摘要
冲击响应谱已经成为冲击问题分析的最有效工具。由于冲击问题的复杂性,如冲击信号一般频带较宽,而且冲击作用时其阻尼较模态实验有较大变化,冲击响应的相位存在很大的随机性等,因此如何得到精度可靠的冲击响应,从而求出其冲击谱;以及如何对计算仿真的冲击响应与实际测量的冲击响进行比较和确认,是冲击响应谱仿真计算的两个关键问题。针对上述问题,作者使用子结构模型修正的方法,根据冲击信号的频域特点,对整体复杂结构的各个子结构进行有限元建模和修正,然后用模态叠加法求解冲击信号的时域响应,进而使用改进的递推数字滤波法求得冲击响应谱;选择冲击谱的均方差和时域响应的主分量分解两种确认准则进行计算与实验比较和确认。通过某车架模型受横向冲击的计算和实验分析对研究方法进行了检验。
Shock Response Spectrum (SRS) is a vital tool in the study of impact problems. The most effective method to calculate a shock response is the finite element analysis when the shock signal is known. As the shock signal spans wide in frequency domain, and the damping value may have some inconsistency with the experimental modal analysis (EMA) result, the phase of shock response changes randomly, so there are some problems that should be overcome. How many modes should be involved in the FE transient analysis? How to validate the FE model with high order modes? In this paper, the FE model is updated and validated hierarchically to predict the shock responses. The principle component analysis (PCA) decomposition and other validation metrics are employed in the model validation. The improved recursive digital filtering method is used in the computation of SRS. And comparative investigation of SRS between the FEA and the test of a complex structure is given.
出处
《振动与冲击》
EI
CSCD
北大核心
2005年第6期32-36,共5页
Journal of Vibration and Shock
基金
博士学科点专项基金(20010287012)资助项目
关键词
冲击响应谱
模型修正
模型确认
主分量分解
Automobile frames
Damping
Finite element method
Modal analysis
Models
Signal filtering and prediction
Structures (built objects)
Time domain analysis
