Static VAr compensator with neural network control

Artificial neural networks (ANNs) have been used in many applications of pattern classification, speech synthesis and recognition, function approximation, associative memory and control. Because of their adaptive nature and parallel computational features, they are promising a lot of hope for future of engineering. In this study, an application of ANN in control has been presented. A static VAr compensator model that has been used to provide reactive energy for a given load was controlled using two multi-layer feed-forward neural networks. ANN control method used in this study was more speedy than classic feedback control system. Because of their adaptivity and generalization features, ANNs provided power factor error less than 2% while inputs contained harmonics in this study. Results show that ANNs could be used to control power electronic circuits in static VAr compensators.