Leak detection in smart water distribution network
Pooja Choudhary, Ankita Modi, B. A. Botre, Sheikh A. Akbar
Abstract
This paper presents an experimental smart water distribution network that addresses the hydraulic parameters of water i.e. flow rate and pressure for leak detection. An experimental test rig was developed in CSIR-CEERI lab and data collected from all the sensors is sent to the SCADA software using PLC. Data of infield sensors (i.e. pressure, flow, temperature, and pump frequency) is collected. Detection of leakage is performed by two techniques (i) by monitoring change in pressure, flow rate, temperature trends using SCADA (ii) by using machine learning techniques - Logistic Regression and Support Vector Machine (SVM).
Topics & Concepts
SCADASupport vector machineLeakLeak detectionComputer scienceLeakage (economics)Pressure sensorVolumetric flow rateReal-time computingEngineeringArtificial intelligenceElectrical engineeringMechanical engineeringEnvironmental engineeringPhysicsMacroeconomicsEconomicsQuantum mechanicsWater Systems and OptimizationWater Quality Monitoring TechnologiesIoT-based Smart Home Systems