摘要
文章主要研究多目标优化问题,为了提高细菌趋药多目标优化算法的收敛速度和解的多样性及弥补逃离局部最小值的不足,提出了一种基于非支配排序的细菌群体趋药多目标优化算法。首先,采用快速非支配排序方法初始化所有细菌的位置;其次,利用细菌群体趋药算法对多目标函数进行优化;最后,采用精英保留策略,避免因算法的随机性而将原本位置较好的点抛弃的情况。实验结果表明,该算法不仅比BCMOA的收敛速度快而且保留了解的多样性。
This paper focuses on the MOOP (Multi -Objective Optimization Problem). To improve the convergence speed and the dis- versity of BCMOA ( Bacterial Chemotaxis Multi - objective Optimization Algorithm) and make up for the shortage of escape from local minimum, this paper proposes an IBCCMOA (Improved Bacterial Colony Chemotaxis Multi -objective Optimization Algorithm). First- ly, Fast Non - dominated Sorting Approach is used to initialize the position of all the bacterias. Secondly, Bacterial Colony Chemotaxis Algorithm is adopted. Thirdly, apply a strategy of Elite Reserve to avoid abandoning the points that the original position is good. Exper- imental results show that the convergence and the diversity solutions of the proposed algorithm are better than that of the existing BCMOA.
出处
《忻州师范学院学报》
2015年第2期10-16,共7页
Journal of Xinzhou Teachers University
基金
山西省自然科学基金项目(2013011017-2)
山西省高校科技创新项目(2013150)
忻州师范学院青年基金项目(QN201408)
忻州师范学院重点学科专项课题(ZDXK201203和XK201308)
关键词
多目标优化
细菌群体趋药
快速非支配排序方法
精英保留策略
Multi - Objective Optimization
Bacterial Chemotaxis
Fast Non - dominated Sorting Approach
Elite Reserve Strategy