Empirical Validation of Bayesian Dynamic Linear Models in the Context of Structural Health Monitoring

Abstract

Bayesian dynamic linear models (BDLMs) are traditionally used in the fields of applied statistics and machine learning. This paper performs an empirical validation of BDLMs in the context of structural health monitoring (SHM) for separating the observed response of a structure into subcomponents. These subcomponents describe the baseline response of the structure, the effect of traffic, and the effect of temperature. This utilization of BDLMs for SHM is validated with data recorded on the Tamar Bridge (United Kingdom). This study is performed in the context of large-scale civil structures in which missing data, outliers, and nonuniform time steps are present. The study shows that the BDLM is able to separate observations into generic subcomponents to isolate the baseline behavior of the structure.