Use of artificial neural networks to evaluate the effectiveness of riverbank filtration.

Riverbank filtration (RBF) is a low-cost water treatment technology in which surface water contaminants are removed or degraded as the infiltrating water moves from the river/lake to the pumping wells. The removal or degradation of contaminants is a combination of physicochemical and biological processes. This paper illustrates the development and application of three types of artificial neural networks (ANNs) to estimate the effectiveness of two RBF facilities in the US. The feed-forward back-propagation network (BPN) and radial basis function network (RBFN) model prediction results produced excellent agreement with measured data at a correlation coefficient above 0.99 for filtrate water quality parameters, including temperature as well as turbidity, heterotrophic bacteria, and coliform removal. In comparison, the fuzzy inference system network (FISN) predicted only temperature and bacteria removal with reasonable accuracy. It is shown that the predictive performances of the ANNs depend on the model structure and model inputs.

[1]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[2]  Martin T. Hagan,et al.  Neural network design , 1995 .

[3]  R. Fujioka,et al.  Association of staphylococcal skin infections and swimming , 1995 .

[4]  Chittaranjan Ray,et al.  Riverbank filtration : improving source-water quality , 2002 .

[5]  Howard John Fallowfield,et al.  The potential of riverbank filtration for drinking water supplies in relation to microsystin removal in brackish aquifers , 2002 .

[6]  R. Smith,et al.  Changes in concentrations of triazine and acetamide herbicides by bank filtration, ozonation, and chlorination in a public water supply , 2002 .

[7]  Y. Lian,et al.  Effect of flood-induced chemical load on filtrate quality at bank filtration sites , 2002 .

[8]  Joo-Hwa Tay,et al.  A fast predicting neural fuzzy model for high-rate anaerobic wastewater treatment systems , 2000 .

[9]  Ni-Bin Chang,et al.  Rough set-based hybrid fuzzy-neural controller design for industrial wastewater treatment. , 2003, Water research.

[10]  J. Vasconcelos,et al.  Consensus method for determining groundwaters under the direct influence of surface water using microscopic particulate analysis (MPA) , 1992 .

[11]  Rodney A. Sheets,et al.  Lag times of bank filtration at a well field, Cincinnati, Ohio, USA , 2002 .

[12]  B. Wett,et al.  Flood induced infiltration affecting a bank filtrate well at the River Enns, Austria , 2002 .

[13]  Ian Flood,et al.  Neural Networks in Civil Engineering. I: Principles and Understanding , 1994 .

[14]  Wolfgang Kuehn,et al.  Riverbank Filtration: An Overview , 2000 .

[15]  J. Schijven,et al.  Removal and inactivation of viruses by drinking water treatment processes under full scale conditions , 1995 .

[16]  Christoph Merz,et al.  Sulfide oxidation and sulfate reduction in a shallow groundwater system (Oderbruch Aquifer, Germany) , 2003 .

[17]  F. Jüttner.  Elimination of terpenoid odorous compounds by slow sand and river bank filtration of the Ruhr River, germany , 1995 .

[18]  J. Schubert.  Hydraulic aspects of riverbank filtration—field studies , 2002 .

[19]  A. Tokar,et al.  Rainfall-Runoff Modeling Using Artificial Neural Networks , 1999 .

[20]  Lakhmi C. Jain,et al.  Fusion of Neural Networks, Fuzzy Sets, and Genetic Algorithms: Industrial Applications , 1998 .

[21]  Michel Detay,et al.  River bank filtration: modelling of the changes in water chemistry with emphasis on nitrogen species , 1997 .

[22]  Marcel G. Schaap,et al.  Database-related accuracy and uncertainty of pedotransfer functions , 1998 .

[23]  H. Börnick,et al.  Laboratory tests for simulating attenuation processes of aromatic amines in riverbank filtration , 2002 .

[24]  P. Martikainen,et al.  Humus transformation at the bank filtration water plant , 1994 .