A spectral method for on‐line computation of the harmonics of symmetrical components in induction machines

Abstract

SUMMARYThis paper describes a procedure for the on-line computation of the harmonics of symmetrical components for voltages, currents and impedances in working induction machines. The algorithm used to compute these components is based on the discrete Fourier transform (DFT) of complex sequences. An experimental test-bed with a 0.12 kW star-connected induction motor has been used to test the procedure. The stability of the numerical results is verified to show the performance of the proposed method.

The estimate of amplitude and phase of harmonics in power system using the extended kalman filter

“…Up to now, there have been several ways to estimate harmonics that are defined as unwanted components in an alternating waveform having distortion. For example, we can name the discrete Fourier transform methods [5], the mode estimation techniques [6], data exploration tools [7], independent component analysis [8], and neural networks [9]. Suresh Kumar et al [10] we used genotype algorithms, the minimum squared genetic algorithms, the optimization of the least squared hybrid particles an adaptive neural network in order to estimate harmonics in the power system.…”

On-line fault detection method for induction machines based on signal convolution

“…Motor Current Signature Analysis (MCSA) is one of the most widely used methods for fault detection analysis in induction machines [4]. MCSA focuses its efforts on the spectral analysis of the stator currents and has been successfully used to detect mechanical faults [5][6][7] (dynamic eccentricity, BRBs, and bearing damage) and electrical faults [8] (single or multiple-phase short circuits, open phases, and abnormal *Correspondence to: Antonio Garcia Espinosa, UPC, Electrical Engineering, Terrassa, Barcelona, Spain.…”