Optimized Clusters for Disaggregated Electricity Load Forecasting

Authors

  • Michel Misiti Laboratoire de Mathématique d’Orsay
  • Yves Misiti Laboratoire de Mathématique d’Orsay
  • Georges Oppenheim Laboratoire de Mathématique d’Orsay
  • Jean-Michel Poggi Laboratoire de Mathématique d’Orsay

DOI:

https://doi.org/10.57805/revstat.v8i2.92

Keywords:

clustering, disaggregation, forecasting, optimization, wavelets

Abstract

To account for the variation of EDF’s (the French electrical company) portfolio following the liberalization of the electrical market, it is essential to disaggregate the global load curve. The idea is to disaggregate the global signal in such a way that the sum of disaggregated forecasts significantly improves the prediction of the whole global signal. The strategy is to optimize, a preliminary clustering of individual load curves with respect to a predictability index. The optimized clustering procedure is controlled by a forecasting performance via a cross-prediction dissimilarity index. It can be assimilated to a discrete gradient type algorithm.

Published

2010-11-11

How to Cite

Misiti , M., Misiti , Y., Oppenheim , G., & Poggi , J.-M. (2010). Optimized Clusters for Disaggregated Electricity Load Forecasting. REVSTAT-Statistical Journal, 8(2), 105–124. https://doi.org/10.57805/revstat.v8i2.92

Most read articles by the same author(s)