A method and system for optical fiber winding tension control based on PLC

Abstract

A kind of PLC-based optical fiber winding tension control method and system, first, the remaining length of optical fiber is obtained according to the set and real-time winding length decreasing model, when the remaining length is less than the threshold value, the winding unit switches to the end preparation state and triggers the motor to slow down, after reaching the end speed, the position information of winding mechanism is obtained by photoelectric sensor, and the motor speed correction coefficient is obtained by position loop speed correction algorithm according to the difference between the target angle and the position information, to generate control instruction.If the angle reaches or exceeds the target angle, the neural network prediction model is started to obtain the shutdown advance amount, and the motor is stopped with low vibration according to the preset path model PLC output shutdown signal. The present application integrates various sensors and intelligent algorithms using PLC, realizes high-precision tension control during the whole winding process, effectively solves the problems of unstable tension control, end problems and other conditions in traditional technology, improves the quality and efficiency of optical fiber winding, and adapts to various types of optical fiber and process requirements.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of optical fiber winding, and particularly relates to a PLC-based optical fiber winding tension control method and system. BACKGROUND

[0002] In the field of optical fiber manufacturing, optical fiber winding is a key production link. Traditional optical fiber winding equipment usually relies on simple mechanical structures and relatively basic control means to realize the winding process. For example, in terms of tension control, mechanical tensioners or simple electrical control circuits are often used. The mechanical tensioner has the problems of limited adjustment precision and poor adaptability to optical fibers of different diameters and materials. The tension adjustment mainly relies on the friction or elastic deformation of mechanical components, and it is difficult to maintain stable and accurate tension output throughout the winding process. Once the diameter, material or winding speed of the optical fiber changes, the mechanical tensioner is difficult to make accurate adjustments in time, thereby easily leading to tension fluctuations in the optical fiber winding process.

[0003] The simple electrical control circuit also lacks the ability to comprehensively monitor and effectively integrate and analyze various dynamic parameters (such as the remaining length of the optical fiber, the change in motor speed, the position of the winding mechanism, etc.) in the winding process. This makes it difficult for the traditional electrical control circuit to meet the precise tension control requirements when facing some special winding tasks, such as high-precision optical fiber product winding with extremely high tension stability requirements, or complex processes that require frequent speed and tension changes during winding. This affects the quality and performance consistency of the optical fiber winding products.

[0004] In addition, the traditional control method also has many drawbacks in the finishing stage of optical fiber winding. Usually, only a fixed deceleration program or the experience of the operator is used to stop the winding motor, and there is a lack of precise control and prediction of the position of the winding mechanism, the tension of the optical fiber and the dynamic characteristics of the entire system. This easily causes sudden changes in tension of the optical fiber during the finishing stage, leading to problems such as optical fiber breakage or irregular winding, which seriously reduces the yield and production efficiency of the products.

[0005] Based on the above analysis, the main technical problem solved by the present application is that the existing optical fiber winding technology has the defects of low precision, poor adaptability and problems in the finishing stage in terms of tension control, and is difficult to meet the needs of the modern optical fiber manufacturing industry for high-quality products. SUMMARY

[0006] The technical problem to be solved by the present application is to provide a PLC-based optical fiber winding tension control method and system, which realizes high-precision tension control in the entire winding process of optical fiber from start to finish through an accurate length calculation model, an advanced motor speed reduction control algorithm, a precise position detection and speed correction strategy, and a neural network-based predictive shutdown control mechanism, effectively reduces tension fluctuations, avoids optical fiber breakage, improves product yield and quality consistency, and can adapt to different types of optical fiber and various winding process requirements, thereby improving the intelligent level and production efficiency of the entire optical fiber winding production process.

[0007] To solve the above technical problems, the technical solution adopted by the present application is:

[0008] A PLC-based optical fiber winding tension control method and system, comprising the following steps:

[0009] S1: According to the pre-set total length of optical fiber and the real-time winding length information fed back by the current encoder, the remaining length of optical fiber is calculated through a length decreasing model;

[0010] S2: If the remaining length of optical fiber is less than or equal to the pre-set end trigger length threshold, the winding control unit switches from the normal winding state to the end preparation state, and an end operation signal is obtained;

[0011] S3: The motor speed reduction control module is triggered according to the end preparation state signal, a pre-set speed reduction curve control algorithm is used to smoothly reduce the motor speed, and it is determined that the motor running speed has reached the end winding speed value;

[0012] S4: When the motor runs at low speed and the wound optical fiber length reaches a value obtained by subtracting a safety margin from the end trigger length, the winding control unit obtains the current position information of the winding mechanism through an optical sensor installed at a pre-set reference position, and determines the angle of the optical fiber relative to the pre-set reference position in the end winding stage;

[0013] S5: According to the difference between the real-time winding mechanism current angle and the target angle, the winding control unit calculates the winding motor speed correction coefficient through a position loop-based speed correction algorithm, obtains the control instruction for the end winding of optical fiber, and accurately controls the position of the end process;

[0014] S6: If the position control calculation model output shows that the current angle of the winding mechanism has reached or exceeded the target angle, the winding control unit starts the pre-set neural network prediction model, the neural network prediction model analyzes and learns the end parking position advance information from the historical operation data, and determines the amount of shutdown signal that needs to be sent in advance by the end control link;

[0015] S7: According to the ending control link shutdown advance quantity given by the prediction model and the preset ending path model, the PLC outputs a shutdown signal to the driver, so that the winding motor stops in a low-vibration deceleration mode, avoids causing the optical fiber to break, and obtains a complete winding product.

[0016] Preferably, the sub-steps of S1 are:

[0017] S2.1 If the optical fiber remaining length data value is less than or equal to the preset ending trigger threshold length value, then

[0018] S2.2 The winding control unit state data is updated to 1;

[0019] S2.3 According to the winding control unit state data, when the winding control unit state data is 1, the winding control unit switches to an ending preparation state;

[0020] S2.4 The winding control unit obtains the ending preparation state data.

[0021] S2.5 The winding control unit enters the ending preparation state;

[0022] S2.6 The winding control unit generates an ending operation pre-trigger signal data;

[0023] S2.7 The winding control unit according to the ending operation pre-trigger signal data;

[0024] S2.8 The winding control unit then runs a self-check by the device self-check module to obtain a device self-check normal signal data;

[0025] S2.9 The device self-check normal signal data is used as a final ending operation precondition;

[0026] S2.10 The ending signal is triggered and output, and the winding device performs an optical fiber ending operation.

[0027] Preferably, the sub-steps of S2 are:

[0028] S2.1 If the optical fiber remaining length data value is less than or equal to the preset ending trigger threshold length value, then

[0029] S2.2 The winding control unit state data is updated to 1;

[0030] S2.3 According to the winding control unit state data, when the winding control unit state data is 1, the winding control unit switches to an ending preparation state;

[0031] S2.4 The winding control unit obtains the ending preparation state data.

[0032] Preferably, the sub-steps of S3 are:

[0033] S3.1 According to the end preparation state signal flip-flop, a high level output is generated, and the motor controller receives the high level to obtain information for control start;

[0034] S3.2 The motor deceleration control module processes the end preparation state signal, and transmits it to the programmable logic controller through the bus to obtain a set of instruction data;

[0035] S3.3 By table lookup method, the function relationship expression y=f(x) of the preset deceleration curve is obtained, the independent variable x represents time, and the dependent variable y represents motor speed, to obtain a set of expected motor speed;

[0036] S3.4 The encoder real-time measurement motor speed information is obtained, and transmitted to the data processor through the serial port to obtain the actual running speed value;

[0037] S3.5 The Kalman filter algorithm is used to denoise the actual running speed value of the motor, and the denoised actual running speed value is used as the feedback input into the PID controller, and the PID parameters are adjusted by iteration to obtain the best PID control instruction;

[0038] S3.6 The PID controller fuses the motor speed value obtained by the preset deceleration curve and the denoised actual running speed value, and if the difference between the two is greater than the set value, the alarm module is triggered for early warning;

[0039] S3.7 A prediction model is established by using the BP neural network algorithm, the prediction model is trained using the actual running speed set obtained by the PID control, and it is judged whether the predicted speed value meets the end winding speed value, and a stop signal is output.

[0040] Preferably, the sub-steps of S4 are:

[0041] S4.1 The encoder installed on the motor periodically records the speed, and generates a data stream combined with the time parameter;

[0042] S4.2 The data stream is temporarily stored through the data buffer pool;

[0043] S4.3 The neural network prediction model adjusts its parameters in real time through a self-learning process according to historical running speed data and establishes a running speed judgment benchmark;

[0044] S4.4 The model compares the current data with the learned judgment benchmark to determine the running speed value and divide different working stages for different values, and the running speed less than a certain value is divided into a low-speed running stage;

[0045] S4.5 The running state value is used as a state label to enter the judgment benchmark data stream for subsequent analysis and processing;

[0046] S4.6 Calculate the number of turns according to the preset parameters of the optical fiber, set the initial number of turns, combine the information collected by the photoelectric sensor to obtain the final number of turns and the circumference of the winding reel, multiply them to obtain the total length at the current time, and then subtract to obtain the actual remaining length;

[0047] S4.7 The comparison length is calculated by subtracting the end trigger length setting value from the safety margin setting value;

[0048] S4.8 Determine the actual remaining length and the comparison length value by using the preset rule to obtain the remaining length in the end trigger range judgment result;

[0049] S4.9 If the length falls within the range, an end trigger instruction is generated;

[0050] S4.10 The preset rule obtains the end trigger instruction, and the end stage control logic obtains the control parameters by analyzing the data packet in the instruction and performs data classification processing, and distributes the classified control instructions to the specified device;

[0051] S4.11 When the photoelectric sensor detects the trigger information, store the trigger information in the cache area;

[0052] S4.12 The photoelectric sensor installed at the preset reference position of the winding device detects the periodic rotation signal of the winding reel and collects the rotation data sequence;

[0053] S4.13 The photoelectric sensor converts the position data into an analog electrical signal with a period of one revolution of the winding device;

[0054] S4.14 Perform frequency domain analysis on the obtained analog electrical signal using Fourier transform to obtain harmonic components and main frequency;

[0055] S4.15 Calculate the total length of the optical fiber and the number of turns of the device, and obtain the length information corresponding to each turn by calculation, and divide the remaining length value of the optical fiber by the length value of each turn to obtain the quotient and the remainder;

[0056] S4.16 Multiply the remainder value by the time corresponding to the circumference of each turn and the phase corresponding to the circumference of each turn to obtain the conversion angle value;

[0057] S4.17 Obtain the total phase angle and convert the remainder part into a radian value under the reference of a specific reference point;

[0058] S4.18 Pre-set a numerical calculation model, which real-time collects the Fourier transform values and clock data as input data, and uses difference calculation to obtain the value change in a period of time;

[0059] S4.19 Collect all data obtained by S4.1-S4.18 to perform data fusion;

[0060] S4.20 The time correlation is obtained by comparing the photoelectric sensor information with the reference time sequence through a time window, and the phase shift value is obtained by combining the reference time sequence with the mechanical structure movement amplitude information, thereby obtaining the time parameter;

[0061] S4.21 The total angle is obtained by calculating the product of the time value and the angular velocity value;

[0062] S4.22 The current motor instantaneous angular velocity and the corresponding fiber length value are obtained by judging the time parameter;

[0063] S4.23 The system updates the trigger length threshold of the end stage in real time according to the fiber length, and uses the fiber type data in the established database as the judgment object;

[0064] S4.24 The end parameter value of each material is obtained, and the end parameter value is transmitted to the motion controller through the data interface;

[0065] S4.25 The mechanical structure characteristic value of the fiber winding equipment is comprehensively compared and analyzed;

[0066] S4.26 The actual position deviation value is determined by the photoelectric sensor output value and the internal set control parameter;

[0067] S4.27 The winding position information of the next period is predicted by the recurrent neural network, and a plurality of future time prediction points are obtained;

[0068] S4.28 The position change trend and change speed are obtained by subtracting the prediction point from the current value, and the trend data is used as feedback control data in real time;

[0069] S4.29 The motor operating parameters are adjusted according to the deviation value through the fuzzy control algorithm, and the photoelectric sensor monitors a preset time value for time and value data synchronization storage in the historical database;

[0070] S4.30 Then the historical database classifies and stores different storage type data, thereby achieving speed adjustment.

[0071] Preferably, the sub-steps of S5 are:

[0072] S5.1 Obtain the current angle information of the winding mechanism and the set target angle information according to real-time detection;

[0073] S5.2 Calculate the angle difference value according to the current angle information and the target angle information, and when the angle difference value exceeds the preset range, enter the early warning database;

[0074] S5.3 reading the angle difference from the early warning database, if the angle difference is lower than the preset lower limit range, skip the speed correction step;

[0075] S5.4 if the angle difference is higher than the preset upper limit range, generate the alarm information corresponding to the angle difference and notify the display module to display the information, control the instruction generation to pause, and at the same time, read the current angle and target angle from the early warning database;

[0076] S5.5 calculating the angle difference by a position loop-based control algorithm to obtain a speed correction coefficient;

[0077] S5.6 correcting the current winding motor output torque according to the speed correction coefficient to obtain the corrected real-time torque control information;

[0078] S5.7 adaptively compensating the real-time torque control information according to the preset winding target length of the optical fiber to obtain the final control instruction;

[0079] S5.8 establishing an environment perception database for the real-time winding environment, and issuing the final control instruction to the winding control execution mechanism, and the execution mechanism receives the final control instruction to control the winding motor to perform the optical fiber end winding work;

[0080] S5.9 if the end state parameter in the final control instruction shows stop, the execution mechanism pauses winding, the control instruction updates pause, then obtains the working data update information in the environment perception database to obtain new winding target length information, and reads the corresponding control instruction information from S5.5 to resend the instruction;

[0081] S5.10 if the end state parameter in the final control instruction does not show stop, continue the operation of this step;

[0082] S5.11 performing online detection according to the real-time data information library of the winding mechanism execution state, determining the motion state of the end winding work and saving.

[0083] Preferably, the sub-steps of S6 are:

[0084] S6.1 obtaining the current angle information of the winding mechanism from the winding mechanism, and judging the difference between the current angle information and the target angle information;

[0085] S6.2 if the difference between the current angle information and the target angle information is greater than a set value, outputting a signal of the winding control unit to the neural network prediction model;

[0086] S6.3 obtaining the signal of the winding control unit by the neural network prediction model, then analyzing the historical operation data from the database;

[0087] S6.4 According to the analysis result, the end parking position advance information is obtained;

[0088] S6.5 According to the end parking position advance information, the end control advance stop signal quantity is determined;

[0089] S6.6 The winding control unit receives the stop signal quantity and controls the winding mechanism to perform corresponding operation;

[0090] S6.7 The database receives and stores the angle information, signal quantity and winding mechanism information.

[0091] Preferably, the sub-steps of S7 are:

[0092] S7.1 Obtain the end stage stop advance quantity data output by the prediction model, and obtain the motor control instruction according to the stop advance quantity data and the preset end stage path model information after fusion;

[0093] S7.2 The winding machine driver obtains the fused motor control instruction, and according to the current winding motor speed data, the driver drives the winding motor to run to the target speed, which is lower than half of the initial winding speed;

[0094] S7.3 Through a preset time sequence, the time sequence parameters come from the database, the optical fiber diameter data in operation is obtained by the optical fiber state sensor group, and the real-time tension data is obtained by the optical fiber tension sensor group, the running state is estimated by using the historical time data, when the change amplitude of the optical fiber diameter in at least one time is greater than the set threshold value, or when the optical fiber tension in at least one time is greater than the threshold value, the optical fiber state monitoring module makes a stop judgment, triggers the optical fiber abnormal state data marking action, determines the specific abnormal marking after the abnormal marking is recognized by the neural network, and then records the optical fiber abnormal state data marking result;

[0095] S7.4 In the end stage, the optical fiber abnormal state data marking result obtained by step S7.3 is collected, and whether the number of all abnormal markings is 0 is analyzed;

[0096] S7.5 In the end stage, a convolutional neural network obtained according to the modeling of the relationship between the optical fiber end process parameters is used to judge whether secondary winding is needed;

[0097] If there is optical fiber abnormal state data marking result and product defect, the secondary winding control link is carried out, a new winding control instruction set is planned through the defect type and position data, and combined with the defect elimination strategy, the actual motor operation control instruction is calculated and generated according to the control instruction set and the product defect position coordinate value;

[0098] Otherwise, no secondary winding operation is performed;

[0099] S7.6 The winding machine driver obtains the control signal after secondary winding;

[0100] S7.7 The signal transmission device transmits the control signal to the winding machine motor drive unit through industrial Ethernet.

[0101] S7.8 The drive unit determines the winding mode and low-vibration mode of the winding machine in the finishing stage for winding processing.

[0102] S7.9 The winding machine executes the control signal in the finishing stage until the product integrity sensor judges that the product is completely wound, and the integral value is calculated through the integrity calculation function, the function is I = ∫t0 to t1 f(x)dx, I represents the integrity, f(x) is the function of the integrity with respect to time, t0, t1 represents the time of the winding stage.

[0103] If the winding result is incomplete, the control information is generated and transmitted to the motor execution unit through the transmitter, and the winding continues.

[0104] S7.10 After obtaining the complete winding product, the actual winding product finishing data is obtained combined with the prediction model.

[0105] S7.11 The data of the finishing winding process is statistically obtained, the actual result of the finishing data is obtained, and a data comparison chart is generated.

[0106] S7.12 Data storage and analysis: store the data of the finishing stage, and use it for subsequent analysis and improvement, ensure that each winding process can learn from historical data, continuously optimize the control algorithm, and improve the quality and production efficiency of the product.

[0107] Preferably, the sub-step of step S7.4 is:

[0108] S7.4.1 If the number of abnormal markers is greater than 0, the finishing control link parameter planning is re-performed based on the abnormal markers.

[0109] S7.4.2 According to the finishing control link shutdown advance amount obtained in S7.4.1, a control signal is generated and sent to the motor control link, and the motor control link adjusts the control curve.

[0110] S7.4.3 If the number of abnormal markers is 0, the control link parameter re-generation mechanism is not triggered, the control link parameter remains unchanged as the result of S7.4.1, and the subsequent operation is continued.

[0111] A PLC-based optical fiber winding tension control system adopts the PLC-based optical fiber winding tension control method.

[0112] The present application can achieve the following beneficial effects:

[0113] The application calculates the residual length of the optical fiber in real time by constructing a length-decreasing model, and controls the winding operation to smoothly transition from normal winding to a tailing preparation state in combination with a preset tailing trigger length threshold. The application uses a deceleration curve control algorithm to reduce the motor speed, and obtains the position information of the winding mechanism through a photoelectric sensor, calculates the difference between the current angle and the target angle, and applies a speed correction algorithm based on the position loop to accurately control the tailing process. A neural network prediction model is also introduced to learn the advance of the tailing parking position from historical data, in combination with a preset tailing path model, to accurately output a parking signal, realize low-vibration deceleration stopping, and ensure the smooth end of the optical fiber winding process and the integrity of the product. BRIEF DESCRIPTION OF DRAWINGS

[0114] The application will be further described below in conjunction with the drawings and embodiments:

[0115] Fig. 1 It is a control logic diagram of a PLC-based optical fiber winding tension control method;

[0116] Fig. 2 It is a signal transmission process diagram of a PLC-based optical fiber winding tension control method. DETAILED DESCRIPTION

[0117] The preferred scheme is as shown in the figure, a PLC-based optical fiber winding tension control method and system, Figs. 1-2

[0118] Embodiment 1:

[0119] A PLC-based optical fiber winding tension control method, the steps are:

[0120] Step S1, according to the preset total length of the optical fiber and the real-time winding length information fed back by the current encoder, the residual length of the optical fiber is calculated through the established length-decreasing model.

[0121] Step S2, if the residual length of the optical fiber is less than or equal to the preset tailing trigger length threshold, the winding control unit is switched from the normal winding state to the tailing preparation state, and the winding control unit is obtained. The signal that the tailing operation is about to be executed.

[0122] The sensor continuously monitors the optical fiber residual length data value; acquires the optical fiber residual length data value and the preset tailing trigger threshold length value. According to the optical fiber residual length data value and the preset tailing trigger threshold length value, the optical fiber residual length data value is compared with the preset tailing trigger threshold length value. If the optical fiber residual length data value is less than or equal to the preset tailing trigger threshold length value, the winding control unit state data is updated to 1.

[0123] ​According to the winding control unit state data, when the winding control unit state data is 1, the winding control unit switches to the end preparation state; end preparation state data is obtained. The winding control unit is in the end preparation state; then the end operation pre-trigger signal data is generated. According to the end operation pre-trigger signal data; then the device self-checking module runs self-checking, and obtains device self-checking normal signal data. The device self-checking normal signal data is used as the final end operation precondition; then the end signal is triggered and output, and the fiber winding device performs the fiber end operation.

[0124] The sensor continuously monitors the excess length data value of the optical fiber at a frequency of 10 times per second and transmits the data to the data processing center in real time. The data processing center pre-sets an end trigger threshold length value, for example, 10 meters. The data processing center uses a sliding average algorithm to average the received 5 consecutive data points to reduce measurement error. The average length at the current time is the average value of the fiber length in the past 5 seconds (including the current time), which obtains a more smooth fiber excess length data value. Assuming that the average calculation of the current average length is 15 meters. Through programming, the data processing center compares the average calculation of the smooth fiber excess length data value 15 meters with the preset end trigger threshold length value 10 meters in real time. Through simple numerical comparison logic, it is determined that the latest monitoring data 15 meters is compared, and it is confirmed that the fiber excess length data value 15 meters is greater than the preset end trigger threshold length value 10 meters, and the winding control unit state data remains 0. Through numerical comparison, if the average calculation of the smooth fiber excess length data value is 8 meters, which is less than the preset end trigger threshold length value 10 meters, the winding control unit state data is immediately updated to 1, and the state switching mechanism is triggered. The system transmits the data to the winding control unit, and the winding control unit controls the motor to reduce the speed based on the updated state data. Specifically, the original speed of the motor is 1000 revolutions per minute, which is now gradually reduced to 50 revolutions per minute, and enters the end preparation state. When the fiber excess approaches the set value, the control system compares the latest data with the threshold value by using a function relationship, which can use a linear or exponential function, triggers and generates a pulse signal with a frequency of 1 Hz. The signal lasts for 5 seconds, and then is processed by the system as end operation pre-trigger signal data.

[0125] The end operation pre-trigger signal activates the device self-checking module, which checks the status of all sensors through a preset diagnostic program and sends a read request to each sensor. All sensors respond within 1 second and feedback their status. If all are normal, the device is in a normal operating state. After the system receives the device self-checking normal signal data, the analysis module can parse this normal signal data, verify whether the parsed data characteristic code meets the expected value, and determine that everything is normal if it does. After confirming that all feedback data from the device self-checking module is normal, a 5 Hz pulse signal with a 50% duty cycle is sent to the winding device as an end signal. After receiving this signal, the winding device performs the preset end operation, such as neatly winding the remaining optical fiber to the predetermined position, completing the entire optical fiber end operation.

[0126] Step S3: Trigger the motor speed control module according to the end preparation state signal, use the preset deceleration curve control algorithm to smoothly reduce the motor speed, and determine that the motor operating speed has reached the end winding speed value.

[0127] According to the end preparation state signal, a high-level output is generated, and the motor controller receives the high-level signal to obtain information for control and startup. The motor speed control module processes the end preparation state signal and transmits it to the programmable logic controller through the bus to obtain a set of instruction data. Through table lookup, the function relationship expression y=f(x) of the preset deceleration curve is obtained, where x represents time and y represents motor speed. The actual running speed value is obtained by measuring the motor speed information in real time through the encoder and transmitting it to the data processor through the serial port. The Kalman filter algorithm is used to denoise the actual running speed value of the motor, and the denoised actual running speed value is used as the feedback input to the PID controller. Through iterative adjustment of the PID parameters, the best PID control instruction is obtained. The PID controller combines the motor speed value obtained from the preset deceleration curve and the denoised actual running speed value. If the difference between the two is greater than the set value, the alarm module is triggered for early warning. A prediction model is established using the BP neural network algorithm, which uses the actual running speed set obtained through PID control for training. If the predicted speed value meets the end winding speed value, a stop signal is output.

[0128] According to the end preparation state signal trigger, a high level output is generated, for example, a voltage higher than 3 volts, and the motor controller receives the high level signal, analyzes the signal as a start instruction, executes the corresponding control logic, and drives the motor to start deceleration operation. The start motor deceleration control module processes the end preparation state signal, for example, the signal is transmitted to the programmable logic controller (PLC) through the CAN bus protocol in a specific message format, the PLC receives the message, analyzes the message content, extracts the key information, such as the current running stage identifier and the start deceleration flag, etc., and stores it in the designated data register, and finally summarizes it as an instruction data set.

[0129] The content of the instruction data set is as follows: including data type code, version number, length, data item, etc. For example, the data in the table is 8 for the end speed, 432 for the distance, 17 seconds for the running time, and 10 for the initial speed. According to the above parameters, the motor start deceleration is set to 0x111 of data item 1, and the start speed is set to 0x110 of data item 2, and the version number is defined as 0x1. Then through the preset deceleration curve table, the function expression that meets the start state is found as:

[0130] y=-5x^2+10, x represents time (unit: second), y represents motor speed (unit: revolution / minute),

[0131] The expected speed set of the motor {5, 8, 5, 2, 0} is calculated, which corresponds to the situations when x = {1, 2, 3, 4, 5}, and the calculation is based on the total running time of 17 seconds. The speed set represents the speed values at 5 different time points during the ideal deceleration and stopping process of the motor. At the same time, the motor speed is collected every 10 milliseconds by the encoder, and a series of speed data such as {4, 1, 6, 2, 1} is obtained and transmitted to the processor through the serial port. Using the Kalman filtering algorithm, the initial estimated value is set to 5, the prediction error covariance is set to 1, and the measurement noise covariance is set to 1. Through the Kalman filtering formula: prior estimation x(k|k-1) = x(k-1|k-1) + 0, state prediction mean square error P(k|k-1) = P(k-1|k-1) + 1, the Kalman gain K(k) = P(k|k-1) / (P(k|k-1) + 1) is calculated, and the motor speed estimation value x(k|k) = x(k|k-1) + K(k)*(z(k) - x(k|k-1)) and the estimation mean square error P(k|k) = (1-K(k))*P(k|k-1) are further calculated. The measured speed is filtered. The filtered speed is fed back to the PID controller, the proportional coefficient is set to 8, the integral coefficient is set to 2, the differential coefficient is set to 1, and the maximum deviation threshold is set to 5. The adjusted PID parameters are 0, 15, and 05, respectively. When the PID control algorithm combines the motor speed 5 and the filtered actual speed 4, the difference is 1, which is less than the set threshold of 5, no alarm is triggered, but when the difference reaches or exceeds 5, the alarm module is started to output an alarm.

[0132] The actual speed set obtained after adjustment is used as a training sample to initialize a BP neural network model, such as containing 3 input nodes (time, target speed, actual speed), 5 hidden layer nodes, 1 output node (predicted speed), using Sigmoid activation function, and performing 500 times of training iteration. If the predicted speed value has reached and stabilized at the rated speed requirement of 5 revolutions per minute in the final stage, output the stop command to stop the motor. The current state data of the device is reported to the monitoring system through the network protocol to ensure that the running state is real-time mastered by the remote monitoring platform, so as to facilitate centralized state supervision and problem diagnosis, and provide real-time and reliable information support for formulating production scheduling scheme and equipment maintenance plan, further optimizing the production process scheduling.

[0133] In step S4, when the motor runs at low speed and the length of the wound optical fiber reaches the value of the end trigger length minus a safety margin, the winding control unit obtains the current position information of the winding mechanism through an optical sensor installed at a specific position, and determines the angle of the optical fiber relative to a specific reference position in the end winding stage.

[0134] According to the above business content and related attributes, the detailed business solution process is proposed as follows: the encoder installed on the motor periodically records the speed and generates a data stream combined with the time parameter. This data stream is temporarily stored through the data buffer pool. The neural network model adjusts its parameters in real time through the self-learning process based on the historical running speed data and establishes the running speed judgment benchmark. The model compares the current data with the learned judgment benchmark to determine the running speed value and divide different working stages for different values. The running speed less than a certain value is divided into a low-speed running stage.

[0135] The running state value enters the judgment benchmark data stream as a state label for subsequent analysis and processing. The number of winding turns is calculated according to the preset parameters of the optical fiber, the initial turn value is set, and the end turn value and the reel circumference are obtained by combining the information collected by the photoelectric sensor. The total length at the current time is obtained by multiplication, and the actual remaining length is obtained by subtraction. The comparison length is calculated by subtracting the end trigger length setting value from the safety margin setting value. The actual remaining length and the comparison length value are compared using the preset rules to determine whether the remaining length is within the end trigger range.

[0136] If the length falls within the range, an end trigger instruction is generated. The preset rule obtains the end trigger instruction, and the end stage control logic obtains the control parameters by analyzing the data packet in the instruction and performs data classification processing. The classified control instruction is distributed to the designated device. When the photoelectric sensor detects the trigger information, the data is stored in the cache area. The photoelectric sensor installed at a specific position of the winding device detects the periodic rotation signal of the reel and collects the rotation data sequence. The photoelectric sensor converts the position data into an analog electric signal corresponding to one rotation period of the winding device.

[0137] The Fourier transform is used to analyze the acquired analog electric signal in the frequency domain to obtain harmonic components and main frequency.

[0138] The total length of the optical fiber and the number of winding turns of the device are calculated, and the corresponding length information of each turn is obtained by calculation. The remainder value is multiplied by the time corresponding to the circumference of each turn and the phase corresponding to the circumference of each turn to obtain the conversion angle value. The total phase angle is obtained, and the remainder part is converted into a radian value based on a specific reference point benchmark. A numerical calculation model is preset, which real-time collects the Fourier transform values and clock data as input data, and uses difference calculation to obtain the value change in a period of time.

[0139] The time sequence collects data through the above multiple steps and performs data fusion. The time correlation is obtained by comparing the photoelectric sensor information with the reference time sequence through the time window, and the phase shift value is obtained by combining the corresponding mechanical structure movement amplitude information of the reference time sequence, so as to obtain the time parameter. The total angle is obtained by calculating the product of the time value and the angular velocity value. According to the time parameter, the current instantaneous angular velocity of the motor and the corresponding optical fiber length value are obtained. The system updates the trigger length threshold of the end stage in real time according to the optical fiber length, and uses the optical fiber category data in the established database as the judgment object. The end parameter value of each material is obtained, which is sent to the motion controller through the data interface. And combined with the mechanical structure characteristic value of the optical fiber winding equipment, comprehensive comparison and analysis are carried out. The actual position deviation value is determined by the photoelectric sensor output value and the internal set control parameter. The next period winding position information is predicted by the recurrent neural network, and multiple future time prediction points are obtained.

[0140] The position change trend and change speed are obtained by subtracting the prediction point from the current value, and the trend data is used as feedback control data in real time. According to the deviation value, the motor operating parameters are adjusted through the fuzzy control algorithm, and the photoelectric sensor monitors a preset time value to store the time and value data in the historical database. Then the historical database stores different types of data. So as to achieve speed adjustment.

[0141] The encoder installed on the motor records the speed at a frequency of 1000 times per second, for example, at a certain moment, the recorded speed is 3000 rpm, and the time parameter at this moment is 10:00:00 to generate a data stream. This data stream is temporarily stored in a data buffer pool with a capacity of 10MB. The neural network model records the data of the past month as training samples at a rate of one record per minute, a total of 43200 data, assuming that the effective data is a total of 40000, these data are imported into the self-built neural network model, and after 1000 cycles of training, a total of 1000 cycles of training, the parameters are adjusted in real time during the training process through the loss function method of cross entropy loss function, and the running speed judgment benchmark is established, for example, a reference value in the benchmark is 2800 rpm. The model compares the current data 3000 rpm with the learned judgment benchmark to determine the running speed value as a normal value. And the neural network will demarcate different working stages for different values, such as less than 1000 rpm, which is demarcated as a low-speed running stage. The running state value enters the judgment benchmark data stream as a state label for subsequent statistics, regression or fault diagnosis.

[0142] The total fiber length is set to 5000 meters. Based on the preset fiber parameters, the required fiber length per winding is set to 2 meters, resulting in a calculated number of windings of 2500. The initial winding value is set to 0. Combined with the per-winding marker information collected by the photoelectric sensor, the final winding value is 2400. For example, if the winding reel circumference is 2 meters, multiplying the final winding number by the reel circumference yields a total fiber length of 4800 meters after 2400 windings. Subtracting 4800 from 5000 yields an actual remaining length of 200 meters. Subtracting the final trigger length setting, for example, 250 meters, from the safety margin setting, for example, 20 meters, yields a comparison length of 230 meters. A comparator then compares 200 with 230. Since 200 meters is less than 230 meters, the remaining length is determined to be within the final trigger range, generating a final trigger command.

[0143] The system generates a finishing trigger instruction, and the control logic parses the data packet in the instruction to obtain control parameters such as length, number of turns, and rotation speed, and performs data classification processing to distribute the control instruction to execution devices such as motors and PLC controllers. When the photoelectric sensor detects the trigger information generated by the winding drum rotating to the preset position, the data is stored in a cache area with a capacity of 5MB. The photoelectric sensor installed at a specific position of the winding device detects the periodic rotation signal of the winding drum, and generates a pulse signal for each turn at a frequency of 100 per second to collect the rotation data sequence. According to the change of the coordinate data of a specific position when the winding rotates one turn, such as a change of 10 centimeters in the x coordinate or a change of 15 centimeters in the y coordinate, the photoelectric sensor converts the detected position coordinate data into an analog electrical signal corresponding to one cycle of the winding device rotation, and then uses a 12-bit analog-to-digital converter to convert the signal. The obtained analog electrical signal is analyzed in the frequency domain using Fourier transform to obtain harmonic components and main frequency, such as using the fast Fourier transform algorithm to analyze and identify the main frequency, harmonic amplitude, and phase data. Using a total fiber length of 5000 meters and dividing by the total number of turns of the winding drum of 2500 turns, the length of the fiber corresponding to each turn is calculated to be 2 meters, and the remaining length of 200 meters of fiber divided by the length of the fiber corresponding to each turn of 2 meters gives a quotient of 100 and a remainder of 0. The remainder value of 0 multiplied by the length corresponding to each turn of 2 meters and the phase corresponding to each turn of the circumference, such as π, gives a conversion angle value of 0. The remainder part of the total phase angle is converted into a radian value under a specific reference point, such as π. A numerical calculation model is preset, which uses the fast Fourier transform values and clock data as input data, uses difference calculation to obtain the value change in a period of time, and then uses methods such as Runge-Kutta to perform numerical calculation in a fixed time window and update the Fourier transform parameters to obtain phase angle and frequency values.

[0144] The time series collects data from the above steps and fuses these data into a unified data stream. Then, by comparing the photoelectric sensor information with the reference time series through a time window of 10 units in length, the time correlation value is obtained. For example, the correlation of two sequences is 95.

[0145] In combination with the reference time series corresponding to the mechanical structure movement amplitude information, such as a maximum movement amplitude of 50 millimeters, the phase shift value is obtained. The product of the time value and the angular velocity value is calculated to obtain the total angle. Then, through the time information read by the encoder, the corresponding angular velocity is 100 revolutions per second, and they are multiplied to obtain the total rotation angle of the winding drum, which is 100 turns. Then, the radian value corresponding to each turn is multiplied by the value to obtain the total rotation angle value.

[0146] According to the time parameter, for example, the time is 2 o'clock in the afternoon, so that the current motor instantaneous angular velocity is obtained, for example, 120 revolutions per second, and the corresponding optical fiber length value is 3 meters. The system updates the trigger length threshold of the tailing stage according to the current optical fiber length 200 meters, and performs numerical calculation on the optical fiber species data in the established database, such as its elastic coefficient, density and the like, as a judgment object, and combines the mechanical structure characteristic value of the optical fiber winding device, such as the diameter of the wire reel, the maximum speed and the like, to perform comprehensive comparison and analysis. The tailing parameter value of each material is obtained, and this value is transmitted to the motion controller through the data interface of the hub and the optical fiber. The motion controller combines the real-time position data, for example, the current is the 100th turn, compares the output value of the optical sensor and the internal set control parameter through the photoelectric sensor, determines the actual position deviation value, for example, the deviation between the current position and the theoretical position is 5 centimeters. The next period winding position information is predicted by using the recurrent neural network, and the future time point position model is established by using the Archimedes spiral line. The recurrent neural network generates data of 10 future time prediction points according to the first 1000 historical data. For example, the predicted coordinates of the next 5 time points are (1, 2), (2, 3), (3, 4), (4, 5) and (5, 6). The predicted points and the current value are subtracted by using the vector calculation method to obtain the position change trend and the change speed, and the trend data is used as the feedback control data in real time.

[0147] According to the 5 centimeter deviation value obtained in the previous step, the motor operating parameters are adjusted by using the fuzzy control algorithm, for example, the proportional, integral and differential coefficients of the speed adjustment are obtained, which are 5, 3 and 2 respectively. The optical sensor monitors a preset time length value, and the time and numerical data are stored in a historical database with a capacity of 1GB. The historical database classifies and stores different types of data, for example, time data is stored in a time database, and numerical data is stored in a numerical database. Through the data, the speed adjustment is guided.

[0148] In step S5, according to the difference between the current angle of the real-time winding mechanism and the target angle, the winding control unit calculates the winding motor speed correction coefficient by using the position loop-based speed correction algorithm, obtains the control instruction for the optical fiber tailing winding, and performs accurate position control on the tailing process.

[0149] According to real-time detection, current angle information and set target angle information of the winding mechanism are obtained. According to the current angle information and the target angle information, an angle difference value is calculated. When the angle difference value exceeds a preset range, the pre-warning database is entered. The angle difference value is read from the pre-warning database. If the angle difference value is lower than a preset lower limit range, a speed correction step is skipped. If the angle difference value is higher than a preset upper limit range, alarm information corresponding to the angle difference value is generated and is notified to a display module to display information. A control instruction is generated to pause, and the current angle and the target angle are read from the pre-warning database. The angle difference value is calculated by a position loop-based control algorithm to obtain a speed correction coefficient.

[0150] According to the speed correction coefficient, the output torque of the current winding motor is corrected to obtain real-time torque control information after correction. The real-time torque control information is adaptively compensated according to a preset winding target length of the optical fiber to obtain a final control instruction.

[0151] An environment perception database is established for a real-time winding environment. The final control instruction is issued to a winding control execution mechanism. The execution mechanism receives the final control instruction to control the winding motor to perform optical fiber end winding work. If the end state parameter in the final control instruction shows stop, the execution mechanism pauses winding, and the control instruction is updated to pause. Then, working data update information in the environment perception database is obtained to obtain new winding target length information, and corresponding control instruction information is read from the fifth step to resend the instruction.

[0152] If the end state parameter in the final control instruction does not show stop, the step is continued. Online detection is performed according to a winding mechanism execution state real-time data information library to determine the motion state of the end winding work and save it.

[0153] The angle of the winding mechanism is detected in real time. For example, the current angle is 30 degrees, and the set target angle is 90 degrees. The angle difference value is calculated to be 60 degrees. Assuming that the preset angle difference value range is plus or minus 5 degrees, and the current angle difference value 60 degrees is far beyond the preset range, the 60-degree difference value data is stored in the pre-warning database. Then, the system reads the difference value from the pre-warning database. The 60-degree difference value is far beyond the preset upper limit range of 5 degrees. Therefore, a specific alarm information is generated, for example: “Warning: angle deviation is too large, actual deviation is 60 degrees, which has exceeded the preset upper limit of 5 degrees!” The information is sent to the display module, and the staff can see the specific warning content.

[0154] At the same time, the generation of new control instructions is suspended, and the current angle of 30 degrees and the target angle of 90 degrees are obtained again from the early warning database for subsequent processing. The angle difference of 60 degrees is calculated through a proportional-integral-derivative control algorithm based on the position loop. For example, the proportional coefficient Kp is set to 5, the integral coefficient Ki is set to 1, and the derivative coefficient Kd is set to 01. According to the control algorithm formula: output = Kp * error + Ki * integral error + Kd * differential error, a speed correction coefficient of 6 is obtained after calculation. According to the speed correction coefficient 6, the current output torque of the winding motor is corrected. Assuming that the current output torque is 10 newtons, the corrected torque is 10 * 6 = 6 newtons, and the corrected real-time torque control information 6 newtons is obtained.

[0155] In combination with the preset target length of the optical fiber winding, such as the target length of 100 meters remaining, adaptive compensation is performed to reduce errors. This can be achieved by adding an adaptive component using the least mean square algorithm (LMS) to adjust the speed coefficient based on the deviation from the target winding length.

[0156] For example, given the current length and target length, we can define the length deviation as follows: length deviation = current length - target length. The current length is 110, so the length deviation = 110 - 100 = 1. Then we will use this length error as the input of the adaptive component. For example, multiply the previously obtained real-time torque control information by an adaptive compensation coefficient of 1 to obtain the final control instruction of 6 newtons * 1 = 6 newtons. Subsequently, the environmental perception database is established, and environmental data such as temperature 25 degrees Celsius, humidity 60%, air pressure 101 kilopascals, as well as control instruction sending time, device number, and other information are recorded. The final control instruction 6 newtons is sent to the actuator. After receiving the torque instruction of 6 newtons, the actuator accurately controls the winding motor to work.

[0157] If the end state parameter in the final control instruction shows stop, the actuator suspends the work of the winding motor, and the control instruction is updated to suspend. The new winding target length is updated to 50 meters from the environmental perception database, so adaptive compensation is performed on the previously obtained corrected real-time torque control information 6 newtons to reduce errors. Assuming that an adaptive compensation coefficient of 2 is obtained, the final control instruction 2 newtons is obtained.

[0158] The updated data is sent to the actuator as the final control command, allowing it to perform the task according to the new parameters. Otherwise, the current step is continued, such as the end state parameter is 1, then maintain the torque output of 6 newton meters. At the same time, the information recorded by the real-time data information base of the winding mechanism is executed, such as motor speed, winding length, winding tension 2 newton, etc. Through these data analysis, it is determined whether the current winding state is normal, and the result of each analysis, for example, "device A on October 27, 2023, 10:30, winding work state is normal, cumulative winding length 500 meters." Is saved.

[0159] Step S6, if the position control calculation model output shows that the current angle of the winding mechanism has reached or exceeded the target angle, the winding control unit starts the preset neural network prediction model. The neural network model learns from historical operation data to analyze and learn the early parking position information of the end parking position, and determines the amount of early parking signal that needs to be sent out by the end control link.

[0160] The current angle information of the winding mechanism is obtained, and the difference between the current angle information and the target angle information is judged. If the difference between the current angle information and the target angle information is greater than the set value, the winding control unit outputs a signal to the neural network prediction model. The neural network prediction model obtains the signal from the winding control unit, then retrieves historical operation data from the database for analysis. According to the analysis result, the early parking position information of the end parking position is obtained. According to the early parking position information of the end parking position, the amount of early parking signal of the end control is determined. The winding control unit receives the parking signal amount and controls the winding mechanism to perform the corresponding operation. The database receives and stores the angle information, signal amount and winding mechanism information.

[0161] The winding mechanism detects its current angle in real time through the built-in angle sensor, for example, the current angle is 80 degrees. The winding control unit has a preset target angle of 180 degrees. The comparator in the control unit calculates the difference between the two values, which is 100 degrees. The system sets a threshold, for example, 10 degrees. Only when the angle difference is greater than the threshold, the next operation will be triggered. In this example, 100 degrees is greater than 10 degrees, so the winding control unit outputs a high-level signal, for example, 5 volts, to the neural network prediction model. After receiving the 5-volt signal, the neural network prediction model activates the model to work. It queries the database to retrieve 1000 historical data of winding operations. These data include the starting angle, target angle, actual parking angle of each winding operation, and the advance of the stop signal corresponding to each operation. Then, through timestamp filtering and analysis, abnormal time data is excluded, including recorded date, time, etc. For example, if it is January 1, 2024, 12:00 noon, the data from 8:00 to 12:00 will be filtered out, for example, 30 data. Through timestamp, each data is added with time decay. The last time before noon is counted as 80% of the latest time period of the winding machine. The second last time before the last time is decayed by 40%. The third last time is decayed by 20%. The fourth last time is decayed by 10%.

[0162] The remaining data are all excluded and then calculated through the neural network model. This model can be established through a long short-term memory network model.

[0163] This neural network prediction model can analyze the parking position of the winding mechanism under similar conditions based on the obtained data, and then analyze the advance of the stop signal to obtain the required prediction value. In this example, it may be found that the stop signal needs to be sent 5 degrees before reaching the target angle to accurately stop at the 180-degree position. Based on this result, the stop advance is set to 5 degrees. Once the winding control unit detects that the current angle reaches 175 degrees (180 degrees minus 5 degrees of advance), it immediately sends a stop signal to the winding mechanism. After receiving the signal, the winding mechanism stops precisely at the target position of 180 degrees through the built-in braking system. The winding machine gradually slows down until it stops precisely at the target angle of 180 degrees, ensuring that the winding angle meets the process requirements. This series of operation data, including the starting angle of 80 degrees, the target angle of 180 degrees, the actual parking angle of 180 degrees, the stop advance of 5 degrees, and the motor current and winding speed of the winding mechanism, are all recorded and stored in the database, providing data support for subsequent prediction and optimization.

[0164] Step S7, according to the end control link shutdown advance given by the prediction model and the preset end path model, the PLC outputs the shutdown signal to the driver, so that the winding motor stops in a low-vibration deceleration mode, avoids causing the optical fiber to break, and obtains a complete winding product.

[0165] According to the prediction model output end stage shutdown advance data, the following steps are performed: obtaining the prediction model output end stage shutdown advance data, and obtaining the motor control instruction according to the fusion of the shutdown advance data and the preset end stage path model information.

[0166] The winding machine driver obtains the fused motor control instruction, and drives the winding motor to operate to a target speed according to the current winding motor speed data, the target speed being lower than one-half of the initial winding speed. Through a preset time sequence, the time sequence parameters are obtained from a database, the optical fiber state sensor group obtains the optical fiber diameter data in operation, and the optical fiber tension sensor group obtains the real-time tension data. The historical time data is used to estimate the running state. When the change amplitude of the optical fiber diameter in at least one time is greater than a set threshold, or when the optical fiber tension in at least one time is greater than a threshold, the optical fiber state monitoring module performs shutdown judgment, triggers the optical fiber abnormal state data marking action, determines the specific abnormal marking after the abnormal marking is recognized by the neural network, and then records the optical fiber abnormal state data marking result. In the end stage, the optical fiber abnormal state data marking result is collected, and whether the number of all abnormal markings is 0 is analyzed. If the number of abnormal markings is greater than 0, the running parameter planning of the end control link is re-performed based on the abnormal marking.

[0167] Based on the shutdown lead time in the finishing control phase, a control signal is generated and sent to the motor control phase, which adjusts the control curve. If the number of abnormality flags is 0, the control phase parameter regeneration mechanism is not triggered, the control phase parameters remain unchanged, and subsequent operations continue. During the finishing phase, a convolutional neural network, modeled based on the relationship between fiber finishing process parameters, is used to determine whether secondary winding is necessary. If abnormal fiber status data is flagged and a product defect is present, the secondary winding control phase is initiated. Based on the defect type and location data and a defect elimination strategy, a new winding control instruction set is planned. The actual motor operation control instructions are calculated based on the control instruction set and the product defect location coordinates. Otherwise, secondary winding is not performed. The winding machine driver receives the control signal after secondary winding and transmits it to the winding machine motor drive unit via a signal transmission device using industrial Ethernet. The drive unit determines the winding mode and low-vibration mode for the winding machine during the finishing phase and performs the winding process. The winding machine executes this control signal during the final phase until the product integrity sensor determines that the product is completely wound. The integral value is calculated using the integrity calculation function: I = ∫ t0 to t1 f(x) dx, where I represents the degree of completeness, f(x) is the function of completeness with respect to time, and t0 and t1 represent the duration of the winding phase. The start and end times of the winding phase are set by a program that includes a timer. If the winding result is incomplete, a control signal is generated and transmitted to the motor execution unit via a transmitter, allowing winding to continue. After obtaining a completely wound product, the prediction model is combined to obtain the final data of the actual wound product. Statistics are collected during the final winding process to obtain the actual final data results, and a data comparison chart is generated.

[0168] Assume that the prediction model outputs a 5-meter lead time for the final stage of the shutdown. First, the lead time data is obtained and fused with the preset path model information for the final stage. For example, the path model information indicates that the motor needs to decelerate according to a specific curve in the final stage. The fused control instruction combines the lead time with this curve. For example, if the initial speed of the winding motor is set to 1000 rpm, the target speed is set to 450 rpm, and the deceleration time is set to 5 seconds, the fused instruction will smoothly reduce the speed to 450 rpm within 5 seconds during the last 5 meters of winding.

[0169] To improve stability and accuracy, smooth transitions are used, such as using S curves, and the deceleration model can be calculated using S curves in control theory, and the formula can be simplified as V(t) = V0 + (Vt - V0) * (1 + exp(-k(t - t0)))^(-1), where V(t) is the speed at time t, V0 is the initial speed, which is 1000 revolutions per minute here, Vt is the target speed, which is 450 revolutions per minute here, k is a constant that can be adjusted according to the sharpness of deceleration, which can be k = 1 here, t0 = 5 seconds, which is used to control the shape of the S curve and the transition period, and V(t) = 1000 + (450-1000) * (1 + exp(-1*(t-5)))^(-1) is an S curve describing the change of speed with time. The winding machine driver obtains this fused motor control command and drives it in combination with the current winding motor speed data. For example, by installing an encoder on the motor shaft to obtain the speed, real-time speed feedback data can be obtained, so that it can be compared with the control curve. Set the preset timing parameters, such as collecting data every 10 milliseconds, and set the control period to 10 milliseconds. The optical fiber state sensor group and the optical fiber tension sensor group monitor the optical fiber diameter and tension in real time.

[0170] When the sensor detects that the fiber diameter changes more than the set threshold of 02 millimeters at a certain moment, or the fiber tension is greater than the set threshold of 5 newtons, the fiber state monitoring module judges that it is abnormal and stops. At this time, the fiber abnormal state data marking action is triggered, such as marking the current data as "abnormal diameter change too large" or "abnormal tension too large". These abnormal marks are identified by a pre-trained neural network model, which can be based on a convolutional neural network or a recurrent neural network, and has been trained on a large amount of historical data. For example, a neural network with an identification accuracy of 98% has been trained and can be used, the neural network training data set is 10000 samples, the neural network includes 1 input layer, 3 convolutional layers, 1 fully connected layer, and 1 output layer. Then further determine the specific abnormal mark, such as refining to "abnormal diameter surge" "abnormal diameter sudden decrease" "abnormal tension spike" and so on. According to the determined specific abnormality, modify the record in the winding database. If the number of abnormal marks is greater than 0, such as 3 "abnormal tension spikes", in order to compensate for the additional length requirement caused by the winding quality, the shutdown advance may be appropriately lengthened. According to the abnormal mark, the running parameters of the finishing control link are re-planned, such as adjusting the shutdown advance to 8 meters, adjusting the parameters in the S curve formula, such as adjusting the formula V(t) = 1000 + (450-1000) * (1 + exp(-5*(t-5)))^(-1). Using a convolutional neural network based on a convolutional neural network, containing 2 convolutional layers, 2 pooling layers, and 1 fully connected layer, a fiber finishing process parameter relationship model has been trained through historical production data, with a prediction accuracy of 99%, the convolutional neural network judges whether there is fiber abnormal state data marking result and product defect. If the judgment is "yes", secondary winding is performed, and the winding machine is controlled to the vicinity of the previously marked abnormal data for repair, such as tighter winding or smaller wire diameter, so as to compensate for the defect and make the final winding product meet the quality standard, or if the defect cannot be repaired, the product will be marked as a defective product. Otherwise, no secondary winding operation is performed, and a prompt message is generated to the main control program. Using industrial Ethernet protocols such as PROFINET or EtherCAT, control signals are transmitted to the winding machine motor drive unit. After receiving the signal, the drive unit controls the winding machine to perform the finishing stage according to the "winding mode" and "low vibration mode". If the winding mode is normal and the low vibration mode is normal. The winding machine executes the control signal until the product completeness sensor judges that the product is complete, and the completeness is calculated by the completeness calculation function integral value, for example, when the integral value I reaches the set value 100, it is considered that the product is complete, and the winding time is from 1 second to 10 seconds.The fullness function assumption about time can be simplified as f(x)=10x, and the integral value calculation is I=∫1 to 10 f(x)dx, and I=495 is obtained, which meets the product winding fullness judgment condition.

[0171] When the product is not complete, the winding is continued. After obtaining the complete winding product, the predicted model and the actual ending data are combined, and a data comparison chart is generated by comparing the differences in the aspects of early stoppage, motor speed, optical fiber diameter, tension and the like.

[0172] The above-described embodiments are only preferred technical solutions of the present application, and should not be regarded as a limitation of the present application. The protection scope of the present application should be based on the technical solutions recited in the claims, including equivalent replacement solutions of the technical features recited in the claims. That is, equivalent replacement improvements within this scope are also within the protection scope of the present application.

Claims

1. A method for controlling the tension of optical fiber winding based on PLC, characterized in that The method comprises the following steps: S1: According to the preset total length of the optical fiber and the real-time winding length information fed back by the current encoder, the residual length of the optical fiber is calculated by a length decreasing model; S2: If the residual length of the optical fiber is less than or equal to the preset ending trigger length threshold, the winding control unit switches from the normal winding state to the ending preparation state, and a winding control unit about to perform ending operation signal is obtained; S3: According to the ending preparation state signal, a motor speed reduction control module is triggered, a preset speed reduction curve control algorithm is used to smoothly reduce the motor speed, and it is determined that the motor running speed has reached the ending winding speed value; S4: When the motor runs at a low speed and the wound optical fiber length reaches a value obtained by subtracting a safety margin from the ending trigger length, the winding control unit obtains the current position information of the winding mechanism through an optical sensor installed at a preset reference position, and determines the angle of the optical fiber relative to the preset reference position in the ending winding stage; S5: According to the difference between the real-time winding mechanism current angle and the target angle, the winding control unit calculates the winding motor speed correction coefficient through a position loop based speed correction algorithm, obtains the control instruction for the optical fiber ending winding, and performs accurate position control on the ending process; S6: If the position control calculation model output shows that the current angle of the winding mechanism has reached or exceeded the target angle, the winding control unit starts the preset neural network prediction model, the neural network prediction model analyzes and learns the ending parking position advance information from the historical operation data, and determines the amount of advance stop signal that needs to be sent by the ending control link; S7: According to the ending control link stop advance amount given by the prediction model and the preset ending path model, the PLC outputs the stop signal to the driver, so that the winding motor stops in a low-vibration speed reduction mode, avoids causing the optical fiber to break, and obtains a complete winding product.

2. The method for controlling the tension of the optical fiber based on the PLC according to claim 1, characterized in that: The sub-steps of S1 are: S2.1: If the optical fiber residual length data value is less than or equal to the preset ending trigger threshold length value; S2.2: The winding control unit state data is updated to 1; S2.3: According to the winding control unit state data, when the winding control unit state data is 1, the winding control unit switches to the ending preparation state; S2.4: The winding control unit obtains the ending preparation state data; S2.5: The winding control unit enters the ending preparation state; S2.6: The winding control unit generates an ending operation pre-trigger signal data; S2.7: The winding control unit generates an ending operation pre-trigger signal data according to the ending operation pre-trigger signal data; S2.8: The winding control unit runs a self-checking module to obtain a device self-checking normal signal data; S2.9: The device self-checking normal signal data is used as the final ending operation precondition; S2.10: The ending signal is triggered and output, and the winding device performs the optical fiber ending operation.

3. The method for controlling the tension of the optical fiber based on the PLC according to claim 1, characterized in that: The sub-steps of S2 are: S2.1: If the optical fiber residual length data value is less than or equal to the preset ending trigger threshold length value; S2.2: The winding control unit state data is updated to 1; S2.3: According to the winding control unit state data, when the winding control unit state data is 1, the winding control unit switches to the end preparation state; S2.4: The winding control unit obtains the end preparation state data.

4. The method for controlling the tension of the optical fiber based on the PLC according to claim 1, wherein: The sub-steps of S3 are: S3.1: According to the end preparation state signal trigger, a high-level output is generated, and after the motor controller receives the high-level, it obtains information for control start; S3.2: The motor deceleration control module processes the end preparation state signal and transmits it to the programmable logic controller through the bus to obtain a set of instruction data; S3.3: Through the table lookup method, the function relationship expression y=f(x) of the preset deceleration curve is obtained, where x represents time and y represents motor speed, and a set of predicted motor speeds is obtained; S3.4: The encoder real-time measurement motor speed information is obtained and transmitted to the data processor through the serial port to obtain the actual running speed value; S3.5: The Kalman filter algorithm is used to denoise the actual running speed value of the motor, and the denoised actual running speed value is used as the feedback input into the PID controller, and the optimal PID control instruction is obtained by iteratively adjusting the PID parameters; S3.6: The PID controller fuses the motor speed value obtained from the preset deceleration curve and the denoised actual running speed value, and if the difference between the two is greater than the set value, the alarm module is triggered for early warning; S3.7: A prediction model is established using the BP neural network algorithm, which is trained using the actual running speed set obtained by PID control, and determines whether the predicted speed value meets the end winding speed value, and outputs a stop signal.

5. The method for controlling the tension of the optical fiber based on PLC according to claim 1, wherein: The sub-steps of S4 are: S4.1: The encoder installed on the motor periodically records the speed and generates a data stream combined with the time parameter; S4.2: This data stream is temporarily stored in the data buffer pool; S4.3: The neural network prediction model adjusts its parameters in real time through a self-learning process based on historical running speed data and establishes a running speed judgment benchmark; S4.4: The model compares the current data with the learned judgment benchmark to determine the running speed value and divide different working stages for different values, and the running speed is less than a certain value, which is classified as a low-speed running stage; S4.5: The running state value is used as a state label in the judgment benchmark data stream for subsequent analysis and processing; S4.6: The number of winding turns is calculated according to the preset parameters of the optical fiber, the initial turn value is set, and the end turn value and the winding reel circumference are obtained by combining the information collected by the photoelectric sensor, and the total length at the current time is obtained by multiplication, and the actual remaining length is obtained by subtraction; S4.7: The end trigger length is set to the difference between the safety margin set value, and the comparison length is calculated; S4.8: The actual remaining length and the comparison length value are compared using the preset rules to determine whether the remaining length is within the end trigger range; S4.9: If the length falls within the range, an end trigger instruction is generated; S4.10: preset rule obtains end trigger instruction, end stage control logic obtains control parameters by analyzing data packet in instruction and performs data classification processing, and distributes classified control instruction to designated device; S4.11: when photoelectric sensor monitors trigger information, store trigger information in cache area; S4.12: photoelectric sensor installed at preset reference position of winding device detects periodic rotation signal of winding disc, and collects rotation data sequence; S4.13: photoelectric sensor converts position data into analog electric signal of one rotation period of winding device; S4.14: frequency domain analysis is performed on obtained analog electric signal by using Fourier transform, and harmonic component and main frequency are obtained; S4.15: total length of optical fiber and winding turns of equipment are calculated, and length information corresponding to each turn is obtained by calculation, and quotient and remainder are obtained by dividing residual length value of optical fiber by length value of each turn; S4.16: remainder value is multiplied by time corresponding to circumference of each turn and phase corresponding to circumference of each turn to obtain conversion angle value; S4.17: total phase angle is obtained, and remainder part is converted into radian value under reference of specific reference point; S4.18: a numerical calculation model is preset, the numerical calculation model collects Fourier transform value and clock data as input data in real time, and a time value change is obtained by using difference calculation; S4.19: time sequence collects all data obtained by S4.1-S4.18, and data fusion is performed; S4.20: time correlation is obtained by comparing photoelectric sensor information and reference time sequence through time window, and phase shift value is obtained in combination with movement amplitude information of mechanical structure corresponding to reference time sequence, so that time parameter is obtained; S4.21: total angle is obtained by multiplying time value and angular velocity value; S4.22: current motor instantaneous angular velocity and corresponding optical fiber length value are obtained by judging according to time parameter; S4.23: system updates trigger length threshold of end stage in real time according to optical fiber length, and uses optical fiber type data in established database as judgment object; S4.24: end parameter value of each material is obtained, and the end parameter value is sent to motion controller through data interface; S4.25: comprehensive comparison and analysis are performed in combination with mechanical structure characteristic value of optical fiber winding equipment; S4.26: actual position deviation value is determined by output value of photoelectric sensor and internal set control parameter; S4.27: next period winding position information is predicted by using recurrent neural network, and multiple future time prediction points are obtained; S4.28: position change trend and change speed are obtained by subtracting current value from prediction point, and trend data is used as feedback control data in real time; S4.29: motor working parameter is adjusted according to deviation value by using fuzzy control algorithm, and time and value data are stored in historical database in synchronization by monitoring preset time value by photoelectric sensor; S4.30: different storage type data are stored in historical database in classification; so as to achieve speed adjustment.

6. The method for controlling the tension of the optical fiber based on the PLC according to claim 1, wherein: The sub-steps of S5 are: S5.1: Obtain current angle information of the winding mechanism and set target angle information according to real-time detection; S5.2: Calculate the angle difference value according to the current angle information and the target angle information, and enter the early warning database when the angle difference value exceeds the preset range; S5.3: Read the angle difference value from the early warning database, and skip the speed correction step if the angle difference value is below the preset lower limit range; S5.4: If the angle difference value is higher than the preset upper limit range, generate the alarm information corresponding to the angle difference value and notify the display module to display the information, control the instruction generation to pause, and read the current angle and target angle from the early warning database; S5.5: Calculate the speed correction coefficient by using the control algorithm based on the position ring for the angle difference value; S5.6: Modify the current winding motor output torque according to the speed correction coefficient to obtain the modified real-time torque control information; S5.7: Perform adaptive compensation on the real-time torque control information according to the preset winding target length of the optical fiber to obtain the final control instruction; S5.8: Establish an environment perception database for the real-time winding environment, and issue the final control instruction to the winding control execution mechanism. The execution mechanism receives the final control instruction to control the winding motor to perform the optical fiber end winding work; S5.9: If the end state parameter in the final control instruction shows stop, the execution mechanism pauses winding, the control instruction is updated to pause, the working data update information in the environment perception database is obtained, the new winding target length information is obtained, and the corresponding control instruction information is read from S5.5 to resend the instruction; S5.10: If the end state parameter in the final control instruction does not show stop, continue the operation of this step; S5.11: Perform online detection according to the real-time data information library of the winding mechanism execution state, determine the motion state of the end winding work, and save it.

7. The method for controlling the tension of the optical fiber based on the PLC according to claim 1, wherein: The sub-steps of S6 are: S6.1: Obtain the current angle information of the winding mechanism from the winding mechanism, and judge the difference between the current angle information and the target angle information; S6.2: If the difference between the current angle information and the target angle information is greater than the set value, output the signal of the winding control unit to the neural network prediction model; S6.3: The neural network prediction model obtains the signal of the winding control unit, and analyzes the historical running data from the database; S6.4: According to the analysis result, the advance amount information of the end parking position is obtained; S6.5: According to the advance amount information of the end parking position, determine the advance amount of the end control stop signal; S6.6: The winding control unit receives the stop signal amount and controls the winding mechanism to perform corresponding operations; S6.7: The database receives and stores the angle information, signal amount and winding mechanism information.

8. The method of claim 1, wherein the method is based on PLC. The sub-steps of S7 are: S7.1: Obtain the end stage stop advance amount data output by the prediction model, and obtain the motor control instruction after fusing the stop advance amount data and the preset end stage path model information; S7.2: The winding machine driver obtains the motor control command after fusion, and drives the winding motor to run to the target speed according to the current winding motor speed data. The target speed is lower than half of the initial winding speed; S7.3: Through the preset timing, the timing parameters come from the database, the optical fiber state sensor group obtains the running optical fiber diameter data, and the optical fiber tension sensor group obtains the real-time tension data. Using historical data to estimate the running state, when the optical fiber diameter change amplitude of at least one time is greater than the set threshold, or when the optical fiber tension of at least one time is greater than the threshold, the optical fiber state monitoring module makes a shutdown judgment, triggers the optical fiber abnormal state data marking action, and determines the specific abnormal marking after the abnormal marking is recognized by the neural network. Then record the optical fiber abnormal state data marking result; S7.4: In the finishing stage, collect the optical fiber abnormal state data marking result obtained by step S7.3, and analyze whether the number of all abnormal markings is 0; S7.5: In the finishing stage, use a convolutional neural network modeled according to the relationship between the optical fiber finishing process parameters to determine whether secondary winding is needed; If there is an optical fiber abnormal state data marking result and there is a product defect, perform the secondary winding control link, plan a new winding control instruction set through the defect type and position data, and combine the defect elimination strategy. According to the control instruction set and the product defect position coordinate value, the actual motor operation control command is calculated and generated; Otherwise, do not perform secondary winding operation; S7.6: The winding machine driver obtains the control signal after secondary winding; S7.7: Through the signal transmission device, use industrial Ethernet to transmit to the winding machine motor drive unit; S7.8: The drive unit determines the winding mode and low vibration mode of the winding machine in the finishing stage to perform winding processing; S7.9: The winding machine executes the control signal in the finishing stage until the product completeness sensor judges that the product is completely wound, and the integral value is calculated through the completeness calculation function, which is I = ∫t0 to t1 f(x)dx, I represents the completeness, f(x) is the function of the completeness with respect to time, t0, t1 represents the time of the winding stage; If the winding result is not complete, generate control information to the motor execution unit through the transmitter, and continue winding; S7.10: After obtaining the complete winding product, combine the prediction model to obtain the actual winding product finishing data; S7.11: Data statistics are performed on the finishing winding process to obtain the actual finishing data and generate a data comparison chart; S7.12: Data storage and analysis: store the data in the finishing stage, and use it for subsequent analysis and improvement, to ensure that each winding process can learn from historical data, continuously optimize the control algorithm, and improve the quality and production efficiency of the product.

9. The method of claim 1, wherein the method is based on PLC. The substep of step S7.4 is: S7.4.1: If the number of abnormal markings is greater than 0, re-plan the finishing control link parameters based on the abnormal markings; S7.4.2: According to the tail control link stop advance quantity obtained in S7.4.1, a control signal is generated and sent to the motor control link, and the motor control link adjusts the control curve; S7.4.3: If the number of abnormal flags is 0, the control link parameter re-generation mechanism is not triggered, the control link parameter remains unchanged as the result of S7.4.1, and the subsequent operation continues.

10. A fiber winding tension control system based on PLC, characterized by: A PLC-based optical fiber winding tension control method according to any one of claims 1-9 is adopted.

Images