An Iot Based Water Monitoring system

Abstract

This project focuses on developing an IoT-based water monitoring system utilising sensors such as the data on water quality is gathered using an ESP32 microcontroller, turbidity sensor, pH sensor, and TDS sensor. By leveraging this data, we build and evaluate various machine learning and deep learning models, including CNN, Random Forest, SVM, KNN, and Decision Tree, to accurately forecast the quality of the water. Our goal is to enhance water resource management and ensure public health safety through precise monitoring. In this project, using a variety of sensors, we created an Internet of Things-based water monitoring system, including the ESP32 microcontroller, turbidity sensor, pH, TDS, and power unit to gather data on the water quality in real time. This data collection forms the foundation for building predictive models using machine learning and deep learning techniques to assess water quality accurately. We implemented and evaluated several models to predict water quality, achieving varying degrees of accuracy. Our CNN model from deep learning demonstrated a high accuracy of 98%. In machine learning, the Random Forest and Decision Tree models both achieved perfect accuracy rates of 100%, showcasing their robustness in handling the dataset. Additionally, the K-Nearest Neighbors (KNN) model yielded an accuracy of 98%, while the Support Vector Machine (SVM) model reached an accuracy of 91%. These results indicate that machine learning models, particularly Random Forest and Decision Tree, can provide highly reliable water quality predictions based on the sensor data collected. The high accuracy rates of our models demonstrate the potential for such IoT-based systems to offer precise and efficient water quality monitoring solutions, contributing to better resource management and public health safety.