Item – Theses Canada

​
OCLC number
426219690
Link(s) to full text
LAC copy
LAC copy
Author
Wu, Jianmou,1977-
Title
CBM optimization models with multivariate observations.
Degree
Ph. D. -- University of Toronto, 2006
Publisher
Ottawa : Library and Archives Canada = Bibliothèque et Archives Canada, [2007]
Description
2 microfiches
Notes
Includes bibliographical references.
Abstract
Modern manufacturing systems have become highly automated. The sophistication of the equipment has increased the impact of unplanned downtime caused by system failures. Identifying a cost effective maintenance program is a key management objective. To express the trade-offs between maintenance costs and benefits in a model, one needs an appropriate maintenance policy and relevant system health measures. Condition-based maintenance (CBM), the policy in which preventive intervention is conducted after identifying a symptom of impending failure with the aid of condition monitoring, provides an effective way to formulate maintenance optimization problems. How to gain robust system health measures and incorporate them into the model in order to gain an appropriate maintenance policy are the main issues in maintenance optimization. This thesis attempts to solve part of the above problems in innovative ways. Firstly, it focuses on VAR modeling of real oil data and application of DPCA in order to gain robust system health indexes from dimensionality reduction and correlation exclusion. Next, multivariate SPC charts (' T'2 chart and DPCA-based (<math> <f> T2<inf>A</inf></f> </math>,'Q') chart) have been applied in an innovative way for condition monitoring of truck transmission. In the last part, the idea of applying SPC chart for maintenance optimization is further developed by designing the SPC chart with considering the unique properties of the maintenance problems. The effectiveness of all the above innovative methodologies is demonstrated by either case studies based on real oil data or simple numerical examples.
ISBN
9780494219638
0494219637