等式约束优化问题SQP算法的超线性收敛充要条件  

Sufficient and Essential Conditions of Super-linear Convergence for SQP Algorithm in Equality Constrained Optimization Proble

作　　者：胡运红[1] 潘美芹[2] 

机构地区：[1]运城学院数学系,山西运城044000 [2]山东科技大学信息科学与工程学院,山东青岛266510

出　　处：《山东科技大学学报（自然科学版）》2006年第3期96-99,共4页Journal of Shandong University of Science and Technology(Natural Science)

摘　　要：对于等式约束问题,Boggs,Tolle和Wang三人将Dennis,Moré的求解无约束优化问题的类似结果加以推广,得到了SQP算法超线性收敛的一个极为重要的充要条件。许多研究学者又作了的改进,进一步减弱假设条件,得到了同样的等式约束问题的SQP算法超线性收敛的充要条件。Boggs, Tolle and Wang extended the analogous result from solving unconstrained optimization problems to the equality constrained problems and obtained a characterized result, which is an important equivalent condition to SQP algorithm＇s super-linear convergence, Subsequently, many reearchers did much work to improve the result and they would try to weaken those assumptions to obtain the same consequence. In this paper, on the basis of previous work, we further weaken the assumptions and obtain the same equivalent conditon to SQP algorithm＇s super-linear convergence.

关 键 词：SQP算法 等式约束问题 超线性收敛 充要条件 

分 类 号：O158[理学—数学]