中国有色冶金

At present, most of electric furnace power setting has been determined by operators according to their experience and onsite operation conditions, it is greatly affected by personal experience and operation, which makes it difficult to guarantee the real-time optimality of given power as to affect the production indicators. As regards smelting furnace of RKEF ferronickel smelting process, this paper proposes an optimization method for online control of electric furnace power. By using this method the online testing is performed for the process variables related to furnace charge, off-gas and furnace cooling system by establishing the electric furnace heat balance model to achieve online calculation of optimum set value of electrode power; by building Error BP network model, the measured data of slag tapping cycle are utilized to produce data sample for training and updating the network model, which realizes prediction and online modification of calculation error of heat balance and enhances the accuracy and adaptability of heat calculation. The effectiveness of this method is verified by simulation experiment based on engineering data. This paper indicates that using this control system can solve the core issue of intelligent smelting by ferronickel furnace. In the next step, the system needs to be combined with other subsystems to form a complete solution of intelligent smelter.