Control of slag and refining conditions in the BOF (Bathfoam) : final report

Control of slag foaming in the BOF has always been problematic. With increased pressure on process yields and productivity, combined with increased demands for accurate control of refining it is important to develop a process control tool that can consistently achieve the optimum performance from a perspective of slopping control as well as refining in the BOF. The project used static and dynamic process data from many sources and related them to process performance by using advanced statistical analysis tools. This was used, combined with data from analytical models, to develop a process control system that pro-actively controls slag formation in the BOF. In the project new relationships for slag foaming behaviour and rheology were derived based on cold and hot physical modelling, measurement of physical properties of slags and the analysis of the slag rheology. Analytical process models were developed to function as soft sensors to generate dynamic process data. Indirect measurements were used with varied success to obtain dynamic temperature, composition and slag foaming data. Process control models running a mix of statistical and analytical process models were developed. These models were successfully tested in open loop mode with one of them also successfully tested in closed loop mode. It was proven that prediction of slopping in the converter is possible using information from dynamic systems combined with static process data. In the short and medium term further development of dynamic process measurements and the required data analysis is required, in order to improve model accuracy and speed.