DocumentCode :
1898413
Title :
Discrimination of faulty compressors and healthy one using Kohonen Maps
Author :
Germen, Emin ; Kaya, Atilla ; Ünlü, Ümit
Author_Institution :
Elektrik ve Elektron. Muhendisligi Bolumu, Anadolu Univ., Eskişehir, Turkey
fYear :
2011
fDate :
20-22 April 2011
Firstpage :
845
Lastpage :
848
Abstract :
Kohonen´s Self Organizing Map (SOM) is a very usable method for the classification problems. In this work SOM is used to classify the possible faults of Hermetic Compressors which are used in the refrigerators. In refrigerators the main cause of the annoying noise/sound is the compressor sound and different faults have different noise types. Classification of the types of the sounds produces valuable information both in the R/D phase and after the production chain. In order to make this classification, the sound pressure values of different compressors with definitive faults have been collected and the healthy one and the faulty compressors are identified using SOM.
Keywords :
acoustic noise; acoustic signal processing; compressors; fault diagnosis; mechanical engineering computing; refrigerators; self-organising feature maps; signal classification; Hermetic Compressors; Kohonen self organizing map; R/D phase; classification problems; compressor sound; faulty compressors discrimination; hermetic compressors; production chain; refrigerators; sound classification; sound pressure values; Argon; Compressors; Conferences; Microwave integrated circuits; Organizing; Refrigerators; Signal processing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing and Communications Applications (SIU), 2011 IEEE 19th Conference on
Conference_Location :
Antalya
Print_ISBN :
978-1-4577-0462-8
Electronic_ISBN :
978-1-4577-0461-1
Type :
conf
DOI :
10.1109/SIU.2011.5929783
Filename :
5929783
Link To Document :
بازگشت