摘要
目前的智能优化算法易陷入本地最优平衡态,并且进化后期的效率低下。为了克服这些缺陷,提出了一种基于正交优化的群智能优化算法。该算法突破了以往正交设计方法仅能用在粒子群初始化和进化前优化搜索过程的局限,基于方差分析和方差比例分析,证实了正交设计方法进一步的搜索方向和范围。使用正交设计的特征在一次阵列计算中寻找包含最优值的间隔,算法可以在优化搜索过程中循环进行方差比例分析。对六峰值驼背函数的仿真分析结果说明,正交智能优化算法相比目前的智能优化算法,计算量更低,搜索时间更短,运行速度更快,且优化搜索过程的精度更高。
Existing intelligent optimization algorithms are easy to fall into local optimal equilibrium states and have low effi- ciency at evolutionary late stage. In order to overcome these shortcomings, this paper proposed a swarm intelligence optimiza- tioin algorithm based on orthogonal optimization. Based on the variance analysis and variance ratio analysis, the algorithm con- firmed the further searching direction and searching range of the orthogonal design method which was previously limited to be used in the swarm initialization and optimization searching before evolution. Making use of the characteristics of orthogonal de- sign to find the interval containing the best solution in one arrayed calculation, the algorithm could circularly make variance ra- tio analysis in the optimization searching process. The simulation analysis of six-hump camel back function shows that com- pared to existing intelligent optimization, the proposed algorithm has lower calculation amount, shorter searching time, more rapid running speed, and higher accuracy of optimization searching process.
出处
《计算机应用研究》
CSCD
北大核心
2015年第1期71-74,共4页
Application Research of Computers
基金
河南省科技攻关项目(112102210500)
关键词
群智能
基于种群的智能优化
蚁群算法
正交设计
方差比例
swarm intelligent
population-based intelligent optimization
ant colony optimization
orthogonal design
vafiance ratio
