Particle Swarm Optimization And It's Application On The Control System Of Continuous Conveyer Of Disc-Tube Assemble

Particle Swarm Optimization is swarm intelligence based biological inspired optimization algorithm. It inspired from the social behavior of biology, and its fundamental ideal is that presents a new approach for optimization problem through emergence intelligence from the cooperation and competition of individuals.Due to its conceptual simplicity, easy implementation and having few parameters to tune, many researchers has been attracted in this area and it has been a hot research topic in the domain of artificial life computation.However, as a new approach of artificial life computation, its theory research and application area should be enlarged. In this thesis we give a deeply research on PSO concerned with its principles, improvements and applications. The detailed content is summarized as follows.a) The paper gives a detailed survey of PSO from its current research, development trend and application area. Three swarm intelligence based optimization algorithms are described and the characters of these algorithms are also analyzed, respectively.b) Give an empirical study of an important parameter of PSO—inertia weight. A new nonlinear dynamical & adaptive selection approach of inertia weigh is presented, saving the user from a tedious trial and error based approach to determine it for each specific problem.c) To solve the problem of trapped in the local minima of traditional PSO, we proposed a new hybrid global optimization algorithm based on PSO and DE. The simulation results illustrates that PSODE is global optimization algorithm with rapid convergence speed and high results precise & robust.d) Set up a nonlinear mathematical model of the control system of Continuous Conveyer of Disc-Tube Assemble. Based on PSODE we designed a new PID controller-PSODE-PID, and then applied it to control the proposed mathematical model. Through the experimental simulation and PSODE-PID has a better control robust and dynamical character when compared with PSO-PID and GA-PID approach.