Water Consumption Monitoring with Fixture Recognition Using Random Forest

• Main Authors: Somontina, James Adrian B, Garcia, Felan Carlo, Macabebe, Erees Queen B
• Format: text
• Published: Archīum Ateneo 2019
• Subjects: fixtures; water resources; monitoring; real-time systems; water conservation; machine learning; IEEE regions; water consumption monitoring; random forest; fixture recognition; Electrical and Computer Engineering
• Online Access: https://archium.ateneo.edu/ecce-faculty-pubs/14; https://ieeexplore.ieee.org/abstract/document/8650112/
• Institution: Ateneo De Manila University

Water is a limited resource that is essential for sustaining life but is often wasted. By monitoring water consumption in real time, users become aware of the amount of water that they utilize. This contribution presents a water monitoring system that determines the volume of water consumed for the following events: use of faucets, toilet flush, and shower. The system utilized a non-intrusive, single point sensing technique. The sensor was interfaced to a microcontroller to monitor fluid dynamics in real-time. Data was collected and processed by a microcomputer that uploads to a graphics user interface for easy viewing and online access. Random forest was used so that the system can identify and determine how much water each faucet or fixture uses. The volume of water used for every event was determined with 92.91 % accuracy and the fixture recognition algorithm resulted in 63% precision.