Title

Application Of Two Individual Data-Driven Based Change/Damage Detection Methods For Bridge Monitoring

Abstract

Fault detection is an important component for Structural Health Monitoring (SHM) applications. Herein, the efficiency of two data-driven based damage detection algorithms for bridge monitoring application will be explored and demonstrated. These algorithms will be based on Robust Regression Analysis (RRA) and Moving Principal Component Analysis (MPCA) as two statistics-based damage detection algorithms, which do not require a mathematical model for implementation. As a result, these methods are classified as data-driven techniques and they are quite effective for practical use in real life as long as the limitations are understood and the uncertainties can be evaluated. These methods will be demonstrated on a phenomenological model developed in the laboratory. This model, the UCF 4-span bridge, is equipped with Fiber Bragg Grating (FBG) sensors at 10 different locations and 2 most common and critical damage scenarios are chosen and induced for fault detection application. In addition to the lab test, the effectiveness of these techniques is tested with a real-life data from a unique structure. © 2013 Taylor & Francis Group, London.

Publication Date

12-1-2013

Publication Title

Safety, Reliability, Risk and Life-Cycle Performance of Structures and Infrastructures - Proceedings of the 11th International Conference on Structural Safety and Reliability, ICOSSAR 2013

Number of Pages

2241-2248

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

Socpus ID

84892418874 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/84892418874

This document is currently not available here.

Share

COinS