Electric traction motors in automotive applications work in operational conditions characterized by variable load, rotational speed and other external conditions: this complicates the task of diagnosing bearing defects. The objective of the present work is the development of a diagnostic system for detecting the onset of degradation, isolating the degrading bearing, classifying the type of defect. The developed diagnostic system is based on an hierarchical structure of K-Nearest Neighbours classifiers. The selection of the features from the measured vibrational signals to be used in input by the bearing diagnostic system is done by a wrapper approach based on a Multi-Objective (MO) optimization that integrates a Binary Differential Evolution (BDE) algorithm with the K-Nearest Neighbor (KNN) classifiers. The developed approach is applied to an experimental dataset. The satisfactory diagnostic performances obtain show the capability of the method, independently from the bearings operational conditions.

Hierarchical k-nearest neighbours classification and binary differential evolution for fault diagnostics of automotive bearings operating under variable conditions

BARALDI, PIERO;CANNARILE, FRANCESCO;DI MAIO, FRANCESCO;ZIO, ENRICO
2016-01-01

Abstract

Electric traction motors in automotive applications work in operational conditions characterized by variable load, rotational speed and other external conditions: this complicates the task of diagnosing bearing defects. The objective of the present work is the development of a diagnostic system for detecting the onset of degradation, isolating the degrading bearing, classifying the type of defect. The developed diagnostic system is based on an hierarchical structure of K-Nearest Neighbours classifiers. The selection of the features from the measured vibrational signals to be used in input by the bearing diagnostic system is done by a wrapper approach based on a Multi-Objective (MO) optimization that integrates a Binary Differential Evolution (BDE) algorithm with the K-Nearest Neighbor (KNN) classifiers. The developed approach is applied to an experimental dataset. The satisfactory diagnostic performances obtain show the capability of the method, independently from the bearings operational conditions.
2016
Bearing diagnostics; Differential Evolution (DE); Feature selection; K-Nearest Neighbours (KNN) classifier; Multi-Objective (MO) optimization; Wavelet Packet Transform (WPD); Wrapper approach; Control and Systems Engineering; Artificial Intelligence; Electrical and Electronic Engineering
File in questo prodotto:
File Dimensione Formato  
submitted_bearing.pdf

accesso aperto

: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 943.36 kB
Formato Adobe PDF
943.36 kB Adobe PDF Visualizza/Apri

I documenti in IRIS 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/11311/1020532
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 102
  • ???jsp.display-item.citation.isi??? 79
social impact