An inverse-problem approach to simulating smart control of domestic electric hot water cylinders using electricity demand time-series data.

Abstract

Smart control of domestic electric hot water cylinders could be used as a relatively inexpensive option to help bridge supply–demand mismatches in highly-renewable electricity grids in many countries. Evaluation of options for the smart control of these cylinders is made difficult by the lack of domestic hot water flow time-series data. In contrast, hot water cylinder electricity demand time-series data, either directly measured or imputed from whole house data, is often much more readily available. In this paper, we propose an inverse-problem approach that uses a physical model of a hot water cylinder to determine hot water flow information from electricity demand data. Once hot water flow information has been determined, it is then possible to simulate smart control of the hot water cylinder element such that it does not interfere with the provision of hot water. We illustrate this approach by using a simple physical model of a hot water cylinder to simulate the hot water flow of 19 houses with electric hot water cylinders from measured time-series electricity data. A smart control case study is then explored where these houses are part of a solar PV microgrid and the cylinder elements are controlled so as to reduce the battery storage needed by the microgrid to reach a certain level of self sufficiency.