Sonocatalytic Tannery Effluent Degradation—Response Surface Methodology and Particle Swarm Optimization Embedded Artificial Neural Network Modeling Approach

Abstract

Experimental studies are performed on the degradation of tannery effluent using an advanced oxidation process combined with sonication. A probe sonicator is used to enhance the advanced oxidation process in the presence of titanium di-oxide (TiO 2 ) catalyst. The effect of initial chemical oxygen demand (COD), ultrasound power intensity, probe diameter, and sonication time are studied. Box-Behnken experimental design is used to study the effect of the process variables on percent COD reduction. 73.6% COD reduction is observed for the optimized condition of 825.71 mg L −1 initial COD of effluent, 542.58 W ultrasound power, 47.4 min of sonication time, and 18 mm probe diameter. The energy requirement calculated for the optimized condition is 2964.21 kWh m −3 . The results of the response surface methodology (RSM) model are compared with a hybrid artificial neural network (ANN) model with particle swarm optimization (PSO). The ANN-PSO model with a 4-11-1 architecture consisting of 15 swarm particles results in the coefficient of determination (R 2 ) value of 0.989 and the mean square error value of 3.88 thereby, predicting the experimental results better than the response surface methodology (RSM) model. An attempt to elucidate the mechanism of degradation by sonication is made by gas chromatography-mass spectrometry and Fourier transform infrared spectroscopy.