摘要
针对多变量时变系统提出了一种基于连续Hop fie ld网络的最优控制系统设计方法.该方法不仅从理论上建立了移动时域上的LQ性能指标与连续Hop fie ld网络能量函数间的等价关系,并在此基础上设计出可求解LQ最优控制问题的连续Hop fie ld网络,而且将滚动优化控制策略引入控制系统,形成了包括连续Hop fie ld网络在内的闭环控制结构,实现了多变量时变系统无限域上的动态最优控制.仿真结果验证了该设计方法的有效性.
An optimal control method based on continuous Hopfield neural network （CHNN） is proposed for multivariable time-varying systems. The equivalence between linear quadratic （LQ） performance index in movingtime horizon and energy function of CHNN is built theoretically, and the CHNN is constructed to solve LQ optimization control problems. Moreover, the receding optimization strategy is adopted to form closed-loop control structure that includes CHNN. Therefore, the dynamic optimization control for multivariable time-varying systems is realized in infinite-time horizon. Simulation results show the effectiveness of proposed method.
出处
《控制与决策》
EI
CSCD
北大核心
2005年第9期1038-1042,1046,共6页
Control and Decision
基金
国家自然科学基金项目(60375017)
国家教育部科学技术研究重点项目(203002)
关键词
连续HOPFIELD网络
LQ控制
滚动优化
多变量时变系统
动态控制
Continuous Hopfield neural network
Linear quadratic control~ Receding optimization
Multivariable time-varying system
Dynamic control
