作　　者： ; ; ; (李森）;

机构地区： 中山大学工学院应用力学与工程系

出　　处： 《振动与冲击》 2010年第4期30-33,共4页

摘　　要： 提出一种改进的广义遗传算法,算法中引入了异种机制以提高种群的多样性,在保证收敛速度的同时防止了早熟收敛。将该方法应用于复杂载荷作用下结构的鲁棒优化问题,并采用Taguchi望目特性的SN比构造了遗传算法的目标函数。数值算例表明,异种机制能够有效地提高广义遗传算法收敛于全局最优解的概率,加快收敛速度;结合了Taguchi鲁棒设计方法的广义遗传算法能够有效地求解复杂载荷作用下带有不确定参数的结构鲁棒优化问题。 An improved generalized genetic algorithm was proposed here.A heterogeneous strategy was introduced to increase population diversity and avoid premature convergence.This algorithm was applied to the robust optimization problems of structures under complicated loading.The objective function of the generalized genetic algorithm was constructed by using the signal-to-noise ratio of Taguchi target.Numerical examples demonstrated that the heterogeneous strategy can raise the convergence probability of the global optimal solution and speed up the process of convergence;combined with the robust design method,the improved generalized genetic algorithm can effectively solve the robust optimization problems of structures with uncertain parameters under complicated loading.

关 键 词： 广义遗传算法 结构动力优化 鲁棒优化 异种机制

领　　域： [建筑科学] [理学] [理学]