Structure-type classification and flexibility-based detection of earthquake-induced damage in full-scale RC buildings

Detecting early damage in civil structures is highly desirable. In the area of vibration-based damage detection, modal flexibility (MF)-based methods have proven to be promising tools for promptly identifying changes in the global structural behavior. Many of these methods have been developed for specific types of structures, giving rise to different approaches and damage-sensitive features (DSFs). Although structural type classification is an important part of the damage detection process, this part of the process has received little attention in most literature and often relies on the use of a-priori engineering knowledge. Moreover, in general, experimental validations are only performed on small-scale laboratory structures with controlled artificial damage (e.g., imposed stiffness reductions). This paper proposes data-driven criteria for structure-type classification usable in the framework of MF-based damage identification methods to select the most appropriate algorithms and DSFs for detecting and localizing structural anomalies. This paper also tests the applicability of the proposed classification criteria and the damage identification methods on full-scale reinforced concrete (RC) structures that have experienced earthquake-induced damage. The considered structures are a seven-story RC wall building and a five-story RC frame building, which were both tested on the large-scale University of California, San Diego-Network for Earthquake Engineering Simulation (UCSD-NEES) shaking table.