EMI Behavioral Model Based CM Noise Prediction Method for DC Power System Considering Multi-Noise Coupling

Abstract

The DC power system with high-proportion power converters has become a trend. The system noise consists of multiple converter noises and is affected by the phase difference between converter noises. However, the phase difference varies with different startup intervals between converters due to the independent operation of converters. The unfixed multi-noise coupling leads to random variation of the system-noise amplitude in the case of multiple startups, which causes the system-noise prediction error. Besides, numerous converters make system EMI modeling more difficult. To address the concerns above, this paper puts forward a novel EMI behavioral model based common-mode (CM) system noise prediction method considering multi-noise coupling. First, the EMI behavioral model is developed for the DC power system, which greatly simplifies system EMI modeling. Since coupling of multiple converter noises forms the system noise, the effect of such multi-noise coupling on system noise prediction is investigated. Finally, the system model is further modified by processing the phase difference between converter noises. The improved CM system noise prediction method considering multi-noise coupling improves the prediction accuracy with an error of 6 dB or less. The effectiveness of this method is verified on an experimental platform of the DC power system with two inverters.