Process optimization for textile industry-based wastewater treatment via ultrasonic-assisted electrochemical processing

The treatment of industrial wastewater at a reasonable cost and in a sustainable approach is a major challenge. The primary objective of this study is to perform textile wastewater treatment by exploring the ultrasonic -assisted electrochemical (UEC) process with parameter optimization. Statistical modeling has been carried out in order to investigate the influence of operating parameters, including pH (4-10), current density (5- 20 mA/cm2), electrolysis time (10-30 min), and ultrasonic power (0-500 watt) on color, chemical oxygen demand (COD), and turbidity removal efficiency. The combination of response surface methodology (RSM) and artificial neural network (ANN) based statistical models are applied to predict and examine the reliance of the operating parameters on removal efficiencies. The results validate that at optimal operating condition (pH 7.3, current density 15.7 mA/cm2, processing duration 21.5 min, and ultrasonic power 420 W), the actual removal efficiencies of color, COD, and turbidity was 97%, 66%, and 79% respectively. The central composite design (CCD) based RSM model predicts the removal efficiencies of color, COD, and turbidity was 93.9 %, 63.6 %, and 77.1 %, respectively. In contrast, the ANN model predicts removal efficiencies of color, COD, and turbidity of 99.3%, 82.1%, and 78.7%, respectively at optimized operating conditions. The results indicated that the ANN model had a greater coefficient of determination (R2) value and smaller mean square error (MSE) than RSM model, showing that ANN is preferable at forecasting response, while RSM accurately predicts impacts of operating parameters. The present manufacturing process with statistical optimization techniques may be envisioned for the large-scale treatment of industrial wastewater.

Automated summary

The study aims to treat textile wastewater through an ultrasonic-assisted electrochemical process and optimize the parameters. The actual removal efficiencies of color, COD, and turbidity were found to be 97%, 66%, and 79% respectively at the optimal operating conditions. The statistical modeling validates that the artificial neural network model outperforms the response surface methodology model in predicting the removal efficiencies of color, COD, and turbidity.