Resumen de Output-only identification of input–output models
Khaled F. Aljanaideh, Dennis S. Bernstein
In applications of system identification where the input signal is unknown, transmissibility identification can be used to estimate the operator between a pair of output signals. A limitation of transmissibility identification is the fact that the numerator and denominator of a transmissibility are numerators of the transfer functions from the input to the sensors. Consequently, transmissibility identification does not capture the poles of the system. To overcome this limitation, this paper introduces transmissibility-based system identification (TBSID), which uses estimates of two transmissibilities to identify the complete dynamics from two unmeasured inputs to two measured outputs. The method assumes that the system can be excited nonsimultaneously by two input signals at two distinct locations.
