摘要
假定任意随机激励信号由白噪声与非白噪声信号组成 ,由此导出线性结构响应之间的相关函数由两部分组成 ,一部分与脉冲响应具有相同的数学形式 ,另一部分为其它形式。利用模态分解法的基本原理 ,把相关函数分解为各个模态函数的叠加与余项之和。这样 ,第一部分信号已经分解为不同的模态函数 ,第二部分中的周期信号也变成了模态函数。这就把非稳态环境激励下多自由度线性结构系统的模态参数辨识问题转化为类似于已知各个单自由度系统的脉冲响应进行模态参数辨识问题。理论和模拟实验表明 ,本文成功地利用模态分解法进行非稳态环境激励下多自由度线性结构系统的模态参数辨识。其主要优点是 :无论是白噪声激励、稳态随机激励还是非稳态随机激励 ,仅根据结构的响应不仅能辨识线性结构的模态参数 ,而且能有效地识别出环境激励中的周期成分。
It is assumed that random signals consist of white noise signal and non white noise signal. Non white noise coefficient is brought forward. The cross correlation functions between responses of linear structures undergoing non stationary random signals consist of two parts and one is the summation of decaying sinusoids of the same form as the impulse response function of the original system. The cross correlation functions are decomposed into modal functions and the residue. Modal parameter identification technique of single degree of freedom(DOF)is applies to each mode function to obtain the natural frequencies, damping ratios and modal shapes. Numerical simulation for a linear system with four DOF demonstrated that the method is effective in identifying parameters of linear structures undergoing non stationary ambient excitation using modal decomposition. There are two advantages of the method: Firstly modal parameters of linear structures undergoing white noise signal, stationary random signal and non stationary ambient excitation can be identified; Secondly periodic signals in random excitation signals can be found.
出处
《振动工程学报》
EI
CSCD
北大核心
2002年第2期139-143,共5页
Journal of Vibration Engineering
基金
国家自然科学基金资助项目 (编号 :19972 0 16 )
