Online condition monitoring and diagnosis of induction motor

Abstract

The induction motor is a working backbone in multiple industries and is widely used in practically almost all aspects of technological applications. To protect any people from hazardous situations, it is essential to make sure the induction motor performs safely and consistently in every system. One of the most frequent defects that can occur in an induction motor is a problem with the stator winding. To provide timely maintenance and condition monitoring, it would be helpful to install and use new technologies, such as artificial intelligence, to check for any premature flaws within the induction motor. This project's suggested method for identifying stator winding defects in induction motors is a non-intrusive Machine Learning approach. Using frequency, real and imaginary impedance magnitude data as my primary input criteria, I can identify the early stages of any stator winding defects. As a result, potential risks are removed, motor downtime is decreased, and maintenance expenses are also decreased.

The testing results of my Neural Network Model will reveal the dependability and accuracy of the proposed strategy.