Rotation Step Of PSO Algorithm And SMVSC Parameter Optimization

Missile control system is a very complex time-varying nonlinear system, in this paper, the simplified local linear time-invariant model was given after the equivalent of dynamic slight changement by using VSC algorithm to control the missile movement. The differences among the local linear model and the real system constitutes a complex nonlinear relationship, which using PSO to identify its parameters.The main problem of PSO is easy to fall in the premature convergence. This paper was studying a variety of improved PSO algorithm including repulsion algorithm from Fan Chaozan, giving the characteristics neighborhood of the funnction, ergodicity property, adhesion property and the detachment property of optimization algorithm, the precocious wells to detect anti-precocious ability, thus the rotation step of PSO was proposed. This algorithm adds mutually exclusive forces between particles, the individual particles can not be gathered in the same spot, so that the entire PSO can not convergence to local optima impossible.Rotation step PSO algorithm can get through the initial precocious trap test,and does not need to assume that the optimal solution in the PSO’s trajectory envelope, it is also better for some function with precocious well character to get the optimal results. Instructions rotation step PSO algorithm can effectively overcome premature problem in the process of optimization.