Model predictive control-based precursor tension control method in carbon fiber drafting process

Abstract

The invention provides a model predictive control-based precursor tension control method in a carbon fiber drafting process, which comprises the following steps of: determining a precursor tension set value in the drafting process according to carbon fiber production process requirements; establishing a nonlinear dynamic model between the driving voltage and the precursor tension in single-stage drafting; based on the dynamic model, constructing a target function taking the combined minimization of the tension tracking error and the driving voltage energy consumption as a target; solving the target function by adopting an optimization algorithm to obtain an optimal driving voltage sequence in a prediction time domain; and selecting a first item in the optimal driving voltage sequence for control, and repeating the steps at the next sampling moment to realize dynamic and stable control of the tension of the precursor. The problems of insufficient model prediction precision and nonlinear coupling of a traditional control mode are effectively solved, and the system control precision is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon fiber production, and in particular to a method for controlling the tension of precursor fibers in a carbon fiber drafting process based on model predictive control. Background Art

[0002] The precursor drawing link is the core process at the front end of the preparation of carbon fiber. Its main function is to draw the initial fibers that have not yet been fully formed through the drawing rollers of different speeds driven by the motor. The filament bundles in contact with them will be drawn at a certain multiple, so that the molecular chains of the initial fibers are arranged in an orderly manner and the crystal structure is denser, which helps to improve the mechanical properties of the produced carbon fiber, such as tensile strength and elastic modulus. Among them, single-stage drawing, as an important stage in the drawing link, is usually completed by a pair of drive rollers consisting of the first roller and the second roller. In this process, the first roller serves as the reference axis to provide a stable linear speed and initial tension; the second roller serves as a dynamic adjustment axis, and its linear speed is dynamically adjusted by the drive voltage. By precisely controlling the drive voltage of the second roller to generate a linear speed difference between the two rollers, the precursor can obtain a stable tension, thereby producing deformation according to the process requirements. A detailed analysis of the single-stage carbon fiber drafting process revealed a significant coupling between the system's controlled variables, precursor speed and precursor tension. This means that controlling a single variable can affect the stability of the other variables. This mutual influence between the controlled variables also introduces nonlinearity into the system. Therefore, the single-stage carbon fiber drafting process is a complex system with both nonlinear and coupled properties. Field production practice has shown that only long-term stable precursor tension can produce high-performance precursors. Therefore, stable control of precursor tension is of great significance.

[0003] Currently, the tension control of the single-stage drawing process of carbon fiber mostly adopts a separate speed regulation method, that is, the motor is used as the main driving source to drive the roller and the initial fiber that is not yet formed on it as the load. The sensor monitors the state changes of physical quantities such as speed and tension in real time. The controller dynamically adjusts the voltage input of the drive motor based on the feedback information provided by the sensor, so that the roller runs at the set speed, thereby ensuring the stability of the original fiber tension. However, this conventional feedback closed-loop control method has difficulty in handling complex control tasks with nonlinear coupling and operational constraints. It requires repeated debugging and adjustment of control parameters based on the characteristics of the dynamic equations of the nonlinear system. It lacks flexibility. In actual field applications, the system response often has large deviations, and the control effect is not ideal.

[0004] Prior art models linking operational variables such as coagulation bath concentration with performance indicators such as draft ratio are constructed to simulate the actual production process of a carbon fiber production line using a water bath. However, this system idealizes the production environment and fails to account for external interference, making it unsuitable for field applications. Secondly, a temperature sensor built into the drafting tank monitors the water temperature in real time to dynamically adjust the draft ratio of the carbon fiber precursor. However, the tank's large size requires time for its temperature to change, and the location of the temperature sensor also affects the real-time nature of temperature monitoring. This results in a delay in the system's response to the temperature signal. Furthermore, based on a tension variation model, a PLC receives surface tension information fed back by the sensor and outputs control commands to the servo motor, thereby controlling the fiber's surface tension. However, this system fails to account for future system state changes. Control actions are only initiated when there is an error between the surface tension measurement and the desired setpoint for the production process, making it difficult to achieve optimal control results. Finally, the amount of desalted water is controlled based on the temperature and pressure measurements fed back by the sensors to achieve stable control of the steam humidity. However, the control system does not consider how to deal with drastic changes in the control quantity and suppress the occurrence of overshoot, making it difficult for the control output to transition smoothly and the system stability is poor. Summary of the Invention

[0005] The present invention proposes a method for controlling the precursor tension in the carbon fiber drawing process based on model predictive control, so as to solve the problem that traditional control methods are difficult to effectively deal with the nonlinear coupling characteristics in the carbon fiber drawing process, which causes significant fluctuations in the process parameter precursor tension.

[0006] To solve the above technical problems, the present invention provides a method for controlling the tension of a precursor in a carbon fiber drafting process based on model predictive control, comprising the following steps:

[0007] Step S1: determining a set value of the precursor tension during the drawing process according to the carbon fiber production process requirements;

[0008] Step S2: establishing a nonlinear dynamic model between the driving voltage and the precursor tension in the single-stage drafting process, wherein the dynamic model includes a single input manipulated variable and two output controlled variables, wherein the input variable is the driving voltage of the dynamic adjustment axis in the single-stage drafting process, and the output variables are the precursor tension and angular velocity of the dynamic adjustment axis;

[0009] Step S3: Based on the dynamic model, constructing an objective function with the goal of jointly minimizing the tension tracking error and the driving voltage energy consumption;

[0010] Step S4: using an optimization algorithm to solve the objective function to obtain the optimal driving voltage sequence in the prediction time domain;

[0011] Step S5: Select the first item in the optimal driving voltage sequence for control, and repeat steps S2 to S5 at the next sampling moment to achieve dynamic and stable control of the original silk tension.

[0012] Preferably, in step S2, the expression of the nonlinear dynamic model is:

[0013]

[0014] Where, u represents the driving voltage of the dynamic adjustment axis during the single-stage drawing process of carbon fiber; x1 and x2 represent the precursor tension and angular velocity on the dynamic adjustment axis, respectively; L1 represents the distance between the two driving motors; E represents the Young's modulus; A is the cross-sectional area of the precursor; R represents the radius of the dynamic adjustment axis; J a represents the moment of inertia; f0 represents the friction coefficient; n represents the transmission ratio between the drive motor transmission rod and the dynamic adjustment shaft; T0 and V0 represent the initial tension value on the reference shaft and the original silk line speed respectively.

[0015] Preferably, the objective function J in step S3 k The expression is:

[0016]

[0017] Where y(k+i|k) represents the predicted value at k+i obtained by the dynamic model at k sampling time; y ref represents the set value of the system's raw wire tension; u(k+i|k) represents the driving voltage at time k+i obtained by the dynamic model at sampling time k; p represents the number of time steps for future prediction, and Q represents a symmetric positive semi-definite matrix.

[0018] Preferably, in step S4, the objective function is solved using an improved particle swarm optimization algorithm, which includes the following steps:

[0019] Step S41: Initialize a particle swarm, where each particle represents a driving voltage sequence in the future prediction time domain;

[0020] Step S42: Calculate the objective function values of all particles in the particle swarm and determine the global optimal particle;

[0021] Step S43: Calculate the angle between the particle and the global optimal particle, and update the particle swarm based on the minimum angle pruning strategy;

[0022] Step S44: dynamically adjusting the inertia weight according to the particle's achievement scalar function value, and updating the particle's velocity and position;

[0023] Step S45: Repeat steps S42 to S44 to iterate until the termination condition is met and then output the optimal driving voltage sequence.

[0024] Preferably, the method for updating the particle swarm based on the minimum angle pruning strategy in step S43 includes: calculating the angle between the particle and the global optimal particle, and replacing it with a random new particle if the angle is less than a threshold, otherwise retaining it.

[0025] Preferably, the expression for calculating the angle between a particle and the global optimal particle is:

[0026]

[0027] Where J(s1) and J(s2) represent the target values of particles s1 and s2; J min represents the global optimal particle p g The corresponding minimum target value; ||·|| represents the Euclidean norm.

[0028] Preferably, the expression of the achievement scalarization function value in step S44 is:

[0029]

[0030] Where, ASF(s i ) represents the i-th particle s in the particle swarm i The achievement scalar function value of J min represents the global optimal particle p g The corresponding minimum target value; J max represents the maximum target value in the particle swarm; ε represents a very small constant; J(s i ) represents particle s i target value.

[0031] Preferably, in step S44, the expression for dynamically adjusting the inertia weight according to the achievement scalar function value of the particle is:

[0032] w(s i )=w min +(w max -w min )·ASF(s i );

[0033] Where, w(s i ) represents the inertia weight of the i-th particle; w min and w max Respectively represent the minimum and maximum inertia weight values in the current particle swarm.

[0034] Preferably, the expression for updating the velocity and position of the particle in step S44 is:

[0035] v′ ij =w(s i )vij +c1r 1j (p ij -s ij )+c2r 2j (p gj -s ij );

[0036] s′ ij =s ij +v′ ij ;

[0037] Where c1 and c2 represent acceleration constants; r 1j and r 2j is a random number uniformly distributed in the interval [0,1]; p ij and p gj Represents particles s i The value of its own historical optimal position and the global optimal position in the particle swarm in the j+1th dimension; v ij and v′ ij Respectively represent the speed of the particle before and after update; s ij and s′ ij Respectively represent the positions of the particles before and after the update.

[0038] The beneficial effects of the present invention include at least:

[0039] 1) This invention constructs a dynamic model reflecting the nonlinear relationship between drive voltage and precursor tension, and uses this model to perform online prediction of the precursor tension output within the future prediction time domain. This method effectively solves the problems of insufficient prediction accuracy and nonlinear coupling in traditional control methods, significantly improving system control accuracy.

[0040] 2) In the objective function design, we introduced the error minimization term between the system output and the set value, as well as the minimization term for the manipulated variable input. During the rolling optimization solution process, we fully considered constraints such as the manipulated variable limit and the controlled variable limit, which can overcome external disturbances and further enhance the dynamic stability of the system.

[0041] 3) At each sampling moment, the system dynamically adjusts the optimal operating variables based on the real-time monitored state information of physical quantities such as speed and tension, as well as the state prediction information for a period of time in the future, continuously reduces the system control deviation, and makes the controlled variables gradually approach and stabilize to the set value, thereby significantly improving the system's response speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention;

[0043] Figure 2Schematic diagram of the variation of the raw yarn tension and roller speed with voltage according to an embodiment of the present invention;

[0044] Figure 3 Schematic diagram of the effect of controlling the tension of the precursor in the carbon fiber drafting process based on model predictive control according to an embodiment of the present invention;

[0045] Figure 4 Schematic diagram of the effect of controlling the tension of the precursor in the carbon fiber drawing process based on PID according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following is a clear and complete description of the technical solutions in the embodiments of the present invention, in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts are within the scope of protection of the present invention.

[0047] like Figure 1 As shown, an embodiment of the present invention provides a method for controlling the tension of a precursor in a carbon fiber drafting process based on model predictive control, comprising the following steps:

[0048] Step S1: Determine the set value of the precursor tension during the drawing process according to the carbon fiber production process requirements.

[0049] Specifically, in this embodiment, according to the production process requirements for the strength, orientation and other performance indicators of the carbon fiber precursor, the set value of the precursor tension in the carbon fiber drawing process is determined to be 0.025 cN, and remains unchanged in subsequent control steps.

[0050] Step S2: Establish a nonlinear dynamic model between the driving voltage and the raw yarn tension in single-stage drawing. The dynamic model includes a single input operating variable and two output controlled variables, where the input variable is the driving voltage of the dynamic adjustment axis in single-stage drawing, and the output variables are the raw yarn tension and angular velocity of the dynamic adjustment axis.

[0051] Specifically, the dynamic model of the nonlinear relationship is expressed as follows:

[0052]

[0053] Wherein, the model represents a single-input two-output control system, the single-input manipulated variable is u, which represents the driving voltage on the dynamic adjustment axis during the single-stage drawing process of carbon fiber, and the two output manipulated variables are x1 and x2, which represent the original fiber tension and angular velocity on the dynamic adjustment axis respectively; L1, E, A, R, J a, f0, n, T0 and V0 are fixed parameters in the prediction model, L1 represents the distance between the two driving motors; E represents the Young's modulus; A represents the cross-sectional area of the precursor; R represents the radius of the dynamically adjustable shaft; J a represents the moment of inertia; f0 represents the friction coefficient; n represents the transmission ratio between the drive motor transmission rod and the dynamic adjustment shaft; T0 and V0 represent the initial tension value on the first roller, i.e., the reference axis, and the original wire speed, respectively.

[0054] For example, the values of key variables are shown in Table 1.

[0055] Table 1

[0056]

[0057]

[0058] When the driving voltage changes in the form of the following piecewise function, the tension changes with the voltage as follows: Figure 2 As shown. Figure 2 (a) shows the change of the original silk tension. It can be seen that the original silk tension decreases with the increase of the voltage input of the roller and then tends to a stable state. For example, when the voltage of the roller increases from 10V to 30V, the original silk tension increases from 1.8×10 -3 cN is reduced to 0.4×10 -3 After cN, it tends to be stable; Figure 2 (b) in the figure shows the change of roller speed. It can be seen that the roller speed increases or decreases synchronously with the voltage input.

[0059]

[0060] Step S3: Based on the dynamic model, construct an objective function with the goal of jointly minimizing the tension tracking error and the driving voltage energy consumption.

[0061] Specifically, first, the constraints such as the manipulated variables and controlled variables are considered, which are defined as follows:

[0062] u min ≤u≤u max ;

[0063] x 1min ≤x1≤x 1max ;

[0064] x 2min ≤x2≤x 2max ;

[0065] Where u max and u min They represent the upper and lower limits of the operating variable u, with values of 10 and -10 respectively; x1max and x 1min They represent the upper and lower limits of the controlled variable x1, with values of 0.03 and 0 respectively; 2max and x 2min They represent the upper and lower limits of another controlled variable x2, with values of 5 and -15 respectively.

[0066] To ensure that the raw yarn tension can be stabilized to the set value while taking into account the minimum energy consumption, the objective function with the goal of synergistically minimizing the tension tracking error and the driving voltage energy consumption is constructed and defined as follows:

[0067]

[0068] Where y(k+i|k) represents the predicted value at k+i obtained by the dynamic model at k sampling time; y ref represents the set value of the system's raw wire tension; u(k+i|k) represents the driving voltage value at time k+i obtained by the dynamic model at sampling time k; p represents the number of time steps for future prediction, and Q represents a symmetric positive semi-definite matrix.

[0069] Step S4: using an optimization algorithm to solve the objective function and obtain the optimal driving voltage sequence in the prediction time domain.

[0070] Illustratively, the optimization algorithms in this embodiment include but are not limited to ant colony optimization algorithm, genetic algorithm, artificial bee colony algorithm and differential evolution algorithm, which will not be described in detail in this embodiment.

[0071] Preferably, this embodiment provides an improved particle swarm optimization algorithm to solve the above objective function, which includes the following steps.

[0072] 1) Algorithm initialization. Randomly generate an initial particle swarm containing n particles, set the speed, position and individual optimal position of each particle. In this algorithm, each particle represents the feasible domain space A potential solution in , that is, a sequence of operating variables within p time steps, is mathematically defined as:

[0073] s i =(u i (k),u i (k+1),…,u i (k+(p-1))),i∈{1,…,n};

[0074] Among them, u i (k+j) represents the value of the manipulated variable of the i-th particle at the future time k+j.

[0075] 2) Target value calculation: Calculate the target values of all particles in the particle swarm, compare the target values of all particles one by one, determine the particle with the smallest target value in the current particle swarm, and update it as the global optimal particle p g .

[0076] 3) Update the particle swarm. In this embodiment, a minimum angle pruning strategy is used to maintain particle diversity. The core idea of this strategy is to measure particle diversity by calculating the angle between the particles in the target space. For two particles s1 and s2 in the decision space, the angle formed by the line connecting their target values in the target space and the current population minimum target value is defined as follows:

[0077]

[0078] Where J(s1) and J(s2) represent the target values of particles s1 and s2; J min represents the global optimal particle p g The corresponding minimum target value; ||·|| represents the Euclidean norm.

[0079] exist Based on the definition, To measure the diversity of particle s1, it is defined as follows:

[0080]

[0081] Where P represents the current particle swarm. The value is less than the threshold θ th , indicating that s1 and s2 are closer in the target space, the more concentrated the distribution of particle s1 in the target space, the worse the diversity of particle s1, and it is replaced by randomly generated new particles to increase the diversity of the particle group and avoid premature convergence; on the contrary, if The value is greater than or equal to the threshold θ th , it means that s1 is far away from other solutions in the target space and its diversity is excellent, so it is retained.

[0082] 4) Inertia weight adaptive adjustment: Calculate the achievement scalar function value of all particles. The calculation formula is as follows:

[0083]

[0084] Among them, ASF(s i ) represents the achievement scalar function value of the i-th particle in the particle swarm, J max Represents the maximum target value in the particle swarm, and ε represents a very small constant, here ε=10 -6 , used to prevent the denominator from being zero.

[0085] The inertia weight is dynamically adjusted according to the particle's achievement scalar function value. The calculation formula is:

[0086] w(s i )=w min +(w max -w min )·ASF(s i );

[0087] Among them, w min and w max Respectively represent the minimum and maximum inertia weight values in the current particle swarm. i ) value is small, indicating that the particle s i is very close to the optimal individual in the particle swarm, the inertia weight should be reduced and the local search ability of the particle should be increased; otherwise, it means that the particle s i Staying away from the optimal individual in the particle swarm should increase inertia and enhance the global exploration ability of the particle.

[0088] 5) Update the speed and position of particles. During the algorithm iteration optimization process, each particle s i According to its own historical optimal particle p i And the global optimal particle p in the particle swarm g To update the current position and velocity. Let v ij and x ij Represent the speed and position of the i-th particle in the j+1-th dimension respectively, then the speed and position update formulas can be expressed as:

[0089] v′ ij =w(s i )v ij +c1r 1j (p ij -s ij )+c2r 2j (p gj -s ij );

[0090] s′ ij =s ij +v′ ij ;

[0091] Among them, w(x i ) represents the inertia weight, c1 and c2 represent the acceleration constants, r 1j and r 2j is a random number uniformly distributed in the interval [0,1], p ij and p gj Represents particles s i The value of its own historical optimal position and the global optimal position in the particle swarm in the j+1th dimension, v′ij and x′ ij Represent the updated velocity and position of the particle respectively.

[0092] 6) Output the current optimal operating variable sequence. Check whether the number of algorithm iterations has reached the maximum number of iterations. If the termination condition is met, end the algorithm optimization process, output the global optimal particle in the particle swarm, and use it as the current optimal control variable sequence. Otherwise, continue to the next iteration.

[0093] Step S5: Select the first item in the optimal driving voltage sequence for control, and repeat steps S2 to S5 at the next sampling moment to achieve dynamic and stable control of the original silk tension.

[0094] Specifically, at the current sampling time k, the optimal operating variable sequence is selected The first item in Acting on the raw yarn tension control system, the remaining control variable sequences are discarded, and the above control steps are repeated according to the latest system status at the next sampling moment, thereby achieving stable control of the raw yarn tension.

[0095] Figure 3 The effect diagram of the precursor tension control in the carbon fiber drawing process based on model predictive control is shown. It can be seen that under the action of the MPC controller corresponding to the model predictive control, by adjusting the operating variable driving voltage, the controlled variable precursor tension can be stabilized to the production process set value of 0.025cN in about 6s. The controlled variable precursor tension output has a good smooth transition, and the controller achieves a stable control effect.

[0096] Figure 4 The figure shows the effect of PID-based precursor tension control in the carbon fiber drawing process. When PID handles the precursor tension control task with nonlinear coupling and operational constraints, this conventional feedback closed-loop control method cannot stabilize the controlled variable precursor tension to the production process set value, resulting in overshoot and unsatisfactory control effect.

[0097] In summary, the present invention proposes a method for controlling the precursor tension in the carbon fiber drawing process based on model predictive control, which can effectively solve the problem of significant fluctuations in the precursor tension of the process parameter caused by the nonlinear coupling characteristics in the carbon fiber drawing process, and significantly improves the control accuracy, dynamic stability and response speed of the system.

[0098] The technical features of the above embodiments may be combined in any manner. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. Only preferred embodiments of the present invention are presented. While the description is relatively specific and detailed, it should not be construed as limiting the scope of the present invention. As long as there are no conflicts in the combination of these technical features, they should be considered to be within the scope of this specification.

[0099] It should be noted that, for those skilled in the art, various modifications and improvements can be made without departing from the scope of the present invention, and these modifications and improvements fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A method for controlling the tension of precursor fibers in a carbon fiber drafting process based on model predictive control, characterized in that: The following steps are involved: Step S1: determining a set value of the precursor tension during the drawing process according to the carbon fiber production process requirements; Step S2: establishing a nonlinear dynamic model between the driving voltage and the precursor tension in the single-stage drafting process, wherein the dynamic model includes a single input manipulated variable and two output controlled variables, wherein the input variable is the driving voltage of the dynamic adjustment axis in the single-stage drafting process, and the output variables are the precursor tension and angular velocity of the dynamic adjustment axis; Step S3: Based on the dynamic model, constructing an objective function with the goal of jointly minimizing the tension tracking error and the driving voltage energy consumption; Step S4: using an optimization algorithm to solve the objective function to obtain the optimal driving voltage sequence in the prediction time domain; Step S5: Select the first item in the optimal driving voltage sequence for control, and repeat steps S2 to S5 at the next sampling moment to achieve dynamic and stable control of the original silk tension.

2. The method for controlling the tension of a precursor in a carbon fiber drafting process based on model predictive control according to claim 1, characterized in that: In step S2, the expression of the nonlinear dynamic model is: Where u represents the driving voltage of the dynamic adjustment axis during the single-stage drawing process of carbon fiber; x1 and x2 represent the precursor tension and angular velocity on the dynamic adjustment axis, respectively; L1 represents the distance between the two driving motors; E represents the Young's modulus; A is the cross-sectional area of the precursor; R represents the radius of the dynamic adjustment axis; J a represents the moment of inertia; f0 represents the friction coefficient; n represents the transmission ratio between the drive motor transmission rod and the dynamic adjustment shaft; T0 and V0 represent the initial tension value and the original silk line speed on the first reference axis respectively.

3. The method for controlling the tension of precursor fibers in a carbon fiber drafting process based on model predictive control according to claim 1, wherein: The objective function J in step S3 k The expression is: Where y(k+i|k) represents the predicted value at k+i obtained by the dynamic model at k sampling time; y ref represents the set value of the system's raw wire tension; u(k+i|k) represents the driving voltage at time k+i obtained by the dynamic model at sampling time k; p represents the number of time steps for future prediction, and Q represents a symmetric positive semi-definite matrix.

4. The method for controlling the tension of a precursor fiber in a carbon fiber drafting process based on model predictive control according to claim 1, characterized in that: In step S4, the objective function is solved using an improved particle swarm optimization algorithm, which includes the following steps: Step S41: Initialize a particle swarm, where each particle represents a driving voltage sequence in the future prediction time domain; Step S42: Calculate the objective function values of all particles in the particle swarm and determine the global optimal particle; Step S43: Calculate the angle between the particle and the global optimal particle, and update the particle swarm based on the minimum angle pruning strategy; Step S44: dynamically adjust the inertia weight according to the particle's achievement scalar function value, and update the particle's speed and position; Step S45: Repeat steps S42 to S44 to iterate until the termination condition is met and then output the optimal driving voltage sequence.

5. The method for controlling the tension of precursor fibers in a carbon fiber drafting process based on model predictive control according to claim 4, characterized in that: The method for updating the particle swarm based on the minimum angle pruning strategy in step S43 includes: calculating the angle between the particle and the global optimal particle, and replacing it with a random new particle if the angle is less than a threshold, otherwise retaining it.

6. The method for controlling the tension of precursor fibers in a carbon fiber drawing process based on model predictive control according to claim 5, characterized in that: The expression for calculating the angle between a particle and the global optimal particle is: Where J(s1) and J(s2) represent the target values of particles s1 and s2; J min represents the global optimal particle p g The corresponding minimum target value; ||·|| represents the Euclidean norm.

7. The method for controlling precursor tension in a carbon fiber drafting process based on model predictive control according to claim 6, characterized in that: The expression of the achievement scalarization function value in step S44 is: Where, ASF(s i ) represents the i-th particle s in the particle swarm i The achievement scalar function value of J min represents the global optimal particle p g The corresponding minimum target value; J max represents the maximum target value in the particle swarm; ε represents a very small constant; J(s i ) represents particle s i target value.

8. The method for controlling precursor tension in a carbon fiber drafting process based on model predictive control according to claim 7, characterized in that: The expression for dynamically adjusting the inertia weight according to the particle's achievement scalar function value in step S44 is: w(s i )=w min +(w max -w min )·ASF(s i ); Where, w(s i ) represents the inertia weight of the i-th particle; w min and w max Respectively represent the minimum and maximum inertia weight values in the current particle swarm.

9. The method for controlling precursor tension in a carbon fiber drafting process based on model predictive control according to claim 8, characterized in that: The expressions for updating the velocity and position of the particle in step S44 are: v′ ij =w(s i )v ij +c1r 1j (p ij -s ij )+c2r 2j (p gj -s ij ); s′ ij =s ij +v′ ij ; Where c1 and c2 represent acceleration constants; r 1j and r 2j is a random number uniformly distributed in the interval [0,1]; p ij and p gj Represents particles s i The value of its own historical optimal position and the global optimal position in the particle swarm in the j+1th dimension; v ij and v′ ij Respectively represent the speed of the particle before and after update; s ij and s′ ij Respectively represent the positions of the particles before and after the update.

Images