二层规划求解的精确罚函数蚁群优化算法

Ant Colony Optimization Algorithm Based on Exact Penalty Function for Bi-level Programming Solver

二层规划问题带有约束条件,求解难度较大,针对此问题提出一种基于精确罚函数的改进蚁群优化算法。首先,设计一种新的精确罚函数处理约束条件,避免了经典罚函数不可微、不光滑的问题;其次,采用改进的蚁群优化算法进行上下层组合求解;再次,编程仿真验证算法的有效性;最后,将其应用于实际价格控制二层规划问题决策中。研究结果表明,基于新的精确罚函数的蚁群优化算法,计算结果更加科学合理,为二层规划、多层规划问题求解提供了一种新的研究思路。

The bi-level programming problem with constraint conditions is difficult to solve.An improved ant colony optimization algorithm based on exact penalty function was presented.Firstly,a new exact penalty function was designed to deal with the constraint conditions,which could avoid the problem that the classical penalty function was not differentiable and smooth,meanwhile.Secondly,the improved ant colony optimization algorithm was used to solve the upper and lower layers.Thirdly,the validity of the algorithm was verified by programming simulation.Finally,it was applied to the decision of bi-level programming problem of actual price control.The results show that the ant colony optimization algorithm based on the new exact penalty function is more scientific and reasonable,which provides a new research idea for the bi-level programming and multi-level programming.