Artificial Intelligence–Based Clean Water and Sanitation Monitoring

Abstract

It is now well-established that good water quality is associated with economic prosperity, reduced incidence on public health and the good functioning of the various ecosystems found in our environment. Water contamination is mostly related to both diffused (agricultural lands and geologic rock degradations) and point sources of pollution. Mauritius has many water resources which depend solely on precipitation for their replenishment. Water parameters which are of relevance include total dissolved solids (TDS), temperature, pH, electrical conductivity, turbidity, dissolved oxygen, dissolved and particulate organic carbon and major cations and anions. The traditional methods of analysis for these parameters are mostly using electrical and optical methods (probes and sensors in the field), while chemical titrations, Flame AAS and High-Performance Liquid Chromatography techniques are carried out in the laboratory. Image Classification techniques using neural networks can also be used to detect the presence of contaminants in water. In addition to basic water quality parameters, the field sensors range have been extended to cover important major ions and can now be integrated with Artificial Intelligence (AI)-based models for the prediction of variations in water quality to better protect human health and the environment, reduce operation costs of water and wastewater treatment plant unit processes.

Citation

Jogee, D., Nowbuth, M.D., Proag, V. and Probst, J.-L. (2024), "Artificial Intelligence–Based Clean Water and Sanitation Monitoring", Fowdur, T.P., Rosunee, S., Ah King, R.T.F., Jeetah, P. and Gooroochurn, M. (Ed.) Artificial Intelligence, Engineering Systems and Sustainable Development, Emerald Publishing Limited, Leeds, pp. 69-80. https://doi.org/10.1108/978-1-83753-540-820241006