Model predictive control and dynamic real-time optimization of steam cracking units

Abstract Unlike other parts of this book, this chapter does not deal with mathematical procedures for modeling complex reaction systems, for example, strategies for construction/reduction of kinetic schemes, approaches to the estimation of their parameters (preexponential factors, activation energies, etc.), and so on. Conversely, it shows how to use accurate mathematical models of chemical processes, based on detailed kinetic schemes, for advanced control and optimization purposes. To this end, it describes two advanced model-based optimization/control strategies, called model predictive control and dynamic real-time optimization, and demonstrates the extent to which they can benefit chemical processes. The process unit, utilized to show the potential of model predictive control and dynamic real-time optimization, is a steam cracking furnace.