رسائل دكتوراة
English
2018
Mathematical Modeling and Detection of Stator Abnormalities in Line Start Permanent Magnet Synchronous Motors
Mohammad Abido, Zakariya Al-Hamouz, لقمان سليمان فايز مراعبة
كلية الهندسة-جامعة الملك فهد للبترول والمعادن · السعودية
In industrial applications, especially in oil and gas plants, induction motors are being gradually replaced by high efficiency permanent magnet motors, called line start permanent magnet synchronous motors (LSPMSMs). LSPMSMs have significant advantages over induction motors, which are: higher torque and power density as well as higher efficiency and operational power factor. As the number of LSPMSMs used in different fields is increasing, the maintenance scheme for these motors becomes important. In practical applications, LSPMSM can experience different types of failures such as broken bars, eccentricity, demagnetization as well as stator windings abnormalities. From the literature, it has been reported that stator winding faults in other machines represent around 36% of motors faults. Such failures threaten the normal manufacturing, interrupt the normal operation, and hence result in a significant loss of revenue. An efficient fault detection technique can reduce the maintenance expenses by preventing the high cost failures and unplanned downtimes. Despite the importance of motors’ fault detection, at the moment there is no scientific manuscript on modeling and diagnosis tool of LSPMSMs under stator winding abnormalities. Accordingly, in this dissertation, novel mathematical and finite element-based models for LSPMSMs under stator winding abnormalities (asymmetrical stator windings and stator inter-turn fault) are developed and implemented. In addition, an experimental setup has been built for validating the developed models. The testing results of the mathematical models are in good agreement with the finite element models and experimental findings. The mathematical models are used to extract representative fault current features which are used in the development of a diagnosing tool for detecting the type and severity of stator winding abnormalities. Testing of the developed diagnosis tool shows a high accuracy of 96% in detecting the type and severity of stator winding abnormalities. Finally, the robustness of the proposed diagnostic tool against motor parameter variations has been investigated. It has been found that the abnormality detection accuracy decreases by no more than 3%.