中文摘要：

提出一类求解无约束优化问题的修正LS共轭梯度法，算法采用一个新的参数公式．在适当条件下，证明算法满足充分下降条件，进而证明在采用广义Wolfe-Powell线搜索和强Wolfe-Powell线搜索时，算法全局收敛．初步的数值试验结果显示方法是有效的．

英文摘要：

This paper propose a modified LS conjugate gradient method in which the new parameter formula is adopted for unconstrained optimization. Under mile conditions, They prove that the method pos- sesses descent property. And the global convergence of the proposed formula with the generalized Wolfe- Powell line search and the strong Wolfe-Powell line search are proved. Preliminary numerical experiments results show that the proposed methods are efficient.