Font Size: a A A

Research On Fuzzy PID Control System Design Base On Particle Swarm Optimization

Posted on:2009-12-25Degree:MasterType:Thesis
Country:ChinaCandidate:M YangFull Text:PDF
GTID:2178360245488736Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
Fuzzy PID control system, as an embranchment of Intelligence control field, is widely applied in the industry control field, which imitates human's control experience instead of the model of the controlled object. In the design of fuzzy controller, the refinement of membership functions and the choice of control rules are key factors to obtainning higher control precision. But there are few ways to get information and no normative and rational methods to establish them. Those facters generally obtained by the experience of experts, can only be obtained by tentative methods.The advantages and disadvantages of Particle Swarm Optimization(PSO) and Fuzzy Logic are studied and rationally combined in this paper. The author uses PSO to find preferable control rules and parameters of membership functions and applies them in the PID controller, which obtains better results and gives a new way to solve many control problems. The main work of this paper include three parts:1)At first, the membership functions and fuzzy control rules are coded together by decimal code. And then, using PSO to search the best values in the wholly question space. The author offers the main frame and algorithm flow of optimizing the fuzzy controller based on PSO;2)Considering on the respective advantages of PSO and Genetic Algorithm(GA), the author offers a method, which is the same with optimization of the fuzzy controller parameters, named Fuzzy Parameter Particle Swarm-Genetic Combined Optimization Algorithm(FPSO-GA). And this method obtains better emulational result than each of them; 3)Because the optimized control rules still contain many redundant information, the author filters the optimized rules by binary PSO and obtains hypo-best result under filtered control rules. Emulational result indicates the validity and practicability of this method.Finally, the work of this paper is summarized, and the prospectiveness of it's juvenility and the future research field are discussed.
Keywords/Search Tags:Particle Swarm Optimization(PSO), Fuzzy PID Controller, Genetic Algorithm(GA), Evolution Algorithm(EA)
PDF Full Text Request
Related items