The Impact of Data Availability on Consumption Flexibility Estimation With Heat Pumps

Abstract

Using the inherent consumption flexibility of electrical systems is becoming increasingly important to help with grid balancing, as renewable energy sources are on the rise. This study investigates how data availability impacts the quality of flexibility predictions and the resulting difference in performance in the task of cost minimization under a dynamic electricity tariff. For a building with a Heat Pump (HP) connected to a water tank, we differentiate the cases where only HP power measurements are available and where either one or both indoor temperature and tank temperature measurements are available. The experiments show a clear advantage of the full data availability case over the others, while the simplest case can compete in the task of cost minimization, when additional assumptions are made.