基于层次遗传算法的非线性双层规划问题求解策略  

A Solution Strategy for Nonlinear Bilevel Programming Problem Based on Hierarchical Genetic Algorithm

作　　者：李昌兵[1] 袁浩[1] 杜茂康[1] 

机构地区：[1]重庆邮电大学电子商务与现代物流重点实验室,重庆400065

出　　处：《系统工程》2011年第4期109-113,共5页Systems Engineering

基　　金：国家自然科学基金资助项目(60905066);重庆市教委科研项目(KJ070509)

摘　　要：双层规划是解决层次决策问题的运筹学工具。当前基于传统的优化思想已经提出了很多算法解决搜索空间已知的双层规划问题。但在双层规划领域仍然存在许多问题无法利用现有算法求解。本文基于进化博弈和多目标优化非支配排序的思想,设计了层次遗传算法并利用其求解非线性双层规划问题。最后通过测试函数验证算法的有效性。Bilevel program（BLP） is an operation researh technique for solving hierarchical decision-making problem.There are numbers of algorithms based on classical optimization methods to solve BLP problems where the search space is known.However,there are number of problems in the BLP which existing algorithms are not sufficiently to slove.This paper,based on the idea of evolution game and multi-objective optimization non-dominated sort,designs a hierarchical genetic algorithm to the model is designed to solve BLP. Finally,the application of the model and its algorithm are illustrated with test functions.

关 键 词：层次遗传算法 双层非线性规划 约束优化 

分 类 号：O221[理学—运筹学与控制论]