Voltage–current locus‐based stator winding inter‐turn fault detection in induction motors

• Published in: International journal of circuit theory and applications Vol. 51; no. 6; pp. 2889 - 2911
• Main Authors: Dongare, Ujwala V., Umre, Bhimrao S., Ballal, Makarand S.
• Format: Journal Article
• Language: English
• Published: Bognor Regis Wiley Subscription Services, Inc 01.06.2023
• Subjects: Algorithms; Coils (windings); Condition monitoring; Critical components; Electric potential; Fault detection; fault severity; fault‐characterizing parameters; impedance correction coefficients; induction motor; Induction motors; Inspection; Insulation; radius of the locus; Short circuits; stator inter‐turn faults; Stators; Voltage; voltage correction coefficients; v–i locus; Winding
• ISSN: 0098-9886; 1097-007X
• DOI: 10.1002/cta.3561

Summary Monitoring the condition of induction motors requires continuous inspection of all motor components. The three‐phase stator windings forming the critical components of the motors and handling heavy currents as high as twice the blocked rotor current under inter‐turn short circuits necessitate a fault detection system that is efficient, appropriate, and worthwhile. The proposed algorithm takes measurements of the three‐phase supply voltages and phase currents and plots their loci for each cycle under various fault levels. The proposed simple, nonintrusive, less expensive, and fault‐sensitive voltage–current loci (£(v–i))‐based technique accurately identifies the faulty phase and detects stator inter‐turn faults (SITFs). In addition, the algorithm corrects supply voltage imbalances, winding asymmetries, and calibration errors of sensing elements. It contributes to examining the insulation condition of the stator windings at the onset and identifying the faulty phase of a motor. The experimental results for a 1 hp, 415 V squirrel‐cage induction motor (SCIM) in a laboratory investigation validate the competence of the proposed technique under various fault levels and load conditions. The proposed online condition monitoring method provides three‐phase stator voltage–current loci (£(v(a,b,c)–i(a,b,c)))‐based stator inter‐turn fault diagnosis and faulty phase discrimination in induction motors at their incipient stage. It applies the appropriate correction coefficients to normalize the three‐phase currents on a digitally unified platform to plot the loci. In the event that the loci do not coincide with the signature loci, the fault is detected and the phase with the greatest deviation in radius of the locus, ρ, is identified as the faulty phase. In industrial applications, the method promotes condition‐based predictive maintenance that avoids reactive maintenance.