Deployment of real time effluent treatment plant monitoring and future prediction using machine learning

Abstract

Industry and civic society are unaware of polluted water’s quality, quantity, and environmental impact. On the other hand, unregulated extraction of groundwater, inefficient use of water at various stages of production, structural challenges in plumbing, lack of low-cost reliable meters, inaccurate data and tampering issues, inability of environmental regulation, and a manpower shortage to inspect the unit at regular intervals across thousands of factories necessitate the development of an automated system for effluent treatment plant monitoring. In this study, we design a cost effective, realistic water quality and quantity monitoring system for different stages of industrial production, with real time data for underground water extraction. All the collected data will be uploaded to a server and displayed on an online dashboard in real-time. The dashboard will be shared by both industries and government officials. We deployed machine learning to provide real-time predictive analytics on water quality and quantity. We automated the effluent treatment plant processes by testing the water quality and quantity in real time and sending appropriate instructions to the respective stakeholders. The industries can be aware of the water quality and quantity in each stage of production by monitoring the data before releasing the water in the environment. This project will help to achieve current and future national and international water compliance, and several sustainable development goals.