摘要
为了提高V型无压载水船的结构优化效率,采用遗传算法结合RBF代理模型对其进行尺寸优化。首先对NOBS油船结构特性进行分析,采用灵敏度分析方法筛选优化设计变量,通过最优拉丁超立方抽样选取样本点,构建设计变量和各响应的高精度RBF代理模型,然后利用遗传算法和全局响应面搜索法进行多目标全局最优化计算。对比分析优化前后的计算结果,优化后舱段重量减少了1.15%,刚度提升了4.3%,符合规范中的要求,并提高了优化设计效率,对V型无压载水船结构优化提供了参考。
In order to improve the efficiency of structural optimization of V-type non ballast water ship,genetic algorithm combined with RBF agent model for size optimization.Firstly,the structural characteristics of non-ballast water ship(NOBS)tanker are analyzed,and the sensitivity analysis method is used to choose the optimal design variables and the sample points are selected by Latin hypercube sampling.And then the high-precision RBF agent model of design variables and responses is constructed.Then,the multi-objective global optimization are calculated by genetic algorithm and global response surface search method.After optimization,the weight of the cabin is reduced by 1.15%,and the stiffness is increased by 4.3%,which meets the requirements of the specification.It also improves the efficiency of optimization design and provides a reference for the structural optimization of V-type non ballast water vessel.
作者
苏绍娟
刘灿波
张祥
王国回
SU Shao-juan;LIU Can-bo;ZHANG Xiang;WANG Guo-hui(College of Naval Architecture and Ocean Engineering,Dalian Maritime University,Dalian 116026,China;Nantong COSCO KHI Ship Engineering Co Ltd,Nantong 226005,China)
出处
《武汉理工大学学报》
CAS
2021年第11期35-42,共8页
Journal of Wuhan University of Technology
基金
国家自然科学基金(51609031)
中央高校基本科研业务费专项基金(3132019320)。
关键词
无压载水船舶
结构优化
径向基函数
代理模型
多目标优化
non-ballast water ship
structural optimization
RBF
surrogate model
multi-object optimization
