一类二次罚函数矫正算法  

A Kind of Revised Penalty Function Self-correcting Algorithm

作　　者：刘芳[1] 单锐[2] 

机构地区：[1]忻州师范学院数学系,山西忻州034000 [2]燕山大学西校区理学院,河北秦皇岛066004

出　　处：《重庆工商大学学报（自然科学版）》2011年第1期8-10,13,共4页Journal of Chongqing Technology and Business University:Natural Science Edition

摘　　要：以优化理论为基础,对求解一般约束优化问题提出一种算法.它对于惩罚因子可以通过算法自我矫正逼近最优项,在一定条件下证明了算法的收敛性.最后给出算例,结合MATLAB数值试验结果验证了这一算法的有效性.This paper gives a new algorithm for both equality and inequality constrained optimization problems based on optimization theory. Being self-correcting of the penalty factor, the algorithm is close to the optimal entry. Then it proves the astringency of the algorithm under certain condition. At last, an example is given, based on numerical test result of MATLAB, the feasibility of this algorithm is proven.

关 键 词：约束最优化 罚函数法 收敛性 

分 类 号：O174[理学—数学]