This thesis presents the simulation and performance optimization of the ammonia stripping column in the ammonia recovery section of the Shiraz Petrochemical Company ammonia production unit using Aspen Plus software. The process aims to recover high-purity ammonia from the sour water stripper feed, resulting in an additional ammonia production of approximately 1642.2 kg/h in the industrial unit.
To select the most appropriate thermodynamic model, several property methods — including NRTL, UNIQUAC, Wilson, ELECNRTL, Peng-Robinson, and Soave-Redlich-Kwong (SRK) — were evaluated and compared against experimental vapor-liquid equilibrium (VLE) data from the NIST database under various operating conditions (polarity, temperature, pressure, etc.). None of the models provided satisfactory agreement with the experimental data in their default form. After regression of binary interaction parameters (k_ij and temperature-dependent coefficients), the SRK model was selected as the final thermodynamic model due to its superior performance.
The stripping column was simulated in Rate-Based mode using IMTP packing. The simulation results showed very good agreement with actual plant data from the industrial unit. The average relative deviation (ARD) for key parameters (ammonia concentration in the bottom product, temperature, and pressure) remained within acceptable industrial limits. The developed model demonstrated high accuracy in predicting stripping efficiency as well as temperature and concentration profiles along the column.
The findings of this study confirm the effectiveness of the regressed SRK model for simulating ammonia stripping processes under moderate to high pressure conditions. The model can serve as a reliable tool for sensitivity analysis, optimization of operating parameters (steam flow rate, reflux ratio, packing height, etc.), and performance prediction in similar industrial units.