Automated process design and optimization of membrane-based CO2 capture for a coal-based power plant

A systematic optimization framework is proposed with the aim to automate the design of multi-stage membrane processes for CO2 capture from flue gas of a coal-fired power plant. This framework utilizes a superstructure approach to determine the optimal configuration of membrane systems and identify the most appropriate operating conditions in a holistic manner. Certain design specifications are satisfied through the use of penalty functions which are used in a Genetic Algorithm (GA) optimization method employed to identify design solutions at or near to the global optimal point. Sensitivity analysis is used to analyze multi-stage membrane designs to understand the effects of different structural and operating parameters on the economics of membrane-based carbon capture. As part of a case study the proposed design framework is applied to design membrane processes for the capture of CO2 from a 600 MWe coal-fired power plant. Fixed membrane permeance and selectivity values are used to analyze sensitivities with respect to costing and structural design parameters. Additionally, the Robeson upper bound correlation between CO2 permeance and CO2/N2 selectivity is used within this framework to identify the optimal membrane properties which give economical separation of CO2 and N2. It is found that membranes having at least 4000 GPU CO2 permeance and over 50 of CO2/N2 selectivity with a commercial available module gave the optimal performance and would be a good guideline for future membrane material development. Also, if different membrane properties are used in each stage (in a multi-stage configuration) then using a higher CO2 permeance for the first stage (e.g. 6000 GPU CO2 permeance and CO2/N2 selectivity of 40) and higher selectivity membranes are used for subsequent downstream membrane stages (e.g. 1334 GPU CO2 permeance and CO2/N2 selectivity of 72) helps to reduce the electricity consumption and product purity which can reduce the overall cost of CO2 capture.