ABSTRACT
High temperature steelmaking processes are inherently complex and their modeling, as seen in earlier chapters entails major efforts. Currently available computational uid dynamics (CFD) platforms yet do not offer “reacting, multiphase turbulent ow, and heat transfer” capability, which is often a prerequisite to the modeling of many steelmaking processes. Developing such models in-house with advanced mesh generation and postprocessing capabilities is complex as well as time intensive and this impedes rigorous process modeling exercise. Therefore, for every conceivable problem, it may not be currently possible to develop models from rst principles. In addition, computational process models, such as those outlined in the preceding chapters, generally involve prolonged run time, which is a serious hindrance to their application particularly when output is desired in real time (i.e., as in industrial process control). Also in many instances, one is concerned with global rather than local behavior. For example, despite thermal inhomogeneties in steel processing vessels, an average rate of change of temperature is often needed to quantify the extent of melt temperature drop or pickup. Rigorous modeling is not important in such context and available simpli ed models do suf ce. In the absence of elaborate differential models, macroscopic as well as input-output models (i.e., Chapter 9) provide effective alternative platforms to investigate steelmaking process dynamics.
