Dual Adaptive Temperature Control Of Magnesium Reduction Furnace

To propel the automation of production and save energy consumption, the dual adaptive control problem of the magnesium reduction furnace is studied in this paper. The detailed temperature control process is analysed and divided into four phases. The control process of one area of the furnace is described by a simplified single-in-single-out (SISO) discrete model with large input delay. With sufficient input/output experiment data, the optimal model structure and parameters are identified offline. The dual generalised minimised variance (GMV) controller which can cope with non-minimum phase plants with time delay is chosen in the temperature control of the magnesium reduction process. The control performance is verified by both simulations and experiments. The results show that the dual GMV controller is superior to a cautious controller or a conventional PID controller.