摘要
研究了连续多面体不确定系统的多目标静态输出反馈控制问题,此目标静态输出反馈控制问题由双线性矩阵不等式(BMI)给出.为了计算静态输出反馈增益,提出了一种基于差分进化算法(DEA)和线性矩阵不等式(LMI)方法的新型混合算法.新算法的基本思想是通过DEA得到控制器的样本,并将BMI问题转化为凸约束优化问题,进而采用LMI方法来最小化系统的性能指标.以多目标H2/H∞控制问题为例,给出了基于静态输出反馈的多目标控制求解的具体方法,并用数值仿真例子说明了该算法的有效性.
The static multi-object output feedback control of continuous polytopic uncertain system was discussed, which was given in terms of bilinear matrix inequalities （BMI）. A new hybrid control approach based on differential evolution algorithm （DEA） and linear matrix inequality （LMI） was proposed to evaluate the static output feedback gain, whose essential ideal is to transform BMI into convex optimization problems employing the controller samples obtained by DEA. LMI optimization method was then adopted to minimize the system output performance. Taking the H2/H∞ control as example, the detailed algorithm of solving multi-object control based on static output feedback was presented. The numerical simulation examples were employed to illustrate the effectiveness of the new hybrid control approach.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2009年第7期49-52,共4页
Journal of Xi'an Jiaotong University
基金
国家自然科学基金资助项目(59705005)
吉林大学"985工程"汽车工程科技创新平台资助项目(450022102002)
关键词
多目标控制
差分进化算法
多面体不确定系统
multi-object control
differential evolution algorithm
polytopic uncertainty system
