To grow high-purity and large sizes MgO single crystals with twin-electrode DC submerged arc furnace
requires that the temperature distribution be well understood and the processing temperature be precisely controlled. For
the complexity of the production of MgO single crystals and the difficulty to measure the temperature inside the furnace,
the temperature distribution was studied by using finite element method (FEM), and the temperature control was realized
by the process control strategies with adaptive neuro-fuzzy inference system (ANFIS). Experiments were carried out to
verify the effectiveness of the method. The result of experiments indicated that using the adaptive neuro-fuzzy control
system can improve the quality and the quantity of the MgO single crystal production.