船舶概念设计阶段的多目标优化与决策

船舶概念设计阶段,需要初步给定多组船型参数方案并从中优选出客观合理的某一方案.以散货船为研究对象,建立了以船舶造价、单位排水量船体阻力、相对回转直径为目标的船舶概念设计优化数学模型;应用改进的非支配解排序的多目标进化算法求解船舶概念设计多目标问题以获得Pareto解集.分析了主观赋权和客观赋权的优劣,提出采用线性叠加在层次分析法、变异系数法这两种主观赋权和客观赋权方法间求取组合赋权向量,将组合向量与TOPSIS法结合对所求Pareto解集进行方案排序.优化决策结果表明,改进的非支配解排序的多目标进化算法DW=36000t散货船概念设计优化能获得多组综合性能优良的船型方案,基于线性叠加的组合赋权TOPSIS决策策略能给出客观、有效的方案序列.这种二阶段的综合方法也能推广应用于船舶优化与决策其他领域.

Multi-objective Optimization and Decision Making of Conceptual Ship Design

A design scheme should be picked from several ship parameters in conceptual design of ships. The model of ship optimization based on empirical formula, speed, maneuverability performance, cost are established. Then, the Non dominated Sorting Genetic Algorithm II , proposed by Deb, is used to perform conceptual design of a bulk carrier. The strengths and weaknesses o{ subjective weight and objective weight are analyzed and a combination weighing method is pro- posed. A bulk carrier example shows that the NSGA-H can get an evenly distributed pareto front. Meanwhile, lhe order made by combination weighting mebod is reasonable. This comprehensive approach can promote the related areas of the ship form optimization.