End-point prediction of basic oxygen furnace (BOF) steelmaking based on improved twin support vector regression

In this paper, a novel prediction method for low carbon steel is proposed based on an improved twin support vector regression algorithm. 300 qualified samples are collected by the sublance measurements from the real plant. The simulation results show that the prediction models can achieve a hit rate of 96 % for carbon content within the error bound of 0,005 % and 94 % for temperature within the error bound of 15 °C. The double hit rate reaches to 90 %. It indicates that the proposed method can provide a significant reference for real BOF applications, and also it can be extended to the prediction of other metallurgical industries