Photodegradation of an industrial Azo dye C.I Basic Red 46, was examined in a semi-pilot scale prototype solar photoreactor under solar radiation. In our study, photodegradation of the dye was optimized using Response Surface Methodology (RSM) based on Box-Wilson approach. The Artificial Neural Network (ANN) was used to establish suitable modeling and optimal conditions for the Solar UV/Immobilized-TiO₂ process in order to evaluate the individual effects of three factors that independently affect the effectiveness of the photodegradation process: (1) initial concentration of the dye, (2) pH, and (3) flow rate. The RSM was in good agreement with the prediction model (R²_Dec = 0.95); meanwhile, the ANN approach revealed that the predicated model fit perfectly with the experimental data to yield the highest value of R² = 0.999. The effects of these three factors could be estimated from a second-order polynomial equation, and the optimal parameters of photodegradation consisted of three main parameters: (1) initial concentration of colorant 10.65 mg.L⁻¹, (2) pH 10.82, and (3) rate of fluid flow of 852 L h⁻¹. The decolorization removal efficiency under these optimal conditions was 99%.

在半辐射规模的原型太阳光反应器中，在太阳辐射下检查了工业偶氮染料CI Basic Red 46的光降解。在我们的研究中，使用基于Box-Wilson方法的响应表面方法（RSM）对染料的光降解进行了优化。人工神经网络（ANN）用于为太阳UV /固定化TiO ₂工艺建立合适的建模和最佳条件，以评估独立影响光降解工艺有效性的三个因素的单独影响：（1）初始染料浓度，（2）pH和（3）流速。RSM与预测模型非常吻合（R ² _Dec = 0.95）；同时，人工神经网络方法表明，该预测模型与实验数据完全吻合，得出的最高R ²  = 0.999。这三个因素的影响可以通过一个二阶多项式方程来估算，并且光降解的最佳参数包括三个主要参数：（1）着色剂的初始浓度为10.65 mg.L ^-1，（2）pH为10.82，以及（3）流体流速为852 L h ^-1。在这些最佳条件下的脱色去除效率为99％。