Continuous carbon fiber length-diameter ratio online regulation and control system based on tension and temperature feedback

Abstract

The invention discloses a continuous carbon fiber length-diameter ratio online regulation and control system based on tension and temperature feedback, and relates to the technical field of fiber manufacturing process control. The sensing detection module is used for acquiring fiber tension and heat treatment temperature; the decision decoupling module is used for obtaining a temperature pre-compensation instruction and a PID feedback control instruction through a tension-temperature coupling model and a controller PID algorithm and superposing the temperature pre-compensation instruction and the PID feedback control instruction to generate a composite control instruction; and the monitoring protection module is used for issuing a driving instruction according to the composite control instruction, acquiring a comprehensive stability coefficient in a driving process, and adjusting the composite control instruction according to a comparison result between the comprehensive stability coefficient and a preset stability threshold value. Therefore, effective decoupling and dynamic cooperation of the two key variables can be realized, the system can stably and quickly respond to various disturbances, the oscillation problem caused by single parameter adjustment can be avoided, and the stability of the production process is fundamentally improved.

Description

Technical Field

[0001] This invention relates to the field of fiber manufacturing process control technology, specifically to a continuous online control system for the aspect ratio of carbon fibers based on tension and temperature feedback. Background Technology

[0002] Carbon fiber, due to its high specific strength, high specific modulus, corrosion resistance, and excellent thermal stability, is widely used in aerospace, rail transportation, new energy vehicles, and high-end sporting goods. In the preparation of continuous carbon fiber reinforced thermoplastic composites (CFRTPs), the fiber aspect ratio (i.e., the ratio of fiber length to its diameter) is one of the key parameters affecting the mechanical properties of the final product. A higher aspect ratio helps improve the tensile strength, impact toughness, and fatigue life of the composite material, while fiber breakage or excessive shearing will significantly reduce its reinforcing effect.

[0003] Currently, continuous carbon fibers are typically composited with thermoplastic resins through methods such as melt impregnation, solution impregnation, or powder blending. This process involves multiple steps including traction, heating, cooling, and winding. However, in high-speed continuous production, fiber bundles are susceptible to uneven tension, localized overheating, or uneven cooling, leading to micro-fractures, fuzzing, and even fiber breakage. This causes the actual aspect ratio to deviate from the design value, severely impacting product quality consistency.

[0004] Chinese invention patent application CN120353121A discloses a cross-linked cable production control method based on an improved PID control algorithm. This method involves establishing a collaborative control architecture for temperature, pressure, and traction speed, with vulcanizing pipe temperature as the primary controlled variable and extruder pressure and traction speed as secondary controlled variables, and constructing an optimal reference model. The primary controller tracks the temperature setpoint in real time and dynamically tunes the PID parameters based on process parameters such as insulation thickness deviation and material cross-linking degree. Precise control is achieved through a three-loop synergistic coupling improved PID controller: the temperature control loop converges the steady-state error between the actual temperature and the setpoint to zero; the pressure and speed control loops converge the fluctuation amplitude to the allowable process range. This invention can improve the temperature stability, pressure uniformity, and traction synchronization in cross-linked cable production, reduce process fluctuations and defect rates, and improve product quality and production efficiency.

[0005] However, the above and similar technical solutions still have the following shortcomings: In the operation of a continuous carbon fiber production line, in order to improve the orientation and strength of carbon fibers, the speed of the drawing rollers is mostly increased to increase fiber tension. However, due to the coupling relationship between fiber tension and heat treatment temperature, the heat treatment temperature will become unbalanced during the adjustment of fiber tension, which will lead to system instability. Summary of the Invention

[0006] The purpose of this invention is to provide a continuous online control system for the aspect ratio of carbon fibers based on tension and temperature feedback, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a continuous carbon fiber aspect ratio online control system based on tension and temperature feedback, comprising: Sensing and detection module: It acquires fiber tension and heat treatment temperature through fiber tension detection unit and heat treatment temperature detection unit; Decision decoupling module: Processes the fiber tension and heat treatment temperature using a tension-temperature coupling model and a controller PID algorithm to obtain temperature pre-compensation commands and PID feedback control commands. Simultaneously, the temperature pre-compensation commands and PID feedback control commands are superimposed to obtain a composite control command, including: SB1: Constructing an internal model: Based on the fiber tension and heat treatment temperature, construct a tension-temperature coupling model; SB2: Feedforward compensation: Based on the tension difference between the preset tension target value and the current fiber tension value, a tension adjustment command is determined, and the tension adjustment command is used as the input of the tension-temperature coupling model to output the corresponding temperature pre-compensation command; SB3: Feedback Control: Based on the tension adjustment between the current fiber tension value and the preset tension target value, and the temperature adjustment between the current temperature value and the preset temperature target value, the PID feedback control command is determined through the controller's PID algorithm; SB4: Coordinated Adjustment: The temperature pre-compensation command and the PID feedback control command are algebraically superimposed to determine the composite control command; Monitoring and protection module: issues drive commands according to the composite control commands, and simultaneously obtains the comprehensive stability coefficient during the drive process. Based on the comparison result between the comprehensive stability coefficient and the preset stability threshold, it adjusts the composite control commands.

[0008] Furthermore, fiber tension and heat treatment temperature are collected, including: SA1: Tension detection setting: A non-contact laser diffraction tension meter is installed at the free section of the fiber between the drafting rollers and rigidly connected to the production line foundation through the mounting base; SA2: Temperature detection setup: Thermocouples are installed at different process sections inside the furnace and fixed to the furnace via flanges or threads, with the measuring end of the thermocouple located inside the furnace. SA3: Real-time monitoring: The fiber tension value is acquired by the non-contact laser diffraction tensiometer, and the heat treatment temperature is acquired by the thermocouple.

[0009] Furthermore, the measuring end of the thermocouple and the outer surface of the protective sheath are both provided with a high-temperature anti-oxidation and anti-carbonization coating, and the compensating wire of the thermocouple is led into the control cabinet through a metal flexible tube or wire trough.

[0010] Furthermore, a tension-temperature coupling model is constructed by physically simplifying the digital twin model, including: WA1: Decomposition Definition: Divide the heat treatment furnace into multiple temperature zones and collect input and output data for each temperature zone; WA2: Unit processing: Based on the input and output data of each temperature zone, construct the thermodynamic and mechanical equations for each temperature zone; WA3: Model Construction: The thermodynamic equations and mechanical equations are transformed into finite difference forms through ordinary differential form to construct the tension-temperature coupled model.

[0011] Furthermore, when the thermodynamic equations and mechanical equations are transformed into ordinary differential forms, the temperature within each temperature zone is set to be uniformly distributed.

[0012] Furthermore, a tension-temperature coupling model is constructed using data-driven machine learning models, including: WB1: Data Acquisition: Process parameters and performance parameters are acquired through a distributed control system or a data acquisition and monitoring control system. The process parameters include the linear speed of the drafting roller, fiber tension value, heating zone power and temperature value. The performance parameters include the tensile strength, elastic modulus and elongation at break of the fiber. WB2: Data Processing: Processing the process parameters and performance parameters through a preset time window to collect and acquire data characteristics; WB3: Model Training: Based on the data features and the set labels, train the neural network model to construct a tension-temperature coupling model.

[0013] Furthermore, the label includes tensile strength and elastic modulus values.

[0014] Furthermore, the temperature pre-compensation command includes a power adjustment amount, the PID feedback control command includes a speed adjustment amount and a power adjustment amount, and the composite control command includes the sum of the speed adjustment amount and the power adjustment amount.

[0015] Furthermore, the composite control command is adjusted, including: SC1: Drive execution: According to the composite control command, set the drive command, and regulate the stretching mechanism drive unit and the heating mechanism drive unit through the drive command; SC2: Index Calculation: Through time series analysis, the collected fiber tension data and heat treatment temperature data are analyzed and processed to obtain the oscillation amplitude and oscillation frequency. At the same time, the comprehensive stability coefficient is determined based on the oscillation amplitude and oscillation frequency. SC3: Adjustment Control: The comprehensive stability coefficient is compared with a preset stability threshold, and the corresponding control mode is determined based on the comparison result, specifically: When the overall stability coefficient is greater than the preset stability threshold, the corresponding control mode is high-performance mode; otherwise, the corresponding control mode is safety mode, and the composite control command is adjusted according to the preset controller parameters and target range.

[0016] Furthermore, the data waveform is obtained through the time series analysis, and the number of peaks and troughs is determined based on the data waveform. The difference between the number of peaks and troughs is obtained, and the oscillation amplitude is determined based on the ratio between the difference and a preset value. The time series analysis is used to obtain the data waveform, determine the number of oscillations within a preset time period, and determine the oscillation frequency based on the ratio between the number of oscillations and the preset time period.

[0017] Compared with the prior art, the beneficial effects of the present invention are: Firstly, this invention utilizes a tension-temperature coupling model to pre-calculate the temperature pre-compensation command required to offset the temperature effects caused by tension changes, based on the tension difference between the preset tension target value and the current actual tension value. Simultaneously, through a classic PID controller, it performs precise closed-loop correction on the deviations between the real-time collected fiber tension and heat treatment temperature and their respective target values, generating PID feedback control commands. The temperature pre-compensation command and the PID feedback control command are algebraically superimposed to generate the final composite control command. This not only achieves effective decoupling and dynamic coordination of the two key variables, enabling the system to respond smoothly and quickly to various disturbances, but also avoids oscillation problems caused by single parameter adjustments, fundamentally improving the stability of the production process. Secondly, this invention divides the complex physical space of the heat treatment furnace into multiple temperature zones, establishes equations based on thermodynamic and mechanical principles for each zone, and transforms them into a differential form suitable for real-time calculation, thereby constructing a digital twin model based on physical mechanisms. At the same time, a neural network model can be trained using historical production data to construct a data-driven machine learning model, which can complement each other, enabling the system to more accurately predict the dynamic relationship between tension and temperature, and thus issue more forward-looking and accurate compensation commands, improving the level of intelligent control. Thirdly, this invention uses time series analysis to calculate the oscillation amplitude and frequency of fiber tension and temperature data in real time, and integrates them into a comprehensive stability coefficient, thereby enabling real-time diagnosis of the system's health status. Attached Figure Description

[0018] Figure 1 This is a system block diagram of the continuous carbon fiber aspect ratio online control system of the present invention; Figure 2 This is a schematic diagram of the fiber tension detection unit acquiring fiber tension in this invention; Figure 3 This is a schematic diagram of the process by which the heat treatment temperature detection unit in this invention acquires the heat treatment temperature. Figure 4 This is a schematic diagram of the process for obtaining composite control commands in this invention; Figure 5 This is a schematic diagram illustrating the construction process of the physically simplified digital twin model in this invention; Figure 6 This is a schematic diagram illustrating the construction process of the data machine learning model in this invention; Figure 7 This is a schematic diagram of the confirmation process for composite control commands in this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] refer to Figure 1This embodiment provides a continuous carbon fiber aspect ratio online control system based on tension and temperature feedback. The system includes a sensing and detection module, a decision decoupling module, and a monitoring and protection module. The sensing and detection module uses a non-contact laser diffraction tensiometer and a multi-point anti-coking thermocouple array embedded inside the furnace to set up corresponding fiber tension detection units and heat treatment temperature detection units to acquire the corresponding fiber tension and heat treatment temperature. The decision decoupling module processes the fiber tension and heat treatment temperature acquired by the sensing and detection module using a set tension-temperature-microstructure correlation model and generates corresponding temperature pre-compensation commands. Simultaneously, the fiber tension and heat treatment temperature acquired by the sensing and detection module are compared with corresponding preset tension target values ​​and preset temperature target values ​​to generate corresponding PID feedback control commands. The generated temperature pre-compensation commands and PID feedback control commands are then superimposed to determine the final composite control command. The monitoring and protection module is used to drive the corresponding physical equipment according to the composite control commands determined by the decision decoupling module. During the driving process, it monitors the real-time data streams of fiber tension and heat treatment temperature and quantifies them into corresponding comprehensive stability coefficients. Based on the comparison between the comprehensive stability coefficients and the preset stability threshold, it adjusts the executed composite control commands in real time.

[0021] In this embodiment, the sensing and detection module acquires the corresponding fiber tension and heat treatment temperature through the configured fiber tension detection unit and heat treatment temperature detection unit. (Reference) Figure 2 and Figure 3 This embodiment provides a sensing and detection method for acquiring fiber tension and heat treatment temperature, as detailed below: Step SA1: Tension Detection Setup. This involves installing the non-contact laser diffraction tensiometer at the free section of the fiber between the two drafting rollers. It is important to note that during installation, a location with minimal vibration, relatively stable fiber shaking, and generally far from the drive and guide rollers should be selected. Specifically, when installing the non-contact laser diffraction tensiometer, a rigid connection is made to the production line foundation using a mounting base such as a steel platform, thereby isolating it from external vibrations that could interfere with the measurement.

[0022] Step SA2: Temperature Detection Setup. This involves selecting the appropriate thermocouple type based on the maximum temperature of the furnace, such as Type S, Type R, or Type B thermocouples. Specifically, during thermocouple installation, corresponding installation points are set at different process sections within the furnace, including the preheating zone, reaction zone, and insulation zone, according to the furnace's length. The thermocouples are also installed at the furnace top, bottom, and sidewalls at the cross-section of each process section to measure the temperature gradient. Simultaneously, corresponding mounting holes are drilled at each installation point on the furnace to secure the thermocouple to the furnace via flanges or threads. The thermocouple's measuring end is positioned inside the furnace to sense the temperature of the fiber region.

[0023] Furthermore, in this embodiment, both the measuring end of the thermocouple and the outer surface of the protective sheath are coated with a high-temperature anti-oxidation and anti-carbonization coating, such as alumina or boron nitride ceramic coating, to ensure long-term measurement accuracy. Simultaneously, the thermocouple's compensating wire is a high-temperature resistant shielded cable, threaded through a metal flexible tube or cable tray, extending from the furnace body to the control cabinet, thus protecting the compensating wire from physical damage and high-temperature radiation during use.

[0024] Step SA3: Real-time monitoring. This involves offline calibration using the non-contact laser diffraction tensiometer set in Step SA1, while the production line is stopped. Specifically, based on the measurement point location set at the free segment of the fiber, a standard weight of known weight is suspended by a pulley, and the signal value acquired by the PLC controller through the non-contact laser diffraction tensiometer is obtained. In other words, based on the weight of the standard weight and the corresponding acquired signal value, the correspondence between the signal value and the weight of the standard weight (i.e., the fiber tension value; it is worth noting that in this embodiment, a 50N weight is equivalent to applying 50N of tension to the fiber).

[0025] Furthermore, based on the correspondence between signal values ​​and fiber tension values, a laser beam is emitted into the running fiber using a non-contact laser diffraction tension meter, thereby determining the corresponding fiber tension value. In other words, the vibration frequency and morphological changes of the fiber can be detected and acquired.

[0026] In this embodiment, a multi-point anti-coking thermocouple array embedded inside the furnace can be constructed based on the multiple thermocouples set in step SA2. That is, the constructed multi-point anti-coking thermocouple array can measure the longitudinal and transverse temperature distribution of the thermal field inside the furnace, obtaining the corresponding operating temperature value.

[0027] In this embodiment, the decision decoupling module generates the final composite control command by superimposing the temperature pre-compensation command and the PID feedback control command. (See reference) Figure 4 This embodiment provides a decision decoupling method for obtaining composite control commands, as detailed below: Step SB1: Constructing the internal model. This involves using the fiber tension and heat treatment temperature collected by the sensing and detection module to construct a tension-temperature-microstructure correlation model, also known as a tension-temperature coupling model. Specifically, in this embodiment, the tension-temperature coupling model includes either a physically simplified digital twin model or a data machine learning model. That is, the corresponding tension-temperature coupling model can be constructed using a physically simplified digital twin model, or it can be constructed based on a data machine learning model.

[0028] In this embodiment, a corresponding tension-temperature coupling model is constructed by physically simplifying the digital twin model, with reference to... Figure 5 This embodiment provides a method for constructing a tension-temperature coupling model based on a physically simplified digital twin model, as detailed below: Step WA1: Decomposition and Definition. Based on the design drawings of the heat treatment furnace, the furnace is divided into multiple continuous temperature zones, including but not limited to preheating, reaction, and holding zones. Simultaneously, during production line operation, input and output data are collected for each temperature zone. Specifically, the input data for each temperature zone includes the fiber temperature, tension, and linear velocity (i.e., mass flow rate) upon entering the zone, as well as the heater power for that zone. The output data for each temperature zone includes the fiber temperature and tension upon leaving the zone.

[0029] Step WA2: Unit Processing. This involves setting up each reaction unit based on the temperature zones defined in Step WA1. In other words, each temperature zone is assigned a corresponding reaction unit. Simultaneously, based on the input and output data of each temperature zone (i.e., the input and output data of the reaction unit), the corresponding thermodynamic and mechanical equations for each reaction unit are determined, specifically: in: For the mass flow rate of the fiber, The specific heat capacity under isobaric pressure of the fiber, The temperature at which the fiber leaves the temperature zone. The temperature at which the fiber enters the temperature zone. The net heat transferred to the fibers by the heater. This refers to the heat lost by the system to the surrounding environment. The tension exerted on the fiber, The elastic modulus of the fiber. For the strain of the fiber, The viscosity coefficient of the fiber. denoted as the strain rate of the fiber.

[0030] Furthermore, in this embodiment, the fiber strain is determined based on the relative elongation of the fiber, specifically as follows: in: For the strain of the fiber, This is the current length of the fiber when it leaves the temperature zone. This represents the original length of the fiber when it enters the temperature zone.

[0031] Step WA3: Model Construction. Based on the thermodynamic and mechanical equations set in each reaction unit in Step WA2, the temperature within each reaction unit is set to a uniform distribution during the configuration process. In other words, the thermodynamic and mechanical equations are simplified to ordinary differential form using the uniformly distributed temperatures within each reaction unit. Simultaneously, according to a preset time step, the ordinary differential form equations are transformed into corresponding difference form equations. Therefore, the obtained difference form thermodynamic and mechanical equations constitute the tension-temperature coupled model constructed from the physical simplified digital twin model.

[0032] It is worth noting that in this embodiment, simplifying the thermodynamic equations and mechanical equations into ordinary differential equations and transforming the ordinary differential equations into the corresponding difference equations are all conventional mathematical transformation techniques, so they are not specifically explained in this embodiment.

[0033] In this embodiment, a corresponding tension-temperature coupling model is constructed using a data machine learning model, with reference to... Figure 6 This embodiment provides a method for constructing a tension-temperature coupling model using a data machine learning model, as detailed below: Step WB1: Data Acquisition. This involves acquiring the corresponding process and performance parameters through the established distributed control system or data acquisition and monitoring control system. Based on these acquired parameters, a corresponding database is then constructed.

[0034] Furthermore, the process parameters in this embodiment include the linear velocity of the drafting roller, fiber tension value, heating zone power, and temperature value. It is worth noting that during the acquisition of each process parameter, the source corresponding to each parameter is determined, namely the corresponding drafting roller ID, real-time fiber tension value, heating zone ID, and temperature measurement point ID.

[0035] Furthermore, the performance parameters in this embodiment include the fiber's tensile strength, elastic modulus, and elongation at break. It is worth noting that during the acquisition of these performance parameters, the performance parameters corresponding to the production stage corresponding to the acquired process parameters are also acquired.

[0036] Step WB2: Data Processing. Based on the process and performance parameters acquired in Step WB1, and within a preset time window, each parameter is processed to acquire corresponding data features. It is worth noting that during the processing of each parameter, corresponding feature data can be obtained according to actual needs; therefore, this embodiment does not elaborate on the acquisition of each data feature.

[0037] Furthermore, in this embodiment, during the processing of various parameters, the corresponding average tension, standard deviation of tension, tension slope, area integral of the temperature curve, maximum temperature, time point of the highest temperature, maximum heating power, and standard deviation of heating power can be collected according to a preset time window. In other words, a corresponding feature vector is constructed based on the collected data features.

[0038] Step WB3: Model Training. Based on the feature vectors constructed in Step WB2, and with corresponding labels (tensile strength and elastic modulus values), the neural network model is trained using these feature vectors and labels to obtain the trained neural network model, which is the corresponding data machine learning model. In other words, this yields the tension-temperature coupling model constructed through the data machine learning model.

[0039] Step SB2: Feedforward Compensation. This involves obtaining the tension difference between the preset target tension value and the current fiber tension value, and determining the corresponding tension adjustment command based on this difference. Simultaneously, the determined tension adjustment command is used as input to the tension-temperature coupling model constructed in Step SB1, and the output is the corresponding temperature pre-compensation command.

[0040] Step SB3: Feedback Control. Based on the fiber tension and heat treatment temperature acquired by the sensing and detection module, the corresponding current fiber tension value and current temperature value are determined. Simultaneously, the current fiber tension value is compared with the preset tension target value, and the current temperature value is compared with the preset temperature target value to determine the corresponding tension adjustment amount and temperature adjustment amount.

[0041] Furthermore, based on the determined tension and temperature adjustment values, the corresponding PID feedback control command is determined using the set controller PID algorithm. For example, if the tension adjustment value is -19N, the corresponding PID feedback control command is "increase speed by 2%". If the temperature adjustment value is -5℃, the corresponding PID feedback control command is "increase power by 1.5%".

[0042] Step SB4: Coordinated Adjustment. This involves algebraically superimposing the temperature pre-compensation command obtained in Step SB2 and the PID feedback control command determined in Step SB3 to determine the corresponding composite control command.

[0043] It is worth noting that the temperature pre-compensation command generated in this embodiment is the corresponding power adjustment amount. The PID feedback control command in this embodiment includes both speed adjustment and power adjustment amounts. That is to say, during the algebraic superposition process, the speed adjustment amount remains unchanged, while the power adjustment amount is superimposed.

[0044] In this embodiment, the monitoring and protection module issues drive commands through composite control commands, simultaneously monitors the overall stability coefficient during the drive process, and adjusts the composite control commands in real time based on the monitoring results. (Reference) Figure 7 This embodiment provides a monitoring and protection method for real-time adjustment of composite control commands, as detailed below: Step SC1: Drive Execution. This involves determining the corresponding drive commands, specifically the speed and power adjustment amounts, based on the composite control commands generated by the decision decoupling module. In other words, the drive units for the stretching mechanism and heating mechanism are adjusted according to these determined speed and power adjustment amounts.

[0045] Specifically, in this embodiment, the operation of the servo motor and servo driver in the drafting mechanism drive unit is adjusted according to the determined speed adjustment amount, so as to accelerate or decelerate the drafting roller motor, thereby linearly changing the tension on the fiber.

[0046] Furthermore, in this embodiment, based on the determined power adjustment amount, the operation of the thyristor power regulator or solid-state relay in the heating mechanism drive unit is specifically adjusted so as to adjust the output power of the heating element by changing the current conduction angle or on / off ratio, thereby adjusting the furnace temperature in the furnace chamber.

[0047] Step SC2: Index Calculation. During the execution of step SC1, fiber tension data and heat treatment temperature data are acquired using a preset sampling frequency. Simultaneously, time series analysis is performed on the acquired fiber tension data and heat treatment temperature data to obtain the corresponding oscillation amplitude and oscillation frequency.

[0048] Specifically, in this embodiment, the number of peaks and troughs is determined based on the data waveform obtained from time series analysis. Simultaneously, the difference between the number of peaks and troughs is determined based on their magnitude. Finally, the corresponding oscillation amplitude is determined based on the ratio between the difference between the number of peaks and troughs and a preset value (which can be specifically set according to actual needs, and therefore is not specifically described in this embodiment).

[0049] Furthermore, in this embodiment, the number of oscillations within a preset time period is determined based on the data waveform obtained from time series analysis, and the corresponding oscillation frequency is determined based on the ratio between the number of oscillations and the preset time period.

[0050] Furthermore, by using a weighted formula, the determined oscillation amplitude and oscillation frequency are combined and calculated to obtain the corresponding comprehensive stability coefficient, specifically: in: For the comprehensive stability coefficient, As the amplitude weighting factor, To standardize the amplitude, For frequency weighting factors, For standardized frequencies.

[0051] Furthermore, in this embodiment, the magnitude of the oscillation amplitude and the magnitude of the oscillation frequency are normalized to obtain the corresponding standardized amplitude and standardized frequency.

[0052] Step SC3: Adjust control. This involves comparing the overall stability coefficient obtained in step SC2 with the preset stability threshold, and determining the corresponding control mode based on the comparison result. Specifically: When the overall stability coefficient is greater than the preset stability threshold, the corresponding control mode is high-performance mode, meaning it operates according to the composite control commands generated by the decision decoupling module. Conversely, when the overall stability coefficient is not greater than the preset stability threshold, the corresponding control mode is safety mode, meaning it operates according to the preset controller parameters and target range. In other words, the composite control commands are adjusted according to the preset controller parameters and target range.

[0053] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended embodiments and their equivalents.

Claims

1. A continuous carbon fiber aspect ratio online control system based on tension and temperature feedback, characterized in that, Including: Sensing and detection module: It acquires fiber tension and heat treatment temperature through fiber tension detection unit and heat treatment temperature detection unit; Decision decoupling module: Processes the fiber tension and heat treatment temperature using a tension-temperature coupling model and a controller PID algorithm to obtain temperature pre-compensation commands and PID feedback control commands. Simultaneously, the temperature pre-compensation commands and PID feedback control commands are superimposed to obtain a composite control command, including: SB1: Constructing an internal model: Based on the fiber tension and heat treatment temperature, construct a tension-temperature coupling model; SB2: Feedforward compensation: Based on the tension difference between the preset tension target value and the current fiber tension value, a tension adjustment command is determined, and the tension adjustment command is used as the input of the tension-temperature coupling model to output the corresponding temperature pre-compensation command; SB3: Feedback Control: Based on the tension adjustment between the current fiber tension value and the preset tension target value, and the temperature adjustment between the current temperature value and the preset temperature target value, the PID feedback control command is determined through the controller's PID algorithm; SB4: Coordinated Adjustment: The temperature pre-compensation command and the PID feedback control command are algebraically superimposed to determine the composite control command; Monitoring and protection module: issues drive commands according to the composite control commands, and simultaneously obtains the comprehensive stability coefficient during the drive process. Based on the comparison result between the comprehensive stability coefficient and the preset stability threshold, it adjusts the composite control commands.

2. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 1, characterized in that, The fiber tension and heat treatment temperature were collected, including: SA1: Tension detection setting: A non-contact laser diffraction tension meter is installed at the free section of the fiber between the drafting rollers and rigidly connected to the production line foundation through the mounting base; SA2: Temperature detection setup: Thermocouples are installed at different process sections inside the furnace and fixed to the furnace via flanges or threads, with the measuring end of the thermocouple located inside the furnace. SA3: Real-time monitoring: The fiber tension value is acquired by the non-contact laser diffraction tensiometer, and the heat treatment temperature is acquired by the thermocouple.

3. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 2, characterized in that, The measuring end of the thermocouple and the outer surface of the protective sheath are both coated with a high-temperature anti-oxidation and anti-carbonization coating. The compensating wire of the thermocouple is led into the control cabinet through a metal flexible tube or wire trough.

4. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 1, characterized in that, A tension-temperature coupling model is constructed using a physically simplified digital twin model, including: WA1: Decomposition Definition: Divide the heat treatment furnace into multiple temperature zones and collect input and output data for each temperature zone; WA2: Unit processing: Based on the input and output data of each temperature zone, construct the thermodynamic and mechanical equations for each temperature zone; WA3: Model Construction: The thermodynamic equations and mechanical equations are transformed into finite difference forms through ordinary differential form to construct the tension-temperature coupled model.

5. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 4, characterized in that, When the thermodynamic equations and mechanical equations are transformed into ordinary differential forms, the temperature in each temperature region is set to be uniformly distributed.

6. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 1, characterized in that, The tension-temperature coupling model is constructed using data-driven machine learning models, including: WB1: Data Acquisition: Process parameters and performance parameters are acquired through a distributed control system or a data acquisition and monitoring control system. The process parameters include the linear speed of the drafting roller, fiber tension value, heating zone power and temperature value. The performance parameters include the tensile strength, elastic modulus and elongation at break of the fiber. WB2: Data Processing: Processing the process parameters and performance parameters through a preset time window to collect and acquire data characteristics; WB3: Model Training: Based on the data features and the set labels, train the neural network model to construct a tension-temperature coupling model.

7. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 6, characterized in that, The label includes tensile strength and elastic modulus values.

8. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 1, characterized in that, The temperature pre-compensation command includes a power adjustment amount, the PID feedback control command includes a speed adjustment amount and a power adjustment amount, and the composite control command includes the sum of the speed adjustment amount and the power adjustment amount.

9. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 1, characterized in that, Adjusting the composite control command includes: SC1: Drive execution: According to the composite control command, set the drive command, and regulate the stretching mechanism drive unit and the heating mechanism drive unit through the drive command; SC2: Index Calculation: Through time series analysis, the collected fiber tension data and heat treatment temperature data are analyzed and processed to obtain the oscillation amplitude and oscillation frequency. At the same time, the comprehensive stability coefficient is determined based on the oscillation amplitude and oscillation frequency. SC3: Adjustment Control: The comprehensive stability coefficient is compared with a preset stability threshold, and the corresponding control mode is determined based on the comparison result, specifically: When the overall stability coefficient is greater than the preset stability threshold, the corresponding control mode is high-performance mode; otherwise, the corresponding control mode is safety mode, and the composite control command is adjusted according to the preset controller parameters and target range.

10. The continuous carbon fiber aspect ratio online control system based on tension and temperature feedback according to claim 9, characterized in that, The time series analysis is used to obtain a data waveform diagram, and based on the data waveform diagram, the number of peaks and troughs is determined, the difference between the number of peaks and troughs is obtained, and the oscillation amplitude is determined based on the ratio between the difference between the number of peaks and troughs and a preset value. The time series analysis is used to obtain the data waveform, determine the number of oscillations within a preset time period, and determine the oscillation frequency based on the ratio between the number of oscillations and the preset time period.

Images