Wyniki wyszukiwania - BazTech

1

Classifying transformer winding fault type, location and extent using FRA based on support vector machine

Ezziane Hassane, Houassine Hamza, Moulahoum Samir, Chaouche Moustafa Sahnoune

Przegląd Elektrotechniczny

|

2022

|

R. 98, nr 1

23--33

EN

In this paper, four common winding faults in power transformers (axial displacement (AD), serial capacitance variation (VSC), ground capacitance variation (VGC), open circuit (OC)) are simulated on a transformer winding model to classify the fault type, location and extent, by applying an intelligent methodology for diagnosing transformer faults, depends on building a comprehensive database by collecting Frequency Responses Analysis (FRA) related to health and faulty conditions and analyzing them using statistical and mathematical indicators, this base that can inventory all possible faults in terms of location and extent, which is used to train a support vector machine (SVM) classifier on the faults included in it, which is then able to classify any new data . The results of the tests showed that the proposed method is characterized by high accuracy in detecting the type of defect, determining its location and the extent of its occurrence, It also contributes to the development of the application of machine learning on transformers.

PL

W tym artykule symulowane są cztery typowe uszkodzenia uzwojeń w transformatorach mocy (przemieszczenie osiowe (AD), szeregowa zmiana pojemności (VSC), zmiana pojemności uziemienia (VGC), obwód otwarty (OC)) na modelu uzwojenia transformatora w celu sklasyfikowania typu zwarcia , lokalizacji i zasięgu, poprzez zastosowanie inteligentnej metodologii diagnozowania uszkodzeń transformatorów, polega na zbudowaniu kompleksowej bazy danych poprzez zbieranie Analizy Odpowiedzi Częstotliwości (FRA) związanej ze stanami zdrowia i wadliwymi oraz analizowanie ich za pomocą wskaźników statystycznych i matematycznych, tej bazy, która może inwentaryzować wszystkie możliwych błędów pod względem lokalizacji i zasięgu, który jest używany do trenowania klasyfikatora maszyny wektora nośnego (SVM) na zawartych w nim błędach, który jest następnie w stanie sklasyfikować dowolne nowe dane. Wyniki badań wykazały, że proponowana metoda charakteryzuje się dużą dokładnością w wykrywaniu rodzaju defektu, określaniu jego lokalizacji oraz zasięgu jej występowania, przyczynia się również do rozwoju zastosowania uczenia maszynowego na transformatorach.

2

Introduction to the Special Section on Intelligent Fault Monitoring and Fault–Tolerant Control in Power Electronics, Drives and Renewable Energy Systems

Malinowski Mariusz, Levi Emil, Orlowska-Kowalska Teresa

Power Electronics and Drives

|

2019

|

Vol. 4 (39)

163--165

EN

This article constitutes an introductory part of the special section on Intelligent Fault Monitoring and Fault-Tolerant Control in Power Electronics, Drives and Renewable Energy Systems. In the current issue of the journal, the first part of this section is published. Accepted articles are focussed mainly on the sensor-fault diagnosis methods for T-type inverter-fed dual- three phase PMSM drives, partial demagnetization, faults of the permanent magnet synchronous generator (PMSG) and online open phase fault detection (FD) in the sensorless five-phase induction motor drive implemented with an inverter output LC filter and third harmonic injection. Also, neural networks (NN) application in the detection of stator and rotor electrical faults of induction motors has been proposed in one of the papers, and the observer-based FD concept for unknown systems using input–output measurements was applied to a brushless direct current motor drive with unknown parameters.

3

Leakage inductance characteristics of power transformer winding fault based on ANSOFT

Wang Na, Xu Ke, Yang Jie, Hu Junzhe

Journal of Power Technologies

|

2019

|

Vol. 99, nr 3

231--238

EN

The development momentum of the power industry is pushed by the increasing demands and continuous progress in all sections of the economy. The improvement of the voltage grade of transmission lines is imminent, and the capacity of a single transformer continuously increases, thereby resulting in the serious problem of magnetic leakage, which that can aggravate the stray loss of transformers and the degree of winding deformation. The simulation model of leakage inductance under the different fault states of transformer winding was proposed on the basis of finite element method to reduce the threat to the safe and stable operation of power grids caused by the winding insulation damage from magnetic leakage. ANSOFT software was used to establish the physical model of 800 kVA and 10 kV true distribution transformers, and the leakage inductance values of transformer winding with a three-phase grounding short circuit, a turn-to-turn fault, and an inter-turn fault were analyzed. Finally, the accuracy of the model was verified through the finite element simulation. Results indicate that the leakage inductance of transformer increases by 30% in a three-phase short circuit, the leakage inductance is negatively correlated with the number of fault turns in a turn-to-turn fault, and the leakage inductance has a nonlinear positive correlation with the number of fault turns in an inter-turn fault. The proposed method provides references for the detection and evaluation of transformer operation fault.