Research On Temperature Control Of Distillation Column Based On Predictive Control

Distillation process is an operation process which has been widely applied in oil refining, petrochemical and other chemical processes. Due to the complexity of internal mechanism, slow response of control function and the parameters of the close correlation relationship, the control of distillation process is highly demanding and difficult. Meanwhile, it is also non-linear, strong-coupling, and difficult to do online test of quality parameters, the automatic control of the distillation column has become a hot spot in the area of process control industry.In this thesis, the background is the shale oil refining production line in shale oil refining factory and the study object is the oil shale distillation column. The thesis analyzes the basic control requirements of the oil shale distillation as well as the basic principles and methods of temperature control of the distillation column. The thesis studies a binary distillation column that has only two kinds of products in the production process. Regular column temperature control is divided into two parts, the first one is to control rectifying section temperature with return flow, and the second is to control distillation section temperature by changing the amount of heating steam which is accessed to the reboiler. According to the actual requirement of the production process, the thesis adopts rectifying section temperature control mode, and it constitutes the cascade control system with return flow as the vice ring and temperature as the main ring. Because temperature can be changed by many factors and the parameters can influence each other, the predictive control is used to design the primary ring temperature controller with PID control as the vice ring in the cascade control system because of its strong robustness to the system parameter variations, which forms the rectifying section temperature cascade control system based on dynamic matrix predictive control.Finally, the thesis analyzes the control performance of the control system designed in the thesis by simulation. Compared with the performance of the conventional PID control, the predictive control system in this thesis not only has better tracking and strong anti-disturbance performance but also can adapt to changes in parameters better, that is, it is quite effective in the rectifying section temperature control system.