基于MMAS的多目标优化算法研究

摘要

针对多目标优化问题求解过程中多个目标相互制约难以求解的特点,为了多目标的协调优化,提出了一种基于最大最小蚁群算法(MMAS)的多目标优化蚁群算法.将蚁群算法的离散搜索机制映射到连续空间,修改了离散蚁群算法的行进规则和信息素的存留策略,使蚁群算法能够应用于解决解空间连续的问题.最大最小蚂蚁系统信息素取值方式的引入,极大地改善了蚁群算法搜索过程中容易陷入停滞的问题,尤其改善了蚁群算法在解空间的全局搜索能力.通过对两组测试函数求解的结果与其它方法比较,仿真结果表明所获得的最优解更多,分布范围更广,所求得的最优解集更加逼近真实的最优前沿.%To meet the special requirements of the multi-objective optimization problem, we propose an improved max-min ant system. The solution space is divided into some subspaces, both the strategy of pheromone updating mechanism and the local search mechanism are modified. Then the discrete search mechanism of the ant colony algorithm is used to the Continuous Space. The possible range of pheromone trail values are limited to improve the global search ability and avoid trapping in stagnation. Simulation results show that the proposed method can efficiently approximate the true Pareto optimal front, and the high requirement of the multi-objective optimization problem is well fulfilled with high accuracy and global search performance.