A Comparison of Spectrum Estimation Techniques for Nonstationary Signals in Induction Motor Drive Measurements

This paper compares different spectrum estimation techniques applied to nonstationary signals in order to determine which is the most suitable in analyzing the spectral harmonic content of signals in a very close frequency band. The comparison examines a specific application in which spectral analysis is applied to nonstationary signals, such as currents or voltages in induction motor drives. A test signal and some performance parameters are defined. Experimental tests were carried out by applying different spectral analysis algorithms in order to compare their behavior in detecting defined harmonic frequencies. The results show that the chirp-Z transform outperforms other techniques especially when a restricted frequency band has to be analyzed.