Non-Invasive Water Flow Sensing for Smart Water Heater Controller

Water flow rate sensors in residential homes have the capacity to revolutionize energy conservation by providing real time, user specific water usage data to the Internet of Things. Water heaters typically heat water to a constant set-temperature throughout the day. This constant heating contributes to about 18% of total home energy usage, making water heaters an especially effective target for potential energy savings. By harnessing the knowledge of hot water usage and flow rate, machine learning processes can determine an optimized water heating schedule for individual users. However, current methods to determine water flow rate involve either a complicated installation process or use of expensive equipment. The work in this paper proposes an economical, non-invasive package to both detect hot water usage and measure the flow rate, by utilizing three temperature sensors. Processing of the data quantitatively correlates temperature change of the incoming/outgoing water pipes to the water flow rate through the pipes. To accomplish this, the principle of energy conservation was applied using transient temperature measurements taken from the outer surface of both the cold inlet and hot outlet pipes. In the process of formulating energy conservation equations, there exist unknowns which will be determined by different self-testing algorithms. Developing transient and steady state equations for the inlet and outlet pipes allowed for calculations of the flow rate through a water heater to be performed. Specified conservation equations applied to both cold inlet and hot outlet pipes will enhance the accuracy and reliability of the proposed method. For verification, experimental setup was built to verify our model by comparing actual usage and flow rate measurements from a household water heater with the calculated usage and flow rate from the temperature change rate.