Employing Artificial Neural Networks for prediction of electrical arc furnace reactive power to improve compensator performance

The time varying nature of electric arc furnace (EAF) gives rise to voltage fluctuations which produce the effect known as flicker. The ability of static VAr compensator (SVC) is limited by delays in reactive power measurements and thyristor ignition. In order to improve the SVC performance, this paper presents a technique for prediction of EAF reactive power for a half cycle ahead. This technique is based on Artificial Neural Networks (ANNs). The procedure uses huge field data, collected from eight arc furnaces in Mobarakeh Steel Industry in Iran. About 90% of the recorded data are used for training the ANN and the rest are used in the test procedure. The performance of the compensator under the case of employing the predicted fundamental reactive power of EAF is compared with that for conventional method by using four indices which are defined based on concepts of flicker frequencies and power spectral density.