当前位置: X-MOL 学术Calcolo › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A scaled three-term conjugate gradient method for large-scale unconstrained optimization problem
Calcolo ( IF 1.7 ) Pub Date : 2019-09-17 , DOI: 10.1007/s10092-019-0333-4
Parvaneh Faramarzi , Keyvan Amini

The moving asymptote method is an efficient tool to solve structural optimization. In this paper, a new scaled three-term conjugate gradient method is proposed by combining the moving asymptote technique with the conjugate gradient method. In this method, the scaling parameters are calculated by the idea of moving asymptotes. It is proved that the search directions generated always satisfy the sufficient descent condition independent of the line search. We establish the global convergence of the proposed method with Armijo-type line search. The numerical results show the efficiency of the new algorithm for solving large-scale unconstrained optimization problems.

中文翻译:

大规模无约束优化问题的按比例缩放三项共轭梯度法

移动渐近线法是解决结构优化问题的有效工具。结合移动渐近线技术和共轭梯度法,提出了一种新的标度三项共轭梯度法。在这种方法中,缩放参数是通过移动渐近线的想法来计算的。证明了生成的搜索方向总是满足足够的下降条件,而与线搜索无关。我们用Armijo型线搜索建立了所提出方法的全局收敛性。数值结果表明,该新算法有效地解决了大规模无约束优化问题。
更新日期:2019-09-17
down
wechat
bug