Artificial neural network based controller for active power line conditioner

Proliferation of non-linear loads resulted in the problems of harmonics, unbalancing, poor power factor. To overcome this active power line conditioner (APLC) is used. In this paper, an artificial neural network (ANN) based controller for active power line conditioner is presented, in which feedforward neural network, trained by levenberg marquardt (LM) algorithm is implemented for generating reference current and pulses to inverter. The simulation results by using MATLAB/SIMULINK of both the conventional and ANN based controller model are discussed. The neutral current has been reduced to maintain system balance.