Title: Modified multi-verse optimiser used for global optimisation

Authors: Dilip Kumar Mishra; Vikas Shinde

Addresses: Government Narmada PG College, Hoshangabad, 461001, India ' Madhav Institute of Technology and Science, Gwalior, 474005, India

Abstract: In this paper, modified version of multi-verse optimiser (MVO) was suggested and tested on numerical optimisation problems. MVO is an innovative optimisation approach which stimulated from the concepts of cosmology; they are named as white hole, black hole and wormhole. Mathematical modelling of this concept has been carried out to acquire exploitation, exploration and local search. Modification in MVO has been made by introducing concept of dynamic variation in population size (universe). Modified multi-verse optimiser (MMVO) was tested on 16 benchmark functions having different complexity. Statistical comparisons of other algorithms outcomes is depicted that MMVO performs better than other algorithms.

Keywords: optimisation; particle swarm optimisation; PSO; genetic algorithm; benchmark functions.

DOI: 10.1504/IJSI.2021.114771

International Journal of Swarm Intelligence, 2021 Vol.6 No.1, pp.65 - 76

Received: 24 Feb 2020
Accepted: 07 Jul 2020
Published online: 05 May 2021 *