Dissertation/Thesis Abstract

Intelligent spot welding quality monitoring using advanced signal processing techniques
by Xingjue, Wang, M.S., National University of Singapore (Singapore), 2015, 142; 10006100
Abstract (Summary)

Resistance spot welding is one of the most important and widely used metal joining techniques in industry. Amongdifferent researches in spot welding, the research on quality evaluation stands out as one of the most important aspects. Traditional quality testing of spot welding is destructive, time-consuming, and expensive. In existing literature, many online non-destructive test schemes have been proposed using neural networks (NN), linear vector quantization, SVM methods, finite element modeling, etc. However, the proposed online monitoring methods require measurements of physical parameters such as electrical signal, electrode displacement, electrode force, etc., and are vulnerable to changes in experiment conditions. In general, the proposed methods either involve complicated physical models which require much computing power or suffer from the black-box drawbacks of NN.

Theaim of this thesis was to find out a fast, convenient, and cheap online monitoring method. (Abstract shortened by UMI.)

Indexing (document details)
School: National University of Singapore (Singapore)
Department: Electrical and Computer Engineering
School Location: Republic of Singapore
Source: MAI 55/03M(E), Masters Abstracts International
Subjects: Electrical engineering
Keywords: Artificial intelligence models, Welding nuggets
Publication Number: 10006100
ISBN: 978-1-339-43962-4
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy