Controlling nitrogen and phosphorus pollution
using neural network tool
D. Zakarya1*, M. Mahi1, M. A. Bahlaoui1, H. El Halouani2
1Université Hassan II- Mohammedia, Faculté des Sciences et Techniques de Mohammedia, BP 146 (20650), Maroc
2Université Mohammed 1er, Faculté des Sciences d’Oujda, B.P 254 (60000) Oujda, Maroc
* Corresponding author. E-mail: email@example.com
Received: 05 April 2006; revised version accepted:14 December 2006
The process of wastewater treatment by high rate oxidation pond (HROP) was studied by modelling the relationships between a set of environmental factors and nitrogen and phosphorus removal based upon one-year sampling program. The models of interaction between variables were tested by using both neural network (NNs) and multiple regression analysis (MRA) methods. Considering the relevant factors obtained from the MRA, a correlation coefficients of 0.98 and 0.91 (n = 52) were obtained with a 6-3-1 and 4-3-1 NN models for NH4-N and PO4-P respectively. The contribution of each factor to the NH4-N and PO4-P removal was evaluated. The best models established were found to be nonlinear. Obtained results suggest that NNs may be used as a practical tool to control the wastewater treatment.
Keywords: Wastewater; Oxidation Pond; Nitrogen; Phosphorus; Neural network.