Comparison of probabilistic performance of calcium looping and chemical solvent scrubbing retrofits for CO2 capture from coal-fired power plant

Carbon capture and storage technologies are seen as crucial for decarbonisation of the power sector because large-scale deployment of these technologies is expected to reduce wholesale electricity prices and decarbonisation costs significantly. An emerging calcium looping process (CaL) was found to be a promising alternative to the amine scrubbing process, as its retrofit has the potential to reduce the efficiency penalty and increase the net power output of the integrated system. Process modelling is often utilised in the development of novel low-carbon power generation systems, but the deterministic nature of the process models does not provide a definitive representation of the actual system's performance. This study provides a framework for comparison of the probabilistic performance of a coal-fired power plant (CFPP) in CO2 capture retrofit scenarios. Having compared the probabilistic performance of the CFPP retrofitted with the mature MEA scrubbing process and emerging CaL process, it was found that the most probable values for the efficiency penalties are 9.5% and 11.5% points in the CaL and MEA retrofit scenarios, respectively. Also, the net power output in the CaL retrofit scenario was found to be 40% higher than that of the reference CFPP and two times higher than that in the case of the MEA retrofit scenario, even considering higher uncertainty in the operating conditions of the CaL process. Finally, the probabilistic approach was found to be capable of producing valuable information on the equipment operating envelope that would help designers assessing the number of equipment trains and their operating limits. Therefore, the probabilistic framework for comparison of CO2 capture retrofit scenarios would provide valuable input to the investment decision-making process, as by considering the impact of uncertainty on the process performance, it generates information that could serve as valuable input to the probabilistic economic analysis and process design. (C) 2016 Elsevier Ltd. All rights reserved.