Monitoring of Induction Machines by Maximum Covariance Method for Frequency Tracking

Several techniques were presented in the literature in order to perform accurate tracking of frequencies for different purposes. Most of them are modified versions of the traditional discrete Fourier transform (DFT) or other spectrum-estimation techniques. This paper presents a procedure based on the statistical analysis of the current signal in the time domain, referred to as maximum covariance method for frequency tracking (MCMFT), which allows to obtain high-frequency resolution independent of the sampling frequency and of the time acquisition period. Owing to the proposed procedure, the spectrum lines related to supply frequency or to rotor-slotting frequency for an induction machine can be detected with extreme accuracy within a wide range of sampled data conditions. Then, an accurate slip computation, or speed estimation, for sensorless control or distribution network diagnosis can be performed. Comparisons between previously existing methods and the proposed one are reported, in order to critically analyze its performances. Two different-sized induction machines were used for the experiments.