Random Walk Mutation-Based De With Eda For Nonlinear Equations Systems

Zuowen Liao,Wenyin Gong,Zhihua Cai,Ling Wang,Yong Wang

DOI: https://doi.org/10.1109/CEC.2019.8790111

2019-01-01

Abstract:Finding multiple roots of nonlinear equations systems (NESs) in a single run is an important yet difficult task. It requires to keep a balance between explorative and exploitative traits. In this paper, we present a random walk mutation-based differential evolution (DE) with estimation of distribution algorithm (EDA) to address this problem. The major characteristics are: i) the random walk mutation is capable of preserving the population diversity, which guides individuals to move toward different promising regions; ii) probability selection is employed to provide suitable parent individuals for evolution; iii) EDA is used to accelerate the convergence and obtains the roots. To evaluate the performance of our approach, 30 NESs with diverse features are selected as test suite. Experimental results indicate that the proposed approach is able to yield better performance compared with other state-of-the-art methods.