基于遗传算法的液压机上梁交互式结构优化

Interactive Structural Optimization for the Upper Cross Beam of a Hydraulic Press Based on Genetic Algorithm

很多结构优化问题的数学模型并不存在或难以求解,传统优化方法难以对这类问题进行准确寻优,需要使用交互式结构优化方法。针对基本遗传算法的缺陷,提出了一种改进型遗传算法。以700T铸造式液压机上梁(简称上梁)为例,建立结构优化模型和遗传算法模型。以上梁的最大变形和最大等效应力为约束条件,以重量为目标函数,基于有限元法和改进遗传算法对上梁进行了交互式结构优化设计。优化结果使上梁变形基本保持不变,最大等效应力降低5.87%,同时使上梁减重12.09%。

The traditional optimization methods cannot find the exact optimal solution of many structural optimization problems because the mathematic models of these types of problems do not exist or can hardly be solved accurately. So interactive method is proposed to cope with the challenge. To overcome the shortcomings of simple genetic algorithms,an improved genetic algorithm( IGA) is proposed. Through analyzing the upper cross beam structure of a 700 T cast hydraulic press,the structural optimization model and IGA model of the upper cross beam are built to undertake the IGA-FEM( Finite Element Method) interactive mechanical structure optimization with the objective of light weight and the constraints of qualified stiffness and strength. The optimization results show that the deformation of the upper cross beam maintains the status quo and the maximum equivalent stress is declined by 5.87% while the weight is decreased by 12. 09% in comparison to the original design.