Load Control Strategies for Electric Water Heaters with Thermal Stratification

This semester project investigates strategies aimed at controlling the aggregate power demand of large groups of domestic electric water heaters. An existing model of a single electric water heater that considers thermal stratification along the height of the hot water tank was tuned and verified using experimental data. The tuned model was then incorporated into an existing model of a water heater population and combined with an existing model representation of typical water draws. This framework was then used as a test-bench for a number of different control strategies. Simulations were carried out on populations of 500 water heaters with heterogeneous parameters. The developed control strategies differ with regard to the assumed amount of information feedback and the assumed capability of the controller with respect to interacting with the internal duty cycle of individual water heaters. It was found that the best overall performance was achieved by an external controller that receives direct temperature information from the location of the device’s internal thermostat and has full control over the heating element’s ON/OFF state. Controllers that are only capable of blocking ON signals provide a similar performance unless the target power profile significantly exceeds the autonomous (uncontrolled) demand for long periods of time. It was also found that external controllers that only receive feedback in the event of temperature limit violations can provide an alternative to controllers that receive constant periodic temperature information, albeit with small sacrifices in terms of control quality and end-user comfort.