摘要
群体智能优化算法利用群体的优势,在没有集中控制并且不提供全局模型的前提下,为寻找复杂的分布式问题的解决方案提供了基础。介绍了两种群体智能算法模型:蚁群算法模型和粒子群算法模型,研究了两种算法的原理机制、基本模型、流程实现、改进思想和方法;通过仿真把蚁群算法与其他启发式算法的计算结果作对比,验证了蚁群算法具有很强的发现较好解的能力,不容易陷入局部最优;微粒群算法保留了基于种群的、并行的全局搜索策略,采用简单的速度-位移模型操作,在实际应用中取得了较高的成功率。
By the use of groups＇advantages, in the absence of centralized control and without providing the overall model situation, swarm intelligence optimum algorithm provides the foundation on finding complex distributed solutions to the problem. Introduces the two swarm intelligence algorithm models：ant colony algorithm model and the particle swarm algorithm model, researches on principle mechanism, the basic model, process realization and improved ideas and methods;and by comparing the calculation results of the ant colony algorithm and other heuristic algorithm through the simulation, proved that the ant colony algorithm has a strong ability to find better solutions, and is not easy for a local optimum. The algorithm which based on the population of reservations, parallel global search strategy, using simplY speed-displacement model operation, has been made in a higher success rate in practical application.
出处
《计算机技术与发展》
2008年第8期114-117,共4页
Computer Technology and Development
基金
国家自然科学基金资助项目(60273043)
安徽省自然科学基金资助项目(050420204)
关键词
群体智能
蚁群算法
粒子群算法
启发式算法
swarm intelligence
ant colony optimization algorithm
particle swarm optimization
heuristic algorithm
