Modeling of glass melting furnaces: Applications to control, design and operation optimization

The present paper describes the strategy for implementation of mathematical modeling of glass melting furnaces in control, design operation optimization and training. Prediction codes capable of computing the three-dimensional characteristics of the aerodynamics, mixing, combustion (single or multiphase), pollutants formation and heat transfer in the combustion chamber of glass melting furnaces exist. Furthermore, models for the heat transfer, chemical reactions, phase change and melting down process occurring in the batch region have been developed. Considerable effort has been placed in the development of glass tank models for the fluid flow and heat transfer inside the tank and including submodels for air bubbling, sand grain dissolution, redox and fining. The present paper describes the state of art of the computer simulation of glass melting furnaces (combustion chamber, batch region and glass tank) and identifies the current drawbacks of such models. Based on the current scenario of research and development, future research priorities are outlined. A mathematical model of combustion chamber, glass tank and batch region is described and its application to an end-oort container glass tank furnace is discussed. A measurements campaign was carried out in this furnace. Measurements have been made for in flame mean gas species concentrations of O-2, CO, CO2, NOx, and mean gas temperature. The results obtained with the model are in good agreement with the measured ones. The paper describes the basic concepts and research strategies involved in an enhanced utilization of modeling technologies for optimized design, operation and control of glass melting furnaces. An innovative procedure for the integration of physically-based and model-based knowledge with advanced optimization algorithms is described. The proposed methodology constitutes a technological contribution towards a new generation of systems able to assist the design, the operation and the control more efficient, cleaner and more intensive glass melting furnaces. Such system may be applicable in operation optimization (if on-line used), design optimization (if off-line used) and operators training (if simulated scenario generation is used). Examples of the implementation of these strategies in industrial environments are presented such as the design optimization of an oxy-fuel furnace, the model-based control and the development of an expert system for operation optimization and training in glass melting furnaces.