Daily peak electrical load forecasting with a multi-resolution approach

In the context of smart grids and load balancing, daily peak forecasting has become a critical activity for stakeholders in energy industry. An understanding magnitude timing is paramount implementation grid strategies such as shaving. The modelling approach proposed this paper leverages high-resolution low-resolution information to forecast demand size timing. resulting multi-resolution framework can be adapted different model classes. key contributions are (a) general formal introduction approach, (b) discussion approaches at resolutions implemented via generalised additive models neural networks, (c) experimental results on real data from UK electricity market. confirm that predictive performance competitive with low- alternatives.