Optimal and robust production of high pressure steam

Steam delivery networks are large energy-consuming processes and an important part in many processing plants. Often, there is a potential for improved operation and reduction in energy consumption by application of advanced control algorithms and real time optimization. In this article, such a solution is investigated by implementing a model predictive controller to control the common header pressure of the steam delivery network. Furthermore, the energy consumption is reduced by incorporating a real time optimizer on top of the model predictive controller providing optimal operating points for the boilers. The potential reduction in energy consumption is examined using data for a real plant, and the application of the model predictive controller with the overlying real time optimization algorithm is examined using a rigorous process simulator.