Adjustable step-size LMS based IRP method for active power filter

摘要

in this paper, an improved instantaneous reactive power (IRP) method has been proposed for computation of the reference current of the active power filter. The adjustable step-size least mean square (LMS) algorithm is integrated with the IRP method to improve its transient behaviour of the APF for compensating the harmonics currents. The proposed scheme removes the requirement of the low pass filter for extracting of the harmonics power. Different algorithms are executed on the active power filter to carry out its comparative simulation study. The IRP method is based on transformation of the signals and subtraction of active power from the low pass filtered active power and consequent computation of the reference current. On the other hand, the second scheme is based on adjustable LMS based IRP method. This is neural network approach, where, harmonics component of the current is estimated online for the active power filter. The MATLAB/Simulink power system block set is used for simulation study of overall active power system.