Application of a new multi-agent Hybrid Co-evolution based Particle Swarm Optimisation methodology in ship design

Cui, Hao and Turan, O. (2010) Application of a new multi-agent Hybrid Co-evolution based Particle Swarm Optimisation methodology in ship design. Computer-Aided Design, 42 (11). 1013–1027. (https://doi.org/10.1016/j.cad.2009.07.005)

[thumbnail of Jrn_CAD_APPLICATION_OF_A_NEW_MULTI-2607.pdf]
Preview
PDF. Filename: Jrn_CAD_APPLICATION_OF_A_NEW_MULTI-2607.pdf
Accepted Author Manuscript

Download (1MB)| Preview

Abstract

In this paper, a multiple objective 'Hybrid Co-evolution based Particle Swarm Optimisation' methodology (HCPSO) is proposed. This methodology is able to handle multiple objective optimisation problems in the area of ship design, where the simultaneous optimisation of several conflicting objectives is considered. The proposed method is a hybrid technique that merges the features of co-evolution and Nash equilibrium with a ε-disturbance technique to eliminate the stagnation. The method also offers a way to identify an efficient set of Pareto (conflicting) designs and to select a preferred solution amongst these designs. The combination of co-evolution approach and Nash-optima contributes to HCPSO by utilising faster search and evolution characteristics. The design search is performed within a multi-agent design framework to facilitate distributed synchronous cooperation. The most widely used test functions from the formal literature of multiple objectives optimisation are utilised to test the HCPSO. In addition, a real case study, the internal subdivision problem of a ROPAX vessel, is provided to exemplify the applicability of the developed method.