Abstract
Carbon dioxide (CO2) capture from natural gas is integral towards meeting pipeline sales gas specifications and avoiding operational problems during natural gas liquefaction. Therefore, it is important to understand how different process parameters would affect the performance of the CO2 capture process plant. In this research, CO2 capture from a typical Nigerian natural gas composition was simulated using ProMax® 4.0. The validated simulation was used to generate 3125 datasets while varying a number of process parameters. A parametric sensitivity analysis was conducted by varying lean amine flow rate (LAF, 3500–4300t/day), lean amine temperature (LAT, 40–60°C), lean amine pressure (LAP, 60–75bar), MDEA–PZ concentration difference (CMDEA–PZ, 36–44wt.%) and heat duty (HD, 50.4–56.52 GJ/h) to determine their effects on CO2 capture efficiency alongside foaming and amine vaporization. The parametric analysis showed that the LAF and MDEA–PZ concentration have the highest effect on the CO2 capture plant. In addition, 70% of the generated datasets was used to train the intelligent model named adaptive neuro–fuzzy inference system (ANFIS) while 30% was used for validation. Results revealed that the ANFIS model accurately predicted the simulation results with 2.4%AAD and RMSE of 4.0E-03, respectively.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call