Abstract
In multimodal optimization, the aim is not only to find one
global optimal solution, but also to find various global optimal solutions. For this purpose, the Species-based Differential Evolution has been proposed previously.
However, this algorithm takes a long time and needs the set of experiential parameter setting to decide the radius of species territory for acquisition of global optima in complicated problems. In this paper, we propose a new method to reduce computational time and to change the radius of species territory adaptively using the shape of the peak of the optimization problems. We compare the proposed method in search performance with the conventional Species-based DE.
