大时滞系统全参数自适应预测控制策略

摘要

针对大时滞系统纯滞后时间长、参数时变的特点,提出一种基于改进的粒子群优化的自适应预测控制算法.利用改进的粒子群优化算法对时变大时滞系统模型的全部参数进行辨识,从而克服预测模型失配对系统控制性能的影响,并且将粒子群优化算法用于预测控制滚动寻优,有效解决系统存在约束条件下的最优值求解问题.仿真结果验证所提方法的有效性和优越性.%A new all parameters adaptive predictive control (APAPC) method based on modified particle swarm optimization (PSO) was presented to solve the control problem in time-varing system with long time-delay. All parameters of system model were identified on-line using PSO to effectively overcome the predictive model mismatch. A PSO based predictive control was proposed in which the PSO was used for iterative optimization, which can solve the complicated optimization with various kinds of constraints. The simulation example proves the effectiveness.