Keywords: active power filters, artificial neural networks, ANNs, current source inverters, CSI, harmonics, unified power quality conditioner, UPQC, power electronics, simulation, ANN compensators
Performance evaluation of CSI-based unified power quality conditioner using artificial neural network
In recent years unified power quality conditioner (UPQC) is being used as a universal active power conditioning device to mitigate both current as well as voltage harmonics at a distribution end of power system network. The performance of UPQC mainly depends upon how quickly and accurately compensation signals are derived. The artificial neural network (ANN) trained with conventional compensator data, can deliver compensation signals more accurately and quickly than conventional compensator at varied load condition. This paper presents performance verification of CSI-based UPQC using artificial neural network. The ANN-based compensation system eliminates voltage as well as current harmonics with good dynamic response. Extensive simulation results using Matlab/Simulink for RL load connected through an uncontrolled bridge rectifier validates the performance of ANN compensator.