Publication detail

Signal Source Location by Means of Neural Network

JIRSÍK, V., ČERNÝ, L.

Original Title

Signal Source Location by Means of Neural Network

Type

conference paper

Language

English

Original Abstract

This work deals with pertinent application of signal source location method by means of neural network. As a source an acoustic emission signal was used. Among the most important parameters participating in the final results was the correct measurement method and signal processing choice. For the source signal location the back propagation-type neural network was used. Thanks to this network‘s ability to ‚learn‘ and adapt to ambient conditions, represented by jamming signals, it was possible to take these into account. The problem dealt with was the choice of suitable signal parameters of all the parameters obtained by measurement. These parameters then represented the neural network inputs and influenced its results. Therefore, it was necessary to make several experimental measurements. As the final ones were selected those, with which the neural network results gave the most accurate values. The accuracy evaluation was done by comparing the neural network outputs with required values of coordinates.

Key words in English

acoustic emission, neural network, location

Authors

JIRSÍK, V., ČERNÝ, L.

RIV year

2002

Released

1. 1. 2002

Publisher

EWGAE

Location

Praha

ISBN

80-214-2174-6

Book

25th European Conference on Acoustic Emission Testing

Pages from

99

Pages to

105

Pages count

7

BibTex

@inproceedings{BUT5352,
  author="Václav {Jirsík} and Luděk {Černý}",
  title="Signal Source Location by Means of Neural Network",
  booktitle="25th European Conference on Acoustic Emission Testing",
  year="2002",
  pages="7",
  publisher="EWGAE",
  address="Praha",
  isbn="80-214-2174-6"
}