Activity calculation model of converter multi-component slag system considering temperature and complex molecular composition changes

Abstract

During the converter smelting process, the gradual increase in slag temperature leads to significant accuracy deviations in traditional slag activity calculation models. Building on the ion-molecule coexistence theory model developed by previous researchers, this study introduces an improved model that incorporates temperature and complex molecular composition changes. The results showed enhanced accuracy, with mean absolute error, mean absolute percentage error, and root mean square error decreasing from 0.113, 0.300, and 0.127 to 0.071, 0.192, and 0.087, respectively, though further improvements are still possible. The study also detailed the effects of temperature and slag composition on slag activity, slag oxidation, and phosphorus distribution ratio. The results indicate that both temperature and slag composition have a significant impact on the performance of slag. Using two sets of typical converter smelting process slag data, the study analysed slag property changes during actual smelting and explored differences between two slag-forming routes. This work provides a new model for real-time slag activity calculation, enabling a mechanism-based characterisation of the converter ‘black box’ process.