.png){kind=icon}
.png){kind=icon}
Return
Improved Differential Evolution With a Modified Orthogonal Learning Strategy
.png){kind=icon}
OA
AI
DOI：10.1109/ACCESS.2017.2705019.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Orthogonal experimental design (OED) is a powerful method for identifying the best combination of factors, and considerably reduces the required number of experimental samples. Researchers have combined OED with evolutionary techniques, such as the genetic algorithm, particle swarm optimization, and artificial bee colony algorithm, resulting in significantly better performance. In this paper, we study the combination of OED and differential evolution (DE). We present a modification to the orthogonal design strategy, and propose a modified orthogonal differential evolution (MODE) technique. Two variants of MODE are developed, one which acts on the crossover operation and a second that operates on the selection stage. These enhance the DE aspect in different ways to improve the discovery of dimensional information during the evolution process. We first construct the basic MODE, which combines the orthogonal design strategy with the basic DE algorithm, and then employ a variant with a self-adaptive parameter strategy. The results of comparative experiments demonstrate the effectiveness of the proposed algorithm.
Keywords:
Orthogonal experiment design
orthogonal learning
differential evolution
dimensional information
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.