Research of Improved Whale Optimization Algorithm based on BFO

Abstract

Whale Optimization Algorithm(WOA) is a new kind of meta-heuristic optimization algorithm that has been applied to many engineering field. It enjoys the advantages of simplicity in principle, small number in parameters, excellent ability in search and global convergence, also suffer from many defects, such as slow and deficient accuracy at convergence as well as to tend to local optimum. In order to overcome the flaws and enhance performance of classical WOA,this paper recognizes the convergence state through making use of slope of straight lines acquired from fitness results, at the same time, integrate some mechanism belonging to BFO with many advantages, names a new improved WOA as BFO-WOA. Cooperation with group communication factor, chemotaxis correction and forced elimination-dispersal makes search precision and capacity of jumping out of local minimum of WOA promote tremendously. Through dual nonlinear variable convergence factor, the progress of search is more well-directed than by original solo factor. Testing results and convergence curves of benchmark functions indicate that exploration and exploitation studied in this paper are competitive with other improved WOA, and it will take positive effects to theoretical research of pattern recognition, artificial intelligence, system control and signal processing, make engineering application meet the time and performance index of practical requirements.