The corona virus search optimizer for solving global and engineering optimization problems

Abstract

In this paper, a new and effective meta-heuristic method named corona-virus search optimizer (CVSO) is proposed inspired based on the movement and search of the corona-virus among different societies, prevalence, and death of individuals. The CVSO has local and global search suitable balance due to the use of evolutionary strategies as well as appropriate social learning. This algorithm has desirable optimization power, easy implementation and requires only a control parameter. The CVSO algorithm is implemented on 30 modern and standard test functions of CEC2014 and also 10 test functions of CEC2019. The performance and capability of CVSO is compared with new and well-known meta-heuristics including particle swarm optimization (PSO), artificial bee colony (ABC) tree growth algorithm (TGA), and Lion optimization algorithm (LOA), bat algorithm (BA) and harris hawks optimization (HHO). The results showed that the CVSO can be more effective than other algorithms to solve the test functions and also various complex real-world engineering optimization problems in terms of optimization accuracy and convergence rate. Also, the superiority of CVSO is demonstrated in the designing of a hybrid energy system by achieving a lower cost of M$ 2.7492 compared to TGA, LOA, ABC, PSO, BA, and HHO algorithms, which achieve costs of M$ 2.7878, M$ 2.7556, M$ 2.8521, M$ 2.9503, M$ 3.0942, and M$ 2.9614, respectively. In addition, due to the optimal balance between local and global search and strength, it is able to avoid getting stuck in locally optimal solutions while obtaining global optimal solutions with a higher convergence rate than other algorithms. The source code of CVSO is publicly available at https://github.com/Irajfaraji/CVSO.