Artificial neural network modeling and optimization of wheat starch suspension microfiltration using twisted tape as a turbulence promoter

Abstract

The aim of this study was the modeling and optimization of the turbulence promoter-assisted starch suspension microfiltration process using artificial neural networks. The main aim was to analyze the influence of process parameters (transmembrane pressure, suspension flow rate, and concentration) on permeate flux (with and without twisted tape) and energy consumption in order to find the optimal combination of process parameters needed for the efficient use of the turbulence promoter. The obtained results showed a very good agreement between artificial neural network predictions and experimental data. Permeate flux was mostly influenced by suspension concentration, while the suspension flow rate had the highest impact on the reduction of the specific energy consumption. Performed optimization, using a genetic algorithm, indicated that microfiltration process should be carried out at the maximum value of transmembrane pressure (0.9 bar), at a suspension flow rate in the range from 80 to 93 L/hr, and at a suspension concentration of 5 g/L. Practical applications Practical application of this research presents the possibility for the wastewater treatment in the wheat starch processing industry that produces a significant amount of wastewater annually, mainly consisting of starch. Adequate treatment of wastewater can recover water-soluble and partly suspended substances from the raw material and at the same time, recycle and purify wastewater prior to releasing to the ecosystem. In the last decades, cross-flow microfiltration has shown many advantages in wastewater treatment operations comparing to conventional methods. In this research, cross-flow microfiltration with twisted tape, as a turbulence promoter, is chosen in order to achieve higher values of permeate flux, simultaneously, by paying attention on energy efficiency. The multi-response optimization, performed by applying the genetic algorithm, enables the determination of process parameters necessary for the accomplishment of economic and cost-effective microfiltration process.