中文摘要：

两层规划问题是一个NP-难问题，这意味着它很难被求解．基于粒子群算法提出了一种求解二层规划问题的方法，通过分离目标函数和约束函数，使每个粒子拥有双适应值，并通过双适应值来决定个体优劣．应用了一种自适应保留不可行个体的策略．数值结果表明了算法的有效性．

英文摘要：

A bilevel programming （BLP） problem is a NP hard problem, which means it is very hard to be solved. In this paper, we introduce a BLP algorithm which separate the objective functions from its constrained functions based on the PSO. Therefore, every particle has double fitness values by which the particle will be decided whether better or not. The strategy to keep a fixed proportion of infeasible individuals is used in this new method. Numerical results show the efficiency and effect of the algorithm.