遗传算法用于控制器参数的满意优化  

Satisfactory Optimization of Controller Parameters Using Genetic Algorithm

作　　者：金炜东[1] 陈立[1] 李岗[1] 

机构地区：[1]西南交通大学电气工程学院

出　　处：《西南交通大学学报》1998年第6期699-704,共6页Journal of Southwest Jiaotong University

基　　金：国家自然科学基金

摘　　要：对于控制器参数的多指标优化问题，提出了一种新型优化设计方法。通过分别设计的各个性能指标的满意度函数，反映出各性能指标对控制系统的具体要求；并用综合满意度函数的设计，反映出对控制目标的综合要求。由此给出了一种优化模型。应用遗传算法自动搜索，实现控制器参数的优化。仿真实验效果良好，表明了本文方法的灵活实用性。A new design method is proposed for multi index optimization of controller parameters. Specific requirement of each performance index of a control system is reflected by a respectively designed 'satisfactory extent function'. The overall requirement of the control system is presented through designing an integrative satisfactory extent function. In this way a model for optimization is obtained. The optimization of controller parameters is realized through automatic search via genetic algorithm. Simulations demonstrate good results, and show its flexibility and practicability.

关 键 词：控制系统 满意优化 遗传算法 优化设计 控制器 

分 类 号：TM571[电气工程—电器] TP273[自动化与计算机技术—检测技术与自动化装置]