Measuring the relationship between time-use and electricity consumption

Panel: 9. Dynamics of consumption

This is a peer-reviewed paper.

Authors:
Philipp Grünewald, ECI - Energy Group. Environmental Change Institute, University of Oxford, United Kingdom
Russell Layberry, Lower Carbon Futures, ECI
University of Oxford, United Kingdom

Abstract

Much hope rests on our ability to reduce, avoid and otherwise manipulate energy consumption. In order to do so, it would be important to understand what we use energy for. Despite the early signs of 'big data' arriving from smart meters and other electricity consumption studies, very little is know about what the resulting demand profiles represent in terms of activities 'behind the meter'. This is especially problematic for peak demand periods, which could increase dramatically in coming years due to changes in technology and practices.

The lack of activity based energy data has two main reasons. First, until recently it didn't matter. Under the 'predict and provide' paradigm the actual service provided by energy is inconsequential. Second, measuring at sub-meter level can be very expensive and labour intensive to instrument.

Both reasons no longer apply. For demand shifting measures to be designed effectively it is essential to understand the activities (not just the appliances) involved in consumption. This paper presents a new and innovative low cost collection method that uses smart phones to collect 1 second resolution household electricity consumption profiles, while also gathering event driven activity information from household participants. The combinations of activity and consumption data bridges a disciplinary gap between social sciences and traditional engineering approaches and allows for a wealth of new insights into energy use practices.

The methodology builds on established time-use research approaches, which can produce statistically significant results on the temporal relationship of activities (e.g. cooking, washing, resting) on electricity consumption profiles. The paper further illustrates how demand responses can by analysed through a practice-theoretical perspective, providing new insights into the origin of demand side flexibility and its limitations.

The methodology itself and empirical results will be presented based on early trials using this method.