论文部分内容阅读
T-S fuzzy inverse model identification based on particle swarm optimization(PSO)algorithm is proposed for a class of nonlinear process,and it is applied to direct adaptive inverse control of nonlinear system.In the first,the fuzzy space is divided by fuzzy grid diagonal method and the Gaussian function with uncertain center and width is used as the membership function,the fuzzy inverse model is obtained by using PSO algorithm to optimize both of the antecedent and consequent parameters in the process of inverse modeling in off-line manner.Afterwards the initial inverse model is connected to the plant in series and the consequent parameters of the inverse model are tuned by the variable-step least mean square(VSLMS)algorithm in on-line manner while its copy is connected to the plant as system controller.This approach realizes the adaptive inverse control of nonlinear system based on PSO algorithm.Simulation results on a hybrid nonlinear system and level control of a spherical tank show the validity of this method.