Neural network for classification of water samples of Shatt Al Arab River

Abstract

In this study, we developed an automated system that uses a collection of attributes to classify water samples from the Shatt al Arab River (EC, Cl, Ca, Mg, Na, and SO4). The water categorization system was involved three steps: first, water samples were collected (monthly) from eight locations along the Shatt Al-Arab River in Basra from October 2009 to September 2020. Second, using established procedures measured and analyzed chemical elements such as electrical conductivity (Ec), chloride (Cl), calcium (Ca), magnesium (Mg), sodium (Na), and Sulphate (SO4). Finally, neural networks with three hidden layers were used for training and testing purposes. The Project was designed by Matlab. The experimental results on the collected database showed that the proposed approach achieves high accuracy in automatic water classification (normal and abnormal).