Powell-Based Bat Algorithm for Solving Nonlinear Equations

Abstract

As the bat algorithm (BA) has defects such as slow convergence and poor calculation precision, it is likely to result in local extremum, and Powell algorithm (PA) is sensitive to the initial value. To resolve the above defects, advantages and disadvantages of PA and bat algorithm are combined in this paper to solve nonlinear equations. The hybrid Powell bat algorithm (PBA) not only has strong overall search ability like bat algorithm, but also has fine local search ability like Powell algorithm. Experimental results show that the hybrid algorithm can be used to calculate solutions to various nonlinear equations with high precision and fast convergence. Thus, it can be considered a positive method to solve nonlinear equations.