A real-time platform dedicated to on-line gear tooth surface damage fault detection in induction machines

Condition monitoring and fault diagnosis of electromechanical systems are considered as vital tasks in several industrial applications. Early indication of incipient faults is the principal objective of any ideal condition monitoring system which avoids catastrophic failures before their occurrence. On-line condition monitoring systems based on digital signal processors (DSP) and field programmable gate arrays (FPGA) helps to accomplish this goal. A non-invasive method based on the stator current space vector analysis has been recently proposed for a gear tooth surface damage fault detection. This paper aims to implement this last technique on a real-time platform which includes both a FPGA and a real-time processor. FPGA resources are used for the computation of stator current space vector while real-time processor resources are dedicated to the computation of a fault index based on the energy computation of fault-related frequencies in the stator current space vector instantaneous spectrum. The proposed algorithm needs only some basic data of the electromechanical system in addition to the mechanical system torsional natural frequency. The performances of the proposed algorithm are evaluated by using a set-up based on a 250W three-phase squirrel-cage induction machine shaft-connected to a single-stage spur gearbox.