Diagnostics of Traction Electric Motors of Electric Rolling Stock Using Artificial Neural Networks

The results of research in the field of diagnostics of a 1DT.003.11 traction electric motor of the ac electric train of the EP3D series by means of modeling of emergency modes of functioning of the kernel neural network are presented. Based on an analysis of the power pseudospectrum graph performed...

Full description

Saved in:
Bibliographic Details
Published in:Russian electrical engineering 2022, Vol.93 (9), p.576-583
Main Authors: Kosmodamiansky, A. S., Inkov, Yu. M., Menshchikov, I. A., Batashov, S. I.
Format: Article
Language:English
Subjects:
Artificial neural networks
Clustering
Diagnostic systems
Electric motors
Engineering
Kernels
Machines
Manufacturing
Matlab
Neural networks
Processes
Rolling stock
Traction
Waveforms
Citations: Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The results of research in the field of diagnostics of a 1DT.003.11 traction electric motor of the ac electric train of the EP3D series by means of modeling of emergency modes of functioning of the kernel neural network are presented. Based on an analysis of the power pseudospectrum graph performed using digital processing of the current curve waveform in the MATLAB program package, discrete values of the frequency components of the logarithmic current spectrum in the armature winding and their corresponding signal power values are obtained in the form of a table, serving for the failures clustering of the traction electric motor of an electric train. The reliability of traction motors with an increased mileage is proposed to be evaluated using a kernel neural network in the MATLAB package. The result of the study of the diagnostic process using clustering of traction motor failures is an assessment of the effectiveness of the practical use of an artificial neural network with a kernel structure in diagnostic tasks, depending on the technical condition of the traction motor at the current moment of operation.
ISSN:1068-3712
1934-8010
DOI:10.3103/S1068371222090097