基于遗传算法的潜艇首端耐压平面舱壁构架分级优化研究

摘要

探讨了潜艇首端耐压平面舱壁的结构优化设计,针对舱壁结构复杂、设计变量多、取值离散规格化等难点,确定两级优化方案,即先后分别对舱壁内部主构架和肘板进行优化,采用遗传算法作为优化方法,并利用ANSYS程序对舱壁结构进行受力分析.对遗传算法进行了改进,并在此基础上用MATLAB语言编程实现遗传算法操作,同时编程与ANSYS结合起来,完成结构优化设计.经过迭代优化,得到一个优化方案,优化部分的结构重量较初始方案降低了18.39%.%This paper investigated the optimum design of a head plane transverse bulkhead of submarine. Due to the structure is complicated and the number of variables are big and the value of variables are discrete, two-lever optimization method is used to optimize, in which the bulkhead frame is optimized firstly and then the proposed new type bracket plate secondly based on genetic algorithm, and the structure is modeled and analyzed with ANSYS. In the paper, the genetic algorithm is improved and the code is developed based on MATLAB and ANSYS,with which struetural analysis is conducted. Numerical results show that the weight of the optimal design scheme can be decreased 18.39% compared to the one of the original initial design scheme after iterative optimization.