Model-based diagnosis of induction motor failure modes

Induction motor failure modes can be instantaneous or progressive. This paper proposes a model-based methodology employed to attempt to identify the root causes of the main fault modes of progressive failures. Utilising indicators from the monitoring of three-phase motor currents, this paper explains how leading-edge and traditional theories could be combined and optimised to provide an integrated software algorithm set that could accurately predict the root causes of the five main failure modes of standard cage type Induction Motors. Utilising mathematical modelling and simulation of each specific fault mode, comparisons can be made between the known symptoms of specific faults and Induction Motor signals obtained from devices operating at full load in the field. Comparison of variances between model states and the state of the operating machine could enable the specific root cause diagnosis to be made.