Liquid/Fluid Usage Control using Machine Learning for Smart City and Industries
Saad Hafeez, University of Engineering and Technology, Taxila, Pakistan.
Farrukh Zeeshan Khan, University of Engineering and Technology, Taxila, Pakistan.
Corresponding Author:
Saad Hafeez (saad.hafeez@students.uettaxila.edu.pk)
Abstract:
Water is wasted needlessly in many parts of the globe because water use is not regulated, measured, or monitored. Uncontrolled water use leads to irregular levels in water reserve tanks, which indirectly raises the cost of energy for the motor that pumps water to the overhead tank. Like Pakistan, most nations have distinct peak and off-peak energy pricing for the cost of units as well as a set municipal corporation water schedule that fills the underground tank. Modifying the old method of float valves to keep the reserve tanks to a required defined level and keeping all the above problems in mind, we propose an ML-based method of making the system smart enough to decide the filling of tanks as per the given consumption data. Based on past related studies and our experiments, we trained and tested the algorithms on a dataset of liquid usage data collected from a smart city and suggested the best-suited algorithm. The system was able to decide the filling as per consumers' time pattern along with a peak-hour restriction placed to save electricity.
Keywords:
Water Usage; Fluid Usage; Consumption Control; Machine Learning; Electricity Saving; Water Saving; Water Smart Metering