Detecting insulation defects in electrical machines’ multi-turn windings based on analysis of transient electromagnetic processes

This study investigates transient processes occurring in a squirrel-cage induction motor with a stator’s multi-turn winding. These processes are used to devise efficient and simple diagnostic methods for assessing the condition of the least reliable element in an electrical machine – the winding insulation. This paper solves the task related to enhancing the operational reliability of electrical machines with multi-turn windings. The proposed high-frequency model of transient processes in multi-turn windings of electrical machines aims to non-destructively diagnose the insulation. This model enables analysis of high-frequency and impulse phenomena with an equivalent frequency of up to tens of MHz, which may arise during the operation of the machine because of switching and atmospheric overvoltage, as well as when powering electric motors with inverters. Underlying the model is the frequency and time domain analyses under above-mentioned influences using a low-power multi-turn induction motor as an example. To construct the model, a multi-segment equivalent circuit was used, represented as a series of connected four-poles with inductive-resistive-capacitive parameters corresponding to sections of the winding phase with conditionally uniform concentrated parameters. The constructed model and the proposed method make it possible to achieve the necessary test overvoltage to detect hidden and undeveloped defects at minimal sparing energy impact, without causing its irreversible destruction. Taking into account the distributed parameters of the winding sections combined with the impulse nature of the excitation ensure non-destructive testing (or non-destructive evaluation) of multi-turn-wound AC machine windings compared to classical testing methods. The proposed method makes it possible to obtain 2–2.5 times higher longitudinal test gradients with a comparable energetic impact at the level of 0.1 J. The results will contribute to improving the reliability of electrical machine operation in industrial settings by integrating the devised method into the system of planned preventive maintenance of electrical equipment.