Hybrid algorithm with fuzzy system and conventional PI control for the temperature control of TV glass furnace

This paper presents a practical application of fuzzy logic to the temperature control of glass melting furnace for television picture tube. Because of the complexity and nonlinearity, temperature control of glass melting furnace is still delegated to human operators. Though the overall characteristics of glass melting furnace are complex and nonlinear, one part of the furnace characteristics can be modeled as a linear system. The linear part of the furnace dynamics is modeled with a first-order-plus-dead-time (FOPDT) system and a proportional integral (PI) controller is applied to the FOPDT model. The remaining complex and nonlinear part of the furnace dynamics is covered by the fuzzy system, i.e., rules of human experts. The PI controller and fuzzy system are combined in cascade. The PI controller is applied to maintain the left-right symmetric operation of the furnace, and a linear regression is used for the temperature trend to determine input to the fuzzy system in order to overcome the high frequency disturbance. The proposed hybrid controller is implemented in Samsung-Corning Company to demonstrate the effectiveness of the proposed control algorithm.