Reliable Detection of Stator Interturn Faults of Very Low Severity Level in Induction Motors
Konstantinos N. Gyftakis, António J. Marques Cardoso
Abstract
The interturn faults are one of the most (if not the most) challenging electrical machine failures to detect online and at incipient severity stages. Past works and experience have shown that this specific fault type will lead to a ground fault and consequently a catastrophic failure of the machine in a very small amount of time. Past works have proposed techniques and methodologies to confront this dangerous fault. However, a common feature in most past efforts is the high level of severity, limited by external resistors to avoid machine breakdowns. Scenarios like the above are not of much use in industry because they lead to the development of diagnostic techniques insensitive to the real fault severity levels that are noncatastrophic and that are by their nature very low. This is the motivation behind this article, which challenges the existing background in this field and offers a reliable solution, which may be adopted in industry. The article studies induction motors with incipient interturn fault severity with many well-known techniques. The experimental results prove many methods unreliable and insensitive to low level interturn fault detection. Finally, the authors propose a novel method that relies on the monitoring of the stray flux at three positions of the machine.