用相干函数改善频响函数偏度估计误差的两种方法

Reducing the Bias Estimation Errors of Frequency Response Function by Two Coherence Functions

本文提出在测量噪声影响较大的情况下,采用(?)(f)和(?)(f)估计可减少实测频响函数估计的偏度误差,分析了(?)(f)和(?)(f)估计的实用范围,并用实验验证了本文方法的正确性。实验结果表明:对频响函数进行偏度误差的改善,其模态参数识别的精度比不改善偏度估计误差的识别结果的精度要高。

This paper presents the methods of (?)(f) and (?) (f) :estimations which can reduce the bias estimation crrors of the measured frequency response under the condition of more influence noise in a measured system, analyses, the practical range of these methods, and their validity is proved by experiments. The ex- perimental results indicate that the accuracy of the model parameter identification is increased when the bias estimation errors of the measured frequency response rune- tions are reduced, otherwise the accuracy is reduced.