Deterioration and damage identification in building structures using a novel feature selection method. (February 2021)

• Record Type: Journal Article
• Authors: Gharehbaghi, Vahid Reza; Noroozinejad Farsangi, Ehsan; Yang, T.Y.; Hajirasouliha, Iman
• Is Part Of: Structures. Volume 29(2021)
• Page Start: 458
• Page End: 470
• Publication Date: 2021-02
• Subjects: Structural Health Monitoring (SHM) -- Deterioration -- Damage identification -- SSFS algorithm; Structural engineering -- Periodicals; 624.1
• DOI: 10.1016/j.istruc.2020.11.040
• Languages: English
• ISSNs: 2352-0124

Abstract

Highlights: A novel feature selection method for detecting defects using only baseline responses. A novel algorithm for assessing slight damages within building structures. Introducing sensitive features for damage and deterioration identification. Comparing signal conditions for damage and deterioration identification. Abstract: Identifying structural defects in complex structures is one of the main objectives in real-world structural health monitoring (SHM) applications. In this article, a signal-based supervised methodology is proposed for detecting deterioration and damage in building structures. This method benefits from a novel feature selection method called signal simulation-based feature selection (SSFS) algorithm, which only relies on baseline signals to extract the most sensitive features from any type of structure. The results showed that the offered methodology is capable of identifying damage and deterioration precisely, and therefore, can be a viable alternative to conventional techniques that require additional information.