摘要
根据工程实际出发,建立离散变量结构拓扑优化模型。针对基本遗传算法(GA)存在的早熟收敛、随机性大、收敛速度慢等不足,从编码方式及遗传算子操作等几个方面对其作了改进,选择操作中引进斐波那契方法,提出一种基于整数编码的改进遗传算法。优化结果表明,这种改进的遗传算法的收敛性得到了很好的改善,提高了计算效率,是有效的结构拓扑优化设计方法。
On the basis of the practical structure design of engineering, a model of structural topology optimization with discrete variables is derived.In view of that the standard genetic algorithm has such frequent shortcomings as premature convergence, over-randomization in iterative process, a integer coded genetic algorithm is proposed, in which from the encoding and genetic operators of its operation is improved, and a Fibonacci method is introduced to improve the genetic operation.The results by exemplification show that the convergent property of improved genetic algorithm is improved and the improved genetic algorithm is an efficient topology optimal design method for engineering structure.
出处
《重庆科技学院学报（自然科学版）》
CAS
2010年第5期158-160,共3页
Journal of Chongqing University of Science and Technology：Natural Sciences Edition
关键词
离散变量
拓扑优化
整数编码
改进遗传算法
discrete variable
topology optimization
integer coded
improved genetic algorithm
