Share

Effects of sound radiation direction in faulty hermetic compressors.

Number: pap. n. C5, 1390

Author(s) : GERMEN E., KAYA A., ÜNLÜ Ü.

Summary

The Neural Networks and especially Self Organizing Map are used in general for the classification problems. The same technique is used in this paper to classify the possible failures, which can be encountered in Hermetic Compressors, during the R/D phase or after the production line. Here, those failures are created deliberately in different compressors, and the sound pressure data has been collected with 10 different microphones. In order to discriminate the faults, SOM and LVQ techniques are used. However the main aim of this paper is to scrutinize the effects of the microphones’ directions and locations while recording the both healthy and the faulty compressors’ sounds. For each microphone, its total corresponding effect has been calculated using Principal Component Analysis (PCA) method and it has been found that, some microphones have superior contribution on the recognition of the faults, but the others have minimal effect. This demonstrates that while getting the sound pressure data, its location has great importance to achieve a successful classification.

Available documents

Format PDF

Pages: 7 p.

Available

  • Public price

    20 €

  • Member price*

    15 €

* Best rate depending on membership category (see the detailed benefits of individual and corporate memberships).

Details

  • Original title: Effects of sound radiation direction in faulty hermetic compressors.
  • Record ID : 30001330
  • Languages: English
  • Source: 2010 Purdue Conferences. 20th International Compressor Engineering Conference at Purdue.
  • Publication date: 2010/07/12

Links


See other articles from the proceedings (31)
See the conference proceedings

Indexing