Design And Simulation Of Nonlinear Controller For Gas Phase Polymerization Reactor

As the industrial plant grow vaster, more integrated and more complicated, the need for controlling devices, management and optimization become harder and harder to meet. Conventional PID controllers are no longer capable of reaching a satisfying controlling performance in modern industry. Gas phase polymerization holds a very important position in chemical industry. It’s also a rather complex process. Since the temperature of the process is directly related with the chain length of the polymer, which affects the quality of the product, it is of great significance. However, as it is of high non-linearity and tends to deviate from the set point over time, it’s rather challenging to be controlled.In this work, a steady state model and a dynamic model of a gas phase polymerization process were built in Aspen Plus in order to acquire an object of the process. Compared with the actual industrial device, the model is of great accuracies in aspects of temperature, pressure, as well as the cumulative chain length distribution.Then, by modeling a CSTR (continuous stirred tank reactor), a polymerization process is simulated in Matlab. A nonlinear controller is then designed based on the model of the CSTR. The controller was then applied to the model and compared to the conventional PID controller to evaluate its performance. Afterward, by designing an interface software for Matlab to communicate with Aspen Plus, the designed nonlinear controller was applied to the polymerization process modeled in Aspen Plus. Compared with a fine-tuned traditional PID controller, the nonlinear controller shows a quicker response to set point change with less overshooting, which is the biggest difficulty in temperature controlling of polymerization.