一种求解不等式约束凸优化问题的内点方法  

An Inerior Point Method for Solving Convex Optimization Problem with Inequality Constraints

作　　者：吴庆丰[1] 

机构地区：[1]淮北师范大学数学科学学院,安徽淮北235000

出　　处：《淮北师范大学学报（自然科学版）》2014年第1期8-11,共4页Journal of Huaibei Normal University：Natural Sciences

基　　金：淮北师范大学青年科研项目(2013xqz11);淮北师范大学教学研究项目(jy110221)

摘　　要：文章给出一种求解不等式约束的凸优化问题的内点方法,此法能保证迭代过程中迭代点仍为内点,解决了IPA每步迭代需要假定迭代点在可行域内部,并不需要罚因子趋于零,从而避免传统内点障碍函数法由于罚因子趋于零导致的病态问题.最后给出了数值实验,实验表明,算法是有效的.In this paper,an inerior point algorithm for solving convex optimization problem with inequality constraints is given.This method can ensure the new iteration point is still an interior point,so as to solve the problem which IPA（interior point algorithm）requires assuming each iterative point in the feasible region.Al-so,the method does not require the penalty factor to be zero,thus avoiding the ill-posed problem of the tradi-tional barrier function method caused by the penalty factor becoming zero.Finally,numerical experiments show that this method is effective.

关 键 词：凸优化 障碍函数法 内点法 

分 类 号：O221.2[理学—运筹学与控制论]