Application of the wavelet and neural technologies for processing of signals obtained during railway tracks diagnostics by the magnetic flux leakage method

Journal Title: Bulletin of the Military University of Technology - Year 2017, Vol 66, Issue 4

Abstract

In this article, the approach for detecting a transverse crack in the rail head via ANN with CWT and application created on its basis are presented. The ways of further development of the ANN for improving its work accuracy and the possibility of identification of other types of defects are also presented. Keywords: defect, transverse crack, CWT, ANN

Authors and Affiliations

Vitalij Nichoga, Liubomyr Vashchyshyn

Keywords

Related Articles

Wyznaczanie funkcji kompensacji momentów podporowych z badań modelu składanej belki ciągłej opartej na podporach stałych

Artykuł obejmuje metodykę wyznaczenia funkcji kompensacji momentów podporowych dla wieloprzęsłowych belek składanych na podstawie badań modelowych. Funkcja ta ma zasadniczy wpływ na rozkład momentów zginających w zależno...

The airport pavements reinforced with Mineral-Asphalt Mixtures’ (MAM) layers

The article presents original solutions, methodology, and procedure for determining the thickness of bituminous overlays for renovation and modernization of the existing airfield pavements.Using the Westergaard method, t...

Ocena właściwości tribologicznych wybranych substancji smarujących na podstawie charakterystyk Herseya–Stribecka

W pracy przedstawiono wyniki badań dotyczących właściwości tribologicznych — na podstawie charakterystyk Herseya–Stribecka, substancji smarujących takich jak syntetyczna baza olejowa PAO-6, olej perfluoropolieterowy PFPE...

Research of speed and accuracy of speech synthesis methods for the purposes of deployment in embedded device

In the article, the methods concerning speech synthesis were analysed, having in mind their usage in an embedded device. Research was carried out on the devices with mixed parameters, and the criteria were accuracy and s...

Fatigue life of materials claded with the method of explosive welding

This paper describes the comparison of fatigue life on bending of two claded materials with the base material. The base material was the carbon steel P3335NH of the thickness of 10 mm and the clad materials was austeniti...

Download PDF file
  • EP ID EP258931
  • DOI 10.5604/01.3001.0010.8332
  • Views 49
  • Downloads 0

How To Cite

Vitalij Nichoga, Liubomyr Vashchyshyn (2017). Application of the wavelet and neural technologies for processing of signals obtained during railway tracks diagnostics by the magnetic flux leakage method. Bulletin of the Military University of Technology, 66(4), 195-201. https://europub.co.uk/articles/-A-258931