简要介绍独立分量分析(ICA)的基本原理,提出将ICA 方法与随机减量法(RDT)结合起来用于随机激励下结构的模态参数识别.结合数值仿真算例和振动试验分析,验证所提出方法用于随机激励下结构模态参数识别的有效性.结果表明,ICA可以准确地从结构随机振动响应信号中分离出各源信号,并同时估计出各阶模态振型向量,源信号与结构模态坐标存在一一对应关系,再结合随机减量法和单模态识别法可识别各阶模态的频率和阻尼比.该方法仅利用振动系统的输出响应进行分析,适用于随机激励下结构的工作模态参数识别.
The basic principle of independent component analysis (ICA) is briefly introduced. The ICA and randomdecrement technique (RDT) are combined and used to identify modal parameters of structures under random excitation. Bothnumerical and experimental results show that the ICA can extract the modal coordinates and estimate the mode shape vectorsfrom the random response signals of the structures directly. Free vibration responses can be obtained from the modal coordinatesby using RDT. Finally, the natural frequencies and damping ratios are calculated by using the classical one-DOF technique.It can be seen that the proposed method is effective since it only uses the output response of the vibration system foranalysis, and is suitable for modal parameter identification of the structures under random excitation