Development of the KR Desulfurization Process Model Centered on Production Rhythm

Abstract

Nowadays, most steelmaking plants still rely on manual experience to perform Kambara Reactor (KR) desulfurization, which makes it difficult to accurately control the endpoint [S] content for different steel grades, and can easily cause the waste of resources. To achieve cost reduction and efficiency increase, a KR desulfurization process model with production rhythm as the core was developed. The model included three desulfurization modes based on different requirements, namely process priority, time priority, and desulfurizer priority. The self-learning function to regress historical data of past heats was used in the modes of time priority and desulfurizer priority to obtain process parameters, while the process priority mode calculated process parameters just in the dependence of working conditions of current heat. After applying the developed model in a steelmaking plant, the average consumption of desulfurizer decreases from 6.41 to 5.27 kg/t. The one-time hit ratio of desulfurization pretreatment increases from 96.47 to 98.63%.