Prediction of Optimal Sulfinol Concentration in Khangiran Gas Treating Unit via Adaptive Neuro-Fuzzy Inference System and Regularization Network

The concentration of H۲S in the inlet acid gas is an important factor that sulfur plant designers must consider when deciding on the right technology or configuration to obtain high sulfur recovery efficiency. Using sterically-hindered solvents such as promoted tertiary amines and various configuration for gas treating unit are several alternatives for acid gas enrichment (AGE) to reduce the concentration of carbon dioxide and heavy aromatic hydrocarbons while enriching the H۲S content of SRU feed stream. The present article uses combinations of Aspen-HYSYS software and two distinct networks (namely, Regularization network and adaptive neuro-fuzzy inference system) to compare the AGE capability of sulfinol-M (sulfolane + MDEA) solvent at optimal concentration to traditional MDEA solution when both of them are used in a conventional gas treating unit (GTU). The simulation outcomes demonstrate that the optimal concentration of Sulfinol-M aqueous solution (containing ۳۷ wt% Sulfolane and ۴۵ wt% MDEA) will completely eliminate toluene and ethylbenzene from the SRU feed stream while removing ۸۰% of benzene entering the GTU process. Furthermore, mole fraction of H۲S in the SRU feed stream increases the conventional ۳۳.۴۸ mole% to over ۵۷mole%. Increased H۲S selectivity of optimal Sulfinol-M aqueous solution will elevate the CO۲ slippage through sweet gas stream at around ۴.۵mole% which is still below the permissible threshold.

کلیدواژه ها:

age ، BTEX ، Regularization network ، MLP ، ANFIS