Abstract :
As a kernel technique of active power filter (APF), harmonic detection and its tracking method are investigated in this paper. Firstly, according to APF discussed here, an adaptive notch filter is presented and the LMS algorithm is examined to meet the crucial requirements of the filter´s tracking behavior under time-varying condition. Secondly, an upgraded method is emphasized on the basis of the traditional LMS algorithm, the variable step-size method is analyzed with both its merits and demerits, in order to guarantee the convergence rate and steady-state performance, a correct-mumax mending method is introduced as the major solution to improve the performance of variable step-size, so as to obtain good tracking reliability as well as rapidity. For better understanding, the simulation results are given out and illustrated, with the friendly toolbox of Matlab
Keywords :
active filters; adaptive filters; harmonic analysis; least mean squares methods; notch filters; power harmonic filters; power supply quality; power system harmonics; power system reliability; tracking filters; LMS algorithm; Matlab simulation; active power filter; adaptive filtering; adaptive notch filter; convergence rate; correct-mumax mending method; filter tracking method; harmonic analysis; harmonic detection; kernel technique; power quality; steady-state performance; time-varying condition; tracking reliability; variable step-size method; Active filters; Adaptive filters; Algorithm design and analysis; Harmonic analysis; Kernel; Least squares approximation; Performance analysis; Power harmonic filters; Power system harmonics; Steady-state; Adaptive filtering; LMS algorithm; harmonics; variable step-size;
