Improved wolf pack algorithm with differential evolution for global numerical optimization

Abstract

Wolf Pack Algorithm (WPA) is a new optimal algorithm with outstanding performance, but there is still room for improvement of its convergence. Aiming at speeding up the convergence of the WPA, a novel hybrid optimal algorithm based on Wolf Pack Algorithm and Differential Evolution (DE) is proposed, which named WPADE for short. Differential Evolution is introduced into the process of the wolves updating to replace the random updating. Four standard benchmark functions are applied to verify the effects of these improvements. Our experiments shows that the performance of WPADE is better than the basic WPA and other evolution algorithm or swarm intelligence optimal algorithm, such as GA, DE, PSO, and ABS.