改进单向搜索遗传算法的工程结构优化设计

本文基于规范规定的约束条件和各项技术标准要求，建立了离散变量结构优化模型。针对遗传算法在迭代过程中经常出现未成熟收敛、振荡、随机性太大和迭代过程缓慢等缺点，提出一种新的遗传算子单亲遗传算子，用于对遗传算法的改进。并提出一种离散变量结构优化设计的单向搜索算法与遗传算法结合在一起解决问题。优化设计结果表明，这种改进单向搜索遗传算法的收敛特性得到了很好的改善，即发挥了单向搜索算法局部搜索能力强的特点，又发挥了遗传算法全局性好的特点。该方法是一种有效的工程结构优化设计方法。

Improved Unidirectional Searching Genetic Algorithm for Optimum Design in Engineering Structures

On the basis of the practical structure design of engineering, sufficiently considering the constraint conditions that are the norm shpulates and the dernands with various engineering standards, a model of building structural optimization with discrete variables was derived. In view of that the standard genetic algorithm (SGA) has such frequent shortcomings as premature convergence, oscillation and over-randomization in iterative process, a new genetic operator named one-parent genetic operator was proposed to improve SGA. Moreover, the unidirectional searching algorithm (USA) for structural optimization with discrete variables was proposed, and was combined with GA to solve the problems. The results by exemplitication show that the convergent property of improved unidirectional searching genetic algorithm is improved and the method has the advantages of saving time, high efficiency, powerfully locally searching capability of USA, and has the advantages of globally searching capability of GA, too. The method is an efficient optimal design method for engineering structure.