Institutional Repository
Technical University of Crete
EN  |  EL

Search

Browse

My Space

Eccentricity impact and detection in permanent magnet generators for renewables

Karampas Stefanos

Full record


URI: http://purl.tuc.gr/dl/dias/C2E72D26-25DA-4FB8-A79F-F70E6FDC2F75
Year 2025
Type of Item Diploma Work
License
Details
Bibliographic Citation Stefanos Karampas, "Eccentricity impact and detection in permanent magnet generators for renewables", Diploma Work, School of Electrical and Computer Engineering, Technical University of Crete, Chania, Greece, 2025 https://doi.org/10.26233/heallink.tuc.102388
Appears in Collections

Summary

This thesis explores the diagnosis of static,dynamic and mixed eccentricity faults in permanent magnet synchronous generators (PMSGs). For the purposes of this thesis, the C-GEN topology, a double rotor, slotless, air-cored Permanent Magnet Synchronous Generator, was employed. Traditional diagnostic methods, such as Motor Current Signature Analysis (MCSA), can lead to false negative results due to the double rotor configuration, air-cored structure and circuit design, which can suppress fault signatures and reduce diagnostic accuracy. The performance and reliability of the C-GEN machine depends on the early identification of eccentricity faults. The machine coils are subjected to electrical and thermal stress as a result of circulating currents in the circuit’s parallel coils caused by the eccentricity fault. This stress can degrade insulation and aggravate faults, leading potentially to secondary failures such as winding short circuits or rotor demagnetization. A Finite Element Method (FEM) Software (Simcenter MAGNET) is used in this study to examine the generator’s behavior, under varying load and fault conditions. The simulated data was post-processed in MATLAB to diagnose each fault condition and analyze the behavior associated with eccentricity faults. The study evaluates multiple diagnostic techniques, such as MCSA, Torque Monitoring, Flux Monitoring, Magnetically Coupled Search Coils, and ZeroSequence Flux (ZSF). ZSF emerged as the most reliable method, offering robust fault identification and high sensitivity.

Available Files

Services

Statistics