Research On Multi-Objective Optimization And Control Of Distillation Process

Distillation has been widely used in refining, petrochemical and chemical industries as a common separating unit. From the perspective of process control, distillation process has the characteristics of strong coupling, multi-disturbance and nonlinear, conventional control strategy has been difficult to improve the quality of the process control. With the development of process industry, distillation process requires not only good control performance, but also good quality of products, energy consumption and steady operation range, so distillation process is actually a complicated process of multi-objective control and optimization.Dynamic Matrix Control (DMC) is a predictive control method, which take into account the industrial requirements, has good performance and strong robustness. Multi-objective optimization method, which based on control algorithm, can solve economic optimization problems of distillation process. This dissertation does some researches on dynamic control and optimization of distillation process based on these two methods. The main works are as follows:1. The equipment and principles of distillation process are briefly introduced. The energy conservation methods of distillation processes are summarized and the difficulties are analyzed. Then the current research and the application of distillation process control and optimization are introduced, also the existing problems are analyzed.2. The characteristics and control objectives of the distillation process control are analyzed. Mainly, the conventional distillation process control and advanced control methods are introduced. Finally, the distillation towers and a single tower system optimization methods and the current optimization software are introduced.3. To deal with the multi-objective optimization and control problem in the distillation process, a two-layered DMC optimal control strategy which faces production and consumption is proposed. The strategy includes the dynamic control layer for zone control along with the steady state optimization layer for the economic target optimization. The simulation result demonstrates:compared with the DMC, the proposed two-layered control strategy can increase economic benefits while assuring the stability of the system compared with only DMC control.4. To deal with the model of the distillation process in refining industry, which has large delay and multi-disturbance, a two-layered DMC optimal control strategy which faces jet fuel productivity is proposed. The simulation result demonstrates:compared with DMC, the proposed two-layered control strategy can increase jet fuel productivity while assuring the stability of the system compared with only DMC control.Finally, it summarized the main research and work, and the future development of distillation control and optimization was prospected.