Fault detection in induction motors powered by inverters operating in nonstationary regimes remains a challenge.
The trajectory in the time–frequency plane of harmonics related
to broken rotor bar develops very in proximity to the path described by the fundamental component. In addition, their energy is much lower than the amplitude of the first harmonic. These two
characteristics make it challenging to observe them. The Dragon
Transform (DT), here presented, is developed to overcome the
described problem. In this article, the DT is assessed with nonlinear
inverter-fed startups, where its high time and frequency resolutions
facilitate the monitoring of fault harmonics even with highly adjacent trajectories to the first harmonic path.
