受随机激励作用结构的演化设计算法

Evolutionary algorithm for solving the structural dynamic optimization with random excitation

提出了研究受随机激励作用的结构动力优化的演化设计方法,演化模型以随机激力作用下结构的位移方差为约束,寻求最优的构件尺寸使结构的重量最轻。演化算法采用多种群遗传与搜索空间收缩策略,并利用高效的虚拟随机激励法进行随机响应重分析和准精确罚函数处理约束,保证了算法稳定而迅速地收敛于最优解,算例显示出本文方法的有效性。

This paper presents an evolutionary design algorithm for structural subjected to dynamic loading with random excitation. The optimization problem is to find the design variable (the sizes of the structural elements) so that the structural weight is minimum under the constraints of the maximum variances of the displacements of the structure nodes. For structural reanalysis, a pseudo-excitation method is applied for the analysis of random responses; for structural optimization, quasi-exactness penalization function is used to deal with constraint, and an iteration scheme in conjunction with narrowing down space technique and multi-population evolutionary algorithm is employed to ensure very rapid and steady convergence. A program for above method is developed, which includes the procedures of the improved genetic algorithm and the pseudo-excitation method. Numerical examples are presented to demonstrate the simplicity and the effectiveness of the proposed method.