Prediction of the Performance of Rapid Heating Furnaces Using Physical and Mathematical Modelling Techniques

Abstract

This thesis shows how a combination of physical and mathematical modelling can be used to predict the thermal performance of tangentially fired rapid billet heating furnaces. The experimental work involves the measurement of mass transfer coefficients in an isothermal scale model of the furnace using a technique based on the measurement of the limiting diffusion controlled current during the electrolysis of an alkaline solution of potassium ferri- and ferrocyanide. Convective heat transfer coefficients are obtained from the mass transfer measurements via the Chilton-Colburn analogy. These coefficients are then combined with a suitable mathematical model of the heat transfer processes in the furnace and the thermal performance predicted. The procedure has first been checked by calculating the heating times to forging temperature of cylindrical copper and aluminium billets and comparing these with results obtained in an actual prototype furnace. Mathematical models are then used to show that the performance of the prototype furnace can be improved by modifications to its design and to demonstrate the significant improvement in transient response of these furnaces compared with traditional design. Finally both the mass transfer measurements and the mathematical analysis are extended to deal with situations in which the heat flux to the load surface is not uniform.