The Research Of Application Of Intelligent PID Based On Fuzzy Logic And PSO

With the development of science and technology and modern industrial manufacturing, the control objects in the practical manufacturing process become more and more complex and the main problems are MIMO, delay, variable, non-linear and couple. Besides that, There are many uncertain and unknown factors of the system and the environment which will have unpredictable influence on the control of the system. While facing the complex system mentioned above, it is very hard for us to acquire the accurate mathematical model, which makes conventional control method unsatisfying. Fuzzy control is an intelligent control method that is not rely on mathematical model, and it will get satisfying control result when meets the above problems. But its application has been narrowed because of bad steady-state performance and lack of systematic parameter design methods. The article combines PID controller with Fuzzy control, and adaptive fuzzy-PID controller is adopted which shows good control effect. Regarding the bad steady-state performance of fuzzy controller, several methods are proposed and the article introduces intelligent integrator to conventional fuzzy controller. The simulation experiment shows that the improved method overcomes weakness of conventional fuzzy controller.PSO(Particle Swarm Optimization) is a new kind of swarm intelligent random optimization algorithm. Compared with conventional algorithm, PSO has faster convergent and computational speed, and it can search the global minimum solution in the given space. It is fit for solving complex optimization problems such as non-linear, non-differential and multi-peaks. Considering that conventional PID tuning methods are not efficient and can not get good control parameters, the article proposes tuning the PID parameters by PSO and the result is satisfying.The research topic of the article comes from the improvement and expand of the ethylene control system of an certain big company and the polypropylene device is selected as the control object. Since the manufacturing process of polypropylene is very complex, large delay and non-linear, and there are many control variables. Conventional control methods such as single loop control and cascade control can not work effectively and guarantee the quality of products.After analyzing the above problems, the article combined theoretical research with engineering application and discussed how to apply fuzzy control and PSO to the practical manufacturing process. The main research work is described as follows:A survey of origin, development, current status and the latest research of fuzzy control is summarized. Several problems that fuzzy control meets are discussed and the development of PSO is also summarized.Deep researches of the manufacturing technics of polypropylene are conducted. The control methods and control difficulties are analyzed and then control project is confirmed.Overall analyse of control system architecture, configuration of workstation and controllers, network and database is conducted.The general principle of fuzzy control is introduced. Considering control difficulties, fuzzy control and PID controller is combined and one kind of fuzzy adaptive PID controller is proposed. The Matlab simulation shows it has good steady-state performance and robust adaptive ability.Three methods of introducing integrator into fuzzy controller is analyzed and inferred to solve the problem that fuzzy control has steady-state error. The simulation experiment of paralleling intelligent integrator with fuzzy controller is conducted and the result shows it has better performance that conventional fuzzy controller.Considering that conventional PID tuning methods can not be auto tuned and are not efficient, a method of tuning the PID parameters by PSO is proposed and the simulation shows it has faster rising speed and good steady-state performance. After the punishing factor is introduced, there is no overshoot.Detailed introduction of COM technology is conducted and an example is made to show the combination programming of Matlab and VC++.