一类绝对值规划的算法  

Algorithm for a class of absolute value programming

作　　者：王爱祥[1] 崔恩华[1] 

机构地区：[1]中国矿业大学理学院,江苏徐州221116

出　　处：《黑龙江科技学院学报》2010年第5期387-390,共4页Journal of Heilongjiang Institute of Science and Technology

摘　　要：针对一类绝对值规划问题,提出对偶规划,给出其弱对偶性及对偶问题的最优性充分条件,并证明对偶间隙也是该类绝对值规划问题的解。同时,引入变量代换,基于线性规划的单纯形法,提出该类绝对值规划问题的全局优化求解算法。算例表明该算法是有效的。Aimed at class of absolute value programming, this paper proposes dual absolute value programming, gives the weak duality theorem and sufficient optimality conditions, and proves that duality gap is the solution of a class of absolute value programming problem. Based on variable substitution and the simplex method for linear programming, the paper also discusses the global optimization algorithm for the absolute value programming. Theoretic analysis and numerical results show the algorithm is efficient.

关 键 词：绝对值规划 对偶问题 单纯形法 

分 类 号：O221[理学—运筹学与控制论]