Assessment of mode shapes based damage detection methods for building structures
Journal
9th ECCOMAS Thematic Conference on Smart Structures and Materials, SMART 2019
Pages
169-176
Date Issued
2019
Author(s)
Abstract
Damage of structures can raise concerns about safety issues to users. Minor damage in structures may not be critical to structural safety; however, users may not exactly know the severity of damage in structures. Thus, structural health monitoring (SHM) becomes a strategy that can estimate current performance and safety of structures. Users can be informed regarding the levels and locations of damage in structures before the damage has been turned into a serious problem. To achieve this goal, dynamic characteristics such as natural frequencies and mode shapes can be employed to derive associated damage indices that inform structural integrity. These characteristics are typically obtained from vibrational measurements of structures through operational modal analysis. Then, damage of structures can be diagnosed from mode shapes. Therefore, the objective of this study is to assess multiple mode shapes based methods for the capability of detecting damage levels and locations. First, mode shapes of buildings are numerically calculated. Because the first mode shape is mostly available from operational modal analysis, the mode shape is then employed in all of mode shapes based damage detection methods. The damage detection methods considered in this study include the mode shape curvature method, the bending energy damage index method, the strain energy damage index method, the continuous wavelet transform method based on mode shape curvatures, and the method of changes in the first mode shape slopes. The capability of these methods for damage location detection is investigated, while the sensitivity of damage levels from these methods are also studied to understand the relationship between the derived damage indices and damage severity. Some modifications are made to the existing methods for better representing damage levels. Moreover, multiple damage locations are considered in this study, and these damage locations are assumed to sequentially occur. A detailed comparison among these methods is carried out. As seen in the numerical results, second derivatives of mode shapes (i.e., mode shape curvatures) can well represent damage of buildings in terms of damage locations and levels. ? 2019 9th ECCOMAS Thematic Conference on Smart Structures and Materials, SMART 2019. All rights reserved.
Subjects
Location; Modal analysis; Plates (structural components); Strain energy; Structural health monitoring; Wavelet transforms; Continuous Wavelet Transform; Damage of structures; Dynamic characteristics; Mode shape curvatures; Natural frequencies and modes; Operational modal analysis; Safety of structures; Structural health monitoring (SHM); Damage detection
Type
conference paper