Equation-Oriented Optimization of Cryogenic Systems for Coal Oxycombustion Power Generation

TitleEquation-Oriented Optimization of Cryogenic Systems for Coal Oxycombustion Power Generation
Publication TypeJournal Article
Year of Publication2014
AuthorsDowling AW, Balwani C, Gao Q, Biegler LT
JournalEnergy Procedia
Type of ArticleJournal Article dcm
Keywordsair seperation unit, carbon capture, CO2 processing unit, coal oxycombustion, optimization

Efficient separation systems are essential to the development of economical CO2 capture system for fossil flue power plants. Air Separation Units (ASU) and CO2 Processing Units (CPU) are considering the best commercially available technologies for the O2/N2 and CO2/N2, O2, Ar separations in coal oxycombustion processes. Both of these systems operate at cryogenic temperatures and include self-integrated refrigeration cycles, making their design challenging. Several researchers have applied sensitivity tools available in the commercial flow sheet simulators to study and improve ASU and CPU systems for oxy-fired coal power plants. These studies are limited, however, as they neglect important interactions between design variables. In this paper, we apply an advanced equation-based flowsheet optimization framework to design these cryogenic separations systems. The key advantage of this approach is the ability to use state-of-the-art nonlinear optimization solvers that are capable of considering 100,000+ variables and constraints. This allows for multi-variable optimization of these cryogenic separations systems and their accompanying multi-stream heat exchangers. The effectiveness of this approach is demonstrated in two case studies. The optimized ASU designs requires 0.196 kWh/kg of O2, which are similar to a “low energy” design from American Air Liquide and outperforms other academic studies. Similarly, the optimized CPU requires 18% less specific separation energy than an academic reference case. Pareto (sensitivity) curves for the ASU and CPU systems are also presented. Finally, plans to apply the framework to simultaneously optimize an entire oxycombustion process are discussed.