The Research And Simulation Of Rosin Distill Column Control System

This thesis is developed on an existing problem for the control of distill column in the process of rosin and turpentine production by continuous vapor distillation. Distill column is the core of rosin and turpentine production, which has the character of non-linear, strong coupling and multivariable, and the control of distill column will influence the quality, output and energy consume of the production directly. The aim of this paper is to control the distill column and make the quality of production up to the mustard.Research methods applied in the thesis are as follows:First: via analyzing the character of distill column we bring forward the control method of feedforward-feedback and subsection control for the upside of the distill column, and self-turning PID decoupling control based on DRNN neural network for the down side of the distill column.Second: via experimental data we get the second-order system model for each control object by using least square curve simulation.Third: the using of self-turning PID controllers with DRNN neural networks overcome the disadvantages of regular PID controller and neural networks.Forth: The results of simulation with MATLAB indicate that the system is not only feasible, but also has the character of lower over adjusting, better anti-jamming.Above all, the research of this paper is not only providing a new idea for the control of distill column in the process of rosin and turpentine production by continuous vapor distillation, but also, by MATLAB simulation, the results proving that the control method we used in this paper is validity. The research method also is creative, practical and has some application and reference value.