摘要
蚁群法是新型的群智能优化法,具有鲁棒性、分布并行机制和易融入启发式信息等特点。尤其通过释放信息素,蚂蚁间相互交流协作,实现正反馈机制,加速全局搜优,提高效率。蚁群法最初只用于离散问题。本文评述蚁群法的起源、改进和发展;重点介绍并分析了连续型蚁群法的几种处理方案和各自特点;全面总结蚁群法已应用在化学化工领域,以及对动态、带约束和多目标问题的处理方式和机制,效果良好,优于其他算法的情况。最后展望蚁群法今后的发展前景和研究方向。
Ant colony algorithm is a novel category of swarm intelligence optimization methods. The positive feedback mechanism is achieved via pheromone, and the global search ability is benefited from mutual communication and cooperation among individuals. Ant colony algorithm has such merits as robustness, parallel distribution mechanism, integration of heuristic information, etc. It has powerful global optimization ability, good efficiency, and its traditional version is just suitable to discrete optimization problems. This paper reviews the origin of ant colony algorithm, and its improvement and development. Several schemes of continuous ant colony algorithm and their respective characteristics are introduced and analyzed. The applications of ant colony algorithm in chemistry and chemical engineering fields are roundly summarized, and the approaches to deal with dynamic, constrained and multi-objective problems are introduced. Finally, the development prospects and future research directions of ant colony algorithm are provided.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2008年第1期1-6,共6页
Computers and Applied Chemistry
基金
国家自然科学基金(20276063)
关键词
蚁群算法
正反馈
离散优化
连续优化
化学化工应用
ant colony algorithm, positive feedback, discrete optimization, continuous optimization, chemistry and chemical engineering applications
