Detection and Identification of Power Device Failures Using Discrete Fourier Transform for Fault-Tolerant Operation of Flying Capacitor Multilevel Converters

Autor:	Xin Yin, Daming Wang, Sai Tang, Z. John Shen, Jun Wang, Chao Zhang, Zhikang Shuai
Rok vydání:	2022
Předmět:	Computer science Harmonics Harmonic Electronic engineering Energy Engineering and Power Technology Fault tolerance Node (circuits) Hardware_PERFORMANCEANDRELIABILITY Electrical and Electronic Engineering Converters Fault (power engineering) Discrete Fourier transform Power (physics)
Zdroj:	IEEE Journal of Emerging and Selected Topics in Power Electronics. 10:5081-5091
ISSN:	2168-6785 2168-6777
Popis:	For a flying capacitor multilevel converter (FCMC), prompt detection of power switch failures is crucial for fault-tolerant operation. This paper presents a new technology for identifying and locating faulty cells in FCMC. Mathematical derivation points out that different fault types and locations will present different high-frequency harmonics. The new technology extracts high-frequency harmonics in the switch node voltage and look up a pre-established table to identify the fault type and location. The discrete Fourier transform (DFT) can be easily performed to analyze the harmonic using the built-in IP core in Field Programmable Gate Array (FPGA). For a FCMC with any number of levels, the new approach requires only one voltage sensor, and can accurately locate the faulty cell after one carrier cycle when the fault occurs. Both simulation and experimental results validate the proposed concept on a 5-level 100 kHz GaN FET based FCMC prototype. Furthermore, we have experimentally demonstrated fault-tolerant operation of the FCMC after a switch short and open fault is detected and identified by the new technique.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::d53a2b7b8675d0d4bd3e2c9df115473a https://doi.org/10.1109/jestpe.2021.3101849 Zobrazit plný text záznamu