Improvement of the /Taguchi/ design optimization using artificial intelligence in three acid azo dyes removal by electrocoagulation

The aim of this research is improvement of the Taguchi design optimization using artificial neural network (ANN) and genetic algorithm (GA) in Acid Orange 7, Acid Brown 14, and Acid Red 18 azo dyes removal by electrocoagulation. For this purpose, 27 tests were undertaken for investigation of five parameters including current density, reaction time, initial dye concentration, dye type, and initial pH by using Taguchi's orthogonal array. Additionally, according to analysis of variance, dye type and reaction time were the most important parameters for responses of dye removal efficiency and operating costs in Taguchi design, respectively. Prediction and modeling of the dye removal efficiency response were also accomplished by ANN. High R2 values (≥97%) indicated that the accuracy of the Taguchi and ANN models are acceptable. In addition, ANN was used in GA for finding the best elimination conditions for the selected dyes according to the Taguchi design. Dye removal efficiencies of 96.79%, 98.12%, and 76.47% were reported for Acid Orange 7, Acid Brown 14, and Acid Red 18, respectively, in the ANN model at the best elimination conditions. © 2015 American Institute of Chemical Engineers Environ Prog, 34: 1568–1575, 2015