Modelling of heat transfer in hot strip mill runout table cooling

The need for accurate prediction and control of cooling profiles of steel strips on runout tables has led to the development of a mathematical model that is able to predict coiler temperatures under any given condition with an accuracy of +/- 14degreesC as well as calculating the entire temperature profile of a steel strip with sufficient accuracy. Comparisons with online strip temperature data at various locations of the runout table, which were obtained by a new experimental procedure, show that the effect of single cooling headers on the thermal response of a steel strip can be predicted. The model takes into account all relevant thermodynamic effects by means of a statistical approach. Heat transfer to the environment, steel thermophysical properties and phase transformation are modelled using B-splines. Model adaptation is realised by fitting calculated and measured coiler temperatures of approximately 40000 strips with a least square method in order to gain optimal base values for the B-spline functions. During model development special attention was paid to the model's capability of being re-adjustable to a large variety of conditions as well as its local behaviour. Therefore, concepts like temperature-dependent heat transfer coefficients, which are applicable only to one specific plant, have been avoided in favour of a more generalised formulation of the model that helps to gain insight into the physics of the processes involved, i.e. heat transfer of subcooled jet impingement boiling and film boiling. It was found that both cooling water and steel surface temperature have a large influence on heat transfer whereas the influence of strip speed can be neglected.