We propose a simple empirical scaling law that describes load forecasting accuracy at different levels of aggregation. The model is justified based on a simple decomposition of individual consumption patterns. We show that for different forecasting methods and horizons, aggregating more customers improves the relative forecasting performance up to specific point. Beyond this point, no more improvement in relative performance can be obtained.
@article{arxiv.1404.0058,
title = {Short Term Electricity Load Forecasting on Varying Levels of Aggregation},
author = {Raffi Sevlian and Ram Rajagopal},
journal= {arXiv preprint arXiv:1404.0058},
year = {2017}
}
Comments
Significant changes from previous version. Extension to full day ahead forecasting, added appendix of methodologies and scaling equality (not previous upper bound). Under review International Journal of Power and Energy Systems