An inductive fuzzy damage classification approach for structural health monitoring

Abstract

Structural health monitoring (SHM) research gained momentum in the last two decades. Early damage detection in infrastructure is the main goal of SHM to enhance structural reliability and safety. SHM has a broader impact by extending service life of structures. A critical part of damage detection is to quantify damage severity in structures. In this article, fuzzy set theory is used in the damage classification process. Using principles of inductive reasoning, fuzzy sets are established to describe damage states in the structure. The proposed method does not rely on a specific damage feature and is applicable to different SHM systems. To demonstrate the ability of the proposed method in damage classification, two case studies are demonstrated. The two cases examine damage detection in a multi-story structure and in a pipeline structure. We establish fuzzy sets based on SHM inducted knowledge and then we implement fuzzy pattern recognition means to identify the unknown states of damage. The efficiency of the proposed method in damage detection is presented.