基于切面上升的狭义遗传算法  

Restricted Genetic Algorithm Based on Ascending of Tangent Planes

作　　者：郑金华[1] 蒙祖强[2] 蔡自兴[2] 

机构地区：[1]湘潭大学信息工程学院,湘潭411105 [2]中南工业大学智能控制研究所,长沙410083

出　　处：《计算机工程与应用》2004年第3期26-28,49,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(编号:69974043);湖南省自然科学基金资助

摘　　要：该文提出了基于切平面上升的狭义算法,并对其全局收敛性进行了证明,其终止条件肯定就是收敛点,很好地解决了文献犤6,8犦中所遇到的问题,即能够方便地求解多元函数的优化问题。它具有收敛速度快,搜索过程稳定性好、可控制性强,控制参数易于选定等特点。The restricted genetic algorithm based on ascending of tangent plane is put forward in this paper.It is proved that the algorithm is global convergent ,and the convergent condition is certainly the convergent place.The problem in document,which is the problem about the optimization of multivariable function,has been solved easily with the algorithm discussed in this paper.The algorithm possesses the following characteristics:higher convergent speed,higher stability,better controllability and easy to choose the control parameters.

关 键 词：遗传算法 狭义遗传算法 多元函数优化 智能计算 切平面上升 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]