Predictive water monitoring: A digital twin approach using decision trees
|
Full Text |
Pdf
|
|
Author |
Pavada Santosh, B. Maha Lakshmi, Md Faizan Bin Haque, Palakonda Dakshayani, Nukapeyyi Tanuja and Kanaka Vishnu Vardhan
|
|
e-ISSN |
1819-6608 |
|
On Pages
|
598-605
|
|
Volume No. |
21
|
|
Issue No. |
9
|
|
Issue Date |
July 10, 2026
|
|
DOI |
https://doi.org/10.59018/052666
|
|
Keywords |
IoT-based monitoring, digital twin, water quality monitoring, machine learning, decision tree classifier, smart water management, cloud-based visualization.
|
Abstract
This work presents a compact IoT-based Digital Twin architecture implemented for water quality monitoring. It is designed to overcome the increasing challenge of pollution and the need for real-time analysis in urban environments. Unlike conventional IoT water-quality systems that basically focus on data collection and cloud display, the proposed architecture introduces key innovations: a multi-sensor model consisting of TDS, ultrasonic, DHT11, and water flow measurements to generate a richer and more accurate representation of physical water conditions. A Digital Twin layer that mirrors real-world water behaviour in real time, enabling early abnormality detection rather than passive monitoring, and an embedded machine-learning pipeline, where a Decision Tree classifier trained on empirically labelled data will predict water quality levels and identify abnormalities through a Python-based interface.
Back