Simulation-optimization of carbon dioxide capture process and the impact of uncertainty on the solution

he increase of global surface temperature has become more severe in recent decades. This is due to an increase in the level of greenhouse gases, mainly carbon dioxide, methane and nitrous oxide, in the atmosphere (Zhang et al., 2009). Carbon dioxide, the dominant of all the GHGs, accounts for around 81% and 83% of the total world and the U.S. GHG emissions, respectively (Energy Information Administration, 2007; International Energy Agency, 2008). The environmental and global warming concerns accelerated the research and development of carbon dioxide capture technologies to control and reduce the emissions of this primary GHG. As a result, there are many different technology options that are suggested for carbon dioxide capture. However, selecting and designing an optimum post combustion process from these candidate technologies is a challenging problem. Although there is much research going on optimizing individual processes for a given flue gas concentration, the research on which technology to select along with its optimum design and operating conditions has been limited (Olajire, 2010).;The objective of this research project is to develop novel simulation-optimization approaches to study the process synthesis, retrofitting and optimization of post-combustion carbon capture technologies. Two simulation-optimization approaches are described and the results obtained from both approaches are compared. The first simulation-optimization technique uses local searches to estimate an appropriate direction to reduce the objective function, i.e., the response surface methodology (RSM). The optimal solution obtained using response surface methodologies will generally be a local optimum. In the second optimization approach, the simulation is used to build a surrogate model, i.e., an artificial neural network (ANN), of the objective function over the whole decision space, and the optimization is performed using this surrogate model. Depending on the accuracy of the surrogate models, the solutions obtained using this approach can be shown to be global within the bounds of the data used to generate the surrogate models. A systematic sensitivity analysis is performed to determine the impact of uncertain parameters on the optimum CO ₂ capture cost, and the feasibility of using a stochastic programming approach to optimize the CO₂ capture process under uncertainty to yield the minimum cost over all possible uncertain outcomes is investigated.;The results of this research project identify lower cost CO₂ capture process designs and accelerate cost reduction for CO₂ capture, and the proposed work has the potential to increase the adaption of these technologies, hence resulting in reduced CO₂ emissions. The RSM is comparable to the ANN approach in terms of the number of simulations and the optimum solution. The results show that the amine-based process with the 48%wt DGA solvent requires the lowest CO₂ capture cost which is around...