Adaptive fundamental component detection approach to power harmonic compensation based on the RLS algorithm

The power harmonic control and compensation in active power filters are sensitive to the phase of reference template and a small displacement of the phase may cause large compensation deviations. Considering the nature of local periodicity of current waveform in power system, the recursive least squares (RLS) scheme used to extract the fundamental harmonic on time is proposed and studied. The performances of the fundamental harmonic estimation are discussed in the presence of time-varying harmonic within a transitional band, and the convergence analysis result of the RLS scheme is given also. Experimental results show that both the RLS and Kalman schemes can trace time-varying harmonic on time within a transitional band with quicker response than FFT method, and the RLS scheme tracks fundamental signal more steadily and with less computational burden than the Kalman filter. The harmonic compensation results using the FFT method are worse than the results associated with RLS scheme due to its harmonic phase detection delay.