Master'sOpen Access

The using of support vector machines in fault diagnosis

2006
0 views
0 downloads
Advisor: Doç.dr. Erhan Akın

Abstract (EN)

ABSTRACTMaster ThesisTHE USING OF SUPPORT VECTOR MACHINES IN FAULT DIAGNOSISSuna YILDIRIMFirat UniversityGraduate School of Natural and Applied SciencesDepartment of Computer Engineering2006, Page: 77Induction motors have been commonly used in the industries.The fault diagnosis in someparts affects all of the system adversely. So, it is very important to determine the fault occured in anypart in a short time and to repaire the all of the system without affecting the all of the system.Although the Support Vector Machines give good results from most problems like textcategorization, image recognition, bioinformatics, it has not been used in fault diagnosis. In this thesis,Support Vector Machines used in fault diagnosis and it has beeen observed that it gave good results.In this thesis, the data received from induction motors used for the input of Support VectorMachines. The classification between healthy motor data and broken rotor bar faulty data has beenclassified. Binary classification and multi-class classification processes are realized. One-against-onemethod is used for multi-class classification.Key Words: Support Vector Machines, Fault Diagnosis, Classification, Kernel Functions.

Author

Dr. Suna Yıldırım

How to Cite

Suna Yıldırım (Master Thesis). The using of support vector machines in fault diagnosis, 2006, Fırat University.

Keywords

License

Tüm Hakları Saklıdır

This work is shared under the specified license terms.

More theses from Fırat University