Artificial Neural Network Controlled Shunt Active Power Filter

This paper presents neural based proportional integral (PI) control applicable for active power filters for single-phase system, which are comprised of multiple nonlinear loads. The system consists of an uncontrolled rectifier and AC controller as the non-linear loads, with an active filter to compensate for the harmonic current injected by the load. The active filter is based on a single-phase inverter with four controllable switches, a standard H-bridge inverter. The AC side of the inverter is connected in parallel with the other nonlinear loads through a filter inductance. The DC side of the inverter is connected to a filter capacitor. The neural PI controller is used to shape the current through the filter inductor such that the line current is in phase with and of the same shape as the input voltage. The spectral analysis of the supply current shows the harmonics produced by the load has been successfully compensated by the active filter. The system is modeled in Matlab Simulink and simulation results prove that the injected harmonics are greatly reduced and system efficiency and power factor are improved