| 摘要: |
| 为了解决有约束的基于共轭梯度二次规划算法的多次迭代问题，结合共轭梯度算法和有效集策略，提出了一个新的算法模型，通过对变量的截取（使用Polak Bibiere 公式）来避免重新开始共轭梯度算法，在大规模的弹性接触问题中，大量的结果表明了这个算法的有效性。 |
| 关键词: 凸规划 条件约束 共轭梯度算法 有效集策略 弹性接触问题 |
| DOI： |
| 分类号: |
| 基金项目: |
|
| Conjugate Gradient Algorithm Based on Elastic Contact |
|
YU Xiao gang, DU Jun huai
|
| Abstract: |
| In order to solve constrained multiple iteration problem based on conjugate gradient for quadratic programming algorithm, by combining conjugate gradient algorithm and effective set strategy, this paper proposes a new algorithm model by truncating variables to avoid restarting conjugate gradient algorithm (by using Polak Bibiere formula). In large scale elastic contact problems, a lot of results show that this algorithm is effective. |
| Key words: convex programming condition constraint conjugate gradient algorithm effective set strategy elastic contact problem |
