IMC-optimization of a direct reduced iron phenomenological simulator

This paper deals with the optimization of a direct reduced iron (DRI) phenomenological static simulator. The technique is based on an internal model controller (IMC) designed using a static approximate reduced (even linear) model. The cost function depends only on the reduced model rendering the minimization much easier. Appropriate corrections are calculated by the means of the IMC structure. Even in presence of a large model mismatch, constraints are respected. The convergence is usually much faster than a direct optimization based of the nonlinear process itself. The algorithm is successfully applied to a direct reduced iron phenomenological simulator, allowing to draw some metallurgical conclusions about the guidance of the process.