An adaptive system for modelling and simulation of electrical arc furnaces

Modelling and simulator development for electric arc furnaces (EAFs) are of significant importance in designing control systems and in performance optimisation of EAFs. This paper presents a method based on adaptive neuro-fuzzy inference systems (ANFIS) for modelling and simulating EAFs with the focus on the regulator loop that is used for positioning the electrodes. The effectiveness of the simulator is shown through experiments by comparing the simulator outputs with actual plant data, using the EAF of Gerdau Ameristeel Whitby (GAW) in Ontario, Canada.