Power system harmonics estimation using different signal processing techniques

Harmonics have been available for quite a while and its presence shapes the execution of a power system. Consequently, harmonics estimation is of principal vitality while analyzing the power system. Emulating the beginning of harmonics, different filters have been formulated to attain an ideal control methodology for constant rejection. This thesis acquaints different algorithms to dissect harmonics in the power system. The target is to estimate the voltage magnitude and phase plot of the power system in the proximity of noise by using various estimation approaches. This thesis has centered the consideration towards the investigation of Kalman filter (KF), Recursive Least squares (RLS), least mean square (LMS) and Variable leaky least mean square (VLLMS) based filter for estimation of harmonics. For a test signal KF, RLS, LMS and VLLMS based calculations have been examined and the results have been looked at. The several algorithms are compared for various signals to noise ratio. The SNR used here are 40 dB, 30dB and 20dB. The proposed estimation methodologies are implemented on a typical power system signal acquired from mechanical burden embodying power electronic converters and arc furnaces