Ideal lowpass filter technique to enhance harmonic spectrum estimation for power electronics circuits

This paper presents a new and accurate method of harmonic analysis that permits to obtain high precision spectrum over wide range of bandwidth in order to mitigate power quality as well as EMI related problems. The proposed method is based on the estimation of intermediate points between the initial samples given by the available data acquisition system; therefore, the Fourier coefficients are estimated more precisely using the fast Fourier transform. As interpolation technique we chose digital to analog conversion with ideal lowpass filter for the reconstruction of the analog signal. The proposed method is tested on experimental signals and validated by adding known harmonic components to the measured signal. The obtained results are compared to traditional FFT as well as spline interpolation, and linear interpolation techniques. The performance of the proposed method and its superiority over the known techniques is discussed and the possible sources of errors are identified