Edge based Device using Machine Learning for Water Quality Management in a Smart Campus

Water contamination is a serious threat imposed with the increased population in a smart city environment. Industrial wastes, leakage from sewage systems, floods and runoff from agriculture disrupt ecosystems and threaten sustainable development. Water quality monitoring is mandate in a smart campus for immediate identification and response, thereby avoiding health issues to the inmates. Manual checking of water quality involves sending water samples for laboratory analysis, which is done on collected samples and not real time. Hence, an attempt is made to provide monitoring of quality drinking water in a smart campus environment through Internet of Things. Sensors continuously gathers data on vital indicators including pH, turbidity, conductivity and total dissolved solids. The collected data is analysed using Decision Tree Machine Learning algorithm running on a central Raspberry Pi which forecasts the likelihood of crossing safety thresholds and even the potability of the water. Also, the real-time updates on the condition of water are provided via a mobile app. Based on the predictions, the authorities can take appropriate actions, such as identifying the origins of contamination or streamlining treatment procedures. Thus, a methodology for safeguarding public health is guaranteed through the proposed technique.