A Type Of Nonlinear Model Predictive Control Based On Improved Particle Swarm Optimization Algorithm

As the nonlinear model predictive control (NMPC) developing up to now, most researches are restricted to the special nonlinear system, the study difficulty for the general nonlinear systems are mainly derived from the research of optimization algorithm, In fact, most reality control systems are nonlinear and likely to contain uncertainty. Therefore, an improved particle swarm algorithm (PSO) to optimize the nonlinear model predictive control is proposed in this paper. The following are the concrete contents:(1) An improved PSO algorithm is proposed. Since the typical PSO algorithm is easy to get trapped in the local optimum and its convergence rate gets slow in the later searching stage, the resampling step of particle filter is introduced into the PSO algorithm, combining the existing method of particle variation. Using four test functions to compare the improved PSO algorithm with the standard PSO algorithm, the results show that the convergence rate and search accuracy of the improved PSO algorithm is better.(2) An improved particle swarm algorithm is studied to apply to a type of nonlinear model predictive control system. A particle swarm optimization algorithm is investigated to optimize nonlinear model predictive control system with bounded random disturbances and probabilistic constraint. In handling probabilistic constraint, the particles that do not meet the constraint are replaced such that the feasible optimal control law can be obtained. The improved algorithm is defined effectiveness by the simulation results.