Unintrusive monitoring of induction motors bearings via deep learning on stator currents

Francesca Cipollini, Luca Oneto, Andrea Coraddu, Stefano Savio, Davide Anguita

Research output: Contribution to journalConference article

Abstract

Induction motors are fundamental components of several modern automation system, and they are one of the central pivot of the developing e-mobility era. The most vulnerable parts of an induction motor are the bearings, the stator winding and the rotor bars. Consequently, monitoring and maintaining them during operations is vital. In this work, authors propose an Induction Motors bearings monitoring tool which leverages on stator currents signals processed with a Deep Learning architecture. Differently from the state-of-the-art approaches which exploit vibration signals, collected by easily damageable and intrusive vibration probes, the stator currents signals are already commonly available, or easily and unintrusively collectable. Moreover, instead of using now-classical data-driven models, authors exploit a Deep Learning architecture able to extract from the stator current signal a compact and expressive representation of the bearings state, ultimately providing a bearing fault detection system. In order to estimate the effectiveness of the proposal, authors collected a series of data from an inverter-fed motor mounting different artificially damaged bearings. Results show that the proposed approach provides a promising and effective yet simple bearing fault detection system.
LanguageEnglish
Pages42-51
Number of pages10
JournalProcedia Computer Science
Volume144
Early online date21 Nov 2018
DOIs
Publication statusE-pub ahead of print - 21 Nov 2018
Event3rd International Neural Network Society Conference on Big Data and Deep Learning, INNS BDDL 2018 - Sanur, Bali, Indonesia
Duration: 17 Apr 201819 Apr 2018

Fingerprint

Bearings (structural)
Induction motors
Stators
Monitoring
Fault detection
Rotors (windings)
Mountings
Deep learning
Automation

Keywords

  • bearings
  • deep learning
  • induction motors
  • monitoring
  • stator currents

Cite this

Cipollini, Francesca ; Oneto, Luca ; Coraddu, Andrea ; Savio, Stefano ; Anguita, Davide. / Unintrusive monitoring of induction motors bearings via deep learning on stator currents. In: Procedia Computer Science. 2018 ; Vol. 144. pp. 42-51.
@article{0111780f3db14540a165bc25221a2d81,
title = "Unintrusive monitoring of induction motors bearings via deep learning on stator currents",
abstract = "Induction motors are fundamental components of several modern automation system, and they are one of the central pivot of the developing e-mobility era. The most vulnerable parts of an induction motor are the bearings, the stator winding and the rotor bars. Consequently, monitoring and maintaining them during operations is vital. In this work, authors propose an Induction Motors bearings monitoring tool which leverages on stator currents signals processed with a Deep Learning architecture. Differently from the state-of-the-art approaches which exploit vibration signals, collected by easily damageable and intrusive vibration probes, the stator currents signals are already commonly available, or easily and unintrusively collectable. Moreover, instead of using now-classical data-driven models, authors exploit a Deep Learning architecture able to extract from the stator current signal a compact and expressive representation of the bearings state, ultimately providing a bearing fault detection system. In order to estimate the effectiveness of the proposal, authors collected a series of data from an inverter-fed motor mounting different artificially damaged bearings. Results show that the proposed approach provides a promising and effective yet simple bearing fault detection system.",
keywords = "bearings, deep learning, induction motors, monitoring, stator currents",
author = "Francesca Cipollini and Luca Oneto and Andrea Coraddu and Stefano Savio and Davide Anguita",
year = "2018",
month = "11",
day = "21",
doi = "10.1016/j.procs.2018.10.503",
language = "English",
volume = "144",
pages = "42--51",
journal = "Procedia Computer Science",
issn = "1877-0509",

}

Unintrusive monitoring of induction motors bearings via deep learning on stator currents. / Cipollini, Francesca; Oneto, Luca; Coraddu, Andrea; Savio, Stefano; Anguita, Davide.

In: Procedia Computer Science, Vol. 144, 21.11.2018, p. 42-51.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Unintrusive monitoring of induction motors bearings via deep learning on stator currents

AU - Cipollini, Francesca

AU - Oneto, Luca

AU - Coraddu, Andrea

AU - Savio, Stefano

AU - Anguita, Davide

PY - 2018/11/21

Y1 - 2018/11/21

N2 - Induction motors are fundamental components of several modern automation system, and they are one of the central pivot of the developing e-mobility era. The most vulnerable parts of an induction motor are the bearings, the stator winding and the rotor bars. Consequently, monitoring and maintaining them during operations is vital. In this work, authors propose an Induction Motors bearings monitoring tool which leverages on stator currents signals processed with a Deep Learning architecture. Differently from the state-of-the-art approaches which exploit vibration signals, collected by easily damageable and intrusive vibration probes, the stator currents signals are already commonly available, or easily and unintrusively collectable. Moreover, instead of using now-classical data-driven models, authors exploit a Deep Learning architecture able to extract from the stator current signal a compact and expressive representation of the bearings state, ultimately providing a bearing fault detection system. In order to estimate the effectiveness of the proposal, authors collected a series of data from an inverter-fed motor mounting different artificially damaged bearings. Results show that the proposed approach provides a promising and effective yet simple bearing fault detection system.

AB - Induction motors are fundamental components of several modern automation system, and they are one of the central pivot of the developing e-mobility era. The most vulnerable parts of an induction motor are the bearings, the stator winding and the rotor bars. Consequently, monitoring and maintaining them during operations is vital. In this work, authors propose an Induction Motors bearings monitoring tool which leverages on stator currents signals processed with a Deep Learning architecture. Differently from the state-of-the-art approaches which exploit vibration signals, collected by easily damageable and intrusive vibration probes, the stator currents signals are already commonly available, or easily and unintrusively collectable. Moreover, instead of using now-classical data-driven models, authors exploit a Deep Learning architecture able to extract from the stator current signal a compact and expressive representation of the bearings state, ultimately providing a bearing fault detection system. In order to estimate the effectiveness of the proposal, authors collected a series of data from an inverter-fed motor mounting different artificially damaged bearings. Results show that the proposed approach provides a promising and effective yet simple bearing fault detection system.

KW - bearings

KW - deep learning

KW - induction motors

KW - monitoring

KW - stator currents

UR - http://www.scopus.com/inward/record.url?scp=85061160245&partnerID=8YFLogxK

UR - http://innsbigdata2018.org/

U2 - 10.1016/j.procs.2018.10.503

DO - 10.1016/j.procs.2018.10.503

M3 - Conference article

VL - 144

SP - 42

EP - 51

JO - Procedia Computer Science

T2 - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -