Simulation of Groundwater Quality Characteristics using Artificial Neural Network

  • O.O. Fadipe
  • L.K. Abidoye
  • J.O. Adeosun
  • B.B. Oguntola
  • O. Adewusi
  • T.E. Okeowo
Keywords: water quality, dug wells, ANN, physico-chemical characteristics, microbiological characteristics


This paper reports the study of groundwater quality assessment in Boluwaduro community, Ofatedo in Osun State. In addition, it utilized the Artificial Neural Network (ANN) tool in MATLAB Software to simulate the water quality parameters/contaminants. Water samples were taken from 18 randomly selected dugwells and subjected to physico-chemicals and microbiological analysis. The mean concentrations of nitrate, nitrite, lead and iron are 20.12 mg/L, 0.78 mg/L, 0.159 mg/L and 0.35 mg/L respectively. Total plate counts range between 27 – 96 cfu/mL with growth in all the water samples. The ANN structure was trained in several rounds till satisfactory output was obtained with correlation value of R2 = 0.97. Simulation of the pH using ANN provides a good match at 10% increment of chloride, nitrate and iron and the pH value of the water sources increased with the corresponding increase in the concentrations of the parameters. The generated model for TDS gave a good prediction with total hardness and magnesium respectively. The concentrations of some metals in the wells are not safe for drinking; it could pose danger to users of the water sources. It is therefore recommended that the wells in the community should be subjected to routine monitoring and treatment of the contaminants should be enforced.


Journal Identifiers

eISSN: 2467-8821
print ISSN: 0331-8443