Neural network based enhancement of power quality in distribution system

Abstract

An application of artificial intelligence is presented in solving power quality problems using a distribution static compensator (DSTATCOM) in the distribution system involving lagging pf loads. A set of three nonlinear neurons is used to obtain the three phase compensating currents. Another set of three layered feed-forward neural network controls the compensating currents. The developed model works accurately under varying load conditions and provides good dynamic response to the step changes in the load currents. A real time performance is achieved using SIMULINKR/Sim-powersystem (SPS) toolboxes and simulated results adhere to the IEEE-519 standard for improvement of power quality.