Abstract :

Water Losses Caused By Hidden Pipeline Leakage Create Serious Environmental, Economic And Operational Challenges In Domestic, Institutional And Municipal Water-distribution Networks.Since Many Pipelines Are Installed Underground Or In Inaccessible Locations, Small Leaks May Continue For Long Periods Before Visible Damage Occurs. This Project Proposes A Machine Learning-Based Smart Water Leakage Detection, Localization And Consumption Forecasting System That Combines Internet Of Things Sensing, Real-time Data Processing And Predictive Analytics. Flow And Pressure Sensors Installed At Selected Points Continuously Collect Pipeline Measurements And Transmit Them Through An ESP32-based Communication Unit To A Centralized Monitoring Platform. The Collected Time-series Data Are Cleaned And Transformed Into Features Such As Flow Imbalance, Pressure Variation, Rate Of Change, Rolling Averages And Consumption Lags. Machinelearning Algorithms, Including Random Forest, Support Vector Machine And Gradient Boosting, Are Evaluated To Distinguish Normal Operating Conditions From Small, Medium And Major Leakage Events. The Probable Leakage Location Is Identified At Zone Level By Comparing Pressure Drops And Inlet–outlet Flow Differences Between Neighboring Sensor Points. In Addition To Leakage Management, The System Forecasts Future Water Consumption Using Historical Usage Patterns, Timebased Variables And Optional Environmental Information. The Forecasting Module Supports Short-term Demand Planning, Pump Scheduling And Efficient Resource Allocation. A Web-based Dashboard Displays Live Sensor Readings, Leakage Severity, Probable Leak Zone, Alerts, Consumption Trends And Predicted Demand. The Proposed System Provides A Low-cost, Explainable And Scalable Framework For Reducing Water Loss, Improving Maintenance Response And Supporting Sustainable Water Management. Keywords: Water Leakage Detection, Leak Localization, Consumption Forecasting, Machine Learning, Internet Of Things, Smart Water Management, Flow Sensor, Pressure Sensor, ESP32.

Published:

17-8-2026

Issue:

Vol. 26 No. 8 (2026)

Page Nos:

956 - 962

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite