Optimal Control Of Model Free Adaptive For Polymerization Process Of PVC

Polyvinyl chloride (PVC) industry occupies an important position in the national economy. The key link in producing PVC is polymerization process, while the polymerizing temperature is an important parameter to reflect the polymerization process. Whether the controlling effect of polymerizing temperature is good or not has a direct influence on the quality of the final products. Therefore, precise control of reacting temperature is particularly important in PVC producing process. Similar to the most industrial processes, polymerization process is featured with large inertia, big time delay and nonlinear. It is difficult to obtain the satisfactory control effect by traditional control methods. Model free adaptive control (MFAC) is a kind of control strategy which gets rid of dependence on the controlled system model. As a result of only using I/O data of the controlled system to design the controller, the "unmodeled dynamics and robustness issues" is essentially overcomed. Particle swarm optimization (PSO) algorithm is a swarm intelligence algorithm in the field of computational intelligence. It is widely used because of its simple algorithm, high computational efficiency and strong robustness.Under the background of PVC producing process of a certain chemical industry group company, some researches have been done based on reading many references about MFAC and PSO algorithm.Firstly, on the basis of a thorough understanding of polymerization production process, influence factors of polymerizing reaction are analized. Polymerizing temperature is determined as the key factor that affects the polymerization process and the final product quality, and cooling water flow can effectively control the polymerizing temperature.Then, with theoretical knowledge about MFAC as a guide, aiming at the general nonlinear and discrete time system, design scheme of MFA controller is given. Simulation experiments are carried out to analyze the influence of controller parameters on controller performance. On the basis of what is analysed above, PSO algorithm is applied to optimize and set the key parameters for controller tuning, then MFA contrlller based on PSO is designed.Finally, taking advantage of the existing actual operational data of polymerization process, according to chemical reaction principle of the thermal equilibrium, least square method is adopted to identify the mathematical model of polymerizing temperature. MFAC algorithm is applied to the PVC polymerization process. Simulation results manifest the controlling effectiveness of MFAC algorithm comparing to traditional PID, while MFAC stragedy based on PSO obtains better tracking results under the same conditions.