基于约束多目标免疫算法的H_∞混合灵敏度加权阵选择  

Weighting function matrix selection for mixed sensitivity in H_∞ robust control based on CMO immune algorithm

作　　者：王杰[1] 王栓[1] 朱晓东[1] 连进灿[1] 

机构地区：[1]郑州大学电气工程学院,河南郑州450001

出　　处：《电力自动化设备》2009年第1期37-40,共4页Electric Power Automation Equipment

基　　金：河南省杰出人才创新基金(074200510013);河南省教育厅自然科学基金(2007520048)~~

摘　　要：在H_∞鲁棒控制中,混合灵敏度加权函数直接反映了系统的稳定性和性能要求。若将加权阵当作设计参数,则控制系统的设计可以表示为一个满足一组不等式约束的多目标优化问题。采用人工免疫算法,利用免疫原理中抗体的多样性产生及保持机理来求解该优化问题,即利用免疫算法在预先设定的搜索域内,找到最满意的解来组成加权阵。在此加权阵基础上,利用Matlab鲁棒控制工具箱得到H_∞最优控制器。以调速系统为例进行仿真,结果表明,该算法可以搜索到满意的加权阵。In H∞ robust control, mixed sensitivity weighting function directly reflects the requirements for system stability and performance. With the weighting function matrix as design parameters, the design of a control system can be expressed as a multi-objective optimization problem with inequation constraints. With the mechanism of producing and preserving the diversity of antibodies, an artificial immune algorithm is applied to it, i. e. searching the most satisfying solutions within preset area to form the weighting function matrix. With Matlab robust control toolbox, an H∞ optimal controller is obtained based on it. With speed control system as an example, simulation shows the satisfying weighting function matrix is effectively obtained.

关 键 词：免疫算法 多目标优化 H∞混合灵敏度 加权阵 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]