Home
Resource
Gantry health monitoring and fault detection based on process status sequence
PDF
PDF format is widely accepted and good for printing.
Plug-in required
PDF-1(1.25 MB)
Citation & Export
View Usage Statistics
Staff View

Citation & Export
Hide

Simple citation

SONG, RUI. Gantry health monitoring and fault detection based on process status sequence. Retrieved from https://doi.org/doi:10.7282/t3-t93d-g977

Export

Click here for information about Citation Management Tools at Rutgers.

Statistics
Hide

Description

TitleGantry health monitoring and fault detection based on process status sequence
NameSONG, RUI (author); Jeong, Myong-Kee (chair); Guo, Weihong (co-chair); Hoang, Pham (internal member); Rutgers University; School of Graduate Studies
Date Created2019
Other Date2019-01 (degree)
SubjectIndustrial and Systems Engineering, Structural health monitoring
Extent1 online resource (38 pages) : illustrations
DescriptionGantry refers to the system that moves the hoist by the machinery house along tracks on the floor level and transfers the material. As a critical asset, gantry has wide applications in many fields such as medical image area, infrastructure, and heavy industry. Mostly, gantry is reliable, however, the loss led by the gantry lockout is inestimable enormous. Moreover, there are limited previous gantry studies concentrate on the statistical quality control to detect the fault not to mention the research that focuses on the algorithms applied to the process status sequence to detect the fault. The categorical process status sequence is hard to obtain the features when dealing with fault identification. This thesis provides a novel method applying texture extraction in image processing to obtain the features of gantry process status sequence. Texture extraction techniques such as the histogram of oriented gradients (HOG) and local binary pattern are applied to the process status sequence. To demonstrate the effectiveness of image-based feature extraction, k-nearest neighbors, support vector machine, linear discriminant analysis, and quadratic discriminant analysis are applied to the time-series gantry process status sequences provided by a leading automobile manufacturer. Result demonstrates that the sequence after the transformation of both texture extraction techniques have improved the accuracy. The process status sequence after HOG transformation has the best performance. Besides, the HOG technique also dramatically reduces the dimension of the process status sequence. This result can help the on-site expert prognosis the fault as well as prepare the corresponding troubleshooting guide to save time and resources.
NoteM.S.
NoteIncludes bibliographical references
Noteby Rui Song
Genretheses, ETD graduate
Persistent URLhttps://doi.org/doi:10.7282/t3-t93d-g977
Languageeng
CollectionSchool of Graduate Studies Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work.
Version 8.5.5
Rutgers University Libraries - Copyright ©2024