2 Step design for sandwich panel structure satisfying with restriction of static stress and eigen frequency

Nihon Kikai Gakkai ronbunshu(2024)

引用 0|浏览0
暂无评分
摘要
The purpose of this paper is to construct a methodology which can produce a lightweight panel structure while being satisfied with more than one constraint by using some simple model. In this study, we chose a problem including bending stiffness and the 1st order eigen frequency as constraints and volume minimization as objective function among basic performances which a vehicle requires. So far, we have been trying to solve the problem by using a method consisting of Fully Stressed Design (FSD) and Genetic Algorithm (GA). FSD is an optimization which evaluates only stiffness of a structure based on relationship between thickness and stress, while GA, a broad array method, can have various number of objective functions and constraints. Prior to this study, the authors found a process was effective that stresses on the elements were to be equalized in the 1st step by applying FSD that made an individual superior on static stiffness. In this study, proposed method, which includes a process of equalization on normalized stresses on eigen vibrations, are added to that of static stresses in the 1st step. In the 2nd step, GA is selected. GA conducts operations, such as select, crossing, and mutation repeatedly until this attains a final solution. As for the model, we chose a sandwich panel which is equipped with top and bottom plates and interlayer lattice stiffeners. The reason for this selection is that the sandwich panel is lightweight and superior to stiffness, dumping, and strength. As a result, the obtained model presents a lightweight structure while being satisfied with the requirements of both stiffness and eigen frequency. Moreover, the proposed method attains optimized results earlier than GA without using FSD.
更多
查看译文
关键词
fully stressed design,genetic algorithm,sandwich panel,bending stiffness,eigen frequency
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要