Key points are not available for this paper at this time.
Dairy wastewater contains a high level of organic matter and nutrients, which can harm the environment if discharged without proper treatment. This study investigated the degradation of pollutants in dairy wastewater using a TiO 2 /NiFe 2 O 4 photocatalytic composite within a parabolic-trough solar reactor. The effects of operational parameters, such as initial chemical oxygen demand (COD), photocatalyst dosage, and solar light intensity, on pollutant degradation efficiency were studied. A new kinetic rate model was proposed based on elementary chemical reactions to describe the dependence of the degradation rate on photocatalyst dosage and light intensity. The model was validated using experimental data. The results showed that when the model was reduced to a pseudo-first-order kinetic rate, it accurately predicted photodegradation efficiency (R 2 = 0.97, MSE = 0.04), showing minimal discrepancy from experiments. This study evaluated the performance of an ANN as a comprehensive empirical kinetic model compared with the kinetic rate equation. The ANN model was trained using both experimental and new synthetic data generated via the mega-trend diffusion technique (MTD). Results showed that the ANN accurately predicts COD removal efficiency under various operating conditions, with a high correlation coefficient (R 2 = 0.98) and low error (MSE = 0.002). The findings suggest that the ANN model, along with the first-order kinetic rate equation, provides a more accurate estimate of the photocatalytic degradation process than the overall kinetic model. Additionally, the ANN model effectively simulated the process under different conditions. • TiO₂/NiFe₂O₄ solar reactor achieved 90 % COD removal in dairy wastewater. • A kinetic rate equation based on chemical pass mechanism was developed. • The catalyst/solar interaction was considered in kinetic rate equation. • ANN outperformed kinetics in predicting COD under changing conditions with R 2 = 0.98 and MSE = 0.002.
Karimi et al. (Wed,) studied this question.