MIMO随机振动试验频响估计中激励和响应的同步方法

频响函数估计是多输入多输出随机振动试验控制算法中的必要环节。在振动环境试验中，通常不采集振动台的实际激励信号，而用实测的响应信号和计算机内存中的激励信号来进行系统频响函数估计。由于计算机内存中的激励信号与实际振动台上的激励信号之间在相位上存在着差异，从而导致实测的响应信号和计算机内存中的激励信号不同步。这给系统频响函数的估计带来了较大困难。为此，根据线性系统中激励和响应之间的关系，结合随机减量法，提出了一种二次相关法用于系统频响函数的估计。用该方法进行频响函数估计只需要采集响应信号。在三轴振动台上进行了对比试验，结果验证了本文所提出的二次相关法的正确性。

Synchronization between excitation and response signals in frequency response function estimation for MIMO random vibration test

The estimation of frequency response functions (FRFs) is necessary in MIMO (Multiple Input and Multiple Output) random vibration test control. The actual excitation signals for the shaker tables are always not sampled during the vibration test. The FRFs are estimated by the actual response signals and the excitation signals in the computer memory. Because there is a phase lag between the signal in the memory and the actual signal on the shaker table, the response signal is not synchronal with the memory excitation signal. So, it is very difficult to estimate the FRFs with these signals. Based on the relationship between the excitation and the response in linear system and the random decrement technique, a Dual Correlation Method (DCM) is presented in the paper to estimate the FRFs without measuring the actual driving force signals. One only needs to sample the response signals to estimate FRFs. A comparing test between DCM and the convention method which needs to sample the excitation and response signals simultaneously is accomplished on a 3-axis vibration table. The result shows that the presented DCM is correct.