A discrete, linear, time-variant, state equation model of the heat transfer processes involved has been applied to describe the behavior of steel ingots during teeming, prior to and following stripping, during soaking pit heating and on extraction. This model gives accurate representations of the solidus formation and of the transient temperature behavior of the ingot while in the molds and while in the soaking pit furnaces. A major application of the resulting model has been to the charging of ingots with partially liquid cores into the soaking pits with a resulting substantial reduction in the energy requirements for the soaking pit operation and a correspondingly large increase in soaking pit throughput or productivity.
INTRODUCTIONOne of the major areas of steel plant operation today is that of ingot processing. The objective in controlling the soaking pits is to achieve suitably high and uniform ingot temperature profiles for rolling while minimizing the total energy consumption in the soaking process.During the past few decades, the mathematical. description of the temperature behavior of steel ingots during solidification, cooling and heating has received the attention of numerous investigators. t.3-20 They have employed various approaches to establish the mathematical models for these ingots. For the development of these ingot models, the following main approaches have been applied to date: lumped parameter linear models, distributed parameter models, on-line optimal estimation, and regression analysis. King et al. (1963), have detailed the basic equations involved and their application to ingot processing modelling.Ingot models can readily be described in the form of partial differential equations (Lausterer, 1978). Ray et al. (1979), have indicated some potential applications of distributed parameter system theory to control of soaking pits. In addition, Soliman (1972) has applied the finite element method to solve the distributed parameter models , Presently Associate Professor of Chemical Engineeringof ingots during the solidification process. In practice, the solution of the partial differential equations of a distributed parameter model with the necessary boundary conditions is a difficult and time consuming task. Most investigators therefore choose to transform the model into a set of linear equations using the integral profile method or a finite difference approximation.In order to simplify the problem further, the resulting models can usually be reduced from three dimensions to one or two by means of some appropriate assumptions. Typical ingot models for this purpose are the two-dimensional model (Sevrin, 1970;Massey et al., 1971), the cylindrical model (Cook et al., 1978) and the spherical model (Maeda, 1975). The cylindrical (one dimensional) form will be followed in this investigation.The objective in controlling soaking pits is to achieve a suitable internal metal temperature profile for rolling, while minimizing surface oxidation and total energy consumption. Unfortunately, in practice, thes...