.png){kind=icon}
返回
Enhancing differential evolution with a target vector replacement strategy
.png){kind=icon}
PRE
AI
DOI：10.1016/j.csi.2022.103631.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Individuals find it difficult to avoid stagnation in the iterative process of the differential evolution (DE) algorithm, and the stagnant individuals have restricted improvement in the population, which will have a negative effect on its performance. This research put forward a target vector replacement strategy (TVRS) for reducing the impact of stagnant individuals on the DE algorithm's performance. For stagnant target vectors, while executing mutation operation and crossover operation, TVRS selects non-stagnant individuals in the population to replace the stagnant target vectors with a specific probability. Because improving stagnant individuals are difficult, TVRS provides more opportunities for improvement to those who are not stagnant, while the opportunities for stagnant individuals to be improved are decreased. For assessing the efficiency of TVRS, TVRS was implemented to six DE algorithms and compared to their original algorithms. Judging from the performance outcomes of TVRS in the CEC 2014 benchmark test set, TVRS can greatly increase the DE algorithm's performance. This study will show the evolutionary opportunity that is obtained by an individual should be related to the individual state. Adjusting the individual evolutionary opportunity based on the individual state is helpful for improving the stagnation problem of the differential evolution algorithm. This study will provide a new way for improving the stagnation problem of the differential evolution algorithm and other evolutionary algorithms.
Keyword:
Stagnation
Differential evolution (DE)
Target vector replacement
期刊
C
IF:
3.1
论文数:
2.3K
被引数:
2.0K
机构
引用论文
Accelerating differential evolution based on a subset-to-subset survivor selection operator.png)
SOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
分享收藏
分享收藏Elite Representative Based Individual Adaptive Regeneration Framework for Differential EvolutionIEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏Modified clustering-based differential evolution with a flexible combination of exploration and exploitationSOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
分享收藏Carbohydrates of pigeon milk and their changes in the first week of secretion鸽乳中的碳水化合物及其在分泌第一周内的变化
PREAI
分享收藏
查看更多内容