Modelling and Multi-Objective Optimization of Continuous Indirect Electro-Oxidation Process for RTB21 Dye Wastewater Using ANN-GA Approach

Abstract

A continuous indirect electro-oxidation (EO) process was developed using graphite electrode to investigate the treatability of reactive turquoise blue RTB21 dye wastewater under specific operating conditions of initial pH, current density, hydraulic retention time (HRT), and electrolyte (NaCl) concentration. The experiments were performed in accordance with the central composite design (CCD), and the findings were used to create a model utilizing artificial neural networks (ANNs). According to the predicted findings of the ANN model, the MSE values for colour and COD removal efficiencies were estimated to be 0.748 and 0.870, respectively, while the R2 values were 0.9999 and 0.9998, respectively. The Multi-objective optimization using genetic algorithm (MOGA) over the ANN model maximizes the multiple responses: colour and COD removal efficiency (%). The MOGA generates a non-dominated Pareto front, which provides an insight into the process's optimum operating conditions.

Keywords

• multi-objective optimization
• artificial neural network
• genetic algorithm
• wastewater
• reactive turquoise blue 21