2012
Vicente, H.; Dias, S.; Fernandes, A.; Abelha, A.; MacHado, J.; Neves, J.
Prediction of the quality of public water supply using artificial neural networks Journal Article
Em: Journal of Water Supply: Research and Technology - AQUA, vol. 61, não 7, pp. 446-459, 2012, ISSN: 00037214, (cited By 38).
Resumo | Links | BibTeX | Etiquetas: Alentejo; Portugal, artificial neural network; concentration (composition); monitoring; numerical model; physicochemical property; prediction; water quality; water supply, Chemical parameters; Formal model; Mean absolute deviations; Mean squared error; Network structures; Public water supply; Sampling frequencies; Surveillance program; Test sets; Training sets; Water quality parameters, Forecasting; Manganese; Neural networks; Nitrates; pH; Potassium; Sodium; Topology; Water quality; Water supply, Quality control
@article{Vicente2012446,
title = {Prediction of the quality of public water supply using artificial neural networks},
author = {H. Vicente and S. Dias and A. Fernandes and A. Abelha and J. MacHado and J. Neves},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-84872189812&doi=10.2166%2faqua.2012.014&partnerID=40&md5=21439d46a149b88ee4c82a7bee0e14ce},
doi = {10.2166/aqua.2012.014},
issn = {00037214},
year = {2012},
date = {2012-01-01},
journal = {Journal of Water Supply: Research and Technology - AQUA},
volume = {61},
number = {7},
pages = {446-459},
abstract = {The Health Surveillance Program was established by the Regional Health Authority of Alentejo to control the quality of public water supply. This authority divides the water quality parameters into three distinct groups, namely P1(pH and conductivity), P2 (nitrate and manganese) and P3 (sodium and potassium), for which the sampling frequency is dissimilar. Thus, the development of formal models is essential to predict the chemical parameters included in group P2 and included in group P 3, for which the sampling frequency is lower, based on the chemical parameters included in group P1. In the present work, artificial neural networks (ANNs) were used to predict the concentration of nitrate, manganese, sodium and potassium from pH and conductivity. Different network structures have been elaborated and evaluated using the mean absolute deviation and the mean squared error. The ANN selected to predict the concentration of nitrate, sodium and potassium from pH and conductivity has a 2-18-14-3 topology while the network selected to predict the concentration of nitrate and manganese has a 2-19-10-2 topology. A good match between the observed and predicted values was observed with the R2 values varying in the range 0.9960-0.9989 for the training set and 0.9993-0.9952 for the test set. © IWA Publishing 2012.},
note = {cited By 38},
keywords = {Alentejo; Portugal, artificial neural network; concentration (composition); monitoring; numerical model; physicochemical property; prediction; water quality; water supply, Chemical parameters; Formal model; Mean absolute deviations; Mean squared error; Network structures; Public water supply; Sampling frequencies; Surveillance program; Test sets; Training sets; Water quality parameters, Forecasting; Manganese; Neural networks; Nitrates; pH; Potassium; Sodium; Topology; Water quality; Water supply, Quality control},
pubstate = {published},
tppubtype = {article}
}
The Health Surveillance Program was established by the Regional Health Authority of Alentejo to control the quality of public water supply. This authority divides the water quality parameters into three distinct groups, namely P1(pH and conductivity), P2 (nitrate and manganese) and P3 (sodium and potassium), for which the sampling frequency is dissimilar. Thus, the development of formal models is essential to predict the chemical parameters included in group P2 and included in group P 3, for which the sampling frequency is lower, based on the chemical parameters included in group P1. In the present work, artificial neural networks (ANNs) were used to predict the concentration of nitrate, manganese, sodium and potassium from pH and conductivity. Different network structures have been elaborated and evaluated using the mean absolute deviation and the mean squared error. The ANN selected to predict the concentration of nitrate, sodium and potassium from pH and conductivity has a 2-18-14-3 topology while the network selected to predict the concentration of nitrate and manganese has a 2-19-10-2 topology. A good match between the observed and predicted values was observed with the R2 values varying in the range 0.9960-0.9989 for the training set and 0.9993-0.9952 for the test set. © IWA Publishing 2012.