In this paper, Neural Network (NN) models are developed and tested to estimate the remaining useful life of electric motors, by means of the features extracted from experimental data derived from motor degradation data. The features were extracted from both time and frequency domain. Both static and dynamic neural networks, with a single-layer and using different combinations of features as inputs, are investigated. The best NN estimates are also compared to the prediction that can be obtained by using a prognostic procedure validated by the authors against the same experimental data.

Assessment of Neural Network Capability to Predict the Remaining Useful Life of Electric Motors

VENTURINI, Mauro
2015

Abstract

In this paper, Neural Network (NN) models are developed and tested to estimate the remaining useful life of electric motors, by means of the features extracted from experimental data derived from motor degradation data. The features were extracted from both time and frequency domain. Both static and dynamic neural networks, with a single-layer and using different combinations of features as inputs, are investigated. The best NN estimates are also compared to the prediction that can be obtained by using a prognostic procedure validated by the authors against the same experimental data.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11392/2327416
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact