Dissertation/Thesis Abstract

Vehicle classification method for use with rapidly emplaced mobile bridges: A sensitivity study
by Rovira, Ricardo E. Basora, M.S.C.E., Purdue University, 2015, 110; 10044106
Abstract (Summary)

A feature detection algorithm is developed to determine which type of vehicle crosses a mobile bridge using acceleration responses. The purpose of this thesis is to examine the results sensitivity of the algorithm to various parameters that influence the ability to correctly classify vehicles. Each of these results will play a role in developing the most suitable procedure.

Using numerical and experimental results, the parameters studied are: bridge length, vehicle speed, noise, sensor filtering, and soil conditions. Each parameter is varied individually to determine how much it affects the ability of the method to classify vehicles traversing the bridge. Consideration is given to how parameters could be controlled under real world conditions to yield reliable results. The investigations demonstrate that results vary slightly to noise levels, the length of the bridge is constant once emplaced, sensor filtering setting can be fixed, soil condition impacts are minimum, and the vehicle speed can be controlled if a ground guide is used.

Based on the observations, a generalized procedure is prepared which consists of: creating a database with multiples parameters, controlling the parameters within realistic constraints, and grouping similar vehicle responses. The procedure aims to provide the best environment to produce reliable detection rates.

Indexing (document details)
Advisor: Dyke, Shirley J.
Commitee: Bruhl, Jakob C., Connor, Robert J.
School: Purdue University
Department: Civil Engineering
School Location: United States -- Indiana
Source: MAI 55/04M(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Civil engineering
Keywords: Acceleration, Dynamic, Feature detection, Mobile bridge, Sensitivity study, Vehicle classification
Publication Number: 10044106
ISBN: 9781339555355
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest