Forecasting Intra-Day Load Profiles under Time-Of-Use tariffs Using Smart Meter Data

Authors: Y. Kiguchi, Dr. M.J. Weeks (Dept. of Economics, University of Cambridge, UK), R. Choudhary (Dept. of Engineering, University of Cambridge, UK), & Heo, Y. (Dept. of Architecture, University of Cambridge, UK)

The installation of smart meters enabling electricity load to be measured half-hourly provides an innovative demand-side management opportunity that is potentially advantageous for utility companies and customers. Time-of-use (TOU) tariffs are widely considered to be a promising solution for optimising energy consumption in the residential sector. In this paper we present the first data driven modelling of residential customer demand response following the adoption of a TOU tariff. Although much of the existing research on demand response in electricity pricing has focused on the evaluation of the impact of TOU tariffs, a practical framework with which to forecast user adaptation under different TOU tariffs has not been fully implemented.

Our modelling framework utilises statistical moments to capture a number of aspects of household lifestyles, which, in combination, characterise the flexibility of potential demand response to the introduction of TOU tariffs. One advantage of this approach is that it enables us to be agnostic about household characteristics. We evaluate our approach by analysing smart meter data from 801 households in Ireland over 6 months pre/postintervention of TOU tariff. The MAPE value in forecasting average load for a group of households with the Decision Tree method investigated is 3.38% for a day period and 2.74% in the peak time period.

Leave a Comment

Your email address will not be published. Required fields are marked *