主催: 一般社団法人 日本機械学会
会議名: Dynamics and Design Conference 2019
開催日: 2019/08/27 - 2019/08/30
Experimental modal analysis is one of the key technologies in structural dynamics analysis. When the resonance frequency is not close to the measured frequencies, it is difficult to identify low damping characteristics due to lack of FRF data around resonance frequency. Thus, the linear fit method which is applicable to systems with very low damping characteristics is proposed in our previous study. In the proposed method, a set of linear functions with respect to the real and imaginary parts of original Frequency Response Function (FRF) is derived by canceling the residue of FRF. This paper focuses on a single-input-multi-output (SIMO) and presents a method of improving the identification accuracy by utilizing the fact that data of the same frequency can be averaged regardless of reference points by erasing the residue terms. First, the theory of identification is explained using the linear fit method and through the application of the SIMO method. Next, a simple simulation is used to validate the method. we conclude that the estimation accuracy of the overall modal parameters can be improved by the proposal method.