Please use this identifier to cite or link to this item: http://localhost:8081/xmlui/handle/123456789/10549
Full metadata record
DC FieldValueLanguage
dc.contributor.authorBhavaraju, Kalyan Manohar-
dc.date.accessioned2014-11-24T08:10:30Z-
dc.date.available2014-11-24T08:10:30Z-
dc.date.issued2010-
dc.identifierM.Techen_US
dc.identifier.urihttp://hdl.handle.net/123456789/10549-
dc.guideHarsha, S. P.-
dc.guideSharma, Satish C.-
dc.description.abstractRotating machines, so called the "man made dynamical systems" in the present days are having increasing demand for their safety and reliability. These requirements mainly alarm the systems in nuclear and thermal power plants, aircraft and automobile industries, and to new and future area of interests such as autonomous vehicles and fast rail systems. Therefore, development of an effective fault diagnosis system is needed in order to avoid the systems shutdown, breakdown and even catastrophes involving human fatalities and material damage. Now a day's development of fault diagnosis is supported by computer technology both in hardware and software. Application of Artificial Intelligence (Al) is found a sort of solution for the same. Al is the intelligence of machines and the branch of computer science which helps in creating an effective fault diagnosis system. Machine Learning has been central to Al research from the beginning. Machine Learning is a scientific discipline that is concerned with the design and development of algorithms.that allow computers to learn based on data, such as from sensor data or databases. A major focus of Machine Learning research is to automatically learn to recognize complex patterns and make intelligent decisions based on data. In the present study a Wavelet based fault diagnosis system is developed. Firstly, six different base wavelets are considered in which three are from real valued and other three from complex valued. Out of these six wavelets, the base wavelets satisfying the quantitative criterion's namely maximum total Energy to total Shannon Entropy ration, Maximum Energy criteria, and Minimum Shannon Entropy and then these criteria are selected to extract statistical features from wavelet coefficients of raw vibration signals. Finally, the bearing faults are classified using these statistical features as input to machine learning techniques. Four machine learning techniques are used for faults classifications, out of which three are supervised machine learning techniques i.e. Support vector machine (SVM) Artificial Neural Network (ANN), Learning Vector Quantization (LVQ) and other one is an unsupervised machine learning technique i.e. Self-Organizing Maps (SOM). The test result showed that the SVM identified the fault categories of rolling element bearing more accurately and has a better diagnosis performance as compared to the ANN, LVQ and SOM.en_US
dc.language.isoenen_US
dc.subjectMECHANICAL INDUSTRIAL ENGINEERINGen_US
dc.subjectFAULT DIAGNOSISen_US
dc.subjectHIGH SPEED ROTOR BEARING SYSTEMen_US
dc.subjectMACHINE LEARNING TECHNIQUESen_US
dc.titleFAULT DIAGNOSIS OF HIGH SPEED ROTOR BEARING SYSTEM USING MACHINE LEARNING TECHNIQUESen_US
dc.typeM.Tech Dessertationen_US
dc.accession.numberG20478en_US
Appears in Collections:MASTERS' DISSERTATIONS (MIED)

Files in This Item:
File Description SizeFormat 
MIEDG20478.pdf5.76 MBAdobe PDFView/Open


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