Co-Reduction of Common Mode Noise and Loop Current of Three-Level Active Neutral Point Clamped Inverters

Journal Article (2021)
Author(s)

Jianing Wang (Hefei University of Technology)

Xiaohui Liu (Hefei University of Technology)

Qiang Peng (Hefei University of Technology)

Yuanwu Xun (Hefei University of Technology)

Shaolin Yu (Hefei University of Technology)

Nan Jiang (Guangdong Academy of Sciences)

Wenbo Wang (University of Twente)

Fengze Hou (TU Delft - Electronic Components, Technology and Materials)

DOI related publication
https://doi.org/10.1109/JESTPE.2020.3043018 Final published version
More Info
expand_more
Publication Year
2021
Language
English
Issue number
1
Volume number
9
Article number
9285288
Pages (from-to)
1088-1103
Downloads counter
260

Abstract

The increased switching frequency and speed of silicon carbide (SiC) MOSFETs lead to higher power density of inverters, but meanwhile resulting in weak electromagnetic interference (EMI). The impedance balance technique is a good way to reduce the common mode (CM) noise by making the voltage across the line impedance stabilization network (LISN) as small as possible. However, a side effect of this technique is the generation of relatively large loop current that circulates in the inverter. It can cause additional losses and cost, which can be a factor that stops the increase of the switching frequency by SiC MOSFET. This article, for the first time, analyzes the relationship between the CM noise and loop current of the three-level active neutral point clamped (ANPC) inverter, and proposes a co-reduction method for both. First, the CM noise and loop current are clarified for the ANPC inverter, and the analytical models for both are established. The conflict between the CM noise and loop current is introduced with a specific case by the existing design method. Then a co-reduction method is proposed and elaborated, which can both suppress the CM noise and the loop current. The extra cost and volume by the proposed method are also analyzed and are negligible. The design guideline is further shown for clarity. Finally, the analysis and proposed method is validated by the experiment.