A stable control method for roller coating

Abstract

The invention discloses a stable control method for roller coating, belonging to the technical field of motor control. The invention first adopts a PI control model with an initial proportional coefficient and an initial integral coefficient to control the power of a roller coating motor in the first stage, establishes a proportional-integral regulation sub-model according to the control situation in the first stage, replaces the initial proportional coefficient and the initial integral coefficient, realizes adaptive adjustment of the proportional coefficient and the integral coefficient, enters the control in the second stage, predicts the future gap value, thereby performing output correction on the PI control model with adjustable proportional-integral, obtains a cascade combination model of the roller coating process, optimizes the cascade combination model, and improves the control accuracy of the roller coating process.

Description

Technical Field

[0001] The invention relates to the technical field of motor control, and in particular to a stable control method for roller coating. Background Art

[0002] In the manufacturing industry, roller coating is a widely used coating technology for applying a uniform coating on the surface of materials. This technology is widely used in fields such as paper, textiles and metal processing. In the roller coating process, the control of multiple roller motors is usually involved, such as the main roller motor, pressure roller motor, tension motor, etc. The precise control of these motors is crucial to ensure the stability and consistency of the roller coating process.

[0003] The main roller motor is responsible for driving the main roller to rotate, thereby driving the material through the roller coating area and completing the coating. The pressure roller motor is responsible for controlling the pressure of the pressure roller, affecting the thickness and uniformity of the coating. The tension motor is used to adjust the tension of the material during the roller coating process to ensure that the material can pass smoothly. The stable operation of these motors directly determines the quality of the final product.

[0004] The existing roller coating motor control usually adopts PID algorithm for stable control. The PID algorithm achieves precise control of the motor by adjusting the proportional coefficient, integral coefficient and differential coefficient. However, the proportional coefficient, integral coefficient and differential coefficient in the PID algorithm are all found through the experimental process to find the optimal proportional coefficient, integral coefficient and differential coefficient, and the coefficients are fixed. When some factors cause the motor operation to fluctuate, the traditional PID algorithm is difficult to quickly adapt to these changes, resulting in low control accuracy. Summary of the invention

[0005] In view of the above-mentioned deficiencies in the prior art, a stable control method for roller coating provided by the present invention solves the problem of fixed coefficients and low control accuracy in the existing PID algorithm.

[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is: a stable control method for roller coating, comprising the following steps:

[0007] S1, adopting the PI control model with the initial proportional coefficient and the initial integral coefficient to control the power of the roller coating motor in the first stage;

[0008] S2. According to the trend of the difference between the actual roller coating power value and the target roller coating power value after the first stage control, a proportional integral control sub-model is established, and the proportional integral control sub-model is added to the PI control model to obtain a PI control model with adjustable proportional integral;

[0009] S3, using a PI control model with adjustable proportional integral to control the power of the roller coating motor in the second stage;

[0010] S4, according to the gap value and real-time fluctuation factor between the actual roller coating power value and the target roller coating power value after the second stage control, the future gap value is predicted, and the proportional integral adjustable PI control model is cascaded to obtain the cascade combination model of the roller coating process;

[0011] S5, using the cascade combination model of the roller coating process to control the power of the roller coating motor in the third stage, and optimizing the cascade combination model;

[0012] S6. According to the output of the optimized cascade combination model, the power of the roller coating motor is stably controlled.

[0013] The beneficial effects of the present invention are as follows: the present invention first adopts a PI control model with an initial proportional coefficient and an initial integral coefficient to perform a first-stage control on the power of a roller coating motor, and according to the control situation of the first stage, establishes a proportional-integral regulation sub-model to replace the initial proportional coefficient and the initial integral coefficient, realizes adaptive adjustment of the proportional coefficient and the integral coefficient, enters the second stage of control, predicts the future gap value, and thereby performs output correction on the PI control model with adjustable proportional-integral, obtains a cascade combination model of the roller coating process, and optimizes the cascade combination model to improve the control accuracy of the roller coating process.

[0014] The present invention realizes more accurate configuration of proportional integral coefficient by adding a proportional integral regulation sub-model, avoids large shocks, improves the stability of the roller coating process, and then corrects the output of the PI control model with adjustable proportional integral through future gap values, further improves the control accuracy and improves the stability of the roller coating process.

[0015] Furthermore, the S2 comprises the following sub-steps:

[0016] S21, calculating the difference value according to the difference between the actual roller coating power value and the target roller coating power value after the first stage control;

[0017] S22, calculating the trend coefficients of multiple gap values;

[0018] S23. According to the trend coefficient, a proportional integral control sub-model is established;

[0019] S24, adding the proportional-integral regulation sub-model to the PI control model to obtain a PI control model with adjustable proportional-integral.

[0020] Furthermore, the formula for calculating the trend coefficient in S22 is: , where ε is the trend coefficient, d 1,t is the gap value at the tth moment after the first stage of control, t is the moment number, N is the length of statistics, and the gap value d 1,tis the difference between the actual roller coating power value and the target roller coating power value after the first stage of control, and | | is the absolute value.

[0021] Furthermore, the proportional integral control sub-model in S23 is: , ,in, is the adjusted proportional coefficient, is the adjusted integral coefficient, ε is the trend coefficient, K p,0 is the initial proportionality coefficient, K I,0 is the initial integration coefficient.

[0022] The beneficial effects of the above-mentioned further scheme are as follows: the present invention calculates the trend coefficient to reflect the trend of the gap value. When the trend coefficient is greater than or equal to 0, the gap value is in a growth state. On the basis of the initial proportional coefficient, the proportional coefficient is increased, the integral coefficient is reduced, and the response speed of the model is enhanced. When the trend coefficient is less than 0, the model is converging, and the actual roller coating power value is approaching the target roller coating power value. Therefore, the initial proportional coefficient and the initial integral coefficient are maintained.

[0023] Furthermore, the S4 comprises the following sub-steps:

[0024] S41, calculating the real-time fluctuation factor of the current gap value according to the gap value between the actual roller coating power value and the target roller coating power value after the second stage control;

[0025] S42, establishing a fluctuation prediction model to predict future gap values ​​based on the real-time fluctuation factor and the gap value;

[0026] S43, adding a conversion proportional coefficient to the future gap value, and cascading it with a PI control model with adjustable proportional integral to obtain a cascade combination model of the roller coating process.

[0027] Furthermore, the formula for calculating the real-time fluctuation factor of the current gap value in S41 is: , where μ is the real-time fluctuation factor of the current gap value after the second stage control, d 2,t is the gap value at the tth moment after the second stage control, f is the normalization function, || is the absolute value, and the gap value d 2,t is the difference between the actual roller coating power value and the target roller coating power value after the second stage control, N is the statistical length, and t is the time number.

[0028] Furthermore, the fluctuation prediction model in S42 is: , where d f is the future gap value, d 2,t is the gap value at the tth moment after the second stage control, d 2,t-1 is the gap value at the t-1th moment after the second stage of control, d2,t-2 is the gap value at the t-2th moment after the second stage of control, a d is the gap value influence coefficient, μ is the real-time fluctuation factor of the current gap value after the second stage control, a μ is the influence coefficient of the fluctuation factor, and t is the time number.

[0029] Furthermore, the cascade combination model in S43 is: , where U(t+1) is the output of the cascade combination model at time t+1, and U PI (t+1) is the output of the PI control model with adjustable proportional integral at the time t+1, and γ is the conversion proportional coefficient.

[0030] The beneficial effect of the above-mentioned further scheme is: in the PI control model, the actual roller coating power value is not completely consistent with the target roller coating power value, but there is oscillation. When the control accuracy is high, the oscillation is small, and when the control accuracy is low, the oscillation is large. Therefore, the present invention establishes a fluctuation prediction model based on the real-time fluctuation factor and the gap value to predict the future gap value, that is, the gap value at the next moment, and corrects the output of the PI control model with adjustable proportional integral to achieve higher-precision control.

[0031] Furthermore, S5 is specifically as follows: using the cascade combination model of the roller coating process to perform the third stage of control on the power of the roller coating motor, setting multiple specific coefficient values ​​for the conversion ratio coefficient in the cascade combination model, adding each coefficient value to the cascade combination model, constructing the objective function, and when the target value is the minimum, the corresponding coefficient value is the optimal conversion ratio coefficient, thereby obtaining the optimized cascade combination model.

[0032] Furthermore, the objective function is: , where L is the target value, d 3,t is the gap value at the tth moment after the third stage control, μ 3,t is the real-time fluctuation factor at the tth moment after the third stage of control, T is the current moment, t is the moment number, and | | is the absolute value.

[0033] The beneficial effect of the above further scheme is: step S5 of the present invention is to find an optimal conversion ratio coefficient. When the objective function composed of the gap value and the real-time fluctuation factor after the third stage control is minimized, the corresponding coefficient value is the optimal conversion ratio coefficient, which minimizes the oscillation of the control process and improves the control accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The present invention is a flow chart of a stable control method for roller coating. DETAILED DESCRIPTION

[0035] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0036] like Figure 1 As shown, a stable control method for roller coating comprises the following steps:

[0037] S1, first stage control: adopt the PI control model with the initial proportional coefficient and the initial integral coefficient to control the power of the roller coating motor in the first stage, wherein the PI control model is a proportional integral controller;

[0038] S2. Constructing a PI control model with adjustable proportional integral: according to the trend of the difference between the actual roller coating power value and the target roller coating power value after the first stage of control, a proportional integral regulation sub-model is established, and the proportional integral regulation sub-model is added to the PI control model to obtain a PI control model with adjustable proportional integral;

[0039] S3, second stage control: using a PI control model with adjustable proportional integral to control the power of the roller coating motor in the second stage;

[0040] S4, constructing a cascade combination model of the roller coating process: according to the gap value between the actual roller coating power value and the target roller coating power value after the second stage control and the real-time fluctuation factor, predicting the future gap value, cascading the proportional integral adjustable PI control model, and obtaining the cascade combination model of the roller coating process;

[0041] S5, third stage control and optimization: using the cascade combination model of the roller coating process to control the power of the roller coating motor in the third stage, and optimizing the cascade combination model;

[0042] S6. Stable control: According to the output of the optimized cascade combination model, the power of the roller coating motor is stably controlled.

[0043] In this embodiment, the types of roller coating motors include: main roller motor, coating motor, traction motor, drying motor, and tension adjustment motor.

[0044] During use, steps S1 to S5 are the parameter adjustment stage, and the optimized cascade combination model after the parameter adjustment is completed is the normal use model of the roller coating process.

[0045] The S2 comprises the following sub-steps:

[0046] S21, calculating the difference value according to the difference between the actual roller coating power value and the target roller coating power value after the first stage control;

[0047] S22, calculating the trend coefficients of multiple gap values;

[0048] S23. According to the trend coefficient, a proportional integral control sub-model is established;

[0049] S24, adding the proportional-integral regulation sub-model to the PI control model to obtain a PI control model with adjustable proportional-integral.

[0050] In this embodiment, the formula for calculating the trend coefficient in S22 is: , where ε is the trend coefficient, d 1,t is the gap value at the tth moment after the first stage of control, t is the moment number, N is the length of statistics, and the gap value d 1,t is the difference between the actual roller coating power value and the target roller coating power value after the first stage of control, and | | is the absolute value.

[0051] In this embodiment, the trend coefficient can also be replaced by the slope, with time as the independent variable and the absolute value of the gap value as the dependent variable. That is, the time and the absolute value of the gap value are fitted to construct a linear function, and the slope of the linear function is used as the trend coefficient to reflect the trend of multiple gap values.

[0052] The proportional integral control sub-model in S23 is: , ,in, is the adjusted proportional coefficient, is the adjusted integral coefficient, ε is the trend coefficient, K p,0 is the initial proportionality coefficient, K I,0 is the initial integration coefficient.

[0053] In step S24, the adjusted proportional coefficient replaces the initial proportional coefficient, and the adjusted integral coefficient replaces the initial integral coefficient, thereby obtaining a PI control model with adjustable proportional and integral.

[0054] The PI control model is a model in which the differential control term is removed from the PID control model. In the present invention, the input error term of the PI control model is the difference between the actual roller coating power value and the target roller coating power value.

[0055] The present invention calculates a trend coefficient to reflect the trend of the gap value. When the trend coefficient is greater than or equal to 0, the gap value is in a growth state. The proportional coefficient is increased on the basis of the initial proportional coefficient, and the integral coefficient is reduced to enhance the response speed of the model. When the trend coefficient is less than 0, the model is converging, and the actual roller coating power value is approaching the target roller coating power value. Therefore, the initial proportional coefficient and the initial integral coefficient are maintained.

[0056] The S4 comprises the following sub-steps:

[0057] S41, calculating the real-time fluctuation factor of the current gap value according to the gap value between the actual roller coating power value and the target roller coating power value after the second stage control;

[0058] S42, establishing a fluctuation prediction model to predict future gap values ​​based on the real-time fluctuation factor and the gap value;

[0059] S43, adding a conversion proportional coefficient to the future gap value, and cascading it with a PI control model with adjustable proportional integral to obtain a cascade combination model of the roller coating process.

[0060] The formula for calculating the real-time fluctuation factor of the current gap value in S41 is: , where μ is the real-time fluctuation factor of the current gap value after the second stage control, d 2,t is the gap value at the tth moment after the second stage control, f is the normalization function, || is the absolute value, and the gap value d 2,t is the difference between the actual roller coating power value and the target roller coating power value after the second stage control, N is the statistical length, and t is the time number.

[0061] In this embodiment, the normalization function may adopt Sigmoid and tanh functions.

[0062] The fluctuation prediction model in S42 is: , where d f is the future gap value, d 2,t is the gap value at the tth moment after the second stage control, d 2,t-1 is the gap value at the t-1th moment after the second stage of control, d 2,t-2 is the gap value at the t-2th moment after the second stage of control, a d is the gap value influence coefficient, μ is the real-time fluctuation factor of the current gap value after the second stage control, a μ is the influence coefficient of the fluctuation factor, and t is the time number.

[0063] In the present invention, the future gap value, that is, the gap value at the next moment, is predicted based on the three actual gap values ​​in the history of the second stage and the real-time fluctuation factor, and the output of the proportional-integral adjustable PI control model at the next moment is corrected.

[0064] In this embodiment, the gap value influence coefficient and the fluctuation factor influence coefficient are specifically set according to experiments.

[0065] The cascade combination model in S43 is: , where U(t+1) is the output of the cascade combination model at time t+1, and U PI (t+1) is the output of the PI control model with adjustable proportional integral at the time t+1, and γ is the conversion proportional coefficient.

[0066] In the PI control model, the actual roller coating power value is not completely consistent with the target roller coating power value, but there is oscillation. When the control accuracy is high, the oscillation is small, and when the control accuracy is low, the oscillation is large. Therefore, the present invention establishes a fluctuation prediction model according to the real-time fluctuation factor and the gap value to predict the future gap value, that is, the gap value at the next moment, and corrects the output of the PI control model with adjustable proportional integral to achieve higher-precision control.

[0067] When the future gap value is greater than 0, the actual roller coating power value is greater than the target roller coating power value, and there is a problem of over-control. PI (t) minus , when the future gap value is less than 0, the actual roller coating power value is less than the target roller coating power value, and there is a problem of insufficient control. PI (t) plus .

[0068] The S5 is specifically as follows: using the cascade combination model of the roller coating process to perform the third stage of control on the power of the roller coating motor, setting multiple specific coefficient values ​​for the conversion ratio coefficient in the cascade combination model, adding each coefficient value to the cascade combination model, constructing the objective function, and when the target value is the minimum, the corresponding coefficient value is the optimal conversion ratio coefficient, thereby obtaining the optimized cascade combination model.

[0069] S5 of the present invention is equivalent to substituting each coefficient value into the cascade combination model, each coefficient value corresponds to an objective function, the objective function obtains a target value, and when the target value is the minimum, the corresponding coefficient value is selected.

[0070] The objective function is: , where L is the target value, d 3,t is the gap value at the tth moment after the third stage control, μ 3,t is the real-time fluctuation factor at the tth moment after the third stage of control, T is the current moment, t is the moment number, and | | is the absolute value.

[0071] Gap value d 3,t is the difference between the actual roller coating power value and the target roller coating power value after the third stage control, and the real-time fluctuation factor μ3,t The calculation method of is the same as the formula of real-time volatility factor in S41.

[0072] Step S5 of the present invention is to find an optimal conversion ratio coefficient. When the objective function composed of the gap value and the real-time fluctuation factor after the third stage of control is minimized, the corresponding coefficient value is the optimal conversion ratio coefficient, which minimizes the oscillation of the control process and improves the control accuracy.

[0073] The present invention firstly adopts a PI control model with an initial proportional coefficient and an initial integral coefficient to control the power of a roller coating motor in the first stage, and establishes a proportional-integral regulation sub-model according to the control situation in the first stage to replace the initial proportional coefficient and the initial integral coefficient, so as to realize adaptive adjustment of the proportional coefficient and the integral coefficient, enter the second stage of control, predict the future gap value, and thus perform output correction on the PI control model with adjustable proportional-integral, obtain a cascade combination model of the roller coating process, and optimize the cascade combination model to improve the control accuracy of the roller coating process.

[0074] The present invention realizes more accurate configuration of proportional integral coefficient by adding a proportional integral regulation sub-model, avoids large shocks, improves the stability of the roller coating process, and then corrects the output of the PI control model with adjustable proportional integral through future gap values, further improves the control accuracy and improves the stability of the roller coating process.

[0075] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A stable control method for roller coating, characterized in that: The following steps are involved: S1, adopting the PI control model with the initial proportional coefficient and the initial integral coefficient to control the power of the roller coating motor in the first stage; S2. According to the trend of the difference between the actual roller coating power value and the target roller coating power value after the first stage control, a proportional integral control sub-model is established, and the proportional integral control sub-model is added to the PI control model to obtain a PI control model with adjustable proportional integral; The proportional integral control sub-model is: , ,in, is the adjusted proportional coefficient, is the adjusted integral coefficient, ε is the trend coefficient, K p,0 is the initial proportionality coefficient, K I,0 is the initial integration coefficient; S3, using a PI control model with adjustable proportional integral to control the power of the roller coating motor in the second stage; S4, according to the gap value and real-time fluctuation factor between the actual roller coating power value and the target roller coating power value after the second stage control, the future gap value is predicted, and the proportional integral adjustable PI control model is cascaded to obtain the cascade combination model of the roller coating process; The cascade combination model is: , where U(t+1) is the output of the cascade combination model at time t+1, and U PI (t+1) is the output of the PI control model at time t+1 with adjustable proportional integral, γ is the conversion proportional coefficient, d f is the future gap value; S5, using the cascade combination model of the roller coating process to control the power of the roller coating motor in the third stage, and optimizing the cascade combination model; The S5 is specifically as follows: using the cascade combination model of the roller coating process to control the power of the roller coating motor in the third stage, setting a plurality of specific coefficient values ​​for the conversion ratio coefficient in the cascade combination model, adding each coefficient value to the cascade combination model, constructing an objective function, when the target value is the minimum, the corresponding coefficient value is the optimal conversion ratio coefficient, and obtaining an optimized cascade combination model; The objective function is: , where L is the target value, d 3,t is the gap value at the tth moment after the third stage control, μ 3,t is the real-time fluctuation factor at the tth moment after the third stage of control, T is the current moment, t is the moment number, and | | is the absolute value; S6. According to the output of the optimized cascade combination model, the power of the roller coating motor is stably controlled.

2. The stable control method of roller coating according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21, calculating the difference value according to the difference between the actual roller coating power value and the target roller coating power value after the first stage control; S22, calculating the trend coefficients of multiple gap values; S23. According to the trend coefficient, a proportional integral control sub-model is established; S24, adding the proportional-integral regulation sub-model to the PI control model to obtain a PI control model with adjustable proportional-integral.

3. The stable control method of roller coating according to claim 2, characterized in that: The formula for calculating the trend coefficient in S22 is: , where ε is the trend coefficient, d 1,t is the gap value at the tth moment after the first stage of control, t is the moment number, N is the length of statistics, and the gap value d 1,t is the difference between the actual roller coating power value and the target roller coating power value after the first stage of control, and | | is the absolute value.

4. The stable control method of roller coating according to claim 1, characterized in that: The S4 comprises the following sub-steps: S41, calculating the real-time fluctuation factor of the current gap value according to the gap value between the actual roller coating power value and the target roller coating power value after the second stage control; S42, establishing a fluctuation prediction model to predict future gap values ​​based on the real-time fluctuation factor and the gap value; S43, adding a conversion proportional coefficient to the future gap value, and cascading it with a PI control model with adjustable proportional integral to obtain a cascade combination model of the roller coating process.

5. The stable control method of roller coating according to claim 4, characterized in that: The formula for calculating the real-time fluctuation factor of the current gap value in S41 is: , where μ is the real-time fluctuation factor of the current gap value after the second stage control, d 2,t is the gap value at the tth moment after the second stage control, f is the normalization function, || is the absolute value, and the gap value d 2,t is the difference between the actual roller coating power value and the target roller coating power value after the second stage control, N is the statistical length, and t is the time number.

6. The stable control method of roller coating according to claim 4, characterized in that: The fluctuation prediction model in S42 is: , where d f is the future gap value, d 2,t is the gap value at the tth moment after the second stage control, d 2,t-1 is the gap value at the t-1th moment after the second stage of control, d 2,t-2 is the gap value at the t-2th moment after the second stage of control, a d is the gap value influence coefficient, μ is the real-time fluctuation factor of the current gap value after the second stage control, a μ is the influence coefficient of the fluctuation factor, and t is the time number.

Images