Epistemic Uncertainty Analysis and Robust Optimization of a Second‐Generation Solvent‐Based Post‐Combustion Carbon Capture Process

Amine‐based carbon capture is widely regarded as a promising avenue for reducing industrial emissions, with the piperazine advanced flash stripper (PZ/AFS) process being a prominent example under active development. While process optimization can be employed to improve its economics, the many inherently uncertain parameters in its process model compromise the reliability of the resulting designs. In this work, the previously developed Pyomo‐based PZ/AFS process model is subjected to a comprehensive uncertainty analysis. The parameters that have the greatest impact on capture performance and costs are determined through sensitivity analysis, followed by a Pedigree analysis to assess knowledge strength for epistemic uncertainties. The chosen uncertainties are characterized through parameter estimation and expressed as a 12‐dimensional uncertainty set. Based on this, a robust optimization problem is solved to mitigate the risk posed by parametric uncertainties in the PZ/AFS flowsheet designs. Uncertainties in CO 2 equilibrium pressure and heat of absorption, along with the reaction rate, have the greatest impact on optimal costs, suggesting that one can benefit from additional efforts to estimate these parameters with better confidence. Compared to the deterministically optimal solution, the most substantial design changes to achieve robustness were noted for the pumps, highlighting the need for their careful design.