Modeling of the CAS-OB Process for Online Use

Abstract

A model of the CAS-OB steelmaking process has been developed and implemented. A lot of effort has been put into developing a model that is fast enough for use in online applications like Kalman filter type estimators and Nonlinear Model Predictive Control (NMPC). Simplifying the reaction kinetics has been an important issue in the model development. The model has been fitted to and verified by process data from Ruukki Steel Works in Raahe, Finland.A Sigma Point Kalman filter and NMPC has been set up for the process with the aim of providing reliable and precise temperature control. The performances of the estimator and controller are evaluated based on a series of tests where a simulated process is used instead of a physical plant. Different kinds of model errors are introduced. The results show good performance for the system, and it is believed that Model Predictive Control may improve the temperature control of the CAS-OB process.