CN103610227B - Cut tobacco dryer head and tail section process variable optimizing control method - Google Patents

Abstract

The invention discloses a process variable optimization control method at the head and tail sections of a silk drying machine. According to the historical data of process variables such as cylinder temperature, wind temperature, and damp damper at the head and tail sections of the silk drying process, a cubic function is used as the radial basis function. The Cubic-RBF-ARX model is used to model the dynamic characteristics of silk drying; the built model has the ability of self-regulation, and can reflect the influence of inlet flow and inlet moisture changes on outlet moisture under different modes, and can be adjusted according to the different modes of the head and tail sections. The change of inlet flow and inlet moisture is used to predict the change of outlet moisture in the future; according to the established model, each process variable is optimized and set, so that the control of the outlet moisture of the head and tail sections can achieve better results. The method of the present invention comprehensively considers the dynamic characteristics between the amount of incoming material and each input variable, can more effectively overcome the influence of incoming material flow and moisture changes on the head and tail sections of the drying process, and is suitable for the inlet flow and inlet flow of shredded silk in different modes. Head and tail section control when wet.

Description

A Method for Optimal Control of Process Variables in the Head and Tail Sections of a Silk Drying Machine

technical field

The invention relates to a process variable optimization control method at the head and tail sections of a silk drying machine.

Background technique

The drying process is the most important process in the production of cigarette shreds. It mainly reduces the moisture content of the shredded leaves by heating and drying the shredded leaves, so that the moisture content and temperature of the shredded leaves after baking are uniform, and controlled. Within a certain range of values to meet the requirements of the production process. The process of drying silk is mainly divided into four parts: preheating, dry head, middle and dry end. In the dry head stage, the inlet flow rate of the shredded leaves is continuously increasing, but there is no detection value of the outlet moisture of the shredded leaves, so it is difficult to carry out feedback control, which will easily cause poor quality of water control at the outlet of the dried head stage and more dry materials; The sudden decrease of the inlet flow rate and the large heat capacity of the drying drum make it difficult for the internal temperature of the drum wall to drop at a specified rate, which may also easily lead to low moisture control performance and a lot of dry material at the outlet during the dry end stage. Therefore, "dry head and dry tail" is the difficulty of outlet moisture control in the current drying process.

The existing control methods for dry start and dry end process mainly include:

(1) Use the materials entering and exiting the drying machine and the drying medium as heat and mass balance objects to establish a mathematical model, and combine the control method of feedforward PID to adjust the cylinder temperature. However, the feed-forward mathematical model only considers the moisture content and flow rate of the feed material, and does not consider other factors that have an important impact on the outlet moisture, such as hot air temperature, and cannot fully reflect the real process, resulting in large fluctuations in the outlet moisture of the head and tail silk dryers. Operators are required to intervene manually, and it is difficult to obtain a satisfactory control effect for incoming materials with different inlet flow rates and inlet moisture in different modes at the head and tail sections.

(2) On the basis of the above-mentioned feed-forward control, a steam spray humidifier is added to the head and tail sections to apply steam water to the head and tail materials to increase the moisture content of the head and tail materials to reduce the amount of dry materials. However, this method only humidifies the surface of the shredded leaf and only increases the humidity of the surface of the shredded leaf, which still causes a reduction in the internal quality of the shredded tobacco and increases the difficulty and stability of outlet moisture control.

(3) Reduce the amount of dry material through multiple tests, seek the best hot air temperature value at the head and tail stages, and adjust technological parameters such as the opening of the tide valve. This method lacks the ability of self-regulation, and cannot guarantee that the process parameters of this group are optimal setting values for incoming materials with different inlet flow rates and inlet moisture in different modes;

(4) On the basis of the PID control strategy, the idea of fuzzy control is applied to the moisture control of the silk drying machine.

Only using a simple two-dimensional fuzzy controller to solve the control problem of outlet moisture in the drying process still cannot obtain the optimal process parameter setting value, and for the changes of inlet flow and inlet moisture in different modes, fuzzy control is still required. The rule table is adjusted, which brings inconvenience to industrial production.

Contents of the invention

The technical problem to be solved by the present invention is to provide a process variable optimization control method at the head and tail section of the silk drying machine in view of the deficiencies in the prior art, so that the moisture at the outlet of the shredded silk in the drying stage can rise as fast as possible and reach a stable state quickly. Make the moisture at the outlet of shredded leaves drop as slowly as possible in the drying and tailing stage, thereby effectively reducing the amount of dry material in the head and tail section and improving the control performance of the drying process; more effectively overcome the impact of incoming material flow and moisture changes on the beginning and tail of the drying process The influence of the segment, avoiding the inconvenience of manually setting and inputting process variable parameters.

In order to solve the above-mentioned technical problems, the technical solution adopted in the present invention is: a method for optimizing the process variables in the head and tail sections of the silk drying machine, the method is:

1) According to the operation process of the silk drying machine, establish the time sequence relationship of the inlet flow of the shredded silk, the inlet moisture, the cylinder temperature, the air temperature, the damp damper, and the outlet moisture in the drying process, and at the same time, according to the drying process, there is no outlet of the shredded silk at the end of the drying stage. According to the characteristics of the moisture detection value, the inlet flow rate and the inlet moisture detection value of the dry-end stage, the cubic function is used as the Cubic-RBF-ARX model of the radial basis function, and the Cubic-RBF-ARX model of the dry-end stage and the dry-end stage of the drying process are respectively established. - RBF-ARX model;

2) According to the historical operation data of the head and tail section of the drying machine, the Cubic-RBF-ARX model of the dry head stage and the dry end stage of the silk drying process were respectively optimized by using the structured nonlinear parameter optimization method;

3) According to the optimized Cubic-RBF-ARX model of the dry end stage and dry end stage of the silk drying process, the double S-type function is used to describe the optimal input curve of the moisture exhaust damper, air temperature and cylinder temperature in the dry end stage; The jump function describes the optimal input curve of the inlet flow in the dry head stage; the exponential function is used to describe the optimal input curve of the tidal discharge damper, air temperature, barrel temperature and barrel motor frequency in the dry end stage;

4) Using the Levi Burger-Nyquilt method, by minimizing the error between the predicted value of outlet moisture and the set value of outlet moisture calculated by the Cubic-RBF-ARX model in the optimized dry head stage and dry tail stage, find out the The parameters of the optimal input curve in the dry head stage and the dry end stage of the drying process are used to adapt to changes in incoming materials and reduce the amount of dry materials in the dry end stage.

In the step 1), the Cubic-RBF-ARX model of the drying stage of the silk dryer is:

in:

Among them, y ^H (t ^H ) represents the outlet moisture of the Cubic-RBF-ARX model in the drying stage of the silk dryer; represent the tidal damper opening, wind temperature, cylinder temperature, inlet flow and inlet moisture of the Cubic-RBF-ARX model in the dry head stage; X ^H (t ^H -1) is the state variable of inlet flow and inlet moisture; np ^H , nq ^H , d ^H and m ^H all represent the order of the Cubic-RBF-ARX model at the dry head stage; Respectively, the center of the RBF neural network of the Cubic-RBF-ARX model output and input items in the dry head stage; is the scalar weight coefficient of the Cubic-RBF-ARX model in the dry head stage; ||·|| _F represents the Frobenius norm of the matrix; ξ ^H (t ^H ) is the modeling error of the Cubic-RBF-ARX model in the dry head stage, Gaussian white noise; T ₀ ^H is the sampling time of the Cubic-RBF-ARX model modeling in the drying stage of the silk dryer, T ₁ is the time from the detection value of the inlet flow to the detection value of the inlet moisture, T ₂ is the time from the inlet moisture The time from the detection value to the detection value of the outlet moisture, _T3 is the time from the detection value of the inlet moisture to the entrance of the drying drum, and _T4 is the drying time of the shredded leaves in the drying drum.

In the step 1), the Cubic-RBF-ARX model of the drying end stage of the silk dryer is:

in:

Among them, y ^T (t ^T ) represents the outlet moisture of the Cubic-RBF-ARX model at the dry end stage of the silk dryer; Respectively represent the barrel temperature, hot air temperature, tide damper opening, inlet flow, inlet moisture and barrel motor frequency of the Cubic-RBF-ARX model in the dry end stage; X ^T (t ^T -1) is the hot air temperature and barrel The state variable of the motor frequency; np ^T , nq ^T , d ^T and m ^T all represent the order of the Cubic-RBF-ARX model in the dry tail stage; are the centers of the RBF neural network of the output and input items of the Cubic-RBF-ARX model in the dry tail stage; is the scalar weight coefficient of the Cubic-RBF-ARX model at the dry end stage; ξ ^T (t ^T ) is the modeling error of the Cubic-RBF-ARX model at the dry end stage, which is Gaussian white noise; T ₀ ^T is the dry end stage of the silk dryer The Cubic-RBF-ARX model models sampling time.

In the step 2), the optimization of the Cubic-RBF-ARX model in the drying stage of the silk dryer is as follows:

$<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

in, is the actual value of the outlet moisture in the drying stage of the drying machine, is the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model at the drying stage of the silk dryer under the actual input; ${\widehat{\mathrm{\&theta;}}}_L^h=\{{\mathrm{\&omega;}}_0^{h,0},{\mathrm{\&omega;}}_{i^h,0}^{{\mathrm{the\; y}}^h},{\mathrm{\&omega;}}_{\mathrm{no},j^h,0}^{u^h},{\mathrm{\&omega;}}_{k^h}^{h,0},{\mathrm{\&omega;}}_{i^h,k^h}^{{\mathrm{the\; y}}^h},{\mathrm{\&omega;}}_{j^h,k^h}^{u^h}|i^h=1,...,{\mathrm{np}}^h;j^h=1,...,{\mathrm{nq}}^h;k^h=1,...,m^h\}$ is the linear parameter of the Cubic-RBF-ARX model in the drying stage of the silk dryer; is the nonlinear parameter of the Cubic-RBF-ARX model in the drying stage of the silk dryer; N ^H is the modeling data length of the Cubic-RBF-ARX model in the drying stage of the silk drying machine.

The optimization of the Cubic-RBF-ARX model in the dry end stage of the silk dryer is as follows:

$<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$

in, is the actual value of the outlet moisture in the drying process of the silk drying machine; is the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model at the dry end stage of the silk dryer under the actual input; ${\mathrm{\&theta;}}_L^T=\{{\mathrm{\&omega;}}_0^{T,0},{\mathrm{\&omega;}}_{i^T,0}^{{\mathrm{the\; y}}^T},{\mathrm{\&omega;}}_{\mathrm{no},j^T,0}^{u^T},{\mathrm{\&omega;}}_{k^T}^{T,0},{\mathrm{\&omega;}}_{i^T,k^T}^{{\mathrm{the\; y}}^T},{\mathrm{\&omega;}}_{j^T,k^T}^{u^T}|i^T=1,...,{\mathrm{np}}^T;j^T=1,...,{\mathrm{nq}}^T;k^T=1,...,m^T\}$ is the linear parameter of the Cubic-RBF-ARX model in the dry end stage of the silk dryer, N ^T is the nonlinear parameter of the Cubic-RBF-ARX model in the dry end stage of the silk dryer; NT is the modeling data length of the Cubic-RBF-ARX model in the dry end stage of the silk dryer.

In step 3):

The expression of the double S-type function used to describe the optimal input curves of the damper, wind temperature and cylinder temperature in the drying stage of the silk dryer is:

${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; sc</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 5</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 6</span>}}}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 7</span>}}}}}$

Among them, t ^s is the input time, the unit is s; λ ₁ , λ ₄ , λ ₅ are the starting point, turning point and end point of the double S-shaped function respectively; λ ₂ , λ ₆ are the two symmetrical axis center position; λ ₃ and λ ₇ are the rising or falling speeds of the double S-shaped function respectively; when λ ₃ and λ ₇ are greater than 0, it means that the S-shaped function is rising, and when λ ₃ and λ ₇ are less than 0, it means that the S-shaped function is falling; c =1,2,3, U _s1 (t ^s ) is the setting value of the damper; U _s2 (t ^s ) is the setting value of the air temperature; U _s3 (t ^s ) is the setting value of the cylinder temperature.

The expression of the step function used to describe the optimal input curve of the inlet flow in the drying stage of the drying machine is:

${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}& {\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ]</span>\\ {\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}& {\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; ]</span>\end{array}\right.<span\; class="notranslate">\; ;</span>$

Among them, t ^T is the input time, the unit is s; κ ₁ , κ ₂ , κ ₃ are the rising speed, rising time and final value of the step function respectively.

In the step 4), the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model in the drying stage of the silk dryer for: ${\stackrel{\&OverBar;}{\mathrm{the\; y}}}^h\left(t^a\right)=f(u_{\mathrm{the\; s}1}\left(t^a\right),u_{\mathrm{the\; s}2}\left(t^a\right),u_{\mathrm{the\; s}3}\left(t^a\right),u_T\left(t^a\right)),$ By substituting the optimized setting curve of each process variable in the dry head stage of the silk dryer into the input variables of Cubic-RBF-ARX in the dry head stage get in. Predicted value of outlet moisture calculated by making dry head stage Cubic-RBF-ARX model The error e _H (t ^a ) with the outlet moisture setting value y _set (t ^a ) is the smallest, that is, the Levi Burger-Nyquilt method is used to solve the optimization problem Find the parameter λ _x of the input curve of the tidal discharge damper, wind temperature and cylinder temperature in the dry head stage, and the parameters κ ₁ , κ ₂ , κ ₃ of the input curve of the inlet flow; among them, x=1,2,...,7; g =1,2,3; M is the duration of the dry head stage.

In the step 3), the expression of the exponential function used to describe the optimal input curve of the damper, wind temperature, cylinder temperature and cylinder motor frequency in the dry end stage is:

$\begin{array}{c}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; zp</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; z</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&times;</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; z</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}\\ \mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2,3,4</span>}\end{array}<span\; class="notranslate">\; ;</span>$

In the formula, U _z1 (t ^z ), U _z2 (t ^z ), U _z3 (t ^z ), and U _z4 (t ^z ) represent the optimal frequency of the damper, wind temperature, cylinder temperature, and cylinder motor frequency in the dry end stage, respectively. Enter the curve.

In the step 4), the predicted value of outlet moisture calculated by the Cubic-RBF-ARX model in the dry tail stage for: ${\stackrel{\&OverBar;}{\mathrm{the\; y}}}^T\left(t^b\right)=f(u_{z1}\left(t^b\right),u_{z2}\left(t^b\right),u_{z3}\left(t^b\right),u_{z4}\left(t^b\right)),$ By substituting the optimized setting curves of each process variable in the dry end stage of the silk dryer into the input variables of the constructed Cubic-RBF-ARX model in the dry end stage Obtained in; the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model of the dry tail stage The error e _T (t ^b ) with the outlet moisture set value y' _set (t ^b ) is the smallest, that is, the Levi Burger-Nyquilt method is used to solve the optimization problem Find the parameter α _pg for the optimal input curve of the damper, air temperature, cylinder temperature and cylinder motor frequency in the dry end stage; where, g=1,2,3; M' is the duration of the dry end stage.

Compared with the prior art, the present invention has the beneficial effects that: the method of the present invention can make the outlet moisture of shredded leaves in the first stage of drying rise as fast as possible and reach a steady state quickly, and can make the outlet moisture of shredded leaves in the tail stage of drying as fast as possible. It may decrease slowly, thereby effectively reducing the amount of dry material in the head and tail sections, improving the control performance of the silk drying process, and having greater economic value; the method of the present invention comprehensively considers the dynamic characteristics between the amount of incoming material and each input variable, It can more effectively overcome the impact of incoming material flow and moisture changes on the head and tail sections of the drying process, and is suitable for the head and tail section control of the shred inlet flow and inlet moisture in different modes; the method of the present invention optimizes the most based on the identification model. Excellent input setting curve avoids the inconvenience of manually setting input process variable parameters.

Description of drawings

Figure 1 is a schematic diagram of the process of the silk drying machine.

Detailed ways

The process of the wire drying machine is shown in Figure 1. Before the shredded leaf enters the drying process, the inlet flow rate u ₄ and the inlet moisture u ₅ of the shredded leaf are detected first. After _T3 time, the shredded leaves arrive at the entrance of the silk dryer. When the shredded leaves are drying in the silk drying cylinder, the system will regularly sample the process variable parameter values such as the opening of the damp damper u ₁ , the air temperature u ₂ , and the cylinder temperature u ₃ of the cylinder. The drying process lasts for _T4 time, and the dried shredded leaves are poured out from the outlet of the drying drum, and the outlet moisture value y of the shredded leaves is measured at the outlet. It takes a long time from the detection value of the inlet flow to the detection value of the outlet moisture, for example, it takes about 340s for a silk drying production line. In addition, there is a large time lag between the input/output variables of the silk dryer.

When an inlet flow is detected, it indicates that the drying process is running. In the initial stage of operation, the silk drying process has the detection value of the inlet flow and inlet moisture of the shredded silk, but there is no detection value of the outlet moisture of the shredded silk. At this time, the drying stage of the shredded shredded process starts. According to the characteristics of the drying stage of the silk drying process, the Cubic-RBF-ARX model structure is established:

in:

Among them, y ^H (t ^H ) represents the outlet moisture of the Cubic-RBF-ARX model in the drying stage of the silk dryer; represent the tidal damper opening, wind temperature, cylinder temperature, inlet flow and inlet moisture of the Cubic-RBF-ARX model in the dry head stage; X ^H (t ^H -1) is the state variable of inlet flow and inlet moisture; np ^H , nq ^H , d ^H and m ^H all represent the order of the Cubic-RBF-ARX model at the dry head stage; Respectively, the center of the RBF neural network of the Cubic-RBF-ARX model output and input items in the dry head stage; is the scalar weight coefficient of the Cubic-RBF-ARX model in the dry head stage; ||·|| _F represents the Frobenius norm of the matrix; ξ ^H (t ^H ) is the modeling error of the Cubic-RBF-ARX model in the dry head stage, Gaussian white noise; T ₀ ^H is the sampling time of the Cubic-RBF-ARX model modeling in the drying stage of the silk dryer, T ₁ is the time from the detection value of the inlet flow to the detection value of the inlet moisture, and T ₂ is the time from the inlet moisture The time from the detection value to the detection value of the outlet moisture, _T3 is the time from the detection value of the inlet moisture to the entrance of the drying drum, and _T4 is the drying time of the shredded leaves in the drying drum.

When the inlet flow rate changes from the normal value to 0, it marks the beginning of the drying process, and when the outlet moisture drops to 3%, it marks the end of the whole drying process of the silk dryer. There is no inlet flow detection value during the dry tail process, but there is an outlet moisture detection value. According to the characteristics of the dry end process of the silk dryer, the following Cubic-RBF-ARX model is established:

in:

Among them, y ^T (t ^T ) represents the outlet moisture of the Cubic-RBF-ARX model at the dry end stage of the silk dryer; Respectively represent the barrel temperature, hot air temperature, tide damper opening, inlet flow, inlet moisture and barrel motor frequency of the Cubic-RBF-ARX model in the dry end stage; X ^T (t ^T -1) is the hot air temperature and barrel The state variable of the motor frequency; np ^T , nq ^T , d ^T and m ^T all represent the order of the Cubic-RBF-ARX model in the dry tail stage; are the centers of the RBF neural network of the output and input items of the Cubic-RBF-ARX model in the dry tail stage; is the scalar weight coefficient of the Cubic-RBF-ARX model at the dry end stage; ξ ^T (t ^T ) is the modeling error of the Cubic-RBF-ARX model at the dry end stage, which is Gaussian white noise; T ₀ ^T is the dry end stage of the silk dryer The Cubic-RBF-ARX model models sampling time.

The present invention uses a structured nonlinear parameter optimization method (SNPOM) method to estimate the model. In order to enable the Cubic-RBF-ARX model constructed above to describe the global dynamic characteristics of the head and tail sections of the drying process, we first use the SNPOM method to optimize the parameters of the model in the case of the minimum one-step prediction error, and use this parameter as the long-term Predict the initial values of the model parameters under the optimization objective. Then, the Levi Burger-Nyquilt method (LMM) is used to optimize the model parameters with the best long-term forecast performance.

The parameter optimization problem of the Cubic-RBF-ARX model (1) in the drying stage of the silk dryer is as follows:

$<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 5</span>}<span\; class="notranslate">\; )</span>$

in, is the actual value of the outlet moisture in the drying stage of the drying machine, is the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model at the drying stage of the silk dryer under the actual input; ${\widehat{\mathrm{\&theta;}}}_L^h=\{{\mathrm{\&omega;}}_0^{h,0},{\mathrm{\&omega;}}_{i^h,0}^{{\mathrm{the\; y}}^h},{\mathrm{\&omega;}}_{\mathrm{no},j^h,0}^{u^h},{\mathrm{\&omega;}}_{k^h}^{h,0},{\mathrm{\&omega;}}_{i^h,k^h}^{{\mathrm{the\; y}}^h},{\mathrm{\&omega;}}_{j^h,k^h}^{u^h}|i^h=1,...,{\mathrm{np}}^h;j^h=1,...,{\mathrm{nq}}^h;k^h=1,...,m^h\}$ is the linear parameter of the Cubic-RBF-ARX model in the drying stage of the silk dryer; is the nonlinear parameter of the Cubic-RBF-ARX model in the drying stage of the silk dryer; N ^H is the modeling data length of the Cubic-RBF-ARX model in the drying stage of the silk drying machine.

The parameter optimization problem of the Cubic-RBF-ARX model (3) in the dry end stage of the silk dryer is as follows:

$<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 6</span>}<span\; class="notranslate">\; )</span>$

in, is the actual value of the outlet moisture in the drying process of the silk drying machine; is the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model at the dry end stage of the silk dryer under the actual input; ${\mathrm{\&theta;}}_L^T=\{{\mathrm{\&omega;}}_0^{T,0},{\mathrm{\&omega;}}_{i^T,0}^{{\mathrm{the\; y}}^T},{\mathrm{\&omega;}}_{\mathrm{no},j^T,0}^{u^T},{\mathrm{\&omega;}}_{k^T}^{T,0},{\mathrm{\&omega;}}_{i^T,k^T}^{{\mathrm{the\; y}}^T},{\mathrm{\&omega;}}_{j^T,k^T}^{u^T}|i^T=1,...,{\mathrm{np}}^T;j^T=1,...,{\mathrm{nq}}^T;k^T=1,...,m^T\}$ is the linear parameter of the Cubic-RBF-ARX model in the dry end stage of the silk dryer, N ^T is the nonlinear parameter of the Cubic-RBF-ARX model in the dry end stage of the silk dryer; NT is the modeling data length of the Cubic-RBF-ARX model in the dry end stage of the silk dryer.

According to the estimated Cubic-RBF-ARX model in the drying stage of the drying process, the optimal input curve of each process variable is designed to adapt to the change of the incoming material and minimize the amount of dry material in the drying stage. The present invention uses a double S-shaped function to describe the optimal input curve of the damper, air temperature, and cylinder temperature in the dry head stage, and uses a step-type function to describe the optimal input curve of the inlet flow.

The double S-curve formula is as follows:

${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; sc</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 5</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 6</span>}}}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 7</span>}}}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 7</span>}<span\; class="notranslate">\; )</span>$

Among them, t ^s is the input time, the unit is s; λ ₁ , λ ₄ , λ ₅ are the starting point, turning point and end point of the double S-shaped function respectively; λ ₂ , λ ₆ are the two symmetrical axis center position; λ ₃ and λ ₇ are the rising or falling speeds of the double S-shaped function respectively; when λ ₃ and λ ₇ are greater than 0, it means that the S-shaped function is rising, and when λ ₃ and λ ₇ are less than 0, it means that the S-shaped function is falling; c =1,2,3, U _s1 (t ^s ) is the setting value of the damper; U _s2 (t ^s ) is the setting value of the air temperature; U _s3 (t ^s ) is the setting value of the cylinder temperature.

The step function formula describing the inlet flow input curve is as follows:

${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}& {\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ]</span>\\ {\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}& {\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; ]</span>\end{array}\right.<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 8</span>}<span\; class="notranslate">\; )</span>$

Among them, t ^T is the input time, the unit is s; κ ₁ , κ ₂ , κ ₃ are the rising speed, rising time and final value of the step function respectively.

Substitute the optimal setting curve (7-8) of each process variable into the input variable of the constructed Cubic-RBF-ARX model (1) , the predicted value of outlet moisture in the dry head stage can be obtained

${\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 9</span>}<span\; class="notranslate">\; )</span>$

Using the Levenberg-Marquardt Method (LMM) method, by minimizing the error between the predicted value of the outlet moisture calculated by the model and the set value of the outlet moisture, find out the tide discharge damper and wind temperature at the dry end stage. , λ _i (i=1,2,...,7) parameters of the optimal input curve of cylinder temperature and κ _j (j=1,2,3) parameters of the optimal input curve of inlet flow. The error between the set value of outlet moisture in the dry head stage and the predicted value based on the dry head dynamic model (9) is:

${\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; set</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; )</span>$

y _set (t ^a ) is the outlet moisture setting value.

The optimization problem of the optimal setting of process variables in the dry head stage is as follows:

$\underset{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; g</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\mathrm{<span\; class="notranslate">\; J</span>}<span\; class="notranslate">\; =</span>\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; a</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 11</span>}<span\; class="notranslate">\; )</span>$

M is the duration of the dry head phase. By solving the above optimization problems, the parameter values of the optimal setting curve can be obtained, so as to design the optimal input curve of each process variable in the drying stage of the silk drying machine.

According to the estimated Cubic-RBF-ARX model in the dry end stage of the drying process, the optimal input curve of each process variable is designed to adapt to the change of incoming materials and minimize the amount of dry material in the dry end stage. An exponential function is used to describe the optimal input curve of the damper, air temperature, cylinder temperature and cylinder motor frequency in the dry end stage. The formula of the exponential curve is as follows:

$\begin{array}{c}{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; zp</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; z</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&times;</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; z</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}\\ \mathrm{<span\; class="notranslate">\; p</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1,2,3,4</span>}\end{array}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 12</span>}<span\; class="notranslate">\; )</span>$

In the formula, U _z1 (t ^z ), U _z2 (t ^z ), U _z3 (t ^z ), and U _z4 (t ^z ) represent the optimal frequency of the damper, wind temperature, cylinder temperature, and cylinder motor frequency in the dry end stage, respectively. Enter the curve. Substitute the optimized setting curve (12) of each process variable into the input variable of the constructed Cubic-RBF-ARX model (3) , the predicted value of outlet moisture in the dry tail stage can be obtained:

${\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; z</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; z</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; z</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; z</span>}\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 13</span>}<span\; class="notranslate">\; )</span>$

Using the Levi Burger Nyquilt (LMM) method, by minimizing the error between the predicted value of the outlet moisture calculated by the model and the set value of the outlet moisture, find out the damper, wind temperature, cylinder temperature and cylinder temperature in the dry end stage α _pg of the optimal input curve of motor frequency; where, g=1,2,3. The error between the set value of outlet moisture in the dry tail stage and the predicted value based on the dry tail dynamic model (13) is:

${\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{{\mathrm{<span\; class="notranslate">\; the\; y</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}_{\mathrm{<span\; class="notranslate">\; set</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 14</span>}<span\; class="notranslate">\; )</span>$

y _set (t) is the set value of outlet moisture.

The optimization problem of the optimal setting of process variables in the dry tail stage is as follows:

$\underset{{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; pg</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}{\mathrm{<span\; class="notranslate">\; J</span>}}^{<span\; class="notranslate">\; \&prime;</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; m</span>}}^{<span\; class="notranslate">\; \&prime;</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; e</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; b</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 15</span>}<span\; class="notranslate">\; )</span>$

M is the duration of the dry tail phase. By solving the above optimization problems, the parameter values of the optimal setting curve can be obtained, so as to design the optimal input curve of each process variable in the drying stage of the silk drying machine.

Claims (3)

1. A process variable optimization control method at the head and tail section of a silk drying machine, characterized in that, the method is: 1) According to the operation flow of the silk drying machine, establish the time sequence relationship of the inlet flow of shredded shreds, inlet moisture, cylinder temperature, wind temperature, moisture discharge damper, and outlet moisture in the shredded shredded drying process, and at the same time, according to the drying process, there is no shredded shredded outlet at the end of the drying stage. According to the characteristics of the moisture detection value, the inlet flow rate and the inlet moisture detection value of the dry-end stage, the cubic function is used as the Cubic-RBF-ARX model of the radial basis function, and the Cubic-RBF-ARX model of the dry-end stage and the dry-end stage of the drying process are respectively established. - RBF-ARX model; 2) According to the historical operation data of the head and tail sections of the silk drying machine, the Cubic-RBF-ARX models of the dry head stage and the dry end stage of the silk drying process were respectively optimized by using the structured nonlinear parameter optimization method; 3) According to the optimized Cubic-RBF-ARX model of the dry end stage and the end stage of the drying process, the double S-shaped function is used to describe the optimal input curve of the damper, air temperature, and cylinder temperature in the dry end stage; The jump function describes the optimal input curve of the inlet flow in the dry head stage; the exponential function is used to describe the optimal input curve of the tidal discharge damper, air temperature, barrel temperature and barrel motor frequency in the dry end stage; 4) Using the Levi Burger-Nyquilt method, by minimizing the error between the predicted value of outlet moisture and the set value of outlet moisture calculated by the Cubic-RBF-ARX model in the optimized dry head stage and dry tail stage, to find out The parameters of the optimal input curve in the dry head stage and the dry end stage of the drying process to adapt to the change of incoming materials and reduce the amount of dry materials in the dry end stage; In the step 1), the Cubic-RBF-ARX model of the dry head stage of the silk dryer is: in: Among them, y ^H (t ^H ) represents the outlet moisture of the Cubic-RBF-ARX model in the drying stage of the silk dryer; represent the tidal damper opening, wind temperature, cylinder temperature, inlet flow and inlet moisture of the Cubic-RBF-ARX model in the dry head stage; X ^H (t ^H -1) is the state variable of inlet flow and inlet moisture; np ^H , nq ^H , d ^H and m ^H all represent the order of the Cubic-RBF-ARX model at the dry head stage; Respectively, the center of the RBF neural network of the Cubic-RBF-ARX model output and input items in the dry head stage; is the scalar weight coefficient of the Cubic-RBF-ARX model in the dry head stage; ||·|| _F represents the Frobenius norm of the matrix; ξ ^H (t ^H ) is the modeling error of the Cubic-RBF-ARX model in the dry head stage, Gaussian white noise; T ₀ ^H is the sampling time of the Cubic-RBF-ARX model modeling in the drying stage of the silk dryer, T ₁ is the time from the detection value of the inlet flow to the detection value of the inlet moisture, and T ₂ is the time from the inlet moisture The time from the detection value to the detection value of the outlet moisture, _T3 is the time from the detection value of the inlet moisture to the entrance of the drying drum, and _T4 is the drying time of the shredded leaves in the drying drum; In said step 1), the Cubic-RBF-ARX model of the drying end stage of the silk dryer is: in: Among them, y ^T (t ^T ) represents the outlet moisture of the Cubic-RBF-ARX model at the dry end stage of the silk dryer; Respectively represent the barrel temperature, hot air temperature, tide damper opening, inlet flow, inlet moisture and barrel motor frequency of the Cubic-RBF-ARX model in the dry end stage; X ^T (t ^T -1) is the hot air temperature and barrel The state variable of the motor frequency; np ^T , nq ^T , d ^T and m ^T all represent the order of the Cubic-RBF-ARX model in the dry tail stage; are the centers of the RBF neural network of the output and input items of the Cubic-RBF-ARX model in the dry tail stage; is the scalar weight coefficient of the Cubic-RBF-ARX model at the dry end stage; ξ ^T (t ^T ) is the modeling error of the Cubic-RBF-ARX model at the dry end stage, which is Gaussian white noise; T ₀ ^T is the dry end stage of the silk dryer Cubic-RBF-ARX model modeling sampling time; In said step 2), the Cubic-RBF-ARX model optimization of the dry head stage of the silk dryer is as follows: $<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; oh</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$ in, is the actual value of the outlet moisture in the drying stage of the drying machine, is the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model at the drying stage of the silk dryer under the actual input; ${\widehat{\mathrm{\&theta;}}}_L^h=\{{\mathrm{\&omega;}}_0^{h,0},{\mathrm{\&omega;}}_{i^h,0}^{{\mathrm{the\; y}}^h},{\mathrm{\&omega;}}_{\mathrm{no},j^h,0}^{u^h},{\mathrm{\&omega;}}_{k^h}^{h,0},{\mathrm{\&omega;}}_{i^h,k^h}^{{\mathrm{the\; y}}^h},{\mathrm{\&omega;}}_{j^h,k^h}^{u^h}|i^h=1,...,{\mathrm{np}}^h;j^h=1,...,{\mathrm{nq}}^h;k^h=1,...,m^h\}$ is the linear parameter of the Cubic-RBF-ARX model in the drying stage of the silk dryer; is the nonlinear parameter of the Cubic-RBF-ARX model in the drying stage of the silk dryer; N ^H is the modeling data length of the Cubic-RBF-ARX model in the drying stage of the silk drying machine; The optimization of the Cubic-RBF-ARX model in the dry end stage of the silk dryer is as follows: $<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; arg</span>}\underset{{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}_{\mathrm{<span\; class="notranslate">\; L</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{\mathrm{<span\; class="notranslate">\; min</span>}}\underset{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{{\mathrm{<span\; class="notranslate">\; N</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{<span\; class="notranslate">\; (</span>{\stackrel{<span\; class="notranslate">\; \&OverBar;</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\stackrel{<span\; class="notranslate">\; ^</span>}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; ot</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; )</span>}^{\mathrm{<span\; class="notranslate">\; 2</span>}}$ in, is the actual value of the outlet moisture in the drying process of the silk drying machine; is the predicted value of the outlet moisture calculated by the Cubic-RBF-ARX model at the dry end stage of the silk dryer under the actual input; ${\mathrm{\&theta;}}_L^T=\{{\mathrm{\&omega;}}_0^{T,0},{\mathrm{\&omega;}}_{i^T,0}^{{\mathrm{the\; y}}^T},{\mathrm{\&omega;}}_{\mathrm{no},j^T,0}^{u^T},{\mathrm{\&omega;}}_{k^T}^{T,0},{\mathrm{\&omega;}}_{i^T,k^T}^{{\mathrm{the\; y}}^T},{\mathrm{\&omega;}}_{j^T,k^T}^{u^T}|i^T=1,...,{\mathrm{np}}^T;j^T=1,...,{\mathrm{nq}}^T;k^T=1,...,m^T\}$ is the linear parameter of the Cubic-RBF-ARX model in the dry end stage of the silk dryer, is the nonlinear parameter of the Cubic-RBF-ARX model in the drying end stage of the silk drying machine, and ^NT is the modeling data length of the Cubic-RBF-ARX model in the drying head stage of the silk drying machine; In the step 3): The expression of the double S-type function used to describe the optimal input curves of the damper, wind temperature and cylinder temperature in the drying stage of the silk dryer is: ${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; sc</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 4</span>}}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 5</span>}}}{\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; e</span>}}^{\frac{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; the\; s</span>}}<span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 6</span>}}}{{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; 7</span>}}}}}$ Among them, t ^s is the input time, the unit is s; λ ₁ , λ ₄ , λ ₅ are the starting point, turning point and end point of the double S-shaped function respectively; λ ₂ , λ ₆ are the two symmetrical axis center position; λ ₃ and λ ₇ are the rising or falling speeds of the double S-shaped function respectively; when λ ₃ and λ ₇ are greater than 0, it means that the S-shaped function is rising, and when λ ₃ and λ ₇ are less than 0, it means that the S-shaped function is falling; c =1,2,3, U _s1 (t ^s ) is the setting value of the damper; U _s2 (t ^s ) is the setting value of the air temperature; U _s3 (t ^s ) is the setting value of the cylinder temperature; The expression of the step function used to describe the optimal input curve of the inlet flow in the drying stage of the drying machine is: ${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\frac{{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}}{{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}& {\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ]</span>\\ {\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}& {\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; T</span>}}<span\; class="notranslate">\; \&Element;</span><span\; class="notranslate">\; [</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>{\mathrm{<span\; class="notranslate">\; \&kappa;</span>}}_{\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; ]</span>\end{array}\right.<span\; class="notranslate">\; ;</span>$ Among them, t ^T is the input time, the unit is s; κ ₁ , κ ₂ , κ ₃ are the rising speed, rising time and final value of the step function respectively; In the step 3), the expression of the exponential function used to describe the optimal input curve of the damper, wind temperature, cylinder temperature and cylinder motor frequency in the dry end stage is: ${\mathrm{<span\; class="notranslate">\; u</span>}}_{\mathrm{<span\; class="notranslate">\; zp</span>}}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; z</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; \&times;</span>{<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; )</span>}^{{\mathrm{<span\; class="notranslate">\; t</span>}}^{\mathrm{<span\; class="notranslate">\; z</span>}}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; \&alpha;</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; 3</span>}}<span\; class="notranslate">\; ;</span>$ p=1,2,3,4 In the formula, U _z1 (t ^z ), U _z2 (t ^z ), U _z3 (t ^z ), and U _z4 (t ^z ) represent the optimal frequency of the damper, wind temperature, cylinder temperature, and cylinder motor frequency in the dry end stage, respectively. Enter the curve. 2. The process variable optimization control method at the head and tail section of the silk drying machine according to claim 1, characterized in that, in the step 4), the optimal setting curve U _sc (t ^s ) of the drying head stage of the silk drying machine, U _T (t ^T ) is substituted into the input variable of the Cubic-RBF-ARX model in the dry head stage of the silk dryer In the process, the predicted value of outlet moisture calculated by the Cubic-RBF-ARX model in the drying stage of the silk dryer is obtained Predicted value of outlet moisture calculated by making dry head stage Cubic-RBF-ARX model The error e _H (t ^a ) with the outlet moisture setting value y _set (t ^a ) is the smallest, that is, the Levi Burger-Nyquilt method is used to solve the optimization problem Find the parameter λ _x of the input curve of the tide discharge damper, air temperature and cylinder temperature in the dry head stage, and the parameters κ ₁ , κ ₂ , κ ₃ of the input curve of the inlet flow; where, x=1,2,...,7; g =1,2,3; M is the duration of the dry head stage. 3. The process variable optimization control method at the head and tail section of the silk drying machine according to claim 1, characterized in that, in the step 4), the optimal setting curve U _zp (t ^z ) for the dry tail stage of the silk drying machine is substituted into The input variables of the Cubic-RBF-ARX model in the dry end stage of the silk dryer In the process, the predicted value of outlet moisture calculated by the Cubic-RBF-ARX model at the dry end stage of the silk dryer is obtained Predicted value of outlet moisture calculated by making dry tail stage Cubic-RBF-ARX model The error e _T (t ^b ) with the outlet moisture set value y' _set (t ^b ) is the smallest, that is, the Levi Burger-Nyquilt method is used to solve the optimization problem Find the parameter α _pg for the optimal input curve of the damper, wind temperature, cylinder temperature and cylinder motor frequency in the dry end stage; where, g=1,2,3; M' is the duration of the dry end stage.