Condition Monitoring of Electric Motors Based on Multiquantity Analysis Under Transient Regimes

During the last decade, many works have proven the important benefits of the analysis of different quantities under transient regimes for enhancing the diagnostic of the condition of electric motors. For instance, the analysis of starting currents and stray fluxes has proven to be very effective to avoid false indications obtained with classical tools, as well as to discriminate among different types of failures. The optimization of signal processing tools and image recognition algorithms that were too computationally expensive years ago has facilitated the proliferation of transient analysis techniques in the condition monitoring area. Moreover, the use of modern artificial intelligence (AI) methods has enabled to start implementing them in autonomous condition monitoring systems. Despite initially applied to induction motors and specific transient regimes and quantities, the use of these methodologies is being expanded to other machine typologies, other transients and even other quantities. This work compiles the recent advances in this area and proposes the extrapolation of startup transient analysis to other quantities such as vibrations. It is shown that the simultaneous analysis of different quantities under transient regimes enables to enhance the diagnostic and even to cover a wider range of faults.