A Fuzzy Logic System for Carbon Estimation in a Basic Oxygen Furnace

In this paper, a fuzzy logic system for metal analysis in a basic oxygen furnace (BOF) is described. This system uses the optical spectral emitted from the BOF mouth as inputs and the carbon content of the molten metal contained in BOF as outputs, in order to provide an efficient decision support application. The membership functions and the rules are defined by utilizing both data analysis and expert knowledge. The optical radiation variables employed to develop the fuzzy logic system are selected from the real production data supported by a dedicated optical sensor of a full scale BOF at Nanjing steels´. In experiments with real data, the fuzzy logic system provides accurate estimates of the carbon content in the converter.