Reverse Osmosis Process Optimization for Acetic Acid Rejection, by Coupling RSM to PSO

Abstract

Acetic acid is one of the main pollutants in ethanol production plants. It appears in the residues of the distillation of fermented musts. Its presence in fermentation inhibits microbial biocatalysts and reduces ethanol production rates. In this study, the separation of acetic acid from vinasse by the reverse osmosis process is studied. This separation is influenced by operatory parameters that must be optimized. Three parameters of feed flow rate, acetic acid concentration, and temperature are considered to maximize the acetic acid rejection. The response surface methodology (RSM) coupled to the particle swarm optimization (PSO) was followed for this purpose, using numerical data obtained, by the reliable model developed in our previous work. The RSM allows satisfactory prediction of acetic acid rejection obtained with a Mean Absolute Percentage Error of about 1%. This model was exploited as an objective function by PSO developed on software Python, to maximize the acetic acid rejection at different feed concentration cases. The results showed that the optimal values obtained, relating to an initial pressure of 17.51 atm and a temperature of 38.65°C could eliminate 99.6% of acetic acid at an energy consumption of 5.67 kWh/m3.