Design environment for structural design: Application to modern multideck ships

Abstract

This paper presents a design environment capable of embedding multiple quality criteria for structural design and to provide the decision support problem (DSP) rationale for the concept design phase. The general mathematical model contains the analysis and the synthesis modules. The analysis module can be decomposed into six meta-systems of which two basic systems provide physical and environmental definitions of the problem/process and the other four are behavioural systems for modelling of response, adequacy, reliability, and quality. Synthesis modules enable the interactive design problem definition, usage of multiple optimization solvers (Monte Carlo-and fractional factorial experiment-based evolution strategies (ES-MC and ES-FFE), sequential linear programming (SLP), multiobjective genetic algorithms (MOGA), and multiobjective particle swarm optimization (MOPSO)) and fivedimensional graphic views of the Pareto frontier in design and attribute spaces. The methodology combines the fast concept exploration of design variants using generic finite element modelling (FEM) models and a two-step decision support procedure (based on topology and scantling optimization) in the concept design phase. The benefits of using the presented approach are demonstrated on the complex problem of multideck ships with hullsuperstructures interaction. The first example shows the good accuracy of the primary response analysis of the cruise ship generic three-dimensional FEM model. The second example briefly presents the complete structural optimization of the International Ship and Offshore Structures Congress (ISSC'06) passenger ship benchmark.

Optimization procedure for liquid natural gas carrier structural design

“…Concept design is defined as the phase in structural design when topology and geometry are open to modifications and structural variants are generated in accordance with the needs of the multiple stakeholders (head designer, owner and so on). 11 Benefits of optimization procedure in this phase are the greatest (CDP controls up to 80% of total assets). Design on the PDP level was the final objective of the project.…”

“…They can be introduced into the design problem definition modules and coupled with different optimization solvers into a multi-attribute multilevel hybrid design procedure. 11,13,14 The problem sequencer should permit flexible and interactive control of the decision-making process for the hierarchically structured design system. Analytical and synthesis modules have been developed and implemented in the practical and efficient DeSSs like MAESTRO and OCTOPUS.…”

Improving the ultimate strength of ship hull

“…Stress of each principal structural element can be obtained from the rapid 3D FEM analysis of the generic coarse mesh hull girder model (Zanic et al 2009). The ratio of this stress and the stress calculated by the means of the Bernoulli-Euler beam idealization can be interpreted as structural efficiency coefficient (Biot et al 2006…”

Application of artificial neural networks to the evaluation of the ultimate strength of uniaxially compressed welded stiffened aluminium plates

“…22 Paik et al 21 defined the fabrication-related initial imperfections of fusion-welded stiffened aluminium plate structures at three levels. Also, Paik 23 29 Zanic et al 30 presented a design environment for the structural design of ships. They showed the usefulness of the environment in the structural design of modern multi-deck ships at the conceptual design stage.…”

“…They finally derived a set of empirical formulations based on regression analyses in order to predict the ultimate strength of stiffened aluminium plates. 29 Zanic et al 30 presented a design environment for the structural design of ships. They showed the usefulness of the environment in the structural design of modern multi-deck ships at the conceptual design stage.…”