Open Circuit Stator Winding Fault Detection of Induction Machines from Transient Data Using Nature Inspired Optimization Algorithms
| Autor: | Andrew Smith, Salaheddine Ethni, Hamza Khalfalla, Muez Shiref |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Computer science
Stator 020208 electrical & electronic engineering Particle swarm optimization 02 engineering and technology Fault (power engineering) Fault detection and isolation Wound rotor motor law.invention law Control theory 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Time domain Sensitivity (control systems) Transient (oscillation) |
| Zdroj: | 2018 53rd International Universities Power Engineering Conference (UPEC). |
| DOI: | 10.1109/upec.2018.8541854 |
| Popis: | This paper investigates the performance of two Nature Inspired Optimization Algorithms (NIOA): Bacterial Foraging Optimization (BFO) and Particle Swarm Optimization (PSO), which are used for early fault detection on Induction Machine (IM) stator windings, to prevent sudden, catastrophic, breakdowns. An open-circuit stator winding fault is experimentally studied. This scheme uses time domain measurements obtained during transients to validate the capability of this technique, and in conjunction with the NIOA, estimates the parameters of the IM mathematical model, detects stator winding faults, and gives information about its type and location. Only stator voltages, currents, and rotor speed are evaluated using experimental data obtained from a wound rotor three-phase IM. The validity and effectiveness of the proposed method using the transient data is verified, showing its accuracy, prediction capability, and sensitivity without the need of prior knowledge of various fault signatures. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|