City University of Hong Kong

CityU Institutional Repository >
3_CityU Electronic Theses and Dissertations >
ETD - Dept. of Building and Construction  >
BC - Master of Philosophy  >

Please use this identifier to cite or link to this item:

Title: Structural health monitoring based on principal components analysis implemented on a distributed and open system
Other Titles: Fen san shi kai fang xi tong xia de ji yu zhu yuan fen xi de jie gou jian kang zhuang kuang jian ce
Authors: Fan, Lili (范麗麗)
Department: Dept. of Building and Construction
Degree: Master of Philosophy
Issue Date: 2006
Publisher: City University of Hong Kong
Subjects: Principal components analysis
Structural analysis (Engineering) -- Mathematical models
Notes: CityU Call Number: TA645.F36 2006
Includes bibliographical references (leaves 97-101)
Thesis (M.Phil.)--City University of Hong Kong, 2006
xiv, 112 leaves : ill. ; 30 cm.
Type: Thesis
Abstract: Structural health monitoring (SHM) and damage detection technologies have been developed for several years and many algorithms were used to achieve the goal. But each algorithm has its own advantages and limitations and different methods are applied to different types of structures for SHM, often making side-by-side comparison of different methods difficult. This thesis presents an algorithm, principal components analysis (PCA) for SHM. Different physical communication media have provided different methods for data transmission. An open and distributed network, namely LonWorks, is employed to implement the system as a control network. PCA has been widely utilized to disciplines including process performance monitoring, pattern recognition and quality control. Recently, PCA also has been used in piezosensor array for damage localization [50]. In this thesis we present another application for structural health monitoring based on principal components analysis. SHM needs critical analysis of raw and processed data from the field and a detailed comparison with the readily available data known to be normal to check whether the real-time data from the field is normal or not. Based on results from an experiment, we have got preliminary analytical finding through the application of PCA method in SHM. Besides the PCA algorithm used in SHM, the main theme of this thesis is to show how to implement such algorithms on hardware to establish a distributed system to achieve the goal of on-line monitoring and real-time damage detection. To increase the reliability of a SHM system, the idea of “decentralization” suggests that the damage detection function performed inside an intelligent device, not in the monitoring and control center, which can reduce the “workload” of the monitoring and control center and shorten delays in processing the data. The advantages of employing an open network involving open protocols, LonWorks in our case, as the medium of data transmission for structural monitoring and damage detection are highlighted. Principal components analysis has been employed to be the main algorithm of SHM in the research project and LonWorks technology provides a distributed and open monitoring system, forming the two key foundations of the SHM system developed. Finally, other than the common physical medium, i.e. twisted pair, the power line carrier, has also been adopted in this SHM system. Although there used to be limitations with power line carrier application as the physical communication medium inside a building, advancement in techniques has solved many problems, making it feasible for this type of medium to be used in the built environment. In this thesis, the power line application in SHM has been demonstrated.
Online Catalog Link:
Appears in Collections:BC - Master of Philosophy

Files in This Item:

File Description SizeFormat
fulltext.html159 BHTMLView/Open
abstract.html159 BHTMLView/Open

Items in CityU IR are protected by copyright, with all rights reserved, unless otherwise indicated.


Valid XHTML 1.0!
DSpace Software © 2013 CityU Library - Send feedback to Library Systems
Privacy Policy · Copyright · Disclaimer