Surrogate-based VSA Process Optimization for Post-Combustion CO2 Capture

Abstract

Abstract Post-combustion CO 2 capture in existing power plants is essential to arrest the current rise in atmospheric CO 2 and the consequent alarming trend of global warming. While absorption and pressure swing adsorption are well-known carbon capture technologies, vacuum swing adsorption (VSA) is a potential candidate. In this work, a comprehensive non-isothermal model is first developed and implemented in the multi-physics software COMSOL to simulate various modes of VSA operation. Our extensive parametric study suggests that even a simple basic VSA cycle can capture CO 2 with high purity & recovery at comparable or lower energy penalty than published data. The rigor of the full transient VSA simulations to reach the cyclic steady state, however, make fully rigorous VSA optimization intractable. To this end, we present a sequential optimization strategy based on response surface models with synergistic combination of COMSOL simulation model with Design and Analysis of Computer Experiments (DACE). Unlike most optimization studies which either focus on maximizing CO 2 purity/recovery or minimizing energy penalty, we use the total-ownership-of-cost approach to rationally drag technology performance, technology economics, energy penalty and environmental impacts to a single basis ($/ton of CO 2 ). The effectiveness of this approach to assess carbon capture economics by combining costing with system analysis is also discussed.