基于遗传算法的肘板结构型式优化

Optimization of Bracket Plate Using Genetic Algorithm

介绍了潜艇平面舱壁上主构架与耐压体连接处的一种新型肘板结构型式,通过ANSYS软件对其应力状况进行分析,利用改进的遗传算法对其结构优化设计,对遗传算法的改进包括设计适应度函数、自适应交叉率、自适应变异率,通过编程实现在遗传算法中调用ANSYS软件,获得计算适应度函数所需的参数。通过计算表明,此类型式肘板能够有效地降低应力集中程度。

We introduced a new structure type of bracket plate at the joint of the main truss on bulkhead and shell in submarine. In this study, we analyzed the stress of this structure with ANSYS soft- ware, and optimized the bracket plate with improved genetic algorithm, in which fitness function, self-adaptive crossover rate, and self-adaptive mutation rate, etc were included. Using ANSYS by programming in genetic algorithm, we get the parameters for calculating fitness function. The results showed that the stress concentration of this type of bracket plate could be well reduced.