Research Unity of Industrial Process Control (UCPI), National Engineering School of Sfax (ENIS), B.P.: W 3038
Sfax-Tunisia
Aroui.tarek@yahoo.fr, Yassine.koubaa@enis.rnu.tn, ahmad.tomi@enis.rnu.tn
Copyright © JES 2007 on-line : http://journal.esrgroups.org/jes/
T. Aroui
Y. Koubaa
A. Toumi
Regular paper
Application of Feedforward Neural
Network for Induction Machine Rotor
Faults Diagnostics using Stator Current
Faults and failures of induction machines can lead to excessive downtimes and generate large
losses in terms of maintenance and lost revenues. This motivates motor monitoring, incipient fault
detection and diagnosis. Non-invasive, inexpensive, and reliable fault detection techniques are
often preferred by many engineers. In this paper, a feedforward neural network based fault
detection system is developed for performing induction motors rotor faults detection and severity
evaluation using stator current. From the motor current spectrum analysis and the broken rotor bar
specific frequency components knowledge, the rotor fault signature is extracted and monitored by
neural network for fault detection and classification. The proposed methodology has been
experimentally tested on a 5.5Kw/3000rpm induction motor. The obtained results provide a
satisfactory level of accuracy.
Keywords Feedforward neural network, Diagnosis, induction motors, rotor faults, single
phase stator current
1. INTRODUCTION
Induction motors play a pivotal role in industry and there is a strong demand for their
reliable and safe operation, and the motor can be exposed to different hostile environments,
misoperations, and manufacturing defects. Internal motor faults ( e.g., short circuit of motor
leads, interturn short circuits, ground faults, bearing and gearbox failures, broken rotor bar
and cracked rotor end-rings), as well as external motor faults (e.g., phase failure,
asymmetry of main supply and mechanical overload), are expected to happen sooner or
later[1]. Furthermore, the wide variety of environments and conditions that the motors are
exposed to can age the motor and make it subject to incipient faults. These incipient faults,
or gradual deterioration, can lead to motor failure if left undetected.
The major faults of electrical machines can broadly be classified according to the main
components of a machine: stator related faults, rotor related faults, bearing related faults
and other faults [2].
Early fault detection allows preventative maintenance to be scheduled for machines
during scheduled downtime and prevents an extended period of downtime caused by
extensive motor failure, improving the overall availability of the motor driven system. With
proper system monitoring and fault detection schemes, the costs of maintaining the motors
can be greatly reduced, while the availability of these machines can be significantly
improved. Many engineers and researchers have focused their attention on incipient fault
detection and preventive maintenance in recent years. There are invasive and noninvasive
methods for machine fault detection, they can be described as [3,4]: