Wavelet Transform-Based EMI Noise Mitigation in Power Converter Topologies

This paper proposes a novel wavelet transform-based approach for electromagnetic interference (EMI) noise mitigation in power electronics topologies. The switching action of power semiconductor devices generates EMI in power converter modules which exceeds the noise limit set by the regulation of CISPR11. In order to cancel the EMI noise, the wavelet transform-based time-frequency localization approach has been used for generating a noise cancellation signal. The effectiveness of the proposed method has been evaluated in the terms of signal-to-noise ratio (SNR). Further, the effect of variance of the noise amplitude has been taken into consideration on the SNR. A relationship between the standard deviation, which is proportional to square root of the variance of the signal, and the SNR has been established. The proposed method has been implemented on the different power electronics topologies such as buck converter, boost converter, three-phase voltage source inverter, and Z-source inverter-based ac drive. The proposed method achieves the reduction of noise level below the safe limit set by CISPR11.