摘要
标准的群搜索优化(GSO)方法是一种适用于解决高维函数优化问题的群智能算法,且简单、高效,易于实现。为了进一步提高其收敛速度和精度,对该方法进行了改进。在保留其"发现者—追随者—游荡者"框架的同时,改进的GSO方法将最大下降方向策略引入发现者行为。在每轮迭代中,发现者不但按照自身方向进行搜索,同时也根据最大下降方向进行搜索。分别通过23个基准测试函数对2种优化方法进行测试,结果表明:改进的GSO方法优于标准群搜索方法。
Standard group searching optimization （GSO）is a swarm intelligence algorithm for high-dimensional function optimization. It is simple and efficient, and easy to implement. In order to enhance its convergence speed as well as precision, an improvement on this method is presented. Inheriting the framework of＂ producer-scroungerranger＂of GSO, the improved GSO （iGSO）introduces the strategy of maximum descendent direction to the behavior of the so-called producer. In each iteration, the producer searches not only according to the direction of itself, but also according to the maximum descendent direction. Tests are carried out through 23 standard test functions on GSO and iGSO independently, the results shows iGSO method is prior to GSO.
出处
《传感器与微系统》
CSCD
北大核心
2012年第9期28-31,共4页
Transducer and Microsystem Technologies
基金
清华大学汽车安全与节能国家重点实验室开放基金资助项目(KF11011)
关键词
群搜索优化方法
函数优化
群智能算法
group searching optimization （GSO）method
function optimization
swarm intelligence algorithm
