Abstract
A model was developed to predict the operation and performance for liquid oxygen storage coupled with the Allam-Fetvedt Cycle (AFC), a direct-fired supercritical carbon dioxide cycle. The cycle uses liquid oxygen to time-shift the major parasitic of the cycle, the air separation unit (ASU), to minimize the economic costs of air separation. This paper presents significant improvements to the model and the techno-economic analysis in the forms of improvements to the assumptions, an exploration of different ASU architectures, potential improvements to the AFC for coupling with liquid oxygen and a sensitivity study on the impact of variables on performance. Several assumptions were updated or added to improve the accuracy of the system. These include adding utilities such as water requirements or fuel usage during ASU operation. In addition, an aerodynamic analysis on three different architectures for the main air compressor (MAC) were compared focusing on the performance at both full load and part load. It was determined that the best architecture for the MAC depended on the underlying economic market. An integrally-geared compressor performed the best in markets with high renewable penetration and frequent short-term cycling due to the improved part load performance. For markets with more seasonal trends, the more efficient axial-radial architectures performed better. In addition to improvements to the charge system, several different modifications were explored to improve the AFC. The first was to use the CO2 stream to vaporize the liquid oxygen, reducing the inlet temperature of the CO2 entering the compressor. Another was to pump the oxygen and CO2 separately to full pressure before mixing the oxidant streams. The potential technology improvements resulted in additional power output of the system that improved the net present value overall. All of the results were compared to each other in a sensitivity analysis that compared how 8 different parameters affected the results for different markets.
