摘要
对闭凸集约束的非线性规划问题构造了一个修正共轭梯度投影下降算法,在去掉迭代点列有界的条件下,分析了算法的全局收敛性.新算法与共轭梯度参数结合,给出了三类结合共轭梯度参数的修正共轭梯度投影算法.数值例子表明算法是有效的.
In this paper,a modified conjugate gradient projection descent method for nonlinear constrained optimization with closed convex set is presented.The global convergence properties of the new method are discussed without assuming that the sequence of iteration is bounded.Combining conjugate gradient scalar with our new method,three new classes of modified gradient projection methods with conjugate gradient scalar are presented.The numerical results illustrate that the new methods are effective.
出处
《应用数学学报》
CSCD
北大核心
2010年第4期640-651,共12页
Acta Mathematicae Applicatae Sinica
基金
国家自然科学基金(10971118)
中央高校基本科研业务费专项资金(09CX04020A)
中国石油大学(华东)研究生创新基金(S10-28)资助项目
关键词
非线性规划
凸约束的非线性规划问题
GLP投影算子
共轭梯度
收敛性
nonlinear programming
nonlinear convex constrained optimization
GLP projection operator
conjugate gradient
convergence
