FPGA based on-line fault diagnostic of induction motors using electrical signature analysis

Preventive maintenance is one of the main concerns in modern industry, in which early failure detection increases the lifecycle of machines. In this paper, electrical signature analysis is employed to indicate the development or existence of faults within the proposed system and this is achieved by embedding a real-time frequency analysis of the motor current. The term in the title electrical signature analysis basically refers to the motor current or voltage attributes are being used as a transducers to detect the changes in their spectrum in both the conditions; healthy and unhealthy. The algorithm used for analyzing the signals in frequency domain is done using Fast Fourier transform. In this work, we have focused on failure of bearing part of single phase induction motor and developed hardware for monitoring conditions (i.e. health of the motor) in run time. Because of the simplicity of this technique the mechanism of fault diagnosis is employed using an FPGA approach that offers re-configurability. This work can be very useful in industrial setup where there are 100 motors working together for some production lines. The findings show promising results which could lead to better reliability performance of the induction motor and lower maintenance costs.