Wavelet-Based Performance Evaluation of Power Converters Operating With Modulated Switching Frequency

It has been demonstrated that modulating the switching frequency of a power converter is a valuable way for reducing the electromagnetic interference (EMI) due to the switching process. Since we are considering a signal whose frequency content varies with time, wavelets are well suited to analyze the performance of such techniques. In this paper, we evaluate the performance of spread spectrum frequency modulation (SSFM) applied to the EMI reduction of a real power converter that uses periodic pattern switching frequency modulation. The performance of the converter under investigation includes the analysis of the switching voltage spectrum (as the main source of EMI) and the output voltage ripple. This evaluation is performed with two coefficients, i.e., maximum energy ratio (MER) and energy dispersion ratio (EDR), which are figures of merit defined in this paper using time-dependent energy density distribution in frequency, obtained from the scalograms of the analyzed signals. Such figures of merit allow comparison in the time-frequency domain of different modulation techniques and the choice of the best solution for each case in terms of reduction of the peak of noise spectrum.