Research Area: Metaheuristic Computing
Marine Predator Algorithm (MPA) is an optimization algorithm inspired by the behavior of predator and prey to catch their own food. MPA is simple and easy to implement. To further improve the performance of MPA, this paper proposes a Multigroup Marine Predator Algorithm (MGMPA). The multigroup mechanism is to divide the initial population into several independent groups. These groups generate the top predator and the Elite matrix based on different strategies and share information after a fixed iteration. Above strategies include the maximum of the same group, the average of the same group, the maximum of different groups and the average of different groups. To verify its performance, the paper compares MGMPA with some classic algorithms such as Particle Swarm Optimization (PSO), Parallel Particle Swarm Optimization (PPSO), Slap Swarm Algorithm (SSA), and Marine Predator Algorithm (MPA). In addition, the proposed MGMPA is also applied to solve Economic Load Dispatch problem (ELD). The experimental results show that the proposed MGMPA has significant advantages under the CEC2013 suite and obtains the minimum cost of power system operation and the maximum economic benefits in the application.
Keywords:
Marine predator algorithm
Particle Swarm Optimization
Slap Swarm Algorithm (SSA)
Author(s) Name: Jeng-Shyang Pan, Jie Shan, Shu-Chuan Chu, Shi-Jie Jiang, Shi-Guang Zheng, Lyuchao Liao
Journal name: Energy Science & Engineering
Conferrence name:
Publisher name: Wiley
DOI: 10.1002/ese3.957
Volume Information: Volume10, Issue6
