基于粒子群算法混合优化的广义预测控制器研究  

Research of Hybrid Optimized Generalized Predictive Controller Based on Particle Swarm Optimization

作　　者：肖本贤[1] 朱志国[1] 刘一福[2] 

机构地区：[1]合肥工业大学自动化研究所,合肥230009 [2]安徽省电力科学研究院,合肥230022

出　　处：《系统仿真学报》2007年第4期820-824,共5页Journal of System Simulation

摘　　要：提出一种基于粒子群算法混合优化的广义预测控制器(generalized predictive control based on particleswarm optimization,简称PSOGPC),将粒子群优化算法(particle swarm optimization,简称PSO)引入到广义预测控制的滚动寻优过程中,有效解决了广义预测控制在被控对象存在约束时难以获得最优预测控制输入及求解复杂的问题。并对普通粒子群优化算法进行了改进,提高了优化过程的求解精度和收敛速度。多种约束情况和对电厂锅炉的主汽温控制系统的仿真结果表明了该方法的有效性和优良的控制性能。A new hybrid optimized generalized predictive control （GPC） based on the PSO technique （PSOGPC） was proposed in which the PSO is used for iterative optimization. The method can solve the complicated solving equation problem when GPC is difficult to obtain the optimum prediction control input because of the constraint of the control process Furthermore, PSO was modified here to improve the solving precision and convergent rates of optimization procedure. The multi example simulation results show the method＇s validity and superior control performance.

关 键 词：广义预测控制 粒子群优化算法 混合优化策略 约束 

分 类 号：TP301[自动化与计算机技术—计算机系统结构] TP391.9[自动化与计算机技术—计算机科学与技术]