Abstract
This paper presents a unique methodology of monitoring the reduction in story stiflhess of a building integrating a filtering technique. In the proposed methodology, identification is conducted based on the subsystem representing the behavior of specific story. The natural frequencies identified with respect to these subsystems give useful information for evaluating the story stiflhess. In estimating such a natural frequency of the subsystem, the subspace identification technique is utilized along with the measurement of only the right upper and lower stories' absolute accelerations. The validity of the proposed methodology is demonstrated by conducting computer simulations and scale-model experiments.
