基于改进蚁群算法的复合材料双向加载十字形试件优化设计

为研究复合材料在双向加载情况下的应力应变特性,基于蚁群算法对双向加载试验中的十字形试件进行优化设计。考虑信息素更新对全局最优解有较大的依赖性,对编码方式和信息素更新公式进行了改进,并通过典型算例对改进后的蚁群算法准确性进行了验证。最后根据复合材料十字形试件设计要求,利用改进蚁群算法与FEM参数化建模相结合的方法对十字形进行了优化设计。由优化结果可知,优化后的试验区应变场均匀度有了很大提高,且优化迭代过程较快。

Optimization of a Biaxial Loading for Cruciform Specimen of Composite Based on Modified ACO

In order to study the complex stress and strain field of composite under biaxial load,a new modified Ant Colony Optimization(ACO) is proposed in this paper for shape and size optimization of composite cruciform specimen.Due to the strong dependence of global best value in the iterations and special requirements of specimen design,the ant′s move method and pheromone updating strategy were modified,and a classical example was also employed to demonstrate validation and versatility of the modified algorithm.A FEM parametric modeling technique was coupled with the modified ACO so as to optimize the geometry of composite cruciform specimen.The optimization result shows that a uniform strain field is obtained in the test zone and the convergence process of the optimization is quick.