An IoT based Smart Water Management System

An IoT based Smart Water Management System

Author Name : Bramesh S M, Puttaswamy B S, Bhoomika M, Megha S, Spoorthy B N, Uday M

DOI: https://doi.org/10.56025/IJARESM.2022.1072907

ABSTRACT

In our daily lives, water is a valuable natural resource. Presently, due to the fluctuating nature of water demand in urban and/or rural areas and also in the context of water resource scarcity, ideal management of water resources is a vital component of sustainable management. On the other hand, with the advent of Machine Learning (ML) and the Internet of Things (IoT), the pursuit of the smart water management system is also gaining momentum. Hence, we aimed at creating an IoT based smart water management system that can monitor and also predict water consumption in real-time. The proposed system consists of two main components, namely an IoT component and a Machine learning component.An IoT component’s role is to collect the water usage data in real-time using several sensors, Arduino UNO, Wi-Fi, and the cloud. The Machine learning component’s role is to analyze the collected real-time water consumption data and then forecast the consumption of the water using the machine learning algorithms (Long Short-Term Memory and Random Forest). Finally, our experimental findings reveal that the Random Forest algorithm performed better when compared with the LSTM algorithm based on R-Squared value (R²), Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and Mean Absolute Error (MAE) metrics, thereby helping us to choose the better prediction algorithm for the ideal management of water resources.

Keywords: Arduino Uno, Long-Short Term Memory (LSTM), Random Forest, Sensors, Water consumption, Wi-Fi.