摘要
对多目标优化问题的固定加权和方法加以改进,采用随机权系数方法将多目标优化问题转换成为单目标优化问题,利用序列二次规划方法获得问题的最优解。由于权系数的随机性,多次运行能够得到多目标优化问题的Pareto优化解和Pareto前沿。和固定权系数相比,这种方法能够找到非凸均衡面上所有的Pareto最优解。采用TOPSIS法进行多属性决策(MADM)研究,对Pareto最优解给出排序。针对船舶概念设计阶段主尺度确定过程的多目标优化问题,给出了一艘散装货船两个优化目标的数值算例。计算结果表明,随机加权和方法可以求出多目标Pareto最优解集,和先验加权方法相比,随机加权方法能为设计者提供更多的选择;和遗传算法相比,现在的方法简便且精度良好。
An improved method of multi-objective optimization based on stochastic weighted sum is presented,and ship’s principal parameters design is studied,and multi-attribute decision making（MADM） is also studied.Sequential quadratic programming（SQP） and design of experiment（DOE） are used to solve the optimization problem.The Pareto solutions can be obtained by running this process a large number of times.A multi-attribute decision making（MADM） approach is adopted to rank these solutions from best to worst and to determinate the best solution in a deterministic environment with a single decision maker.Pareto solutions and frontier are given for a bulk carrier example,with six-parameter,three criterion and 14-constraint.The ranking of Pareto solution is based on TOPSIS method.The results using present method are compared with those from modern multi-objective problems.Results show that the hybrid approach has the great potential for handling multi-objective optimization problem.
出处
《中国造船》
EI
CSCD
北大核心
2010年第2期99-106,共8页
Shipbuilding of China
关键词
船舶、舰船工程
多目标优化
随机权系数
多属性决策
信息熵
ship engineering
multi-objective optimization
stochastic weighted sum method
multi-attribute decision making
entropy weights
