基于遗传算法和参数化建模的非线性结构优化

提出了一种对存在接触关系的非线性结构(装配体)进行优化设计的新方法。该方法将遗传算法与结构几何及有限元参数化建模方法相结合，在通用CAE软件的二次开发编程环境中实现对带接触的结构装配体进行结构尺寸和形状优化设计。文中利用该方法对某浮动式闭气结构的重要结构参数和关键构件形状实施了优化设计，使其闭气性能得到大幅度提高，体现了本文方法在解决这类优化问题中的优势。本文的方法有利于拓宽结构优化技术在机械设计领域的应用范围。

Optimization of nonlinear structure based on genetic algorithm and parametrized model

The application potential of structural optimization has received wide attention from a large variety of disciplines. The current research work is mainly focused on the single continuum component structures or discrete truss or/and frame structures. The paper proposes a new method for optimized size and shape design of nonlinear structures with multiple components contacting with each other (assembly structure). The method is based on a combination of Genetic algorithm (GA) and the parameterized model of geometry and finite element of such structures. The physical and geometric properties of the structures are parameterized into a set of variables in design space, and the optimal or satisfied combination of these design variables is obtained through the search procedure in genetic algorithm. The design variables of optimized assembly structure are coded into the chromosome of an individual of GA in the floating-point number before the evaluation of each individual in a GA search process, the chromosome is decoded and converted into a finite element model, and the information required in fitness evaluation in GA is provided by finite element analysis. The algorithm and the process of modeling and analysis above are implemented in the programming environments for secondary development of MSC.PATRAN and MARC. As a demonstration, an optimized parameter design for the obturator structure in the floating breech mechanism of an automatic mortar is presented; the result indicates that its obturating performance is substantially improved. The present method helps to widen the application of structural optimization in the field of mechanical design.