PH modelling by neural networks. application of control and validation data series in the middle Loire river
MOATAR ; FESSANT ; POIREL
Type de document
ARTICLE DE PERIODIQUE
Langue
anglais
Auteur
MOATAR ; FESSANT ; POIREL
Résumé / Abstract
Artificial neural networks are applied to estimate the ph of the middle Loire river. The ph is modelised as a function of hydro-meteorological data such as discharge and solar radiation. The model, which was adopted for its generality and its simplicity, and also because of the availability and reliability of the significant input variables, was integrated into a system of modelling tools which facilitate the critical analyses and validation of physical chemical measurements. This system of modelling tools is currently in the process of being put into service on line by EDF to allow them to follow and critically evaluate water quality parameters with respect to hydro-meteorological conditions.
