Abstract
In this investigation, the adaptive network-based fuzzy inference system (ANFIS) has been successfully developed as a method of estimating the heat capacity (CP) of certain amine-based solvents, using concentration of the solvent, system temperature, and an apparent molecular weight as input parameters. To develop a modelling system, 1,253 and 278 experimental data points were extracted from the published studies for binary and ternary systems respectively. Comparison of the model results with experimental data was done by investigating some statistical parameters like mean square deviation (MSD), R-squared (R2) and average relative deviation (ARD) for assessing the ability and veracity of the developed model. The developed model's ARD for binary and ternary systems are -0.937% and -0.036% respectively. Outcomes revealed that the developed hybrid-ANFIS model could be applied as a practical method in simulation of CO2 removal processes by amine-based solvent for both binary and ternary systems. [Received: April 16, 2020; Accepted: April 22, 2021]
Sign-in/Register to access full text options 
Free) 
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
