Flower Pollination Algorithm Combined with Multiple Strategies of Opposition–Based Learning

Abstract

Flower Pollination Algorithm (FPA) has been widely used to solve optimization problems. However, it faces the problem of stagnation in local optimum. Several approaches have been proposed to deal with this problem. To improve the performance of the FPA, this paper presents a new variant that combines FPA and two variants of the Opposition Based Learning (OBL), such as Quasi OBL (QOBL) and Elite OBL (EOBL). To evaluate this proposal, 10 benchmark functions were used. In addition, the proposed algorithm was compared with original FPA and three variants such as FA–EOBL, SBFPA and DE–FPA. The proposal presented significant results.