Home > Archive > 2020 > Volume 10 Number 4 (July 2020) >
IJMLC 2020 Vol.10(4): 594-598 ISSN: 2010-3700
DOI: 10.18178/ijmlc.2020.10.4.978

Leak Detection on Air reservoir via Acoustic Models with TensorFlow Based

Naparat Pairin and Ramil Kesvarakul

Abstract——Fluid, both gas and liquid, is a widely used substance which can be used in pneumatic and hydraulic system. However, the pneumatic system with compressed air system integrated has a flaw that leakage air during power transmission cost a lot of loss both resources and performance. Leak detection is one of the main solution to plug the flaw. In this research, we use acoustic signal to detect the leakage by using it as an input for model. Artificial Neural Network (ANN) is used in our model to achieve deep learning property via Tensorflow. Acoustic signal is recorded in different situation and is used as a model input. So, our model can be trained with leak data and predict the leakage in pneumatic system. We evaluate model using test data and shows the leakage prediction in probability distribution.

Index Terms—Artificial neural network, leak detection, Tensorflow.

The authors are with King Mongkut’s University of Technology North Bangkok/Production Engineering, Bangkok, Thailand (e-mail: naparat.pai@gmail.com, ramil.k@eng.kmutnb.ac.th).

[PDF]

Cite: Naparat Pairin and Ramil Kesvarakul, "Leak Detection on Air reservoir via Acoustic Models with TensorFlow Based," International Journal of Machine Learning and Computing vol. 10, no. 4, pp. 594-598, 2020.

Copyright © 2020 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).

General Information

  • ISSN: 2010-3700 (Online)
  • Abbreviated Title: Int. J. Mach. Learn. Comput.
  • Frequency: Bimonthly
  • DOI: 10.18178/IJMLC
  • Editor-in-Chief: Dr. Lin Huang
  • Executive Editor:  Ms. Cherry L. Chen
  • Abstracing/Indexing: Inspec (IET), Google Scholar, Crossref, ProQuest, Electronic Journals Library.
  • E-mail: ijmlc@ejournal.net


Article Metrics