Prediction of Temperature of Liquid Steel in Ladle Using Machine Learning Techniques

Abstract

Maintaining the temperature of liquid steel in the ladle in the required range affects the quality of casted billets, reduces energy consumption, and guarantees smooth control of the melting sequence. Measuring its temperature is a challenging task in industrial settings, often hindered by safety concerns and the expensive nature of equipment. This paper presents models which enable the prediction of the cooling rate of liquid steel for variable production parameters, i.e., steel grade and weight of melt. The models were based on the FEM solution of the Fourier equation, and machine learning approaches such as decision trees, linear regression, and artificial neural networks are utilized. The parameters of the model were identified using data from the monitoring system and inverse analysis. The results of simulations were verified with measurements performed in the production line.