具有等式与不等式约束条件拟可微优化的Lagrange乘子型最优性条件(英文)  

Optimality Conditions in Lagrange Multiplier Type for Quasidifferentiable Optimization with Equality and Inequality Constraints

作　　者：高岩[1] 

机构地区：[1]大连理工大学应用数学系,大连116024

出　　处：《运筹学学报》1999年第4期47-54,共8页Operations Research Transactions

基　　金：The project supported by the Natural Science Foundation of Liaoning Province of China.

摘　　要：讨论了不等式约束优化问题中拟微分形式下Ｆｒｉｔｚ Ｊｏｈｎ必要条件与 Ｃｌａｒｋｅ广义梯度形式下Ｆｒｉｔｚ Ｊｏｈｎ必要条件的关系．在较弱条件下给出了具有等式与不等式约束条件的两个Ｌａｇｒａｎｇｅ乘子形式的最优性必要条件，在这两个条件中等式约束函数的拟微分和Ｃｌａｒｋｅ广义梯度分别被使用。The relation between Fritz John necessary optimality condition in terms of quasidifferential and that in terms of Clarke generalized gradient for the inequality constrained nonsmooth optimization, is discussed. Two necessary optimality conditions in Lagrange multiplier type for a quasidifferentiable optimization with equality and inequality constraints, is proposed under mild assumptions. In these two optimality conditions, quasidifferentials and Clarke generalized gradients of equality constraint functions are employed, respectively.

关 键 词：最优性条件 不等式约束 等式约束 拟可微优化 

分 类 号：O224[理学—运筹学与控制论]