Toggle Main Menu Toggle Search

Open Access padlockePrints

A Hybrid Optimization Technique for Marine Structures

Lookup NU author(s): Dr Arun DevORCiD

Downloads

Full text for this publication is not currently held within this repository. Alternative links are provided below where available.


Abstract

A structural optimization method is proposed for marine structural design. A single objective optimization can no longer meet more and more design objectives which may be conflicting. Multi-objective optimization becomes a powerful tool in marine design to help find an optimal solution for conflicting objectives. The structure is broken down into panels following structure production plan of building a ship or an offshore structure. The constraint and objective functions in terms of the design variables are formulated in the Spreadsheet. A multiobjective optimization engine is employed to generate a set of Pareto solutions. These solutions are evaluated and ranked by using a “technique for order preference by similarity to ideal solution” technique (TOPSIS). The optimum structure which gives a balance for those trade-off objectives with the designer’s preference will be selected. The objectives can be minimum lightweight, minimum cost, minimum environment impact during its life cycle, crashworthiness, maximum fatigue coefficient, etc. The proposed optimal design will be verified by using FEA. At this stage, a topology optimization will be applied to further optimize the mass of the structure by reducing the extra material on primary structure which doesn’t affect the structural integrity. A case study is performed on a general cargo ship which is one of the most common vessels in world fleet.


Publication metadata

Author(s): Wei X, Dev AK

Publication type: Conference Proceedings (inc. Abstract)

Publication status: Published

Conference Name: Hyperworks Technology Conference (HTC 2012)

Year of Conference: 2012

URL: http://www.altairatc.com/india/previous-events/2012/papers-2012/O_Asean-Newcastle.html


Share