XGBoost-based converter endpoint prediction model

• 1、School of Metallurgy and Ecological Engineering, University of Science and Technology Beijing, Beijing 100083, China
• 2、School of Computer and Communication, University of Science and Technology Beijing, Beijing 100083

Abstract：In the production process of converter smelting, the hit rate of the end point is very important to the converter. The accuracy of the end point prediction will have a certain impact on the quality of the product and the cost of smelting. In order to improve the hit rate of the converter end point, this paper uses the XGBoost algorithm to build a converter end point prediction model. By analyzing the influence weight of each influencing factor on the result, it is assigned XGBoost-based converter endpoint prediction modelto the corresponding branch node in the XGBoost algorithm, using 198 heats from a steel plant The process data is used as the data to be tested. The carbon content and temperature value of the converter endpoint are predicted by the XGBoost-based converter endpoint prediction model, and the endpoint temperature prediction is achieved by using the converter endpoint prediction model based on case-based reasoning as a comparative experiment. The results show that the converter endpoint prediction model based on XGBoost has high accuracy for both the temperature and carbon content predictions, which basically meets the production requirements.

Keywords： ferrous metallurgy converter end point prediction model XGBoost algorithm end point carbon content end temperature