Variability Evaluation of Measured FRF and the Effect on Force Identification

- The Effect of Systematic Error Caused by Set-up for the Experiment -

Abstract

This paper evaluates variability of experimentally measured Frequency Response Function (FRF). The FRF measurements are repeated with and without resetting the accelerometer and road cell. The difference shows the variability due to the sensor set-up for the experiment. After that, we discuss the effect of FRF variation on identified forces in Transfer Path Analysis, which is a key technology for solving the noise and vibration problems. The Matrix Inversion Method is used for force identification. The effect of the number of accelerometers on the variation of the identified forces is examined by changing the number of adopted sensors. It is turned out that the variation of identified force becomes smaller as the number of sensors is increased.