Experimental data driven thermodynamic modelling and process simulation for biogas upgrading

Data driven thermodynamic models has a great potential to add value to the accuracy of simulation processes. In this study experimental data driven e-NRTL activity coefficient model is used to evaluate the behavior of MEA/AMP and MDEA/AMP aqueous systems for biogas upgrading. The vapor liquid equilibrium (VLE) experimental data is used for regression of the electrolyte pair parameters of e-NRTL activity coefficient model, which were then used to design the process for biogas upgrading by using Aspen Plus in combination with MATLAB. Experiments have been carried out for CO2 solubility as a function of CO2 partial pressure (PCO2) at different amine blending ratios such as 9/21/70, 15/15/70, and 21/9/70 (w/w/w percent) for each MEA/AMP/H2O and MDEA/AMP/H2O systems at a wide temperature range of 323.15–383.15 K. The importance of regressed parameters was confirmed through global sensitivity analysis (GSA) using Monte Carlo simulation. The local sensitivity analysis (LSA) and GSA for biogas upgrading process based on data driven thermodynamic modeling have been carried out in terms of the biomethane purity, reboiler heat duty, and methane slip. Finally, the optimal process parameters were found through multi-objective optimization (MOO) for developed biogas upgrading process.