The harmonics detection method based on neural network applied to harmonics compensation

Abstract

Several different methods have been used to sense load currents and extract its harmonic component in order to produce a reference current in shunt active power filters (SAPF), and to contribute significantly in determining the SAPF performance. Consequently, many structures based on artificial neural network (ANN) have been developed in the literature, The most significant have used at least two Adaline networks for introducing a controller which serves to keep the terminal capacitor voltage stable. This paper presents a novel method and design that utilises one Adaline network. The latter combines both the strategies for extracting the reference currents and controlling DC link voltage which ensure suitable transit of powers to supply the inverter. To investigate the performance of this identification method, the study has been accomplished using simulation with MATLAB Simulink Power System Toolbox. The simulation study results of this novel technique compared to other similar methods are found quite satisfactory by assuring good filtering characteristics and high system stability.

Keywords: Artificial Neural Networks (ANN), p-q theory, (SAPF), Harmonics, Total Harmonic Distortion.