and system for eccentric yarn wrapping of the present invention are as follows:

[0055] This invention acquires real-time data on the tension of the wrapping material on both the eccentric and non-eccentric sides, the core speed, the load and speed of the servo motor and variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder. Using a pre-trained LSTM model, it predicts the speed deviation at the next moment based on the power supply voltage, load, and speed of the variable frequency motor. It then calculates the core speed estimate by combining historical core speed and transmission ratio. Based on the core speed estimate, the current core speed, and the variable frequency motor speed, it adjusts the inverter's power supply frequency. It adjusts the servo motor speed according to the encoder feedback delay time and speed. Simultaneously, it dynamically adjusts the speeds of the servo motor and variable frequency motor based on the tension difference and safety range between the eccentric and non-eccentric sides, according to the positive or negative value of the core speed deviation and tension difference.

[0056] This invention solves the problems of speed jumps in servo motors due to encoder feedback delay during sudden load changes and periodic speed fluctuations in variable frequency motors caused by power supply voltage fluctuations. It achieves stable tension control during the eccentric wire winding process, ensuring balanced force on the winding material, avoiding breakage, loosening or accumulation of the winding material, and improving product quality and production efficiency. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of an automatic tension control system for eccentric yarn wrapping according to an embodiment of the present invention;

[0058] Figure 2 This is a flowchart of an automatic tension control method for eccentric yarn wrapping according to an embodiment of the present invention;

[0059] Figure 3 This is a flowchart illustrating a method for outputting a corresponding power supply frequency from a frequency converter to a frequency converter based on the positive and negative values ​​of the wire core speed deviation and the speed fluctuation of the frequency converter motor, according to an embodiment of the present invention.

[0060] Figure 4 This is a flowchart illustrating a method for adjusting the speeds of the servo motor and the variable frequency motor at the next moment based on the positive and negative values ​​of the wire core speed deviation and tension difference, according to an embodiment of the present invention.

[0061] Figure 5 This is a schematic diagram of the training method for the speed prediction model according to an embodiment of the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be described in detail, clearly, and completely below with reference to the accompanying drawings. It should be particularly noted that the specific embodiments described below are only for better illustrating and explaining the technical solutions of the present invention, and are intended to enable those skilled in the art to better understand and implement the present invention, and should not be construed as limiting the scope of protection of the present invention. Without departing from the spirit and substance of the present invention, those skilled in the art can modify, adjust, or make equivalent substitutions based on the content disclosed in the present invention, and these should all be considered within the scope of protection of the present invention.

[0063] Example 1

[0064] Please see Figure 1 As shown, this embodiment discloses an automatic tension control system for eccentric yarn wrapping, including modules that are connected by wires and / or wirelessly to achieve data transmission.

[0065] The data acquisition module is used to acquire in real time the tension of the wrapping material on the eccentric and non-eccentric sides, the core speed, the load and speed of the servo motor and the variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder.

[0066] The tension of the wrapping material on both the eccentric and non-eccentric sides requires the installation of tension sensors, such as strain gauge tension sensors, along the material's travel path on both sides. These sensors are connected in series between the wrapping material and the guide wheel, ensuring that the tension of the wrapping material directly acts on the sensor's sensing end. The core speed is acquired by installing an incremental encoder on the driven wheel of the core traction mechanism. The incremental encoder is coaxially connected to the driven wheel. When the core drives the driven wheel to rotate, the incremental encoder outputs a pulse signal. The real-time core speed is calculated based on the pulse frequency and the driven wheel's circumference. The servo motor's speed is obtained through its built-in absolute encoder, which outputs real-time angular position pulses of the motor shaft. The speed is obtained by calculating the number of angular position pulses per unit time. The servo motor's load is obtained through the servo driver. The driver monitors the servo motor winding current and calculates the load percentage based on the servo motor's rated current: Load = Real-time Current / Rated Current × 100%. The speed of the variable frequency motor is detected in real time by the frequency converter. The frequency converter calculates the speed based on the power supply frequency and the number of pole pairs of the motor, i.e., speed = 60 × power supply frequency / number of pole pairs. Simultaneously, it calculates the load percentage by detecting the motor input current and referring to the rated current. The power supply voltage of the variable frequency motor needs to be obtained by connecting a voltage sensor, such as a Hall effect voltage sensor, in parallel with its power supply line. The feedback delay time of the servo motor encoder is measured by a timer. The time difference between sending the speed command and receiving the encoder feedback signal is the feedback delay time.

[0067] Real-time acquisition of the tension of the wrapping material on both the eccentric and non-eccentric sides allows for direct monitoring of whether the stress difference between the two sides is in equilibrium. The eccentric structure inherently leads to a natural difference in stress on the materials on both sides, and a sudden increase or decrease in tension is a direct manifestation of tension imbalance. Real-time tension data can capture this anomaly immediately, providing a basis for determining whether the tension exceeds the safe range and whether motor adjustments are needed. This prevents excessive stretching of the wrapping material on the eccentric side or accumulation of the wrapping material on the non-eccentric side due to excessive stress differences.

[0068] Real-time data on the core wire speed is the core benchmark for tracking speed fluctuations. Dynamic matching between the core wire speed and the supply speed of the wrapping material is a prerequisite for tension stability, while abnormal fluctuations in the core wire speed are a direct cause of tension imbalance. Real-time acquisition of the core wire speed allows for timely detection of deviations from expected values, providing a monitoring basis for subsequent adjustments to the speeds of the variable frequency motor and servo motor to stabilize the core wire speed. This prevents speed anomalies from going undetected and amplifying tension fluctuations, such as a sudden increase in speed causing a sharp increase in tension in the eccentric wrapping material.

[0069] By combining the load and speed data of the servo motor with the feedback delay time of the servo motor encoder, the problem of speed jumps caused by encoder feedback delay during sudden load changes can be specifically solved. Sudden load changes in the servo motor cause rapid changes in the actual speed, while encoder delay causes the system to adjust based on old data, resulting in a disconnect between the command and the actual demand. By monitoring the load and speed of the servo motor in real time and calculating the speed lag based on the delay time, the speed command can be corrected in advance, ensuring precise synchronization between the servo motor speed and the wire core movement, reducing the interference of speed jumps on the wire core speed, and preventing the tension balance from being disrupted by the jumps.

[0070] The load, speed, and supply voltage data of variable frequency motors are key to solving the problem of periodic speed fluctuations caused by supply voltage fluctuations. The supply voltage directly reflects the fluctuation of the power grid, while the load and speed data record the response pattern of the variable frequency motor under dynamic disturbances, such as the correlation between increased load and decreased speed when the voltage drops.

[0071] The feedback delay time of the servo motor encoder is used to quantify the impact of delay on servo motor control. Encoder delay can cause the system to collect historical speed data, leading to a disconnect between the adjustment command and actual demand during sudden load changes, resulting in speed jumps. By acquiring the delay time in real time, the speed lag can be calculated, correcting the servo motor speed command and synchronizing the adjustment action with the actual motor state. This avoids speed jumps caused by delay and ensures precise matching between the winding material supply speed and the wire core speed.

[0072] The speed prediction module is used to input the power supply voltage, load and speed of the variable frequency motor per unit time into the pre-trained speed prediction model, output the estimated value of the speed deviation of the variable frequency motor at future time, and calculate the estimated value of the core speed by combining the historical core speed and transmission ratio within a set time window; the transmission ratio is the conversion coefficient between the speed of the variable frequency motor and the core speed.

[0073] In actual production, the core speed prediction cannot be achieved through direct presetting. Fluctuations in the power supply voltage of the variable frequency motor will change its output torque. Uneven core material in the wrapping material, sudden changes in wrapping tension, and other load variations will cause nonlinear responses in the motor's speed. Even mechanical clearances in the transmission mechanism can cause slight fluctuations in core speed. The power supply voltage, load, and speed data of the variable frequency motor are random and real-time. Directly presetting a fixed core speed prediction will result in a continuous deviation from the actual core speed. When the actual core speed is higher than the preset core speed, the wrapping tension will suddenly increase, causing the core to tighten; when the actual core speed is lower than the preset core speed, the tension will loosen, causing the core to sag, ultimately leading to wrapping quality defects such as uneven wire diameter or broken wires. Therefore, it is necessary to capture the real-time trend of the core speed through dynamic prediction, rather than relying on presets.

[0074] Meanwhile, changes in the core speed directly affect the tension of the wrapping material. Increased core speed leads to increased tension, while decreased core speed leads to decreased tension. However, motor control involves physical delays, such as inverter response time or mechanical inertia when adjusting the speed of a variable frequency motor. By predicting the core speed, future trends can be anticipated: if deceleration is predicted, the variable frequency motor speed can be increased in advance to offset the speed decrease; if acceleration is predicted, the variable frequency motor speed can be decreased in advance to suppress sudden tension changes.

[0075] Furthermore, the conductor wire is connected to the variable frequency motor via a transmission mechanism such as gears or rollers. When the transmission ratio is fixed, the conductor wire speed is directly proportional to the variable frequency motor speed, i.e., conductor wire speed = variable frequency motor speed × transmission ratio. Therefore, the motor speed itself is a direct mapping of the conductor wire speed. Supply voltage and load are key variables affecting motor speed. Fluctuations in supply voltage will change the output torque of the variable frequency motor. When the supply voltage decreases, the torque is insufficient, and the speed tends to drop. Increased load will force the variable frequency motor speed to decrease. Voltage, load, and speed data per unit time record the motor's response pattern under dynamic disturbances. By training a speed prediction model, such as an LSTM neural network, the temporal correlation between voltage-load-speed changes and conductor wire speed changes in historical data can be learned. Based on real-time data within the current unit time, the trend of conductor wire speed in the future, such as within 1-3 seconds, can be inferred, ultimately yielding a predicted conductor wire speed.

[0076] Based on the estimated rotational speed deviation, the historical core speed and transmission ratio within a set time window are used to calculate the estimated core speed, as follows:

[0077] Obtain the estimated value of the speed deviation of the variable frequency motor output by the pre-trained speed prediction model; calculate the average value of the historical core speed within a set time window as the historical reference value of the core speed; multiply the estimated value of the speed deviation by the transmission ratio to obtain the estimated value of the core speed deviation; add the historical reference value of the core speed to the estimated value of the core speed deviation to obtain the estimated value of the core speed.

[0078] Specifically, the time window setting must match the fluctuation frequency of the wire core speed. If the wire core speed exhibits high-frequency, small-amplitude fluctuations due to process characteristics such as material tension changes and motor load fluctuations (e.g., more than 10 fluctuations per second), the time window should be smaller, such as 0.1-0.5 seconds, to avoid including too much outdated data that would cause the mean to lag behind the actual changes. If the speed changes smoothly, such as fluctuating a few times per minute, the window can be appropriately increased, such as 1-3 seconds, to smooth out random noise with more historical data and ensure the stability of historical reference values. Furthermore, the time scale of the pre-trained speed prediction model must be adapted. The pre-trained speed prediction model is trained based on samples at specific time intervals, such as collecting data every 0.01 seconds. The time window selection must match the input time step of the speed prediction model. For example, if the model input is data from the past 10 time steps, the window can be set to 0.01 seconds × 10 = 0.1 seconds to ensure consistency between the estimated speed deviation and the historical reference value of the wire core speed in the time dimension, avoiding increased prediction errors due to time scale mismatch.

[0079] Please see Figure 5 As shown, the specific training method for the speed prediction model is as follows:

[0080] Historical power supply voltage, historical load, and historical speed of the variable frequency motor are collected to form a training set. The training set undergoes outlier removal preprocessing. The power supply voltage, load, and speed of the variable frequency motor at consecutive time points are used as input samples, and the predicted speed deviation at the corresponding time point is used as the output label, resulting in preprocessed data. A temporal network architecture based on LSTM is built, and corresponding hidden layers, neurons, and activation functions are set to obtain the initial model. For example, two hidden layers can be set: the first layer contains 128 LSTM neurons to capture the long-term correlation between the power supply voltage, load, and speed of the variable frequency motor; the second layer contains 64 LSTM neurons to extract short-term dynamic features. Both layers use the tanh activation function to adapt to the internal control mechanism of the LSTM. The numerical range is set, and the output layer has one neuron with a linear activation function. Since the speed deviation prediction is a continuous value, no nonlinear mapping is needed. A dropout layer is added between the two hidden layers to prevent overfitting. The preprocessed data is divided into training, test, and validation sets proportionally. The initial model is trained using backpropagation, and the validation set error is monitored in real time. Training stops when the error no longer decreases. The performance of the initial model is evaluated using the test set. If the prediction error exceeds a set range, the initial model parameters are adjusted and retrained until the preset accuracy requirement is met, resulting in a well-trained speed prediction model. For example, the mean square error (MSE) between the predicted speed deviation and the actual speed deviation on the test set can be used for evaluation. If the set accuracy requirement is MSE ≤ 0.5 rpm... 2 The initial model's MSE on the test set after the first training was 0.8 rpm. 2If the initial model parameters need to be adjusted, such as increasing the number of neurons in the first layer to 150 to enhance feature extraction capability, reducing the learning rate from 0.001 to 0.0005 to avoid training oscillations, and increasing the number of training epochs to 200, then after retraining, if the MSE on the test set drops to 0.3 rpm... 2 Then, a well-trained speed prediction model is obtained.

[0081] The motor adjustment module adjusts the power supply frequency of the inverter for the variable frequency motor based on the estimated core speed, the current core speed, and the current speed of the variable frequency motor; and adjusts the speed of the servo motor at the next moment based on the feedback delay time of the servo motor encoder and the current speed of the servo motor.

[0082] Based on the estimated core speed and the corresponding core speed at that moment, as well as the rotational speed of the variable frequency motor, the power supply frequency of the variable frequency motor's inverter is adjusted as follows:

[0083] The difference between the estimated core speed and the current core speed is calculated to obtain the core speed deviation; the difference between the current variable frequency motor speed and the set variable frequency motor speed is calculated to obtain the variable frequency motor speed fluctuation; the set variable frequency motor speed is determined based on the specifications of the manufactured electromagnetic wire. Different specifications of electromagnetic wire correspond to different target core travel speeds. The core speed and variable frequency motor speed are related through a fixed transmission ratio. Therefore, it needs to be calculated based on the target core speed and transmission ratio to adapt to the wide range of speed adjustment requirements of the core main drive system and ensure that the core can travel stably at the basic speed required by the process; the difference between the current servo motor speed and the set servo motor speed is calculated. The difference is used to obtain the servo motor speed fluctuation. The set servo motor speed is determined based on the type and density requirements of the wrapping material and needs to match the core speed to ensure that the supply speed of the wrapping material is precisely synchronized with the core speed, thereby maintaining the basic tension balance of the wrapping material and avoiding tension abnormalities caused by mismatch between the supply and core speed. Its value needs to be combined with the stress characteristics such as the breaking tension and elongation of the wrapping material to ensure that the wrapping material will not be overstretched due to too slow supply or accumulate due to too fast supply at the set servo motor speed. The weighted sum of the absolute value of the core speed deviation and the absolute value of the frequency converter motor speed fluctuation is calculated to obtain the compensation frequency.

[0084] The weights of the absolute values ​​of the conductor speed deviation and the absolute values ​​of the variable frequency motor speed fluctuation are determined by collecting the correspondence between the conductor speed deviation, the variable frequency motor speed fluctuation, and the actual required frequency adjustment under different operating conditions. For example, when the conductor speed deviation is 0.2 m / s, an adjustment of 0.004 Hz is required to offset the speed deviation; when the variable frequency motor speed fluctuation is 5 rpm, an adjustment of 0.05 Hz is required to stabilize the speed. The contribution weights of the conductor speed deviation and the variable frequency motor speed fluctuation to frequency compensation are obtained by fitting the least squares method.

[0085] Please see Figure 3 As shown, when the absolute value of the wire core speed deviation is greater than the set speed deviation threshold, based on the positive or negative sign of the wire core speed deviation and the speed fluctuation of the variable frequency motor, the inverter outputs a corresponding power supply frequency to the variable frequency motor, as follows:

[0086] The speed deviation threshold is determined by combining the stress characteristics of the wrapping material and the production process requirements. Firstly, considering the elongation and breaking tension of the wrapping material, the speed fluctuation range that the material can withstand directly determines the upper limit of the speed deviation threshold, avoiding excessive stretching or accumulation of the wrapping material due to excessive deviation. Simultaneously, based on the specifications of the produced electromagnetic wire, different specifications correspond to different core speeds. The speed deviation threshold must be adapted to the core speed; when the core speed is high, the speed deviation threshold can be appropriately relaxed, while when it is low, it needs to be tightened to ensure accuracy. During setting, trial production tests are conducted, gradually adjusting the deviation value to observe whether the core speed at this deviation will cause the tension difference to exceed the safe range. The maximum deviation that does not lead to tension imbalance is used as the speed deviation threshold, while also considering adjustment stability to avoid fluctuations caused by frequent adjustments of the variable frequency motor due to an excessively small speed deviation threshold, or adjustment lag due to an excessively large threshold.

[0087] If the conductor speed deviation is positive and the variable frequency motor speed fluctuation is negative, the inverter outputs the sum of the current supply frequency and the compensation frequency to the variable frequency motor. This is because a positive conductor speed deviation indicates a higher predicted conductor speed than the current actual conductor speed, suggesting a future upward trend in conductor speed. Conversely, a negative variable frequency motor speed fluctuation indicates a lower actual speed than the set speed. Without intervention, the conductor speed will be insufficient to reach the predicted value due to insufficient motor power, resulting in insufficient speed. Therefore, it is necessary to directly increase the inverter's supply frequency by outputting the sum of the current supply frequency and the compensation frequency. Since the variable frequency motor speed is positively correlated with the supply frequency, an increase in the supply frequency will immediately increase the motor speed. The motor speed, through a fixed transmission ratio, directly affects the conductor, causing the conductor speed to increase synchronously with the motor speed. The advance adjustment time is matched with the prediction cycle of the speed prediction model. When the core speed increases as predicted in the future, the variable frequency motor has already completed the speed reserve by increasing the speed in advance, thereby offsetting the speed deficiency that may have been caused by the low speed of the variable frequency motor, and ensuring that the core speed smoothly transitions to the predicted value.

[0088] If the conductor speed deviation is negative and the variable frequency motor speed fluctuation is positive, the inverter outputs the difference between the current supply frequency and the compensation frequency to the variable frequency motor. This is because a negative conductor speed deviation (meaning the predicted conductor speed is lower than the current actual conductor speed) indicates a future decreasing trend in conductor speed, while a positive variable frequency motor speed fluctuation (meaning the current actual motor speed is higher than the set speed) means that without intervention, the conductor speed will exceed the predicted value due to excess motor power, resulting in excessive speed. Therefore, it is necessary to output a command to the inverter to reduce the inverter's supply frequency by specifying the difference between the current supply frequency and the compensation frequency. A lower supply frequency directly leads to a decrease in motor speed, which in turn reduces the conductor speed synchronously through the transmission ratio. When the conductor speed decreases according to the predicted value in the future, the motor has already reduced its power output by slowing down in advance, preventing the conductor speed from exceeding the predicted value due to excessive speed and ensuring that the conductor speed fluctuates stably around the predicted value.

[0089] For cases where the conductor speed deviation is positive and the variable frequency motor speed fluctuation is positive, and for cases where the conductor speed deviation is negative and the variable frequency motor speed fluctuation is negative, no adjustment is required because their physical meaning does not match the adjustment target, as detailed below:

[0090] A positive core speed deviation indicates that the estimated core speed is higher than the actual core speed. If the variable frequency motor speed fluctuation is positive at this time, it means that the actual speed is already higher than the set speed, indicating that the current output of the variable frequency motor is already too strong. The core speed will naturally increase due to the high speed, and there is no need to increase the frequency. Otherwise, it will exacerbate the excessively high speed. A negative core speed deviation indicates that the estimated core speed is lower than the actual core speed. If the variable frequency motor speed fluctuation is negative at this time, it means that the actual speed is already lower than the set speed, indicating that the current output of the variable frequency motor is already too weak. The core speed will naturally decrease due to the low speed, and there is no need to decrease the frequency. Otherwise, it will exacerbate the excessively low speed.

[0091] Based on the feedback delay time of the servo motor encoder and the current servo motor speed, the servo motor speed is adjusted for the next moment, as follows:

[0092] Based on the current servo motor speed, the speed at the previous moment, and the feedback delay time of the servo motor encoder, the speed lag is obtained; the speed lag is calculated as (real-time speed of the servo motor - speed at the previous unit moment) × feedback delay time of the servo motor encoder.

[0093] The corrected speed is obtained by adding the speed lag to the current speed of the servo motor; if the speed lag is not zero, the corrected speed is output to the servo motor.

[0094] The encoder feedback delay of a servo motor can cause the collected servo motor speed to be historical data from a later point, rather than a real-time value. When the load changes abruptly, the actual servo motor speed has already begun to change rapidly, but the system still adjusts based on the old data due to the encoder feedback delay. This leads to a disconnect between the adjustment command and the actual demand. For example, when the load suddenly drops, the servo motor may have already accelerated, but the system may still be outputting a command to maintain the original speed, ultimately causing excessive speed jumps in the servo motor.

[0095] The calculation of speed lag is precisely to quantify the impact of the encoder feedback delay of the servo motor. For example, if the encoder feedback delay of the servo motor is 0.02s, and the servo motor speed increases from 1000rpm to 1005rpm in 0.001s, then the speed lag is 0.02 × 5000 = 100rpm. This means that the system considers the servo motor speed to be 1000rpm, while it has actually reached 1100rpm, resulting in a 100rpm lag error.

[0096] The speed correction based on hysteresis essentially compensates for speed deviations in advance: when the speed hysteresis is positive, meaning the actual servo motor speed is higher than the system's perceived value due to the encoder feedback delay, a corrected speed is output to the servo motor, allowing the system to output a higher speed command in advance to offset the accumulated speed deficiency during the delay period. When the speed hysteresis is negative, meaning the actual speed is lower than the system's perceived value due to the delay, a corrected speed is output to the servo motor, reducing its speed and preventing potential excessive speed drop. This advance correction is equivalent to completing the servo motor speed adjustment within the encoder feedback delay time, synchronizing the system speed with the actual servo motor speed and avoiding excessive speed jumps caused by the encoder feedback delay.

[0097] The speed adjustment module adjusts the speed of the servo motor and the frequency converter motor at the next moment based on the estimated core speed, the current core speed, and the tension of the wrapping material on the eccentric and non-eccentric sides.

[0098] Based on the estimated core speed, the current core speed, and the tension of the wrapping material on both the eccentric and non-eccentric sides, the speeds of the servo motor and the variable frequency motor are adjusted for the next moment, as follows:

[0099] The tension difference is obtained by calculating the tension difference between the wrapping material on the eccentric side and the non-eccentric side.

[0100] Based on the material of the wrapping material, a safe range for the tension difference is pre-set: the core physical parameters of the wrapping material need to be collected first, including the breaking tension, which is the maximum tension the material can withstand, the elastic modulus, which reflects the tensile strength of the material, and the elongation, which is the maximum elongation ratio of the material before it breaks. These parameters can be obtained from the material manufacturer's instructions or measured through material mechanics tests. For example, the breaking tension of glass fiber is usually 5-8N, the breaking tension of film is 3-5N, and the elongation is lower than that of glass fiber. Subsequently, the eccentric wire wrapping equipment is started under no-load conditions. The wrapping material within the safety range of the tension difference to be set is threaded into the wrapping mechanisms on the eccentric and non-eccentric sides respectively. The initial tension on both sides is gradually adjusted to the basic value required by the process, such as 3N on the eccentric side and 2N on the non-eccentric side. Then, the tension on one side is slowly changed through the tension adjustment device, so that the tension difference is gradually increased from 0, for example, by 0.2N each time. At the same time, the state of the wrapping material is observed in real time: if the wrapping material shows local tensile deformation, such as film wrinkling or micro-cracks in the glass fiber, the tension difference value at this time is recorded as the warning threshold; if the wrapping material breaks or becomes significantly loose, such as yarn breakage, the tension difference value at this time is recorded as the limit threshold. The above test is then repeated under loaded production conditions. Because the friction force generated by the movement of the wire core will change the actual stress state of the material, the warning threshold and the limit threshold need to be corrected in combination with the wire core speed. For example, when the wire core speed increases by 10%, the warning threshold of the glass fiber needs to be reduced by 0.3N to avoid brittle fracture at high speed. Finally, the range between the corrected warning threshold and the limit threshold is determined as the safe range of the tension difference of the material. For example, the range is set to -0.5N to +0.8N for glass fiber and -0.3N to +1.0N for film. The negative sign indicates that the tension on the non-eccentric side is greater than that on the eccentric side.

[0101] Please see Figure 4 As shown, if the tension difference exceeds the safe range, the speeds of the servo motor and the variable frequency motor will be adjusted at the next moment based on the core speed deviation and the sign of the tension difference, as follows:

[0102] Based on the core speed deviation, current tension difference, and current load of the servo motor and variable frequency motor, the speed correction is calculated. Specifically, by using historical data on core speed deviation, tension difference, and load, along with the corresponding speed correction, speed influence coefficients, tension influence coefficients, and load influence coefficients are fitted. For example, 1000 sets of historical data from the same wrapping material production process over the past 3 months are selected. Each set of data includes core speed deviation (unit: m / s), tension difference (unit: N), variable frequency motor load (unit: % of rated load), and the corresponding actual speed correction (unit: rpm). Some data are as follows: when the core speed deviation is 0.3 m / s, the tension difference is 0.5 N, and the load is 60%, the corresponding speed correction is 2.8 rpm; when the core speed deviation is 0.2 m / s, the tension difference is 0.3 N, and the load is 50%, the correction is 1.9 rpm; when the core speed deviation is 0.4 m / s, the tension difference is 0.6 N, and the load is 70%, the correction is 3.5 rpm. The parameters that minimize the error between the predicted and actual speed corrections are calculated using the least squares method: If the calculated values ​​are k1 = 2 rpm / (m / s), k2 = 1.5 rpm / N, and k3 = 0.02 rpm / %, then it means that every 1 m / s deviation in wire core speed requires a speed correction of 2 rpm, every 1 N tension difference requires a correction of 1.5 rpm, and every 1% load requires a correction of 0.02 rpm.

[0103] Among them, the speed influence coefficient is the speed correction amount corresponding to each unit of core speed deviation; the tension influence coefficient is the speed correction amount corresponding to each unit of tension difference; the load influence coefficient is the speed correction amount corresponding to each unit of load; the speed correction amount = speed influence coefficient × absolute value of core speed deviation + tension influence coefficient × absolute value of current tension difference + load influence coefficient × current load.

[0104] If the core speed deviation is positive, meaning the estimated core speed is higher than the current actual core speed and there is a future acceleration trend, and the tension difference is positive, it indicates that the tension of the eccentric side wrapping material is greater than that of the non-eccentric side wrapping material. This could lead to the eccentric side wrapping material breaking due to excessive stretching. In this case, the speed of the eccentric side servo motor is corrected to the sum of the current speed and the speed correction amount. Increasing the speed of the eccentric side servo motor can increase the speed at which it releases the wrapping material, matching the future acceleration trend of the core. This prevents the tension difference from further widening due to insufficient supply of wrapping material on the eccentric side during core acceleration, thus balancing the tension on both sides. At the same time, if the speed of the variable frequency motor is lower than the set variable frequency motor speed, it indicates that the current power is insufficient to support core acceleration. In this case, the speed of the variable frequency motor is corrected to the sum of the current speed and the speed correction amount. By increasing the power, it ensures that the core can accelerate smoothly according to the estimated value, avoiding tension fluctuations caused by insufficient power leading to core speed lag.

[0105] If the core speed deviation is positive and the tension difference is negative, meaning the tension of the non-eccentric side wrapping material is greater than that of the eccentric side wrapping material, the non-eccentric side wrapping material is prone to breakage due to excessive tightness. In this case, the speed of the non-eccentric side servo motor is corrected to the difference between the current speed and the speed correction amount. This is because reducing the speed of the non-eccentric side can reduce the release speed of its wrapping material, preventing the non-eccentric side wrapping material from becoming further tensile due to excessive release when the core accelerates, thereby reducing the tension difference. At the same time, if the load of the variable frequency motor is greater than the preset ratio of the rated load, it indicates that the current load of the variable frequency motor is too high and the core traction force is too strong, which will aggravate the stress on the non-eccentric side wrapping material. In this case, the speed of the variable frequency motor is corrected to the difference between the current speed and the speed correction amount. By reducing the speed, the core traction force is reduced, alleviating the tension pressure on the non-eccentric side wrapping material and preventing damage to the wrapping material caused by excessive load of the variable frequency motor.

[0106] If the core speed deviation is negative, meaning the estimated core speed is lower than the current actual core speed, indicating a future deceleration trend, and the tension difference is negative (meaning the tension of the non-eccentric side wrapping material is high, and the non-eccentric side wrapping material tends to accumulate during core deceleration), then the speed of the eccentric side servo motor is corrected to the difference between the current speed and the speed correction amount. This is because reducing the speed of the eccentric side servo motor can reduce the release of its wrapping material, preventing the accumulation of eccentric side wrapping material due to oversupply during core deceleration, and simultaneously reducing the tension advantage of the non-eccentric side wrapping material. Simultaneously, if the speed of the variable frequency motor is less than the preset ratio of the set variable frequency motor speed, it indicates that the current variable frequency motor power is too weak, and the core may decelerate excessively. Therefore, the speed of the variable frequency motor is corrected to the sum of the current speed and the speed correction amount. By appropriately increasing the speed of the variable frequency motor, the core speed is maintained, preventing more severe accumulation of non-eccentric side wrapping material due to excessive deceleration, and ensuring wrapping uniformity.

[0107] If the core speed deviation is negative and the tension difference is positive, meaning the tension of the eccentric side wrapping material is high, the eccentric side wrapping material is prone to becoming too tight when the core decelerates. In this case, the speed of the non-eccentric side servo motor is corrected to the difference between the current speed and the speed correction amount. This is because reducing the speed of the non-eccentric side servo motor can reduce the release of its wrapping material, preventing insufficient release of the non-eccentric side wrapping material during core deceleration and exacerbating the tension advantage of the eccentric side wrapping material, thus balancing the forces on both sides. At the same time, if the load of the variable frequency motor is less than the preset ratio of the rated load, it indicates that the current load of the variable frequency motor is too light, resulting in insufficient core traction and potentially leading to excessively rapid deceleration. In this case, the speed of the variable frequency motor is corrected to the difference between the current speed and the speed correction amount. By appropriately reducing the speed of the variable frequency motor to match the core deceleration trend, a sudden drop in core speed due to an excessively light load on the variable frequency motor is avoided, preventing the eccentric side wrapping material from becoming overly tight due to a slow supply speed, and ensuring the stability of the wrapping material.

[0108] The preset ratio of rated load needs to be determined through trial production testing, taking into account the load-bearing capacity of the wrapping material and the rated operating characteristics of the variable frequency motor. First, based on the physical parameters of the wrapping material, such as breaking tension and elastic modulus, the stress limit of the material under different loads is determined. For example, glass fibers are prone to breakage due to excessive core traction when the load exceeds 80% of the rated load, while films are prone to overstretching when the load exceeds 70%. At the same time, the rated load parameters of the variable frequency motor are referenced to ensure that the ratio does not exceed the safe operating range of the motor, avoiding overheating or sudden speed drop of the motor due to excessive load.

[0109] Example 2

[0110] Please see Figure 2 As shown, this embodiment provides a method for automatic tension control of eccentric yarn wrapping, including:

[0111] The system acquires in real time the tension of the wrapping material on both the eccentric and non-eccentric sides, the core speed, the load and speed of the servo motor and the variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder; the servo motor is used to adjust the tension of the wrapping material; the variable frequency motor is used to adjust the core speed.

[0112] The power supply voltage, load, and speed of the variable frequency motor per unit time are input into a pre-trained speed prediction model, and the estimated value of the speed deviation of the variable frequency motor at the next moment is output. Combined with the historical core speed and transmission ratio within a set time window, the estimated value of the core speed is calculated. The transmission ratio is the conversion coefficient between the speed of the variable frequency motor and the core speed.

[0113] Based on the estimated core speed, the current core speed, and the current speed of the variable frequency motor, adjust the power supply frequency of the variable frequency motor's inverter; based on the feedback delay time of the servo motor encoder and the current speed of the servo motor, adjust the speed of the servo motor at the next moment.

[0114] Based on the estimated core speed, the current core speed, and the tension of the wrapping material on the eccentric and non-eccentric sides, adjust the speed of the servo motor and the variable frequency motor at the next moment.

[0115] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0116] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for automatically adjusting the tension of an eccentric wire wrapping, characterized in that, include: Real-time acquisition of tension of the wrapping material on the eccentric and non-eccentric sides, core speed, load and speed of servo motor and variable frequency motor, power supply voltage of variable frequency motor and feedback delay time of servo motor encoder; Servo motors are used to adjust the tension of the wrapping material; Variable frequency motors are used to adjust the speed of the wire core; The power supply voltage, load, and speed of the variable frequency motor per unit time are input into the pre-trained speed prediction model, and the output is the speed deviation estimate of the variable frequency motor at the next moment. Combined with the historical core speed and transmission ratio within the set time window, the core speed estimate is calculated. The transmission ratio is the conversion coefficient between the speed of the variable frequency motor and the speed of the wire core; Methods for calculating the estimated core velocity include: The speed deviation estimate of the variable frequency motor is obtained from the output of the pre-trained speed prediction model. The speed deviation estimate of the variable frequency motor is the difference between the predicted speed and the set speed. Calculate the average historical core velocity within a set time window, and use it as a historical reference value for core velocity; Multiply the estimated rotational speed deviation by the transmission ratio to obtain the estimated core speed deviation. The estimated core speed is obtained by adding the historical reference value of the core speed to the estimated core speed deviation. Based on the estimated core speed, the current core speed, and the current speed of the variable frequency motor, adjust the power supply frequency of the variable frequency motor's inverter; based on the feedback delay time of the servo motor encoder and the current speed of the servo motor, adjust the speed of the servo motor at the next moment. Based on the estimated core speed, the current core speed, and the tension of the wrapping material on the eccentric and non-eccentric sides, adjust the speed of the servo motor and the variable frequency motor at the next moment.

2. The method for automatic tension control of eccentric wire wrapping head according to claim 1, characterized in that... Based on the estimated core speed, the corresponding core speed, and the rotational speed of the variable frequency motor, methods for adjusting the power supply frequency of the variable frequency motor's inverter include: The difference between the estimated core speed and the current core speed is calculated to obtain the core speed deviation; Calculate the difference between the current speed of the variable frequency motor and the set speed of the variable frequency motor to obtain the speed fluctuation of the variable frequency motor; Calculate the difference between the current servo motor speed and the set servo motor speed to obtain the servo motor speed fluctuation. The compensation frequency is obtained by calculating the weighted sum of the absolute value of the core speed deviation and the absolute value of the speed fluctuation of the variable frequency motor. When the absolute value of the core speed deviation is greater than the set speed deviation threshold, the inverter outputs the corresponding power supply frequency to the variable frequency motor based on the positive and negative values ​​of the core speed deviation and the speed fluctuation of the variable frequency motor.

3. The method for automatic tension control of eccentric wire wrapping head according to claim 2, characterized in that, Based on the positive and negative values ​​of wire core speed deviation and variable frequency motor speed fluctuation, methods for outputting the corresponding power supply frequency from the frequency converter to the variable frequency motor include: If the core speed deviation is positive and the variable frequency motor speed fluctuation is negative, then the variable frequency drive outputs the sum of the current power supply frequency and the compensation frequency to the variable frequency motor. If the core speed deviation is negative and the variable frequency motor speed fluctuation is positive, then the variable frequency drive outputs the difference between the current power supply frequency and the compensation frequency to the variable frequency motor.

4. The method for automatic tension control of eccentric yarn wrapping according to claim 1, characterized in that, Methods for adjusting the servo motor speed at the next moment based on the feedback delay time of the servo motor encoder and the current servo motor speed include: Based on the current servo motor speed, the previous speed, and the feedback delay time of the servo motor encoder, the speed lag is obtained. The corrected speed is obtained by adding the speed lag to the current speed of the servo motor. If the speed lag is not zero, a corrected speed is output to the servo motor.

5. The method for automatic tension control of eccentric yarn wrapping according to claim 1, characterized in that, Based on the estimated core speed, the current core speed, and the tension of the wrapping material on both the eccentric and non-eccentric sides, the methods for adjusting the speeds of the servo motor and the variable frequency motor at the next moment include: The tension difference is obtained by calculating the tension difference between the wrapping materials on the eccentric side and the non-eccentric side; Based on the material of the wrapping material, a safe range for tension difference is pre-set; If the tension difference exceeds the safe range, the speed of the servo motor and the variable frequency motor will be adjusted at the next moment based on the deviation of the wire core speed and the sign of the tension difference.

6. The method for automatic tension control of eccentric yarn wrapping according to claim 5, characterized in that, Based on the sign of the wire core speed deviation and tension difference, the methods for adjusting the speed of the servo motor and the frequency converter motor at the next moment include: Calculate the speed correction amount based on the core speed deviation, the current tension difference, and the current load of the servo motor and the variable frequency motor; If the core speed deviation is positive and the tension difference is positive, then the speed of the eccentric side servo motor will be corrected to the sum of the current speed and the speed correction amount. If the speed of the variable frequency motor is less than the set speed of the variable frequency motor, then the speed of the variable frequency motor will be corrected to the sum of the current speed and the speed correction amount. If the core speed deviation is positive and the tension difference is negative, the speed of the non-eccentric servo motor will be corrected to the difference between the current speed and the speed correction amount. If the load of the variable frequency motor is greater than the rated load of the preset ratio, the speed of the variable frequency motor will be corrected to the difference between the current speed and the speed correction amount. If the core speed deviation is negative and the tension difference is negative, the speed of the eccentric servo motor will be corrected to the difference between the current speed and the speed correction amount. If the speed of the variable frequency motor is less than the speed of the variable frequency motor set by the preset ratio, the speed of the variable frequency motor will be corrected to the sum of the current speed and the speed correction amount. If the core speed deviation is negative and the tension difference is positive, the speed of the non-eccentric servo motor will be corrected to the difference between the current speed and the speed correction amount. If the load of the variable frequency motor is less than the preset ratio of the rated load, the speed of the variable frequency motor will be corrected to the difference between the current speed and the speed correction amount.

7. The method for automatic tension control of eccentric yarn wrapping according to claim 6, characterized in that, The methods for calculating the speed correction amount based on the core speed deviation, current tension difference, and current load of the servo motor and frequency converter motor include: By fitting the wire core speed deviation, tension difference, and load from historical data, along with the corresponding rotational speed correction, the speed influence coefficient, tension influence coefficient, and load influence coefficient are obtained. The speed influence coefficient is the speed correction amount corresponding to each unit of wire core speed deviation; the tension influence coefficient is the speed correction amount corresponding to each unit of tension difference; the load influence coefficient is the speed correction amount corresponding to each unit of load. Rotational speed correction = speed influence coefficient × absolute value of core speed deviation + tension influence coefficient × absolute value of current tension difference + load influence coefficient × current load.

8. The method for automatic tension control of eccentric yarn wrapping according to claim 1, characterized in that, Training methods for speed prediction models include: Collect historical power supply voltage, historical load, and historical speed of the variable frequency motor to form a training set; The training set is preprocessed by removing outliers. The power supply voltage, load and speed of the variable frequency motor at multiple consecutive time points are used as input samples according to the time series, and the speed deviation estimate at the corresponding time point is used as the output label to obtain the preprocessed data. A temporal network architecture is built based on LSTM, and the corresponding hidden layers, neurons and activation functions are set to obtain the initial model; The preprocessed data is divided into training set, test set and validation set according to the proportion. The initial model is trained by backpropagation algorithm and the error of the validation set is monitored in real time. Training is stopped when the error no longer decreases. The initial model performance is evaluated using a test set. If the prediction error exceeds the set range, the initial model parameters are adjusted and the model is retrained until the preset accuracy requirements are met, resulting in a well-trained speed prediction model.

9. An automatic tension control system for eccentric yarn wrapping, used to implement the automatic tension control method for eccentric yarn wrapping as described in any one of claims 1-8, characterized in that, include: The data acquisition module is used to acquire in real time the tension of the wrapping material on the eccentric and non-eccentric sides, the core speed, the load and speed of the servo motor and the variable frequency motor, the power supply voltage of the variable frequency motor, and the feedback delay time of the servo motor encoder. The speed prediction module is used to input the power supply voltage, load and speed of the variable frequency motor per unit time into the pre-trained speed prediction model, output the speed deviation estimate of the variable frequency motor at the next moment, and calculate the core speed estimate by combining the historical core speed and transmission ratio within the set time window. Methods for calculating the estimated core velocity include: The speed deviation estimate of the variable frequency motor is obtained from the output of the pre-trained speed prediction model. The speed deviation estimate of the variable frequency motor is the difference between the predicted speed and the set speed. Calculate the average historical core velocity within a set time window, and use it as a historical reference value for core velocity; Multiply the estimated rotational speed deviation by the transmission ratio to obtain the estimated core speed deviation. The estimated core speed is obtained by adding the historical reference value of the core speed to the estimated core speed deviation. The motor adjustment module adjusts the power supply frequency of the inverter of the variable frequency motor based on the estimated core speed, the current core speed, and the current speed of the variable frequency motor; and adjusts the speed of the servo motor at the next moment based on the feedback delay time of the servo motor encoder and the current speed of the servo motor. The speed adjustment module adjusts the speed of the servo motor and the frequency converter motor at the next moment based on the estimated core speed, the current core speed, and the tension of the wrapping material on the eccentric and non-eccentric sides.

Images