A Two-Stage Bayesian Approach for Ambient Modal Analysis using Wireless Sensor: Case of Separated Modes

Abstract eng:
As a promising alternative to the traditional wired sensors, wireless sensors are attracting recently widespread attention in the field of structural health monitoring (SHM). In this study, a two-stage Bayesian approach is proposed for ambient modal analysis employing wireless sensor networks (WSN). Herein we consider the case of a system with well-separated modes. During the first stage of the approach, spectral variables (i.e., eigen-frequencies, damping ratios as well as the spectral intensities of modal excitation and prediction error) and their associated uncertainties are identified by a so-called 'Fast Bayesian Spectrum Trace Approach' (FBSTA). Next, during the second stage of the approach and given the modal parameters identified in the first stage, the spatial variables (i.e., mode shapes) as well as their uncertainties are extracted instantaneously through a so-called 'Fast Bayesian Spectrum Density Approach' (FBSDA). A distributed computing strategy that improves the computational efficiency and reduces the required data transmission as well as the required memory space is also proposed to realize the proposed algorithm in the environment of WSN. A numerical example is presented to illustrate the efficiency of the proposed method.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket