Prediction of Molten Iron Temperature in the Transportation Process of Torpedo Car

Abstract

Molten iron temperature prediction during hot metal transportation plays a key role in reducing energy consumption and improving the quality of steel-making products. In this paper, first, a mechanism model based on the physical laws of temperature drop principle and heat transfer relationship is established. Aiming at the problems of difficult detection of molten iron temperature in torpedo tank and incomplete measurement information caused by limited measurement information, a prediction method of molten iron temperature based on data analytics was proposed through the historical data obtained from production process, and a prediction model of molten iron temperature was established. Due to the uncertainty in the process of molten iron in the transport, and the influence of waiting time, lead to on-line temperature prediction accuracy of the independent mechanism model or data analytics model is not high. In response to this problem, a data fusion prediction method based on Kalman filtering is proposed to meet the needs of actual production.