Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/15082
Title: Statistical pattern recognition and data clustering techniques in industrial plant
Researcher: Kiran Jyoti
Guide(s): Satyaveer Singh
Keywords: Computer Science
Upload Date: 15-Jan-2014
University: Shri Jagdishprasad Jhabarmal Tibarewala University
Completed Date: 27-11-2012
Abstract: In modern day collection of information from different sources, classification of information according to their property, traits etc and arranging the information according to the sub group is very important in nature. The whole task of collection, classification and clustering (arranging the sub group of data) is very challenging in nature. Different researchers are working to find a better way to classify data, recognize the data and cluster the data. The data collection, classification and clustering is of prime importance of different industry. newlineIn industrial plant, where there are thousands of sensors and transducer to provide online streaming of process variable data, data collection, data processing is a very challenging job. Many a times the data is crippled and distorted by different kind of noises. So different noise resistant classification, pattern recognition algorithms and clustering algorithms has to be used. The data processing done using these steps in the industry helps the management to see the overall performance of the plant and act on the drawbacks of the system to fine tune the system. In industrial plant, these steps are useful in statistical process control application. newlineThis research studies different aspect of pattern classification, pattern recognition and data clustering algorithms in general. A industrial process control application is taken in to consideration for implementation of different classification, recognition and clustering algorithms. Different noise resistant classification and recognition algorithms are also studied as these algorithm finds wide spread use in newlineviii newlinenoisy industrial environment. A PSO (Particle Swarm optimization method) based clustering algorithm is proposed in this thesis, which can cluster the subgroup of data in an efficient manner. newline
Pagination: 
URI: http://hdl.handle.net/10603/15082
Appears in Departments:Faculty of Computer Science & Engineering

Files in This Item:
File Description SizeFormat 
01_cover page.pdfAttached File210.24 kBAdobe PDFView/Open
02_declaration by the candidate.pdf30.37 kBAdobe PDFView/Open
03_certificate.pdf31.4 kBAdobe PDFView/Open
04_aknowledgement.pdf81.68 kBAdobe PDFView/Open
05_table of contents.pdf70.85 kBAdobe PDFView/Open
06_list of tables.pdf82.89 kBAdobe PDFView/Open
07_list of figures.pdf57.03 kBAdobe PDFView/Open
08_abstract.pdf53.01 kBAdobe PDFView/Open
09_chapter-1.pdf160.07 kBAdobe PDFView/Open
10_chapter-2.pdf118.26 kBAdobe PDFView/Open
11_chpater-3.pdf119.13 kBAdobe PDFView/Open
12_chapter-4.pdf1.14 MBAdobe PDFView/Open
13_chpater-5.pdf1.37 MBAdobe PDFView/Open
14_chapter-6.pdf998.91 kBAdobe PDFView/Open
15_chapter-7.pdf75.2 kBAdobe PDFView/Open
16_appendix.pdf33.3 kBAdobe PDFView/Open
17_references.pdf62.31 kBAdobe PDFView/Open


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