摘要
本文研究了下层以最优解返回上层的二层线性规划问题的遗传算法。在提出可行度概念的基础上,构造了二层线性规划上层规划问题的适应度函数,由此设计了求解二层线性规划问题遗传算法。为了提高遗传算法处理约束的能力,在产生初始种群时将随机产生的初始种群变为满足约束的初始种群,从而避免了使用罚函数处理约束带来的困难,最后用实例验证了本文提出的二层线性规划的遗传算法的有效性。
For the linear bilevel programming problem, we propose the definition of the feasibility, and construct the fitness function to use the genetic algorithms to solve the linear bilevel programming problem. To improve the capacity of the proposed algorithm to deal with the constraints, a randomly generated initial population is changed into an initial population satisfying the constraints, avoiding the use of penalty function to deal with the constraints. Finally, the examples are adopted to verify the effectiveness of the proposed method.
出处
《运筹与管理》
CSCD
2005年第2期54-58,共5页
Operations Research and Management Science
基金
国家自然科学基金资助项目(70371032
60274048)
中国教育部博士点专项基金资助项目(20020486035)
关键词
运筹学
二层线性规划
遗传算法
适应度函数
operational research
linear bilevel programming
genetic algorithms
fitness function
