摘要
提出一种求解一般双层规划问题的层次粒子群算法.和传统的针对特定类型的问题或者基于特定假定假设条件所设计的算法不同,所提出的算法是一个层次算法框架,它通过模拟双层规划的决策过程来直接求解一般双层规划问题.层次粒子群算法将求解一般双层规划问题转化为通过两个变形粒子群算法的交互迭代来求解上下两层规划问题.同其它算法的实验结果比较表明层次粒子群算法是一个有效的求解一般双层规划问题的方法.
In this paper we propose a hierarchical particle swarm optimization method to solve general bilevel programming problems. Unlike traditional algorithms designed for solving specific types of bilevel programs, the proposed approach provides a hierarchical algorithm framework, which solves the general bilevel programs by simulating its decision process. In the proposed algorithm, solving for general bilevel programs is transformed into iteratively solving for the upper-level and lower-level programming problems using two variants of particle swarm optimization. The experimental results of the proposed algorithm, as compared with those of other algorithms, show that the proposed hierarchical particle swarm optimization is another effective algorithm for solving general bilevel programming problems.
出处
《管理科学学报》
CSSCI
北大核心
2008年第5期41-52,109,共13页
Journal of Management Sciences in China
关键词
粒子群算法
现代启发式算法
双层规划问题
约束优化
particle swarm optimization
metaheuristic
bilevel programming problem
constrained optimization
