Wang-Ji Yan

Mode Shape Assembly for Ambient Modal Analysis using a Two-stage Bayesian Spectral Density Approach

Abstract. A two-stage Bayesian spectral density approach was formulated for ambient modal analysis recently. The interaction between spectrum variables (e.g., frequency, damping ratio as well as the magnitude of modal excitation and prediction error) and spatial variables (e.g., mode shape components) can be decoupled so that they can be identified separately. The proposed method can be implemented in the environment of a wireless sensor network through a distributed computing strategy so that local mode shapes as well as their uncertainties confined to different clusters can be identified. However, the difficulty on how to assemble these local mode shapes estimated from multiple clusters is still a problem required to be resolved properly. In this study, a Bayesian mode shape assembly methodology is proposed so that the weight for different clusters is accounted for properly according to their data quality. The optimal values for the global mode shape components corresponding to all measured dofs can be obtained by a fast iterative scheme, while the associated uncertainties can be derived analytically. There is no need to share the same set of reference dofs for all clusters for scaling purpose when using ambient vibration data. The proposed mode shape assembly method is investigated with a shear building model. Results show that the overall mode shape can be effectively identified by the proposed method.