An Improved Cuckoo Search Algorithm Utilizing Nonlinear Inertia Weight and Differential Evolution for Function Optimization Problem

Cheng-Xu Zhang,Kai-Qing Zhou,Azlan Mohd Zain,Shao-Qiang Ye

doi:10.1109/access.2021.3130640

Cheng-Xu Zhang, Kai-Qing Zhou + Show 2 more

Open Access

https://doi.org/10.1109/access.2021.3130640

Copy DOI

Abstract

An Improved Cuckoo Search Algorithm Utilizing Nonlinear Inertia Weight and Differential Evolution for Function Optimization Problem

Highlights

Swarm intelligence (SI) algorithm is a new optimization method to obtain the optimal solution of complex optimization problems by simulating social animals' group behavior and utilizing the information transmission and cooperation among individuals in the population
To overcome the inherent bottlenecks of Cuckoo search (CS), such as low convergence accuracy, poor local optimization ability, and easy to fall into local optimization, a series of improvement measures have been used to improve the performance of standard CS
Cheng et al present a revised CS by adding a random movement operator and adjust the parameters by an improvement rate to the standard CS for dynamical selecting and updating the rules of the population. It has been tested with six CS variants on 42 benchmark functions over different dimensions, and the results show that the proposed algorithm is a competitive method [21]

Summary

INTRODUCTION

Swarm intelligence (SI) algorithm is a new optimization method to obtain the optimal solution of complex optimization problems by simulating social animals' group behavior and utilizing the information transmission and cooperation among individuals in the population. Cheng et al present a revised CS by adding a random movement operator and adjust the parameters by an improvement rate to the standard CS for dynamical selecting and updating the rules of the population It has been tested with six CS variants on 42 benchmark functions over different dimensions, and the results show that the proposed algorithm is a competitive method [21]. The mutation and cross-selection mechanisms of DE are selected to perform mutation operation on its position after updating the bird's nest and seeking the optimal fitness value using the cross-select operation This proposed WCSDE makes up for the shortcomings of information exchange between populations in the standard CS algorithm and enhances information utilization to obtain better convergence accuracy.

STANDARD CS AND DE ALGORITHMS

FUNCTION OPTIMIZATION USING THE WCSDE

D F3 xi2 10cos 2 xi 10 i 1

CONCLUSION