Abstract
In this study, a sequencing batch biofilm reactor combined with a vertical flow constructed wetland(SBBR-VFCW) system was constructed and applied to the wastewater treatment. Thalia dealbatas wereplanted in the VFCW. The artificial neural network (ANN) was used to simulate and predict the performance of SBBR- VFCW. The results showed that when the concentrations of COD, NH4+-N, TN and TP in thewastewater were 200.22 mg/L, 48.11 mg/L, 48.11 mg/L and 6.11 mg/L respectively, the removal efficiencies were 97.0%, 98.5%, 91.5% and 88.5%, correspondingly, which indicated that the SBBR-VFCW systemcan treat the wastewater effectively. According to the results of the ANN simulation analysis, the correlation coefficients (R-2) were all higher than 0.99, and the root mean squared errors (RMSE) were lowerthan 0.0782. The concentrations of DO, NH4+-N and TP in the influent exhibited strong impacts on theeffluent. This study reveals that the ANN can efficiently reflect the nonlinear function of each factor, andis suitable for the dynamic monitoring of SBBR-VFCW treatment for wastewater in various conditions. (c) 2013 Elsevier B.V. All rights reserved.