Modeling and optimization by response surface methodology and neural network–genetic algorithm for decolorization of real textile dye effluent using Pleurotus ostreatus: a comparison study

Abstract

This study focuses on the modeling and optimization of the decolorization procedure of real textile dye. The percentage of decolorization of effluent in the Erlenmeyer flask level, as obtained by both response surface methodology (RSM) and artificial neural network (ANN), was determined and subjected to comparative evaluation. The effect of independent variables such as pH (5–8), self-immobilized Pleurotus ostreatus, bead volume (30–50%) (V_b/V_r), and initial effluent concentration (50–100%) was examined using three-level Box–Behnken design. A similar design was utilized to train a feed-forward multilayered perceptron with back-propagation algorithm. Errors were computed using error functions, and the values obtained for RSM and ANN were compared. The maximum percentage decolorization and COD reduction of effluent under optimized conditions over a 24-h period were observed as 89 and 72%, respectively. The parameters optimized in the flask level were adapted in an inverse fluidized bed bioreactor of 6 l working volume, in which the quantity of decolorization and COD reduction over a 24-h period was observed as 92 and 76%, respectively.

Keywords:

• Artificial neural network
• Response surface methodology
• Genetic algorithm
• Industrial dye effluent
• Inverse fluidized bed bioreactor
• P. ostreatus

Acknowledgment

The authors are very grateful to the management of SRM University, Chennai, India, for providing the necessary research facilities, support, and constant encouragement.