Water quality monitoring Using Internet of Things (IoT) and Machine Learning (ML) for domestic application

Abstract

Water is one of the greatest blessings that nature has to offer, and it is necessary for the survival of all living things, including humans, animals, and plants. At this time, challenges are being faced all around the world in meeting the requirements for water that can be consumed. The most severely impacted by this crisis are those in underdeveloped countries. Because there is not enough adequate monitoring of water quality and quantity, the current water issue is just going to get worse as the population continues to grow. With the recent advancement in information and communication system technology, there is a growing interest in the development of smart and cost-effective solutions for water quality monitoring system (WQMS). To get a lasting solution to the above problem, we built a system containing eight sensors (PH, Temperature, TDS, Turbidity, Pressure, Volume, Color, and Flow) that can monitor the quality and quantity of the household water. All these sensors were connected with an esp32 Wi-Fi module, and all the data collected from the sensors were stored into a cloud-based server. Afterwards water quality data was visualized in the Thingspeak server in real time. Additionally, we employed two machine learning algorithms such as FB prophet and SARIMA ARIMA and predicted the future values. The primary goal of this project is to develop a water quality monitoring system (WQMS) so that the consumer can monitor the water quality in real-time and can use the safe water for drinking. This low-cost monitoring system uses the new technology Internet of Things, and Machine Learning, all of which have the potential to replace the conventional methods of water quality monitoring. Customers, vendors, and the government all can benefit from this system as communication can move in both directions. This approach will enable the government not only to obtain taxes but also help to regulate the water situation, which will lead to an increase in customer trust in the government and the suppliers. The developed system can surely use for water quality monitoring systems. To check the validation of the developed system I use two different sources (normal water source and contaminated water source) the result was as expected different sources have different results.