微分演化算法在结构参数识别中的应用

唐和生;周进;薛松涛;范存新

振动与冲击 ›› 2010, Vol. 29 ›› Issue (9) : 42-46.

PDF(1454 KB)
PDF(1454 KB)
振动与冲击 ›› 2010, Vol. 29 ›› Issue (9) : 42-46.
论文

微分演化算法在结构参数识别中的应用

  • 唐和生1;周进1;薛松涛1,2;范存新3
作者信息 +

Application of a Differential Evolution Algorithm in Structural Parameter Estimation

  • TANG Hesheng1;ZHOU Jin1;XUE Songtao1, 2;FAN Cunxin3,
Author information +
文章历史 +

摘要

基于微分演化算法(Differential Evolution, DE)提出了一种新的结构参数识别方法。通过对参数识别反问题转化为一个优化问题,根据实际观测结构响应数据与数值模型系统输出之间的差异建立识别问题的目标函数。利用DE求解该目标函数的全局最小,从而得到最优参数解。DE算法是一种新颖的随机搜索进化算法,通过采取全局优化的策略确保算法得到合理的解。DE算法具有算法简单、编程计算方便、同时收敛速度快、计算结果精度高、和鲁棒性强的优点。本文通过数值模拟及该识别方法在真实结构参数识别中的应用验证该方法的有效性。

Abstract

A new approach to parameter estimation in structural system is developed using differential evolution (DE) algorithm. The inverse problem of parameter identification is formulated as an optimization problem. Based on the information from the observed dynamic time-history and calculated dynamic time-history, an objective function for inverse problem is defined. DE is used to find global minimum of the objective function, which in turns is the solution for the system parameters. DE is a new robust stochastic evolutionary computational algorithm. This novel heuristic technique aims to ensure the proper solution of this problem by adopting a global optimization approach. The potentialities of DE are its simple structure, easy use, convergence property, quality of solution, and robustness. The effectiveness of the proposed method is evaluated through the numerical simulations and an application to a real building.

关键词

微分演化 / 参数识别 / 优化

Key words

Differential Evolution / Parameters Estimation / Optimization

引用本文

导出引用
唐和生;周进;薛松涛;范存新. 微分演化算法在结构参数识别中的应用[J]. 振动与冲击, 2010, 29(9): 42-46
TANG Hesheng;ZHOU Jin;XUE Songtao;;FAN Cunxin;. Application of a Differential Evolution Algorithm in Structural Parameter Estimation[J]. Journal of Vibration and Shock, 2010, 29(9): 42-46

PDF(1454 KB)

1574

Accesses

0

Citation

Detail

段落导航
相关文章

/