Fault Detection of Power Electronic Circuit using Wavelet Analysis in Modelica

Abstract

In more electric aircrafts (MEA) the electric power network is important for the reliability. To prevent severe faults it is the key issue to identify the faults in the early stage before a complete failure happens. In this paper an early stage fault detection method using wavelet multi-resolution analysis (MRA) for a regulated buck DC-DC converter is studied. Specifically, the electrolyte input capacitor is diagnosed. The study was carried out using simulation with Modelica / Dymola. The fault features that were extracted from different levels of wavelet decomposition provided clear information for both fast and slow occurring faults. This method showed significant advantages compared with filter techniques. It is concluded that wavelet transform is a suitable tool for early stage fault detection of the power electronics in MEA. In addition, the simulation language Modelica provides a convenient possibility for the quick design of fault detection strategy.

Electrical protection solutions enabling integrated electrical power and composite structure systems for aircraft

“…Wavelet transform [28] LG Ungrounded Aerospace Wavelet multi-resolution analysis (MRA) [29] LG Solidly grounded Land-based grid DWT [30] LG IT Land-based grid S-Transform [31] HIF High resistance Shipboard Differential technique/STFT [32] LG High resistance location of the fault using over current and/or under voltage detection. However higher resistance rail to ground faults in TN-C-S grounded system result in a much lower fault current, which may be difficult to locate and detect.…”

Identification of the Capacitance Changes in DC-DC Converter Using Wavelets

“…Робота [3] IV. ПОБУДОВА МОДЕЛІ Основні електричні параметри моделі понижуючого DC-DC перетворювача вказані в табл.…”

Wavelet Library for Modelica

“…Proceedings of the 10 th International ModelicaConference March 10-12, 2014, Lund, Sweden ory will be given [1] [2]. The theoretical part of [9] is roughly repeated here but with improvements for better understanding of common readers.…”

“…In the paper of Ji [8] in 2010 about a Modelica signal analysis tool towards the design of more electric aircrafts the authors expressed the wish to do the analysis using wavelet transform directly within Modelica. The first application for the failure analysis in the power system using the prototype of the Modelica wavelet library was already introduced in 2012 [9]. After the successful development, the final Modelica wavelet library is introduced in this article.…”