Open Collections

Abstract

Optimization of marine structures involves many nonlinear constraints safeguarding against var-ious limit states. Metaheuristic optimization algorithms are often the preferred choice for optimization, since they can handle such constraints, conflicting objectives and discrete design variables. Yet, their constraint han-dling technique (CHT) can significantly affect their performance. This article demonstrates the effect of con-straint handling on the performance of a few prominent swarm and evolutionary algorithms. Beside genetic algorithms, a few recent swarm optimization algorithms are tested. Constraint handling is performed using a few prominent approaches and a few recently proposed techniques. Case study is a 180 m long chemical tanker which needs to fulfil class society’s requirements. Main frame is optimized considering structural weight and deck adequacy as two concurrent objectives. Optimization of deck adequacy leads to decrease of stresses in the deck. All algorithms are run for 30 times and the statistical results are presented. Results are compared in terms of objective values and speed of convergence. Results reveal that recently proposed swarm algorithms perform worse than well-known evolutionary algorithms in terms of the convergence rate and spread of the non-domi-nated front. In addition, algorithms’ performance is strongly influenced by the constraint handling approach used. Among the ones tested, the best CHT is the repair method.