摘要
针对传统蚁群算法在多目标优化问题中容易陷入局部最优的缺点,提出一种采用直接学习机制的改进蚁群算法。该算法通过采用模拟蚂蚁用触角交流信息过程的直接通信学习机制,用以改进信息素的更新规则,从而维持群体的多样性。通过两组多目标基准函数验证算法性能,仿真结果表明该算法所获得的Pareto解具有多样性以及均匀分布性,有效地提高了蚁群算法全局寻优的能力。
In order to solve the local optimum problem of the traditional ant colony algorithm for multi-objective optimizations an improved ant colony algorithm with a direct learning mechanism is proposed in this paper. Aiming to maintain the diversity of the groups, this algorithm improves the pheromone updating rules by learning from the true ants exchanging information through their antennae. Finally, two benchmark cases are used to test the proposed algorithm, the simulation results show that this algorithm possesses high a searching efficiency, and can efficiently find multiple Pareto optimal solutions.
出处
《江南大学学报（自然科学版）》
CAS
2013年第4期394-398,共5页
Joural of Jiangnan University (Natural Science Edition)
基金
国家自然科学基金项目(61273131)
江苏高校优势学科建设工程项目
关键词
多目标优化
蚁群算法
信息素
multi-objective optimization, ant colony algorithm, pheromone
