Machine Learning based Real Time Water Quality Monitoring System

Deterioration of natural water systems including lakes, streams, and estuaries is a major problem that has to be addressed immediately. Water pollution has far-reaching consequences that touch on every facet of life. Effective management of water resources is essential to maintain pristine water quality. If data are continually examined and groundwater quality is forecasted in advance, the consequences of water pollution may be lessened. There have been a number of studies done on the topic but more are required to confirm that the present approaches to water quality management are reliable and successful. The main goal of this study is to use ANN in tandem with an analytical strategy that accounts for water quality parameters to create a model for estimating groundwater supplies. The water quality data used in this study was gathered in 2014 at a 6-minute interval. The USGS uses the online National Water Information System (NWIS) to compile the data. Quantitative data on four factors that affect water quality are presented here. Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and regression analysis are all methods used for assessing performance. Additionally, this research assesses existing water quality forecasting work and suggests ways forwarded. The most accessible and economical sources of drinkable water are those found at or near the earth’s surface and in underground water tables.