On-line estimation of inductance of flexible cables and electrode arms in an electric arc furnace by artificial neural network

This paper proposes a code which uses the geometrical parameters of flexible cables and electrode arms as input, and calculates the inductance of these parts in an electrical arc furnace (EAF) by 3-D numerical method at any moment. Exact calculation of the momentarily inductance of flexible cables and electrode arms can result in improving the control system of EAF and satisfactory study of the effects of EAF on power system. Because of using numerical method, calculation of inductance is a time consuming process and cannot be used to improve the control system of EAF. So an artificial neural network (ANN) is proposed for a fast estimate of inductance of flexible cables and electrode arms. The ANN is trained using the data obtained by proposed code.