Damage Detection for Structural Health Monitoring of Bridges as a Knowledge Discovery in Databases Process

Abstract

The structural health monitoring (SHM) field is concerned with the increasing demand for improved and more continuous condition assessment of engineering infrastructures to better face the challenges presented by modern societies. Thus, the applicability of computer science techniques for SHM applications has attracted the attention of researchers and practitioners in the last few years, especially to detect damage in structures under operational and environmental conditions. In the SHM for bridges, the damage detection can be seen as the end of a process to extract knowledge regarding the structural state condition from vibration response measurements. In that sense, the damage detection has some similarities with the Knowledge Discovery in Databases (KDD) process. Therefore, this chapter intends to pose damage detection in bridges in the context of the KDD process, where data transformation and data mining play major roles. The applicability of the KDD for damage detection is evaluated on the well-known monitoring data sets from the Z-24 Bridge, where several damage scenarios were carried out under severe operational and environmental effects.