Neural network for classification waters sample of shatt al Arabe River

Abstract

In this study, we develop an automated system that uses a collection of attributes to classify water samples from the Shatt al Arabe River (EC, Cl, Ca, Mg, Na, and SO4). There are three steps to the water categorization system: First and foremost, in this manner From October 2009 to September 2020, water samples were taken monthly from eight locations along the Shatt Al-Arab River in Basra. Qurna, Tubular Bridge, Al Muzairah (station 1), Saad Birdge (station 2), Al-Karma (station 3), Al-Sandbad (station 4), Al-Ashar (station 5), Abo Al-Kasseb (station 6), Al-Seba (station 7) and Al-Foa (station 8) are among these locations (station 8). 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, there are neural networks with three hidden layers are used for training and testing purposes. The experimental results on the collected database show that the proposed approach achieves high accuracy in automatic water classification (normal and abnormal). The Project is designed by Matlab.