遗传算法和强度评判结合作空间网架截面优化分析

Genetic Algorithm and Strength Judgment for Design Optimization of Space Truss Structure

尝试将遗传算法与强度评判结合，用于空间网架的截面积优化，以得到最轻结构．这样，既可以对多峰性问题突破局部优化解的局限，又可以弥补遗传算法概率探索的盲目性，加快收敛速度．用型材构造的空间网架，将离散的截面参数用二进制的遗传因子表示，利用染色体的遗传繁殖和自然淘汰等特点，通过相当数量的染色体集团的交叉和变异，在广阔的参数域内对优化解进行搜索．算例表明，此方法可以在不大的计算量下获得相当良好的优化解．

The paper attempts to optimize sectional area of a space truss structure using a combined method of genetic algorithm and strength judgment.This method can not only break through confine of a local solution for the multi peak problems,but also make up for a blindness of probability search of the genetic algorithm and accelerate convergence.For a space truss consisting of the mold material,separable sectional parameters are replaced by the genetic coding of binary number.By means of characters of genetic reproduction and natural selection as well as crossover and mutation of a large number of chromosome population,optimal solution is searched on within design parameter region.The computational example indicates that this method can obtain better solution spending less cost.