基于贝叶斯谱密度法的非线性桥梁结构动力系统模型参数修正  

Updating Parameters of Bridge Structural Nonlinear Dynamic System Model Based on Bayesian Analysis to Spectral Density

作　　者：唐洪亮[1] 马骎[1] 冯良平[1] 李娜[1] 

机构地区：[1]中交公路规划设计院有限公司,北京市100088

出　　处：《公路》2012年第12期19-22,共4页Highway

摘　　要：采用贝叶斯概率方法对在不确定输入作用下随机模型的模型参数进行定量化。对系统响应信号运用谱密度识别法进行模型修正。贝叶斯谱密度法运用快速傅里叶变换(FFT)对非线性动力系统结构参数的概率密度函数进行修正,不仅符合系统响应信号的概率分布,同时也适用于非线性系统参数识别。利用贝叶斯谱密度法,对两跨连续梁桥非线性系统结构模型进行修正。A Bayesian probabilistic approach is employed to quantify the uncertain model parameters of nonlinear dynamic systems. A spectral density approach for the identification is used to update parameters by system response measurements. The proposed approach utilizes Fast Fourier Transform which respects the probability distribution of the response signal to obtain the updated probability density function for the nonlinear dynamic system parameters. The proposed approach is well suited to the nonlinear systems iden- tification. For illustrating the proposed approach, a two-span continuous beam bridge nonlinear dynamic system will be employed.

关 键 词：贝叶斯分析 非线性 动力系统 谱密度函数 

分 类 号：U441.5[建筑科学—桥梁与隧道工程]