Wolfe线搜索下具有充分下降性的混合共轭梯度法  

A Hybrid Conjugate Gradient Method with Sufficient Descent Property under Wolfe Line Search

作　　者：房明磊 丁德凤 王敏 FANG Minglei;DING Defeng;WANG Min(School of Mathematics and Big Data,Anhui University of Science and Technology,Huainan 232001,China)

机构地区：[1]安徽理工大学数学与大数据学院,安徽淮南232001

出　　处：《长春大学学报》2022年第8期16-22,共7页Journal of Changchun University

基　　金：安徽省高校自然科学基金项目(KJ2021A0451);安徽省自然科学基金项目(2008085MA01)。

摘　　要：共轭梯度法存储量低,运算简洁,对于求解大规模无约束优化问题非常有效。通过对PRP算法进行修正,提出一种新的混合共轭梯度法,在Wolfe线搜索下,每一步迭代都产生充分下降方向,在常规的假设条件下证明提出的算法具有全局收敛性。实验结果表明提出的算法对解决优化测试问题是有效的。Conjugate gradient method has low storage capacity and concise operation, which can solve large-scale unconstrained optimization problems effectively. Through revising PRP algorithm, a new hybrid conjugate gradient method is presented, which can generate sufficient descent directions at every iteration under Wolfe line search and it is proved to have global convergence under suitable conditions. Experimental results show that the proposed algorithm is efficient in solving optimization and test problems.

关 键 词：无约束优化 混合共轭梯度法 充分下降性 WOLFE线搜索 全局收敛性 

分 类 号：O221.2[理学—运筹学与控制论]