Abstract:

We present a Monte Carlo based estimation of the probability that a given day will be a “CP day”, and the probability that each given hour be the hour during which the peak occurs. The algorithm is based on a statistical model fitted to a minimal amount of publicly available data. We explain how to use it when forecasts are not directly available. In the second part of the talk, we show how to use these estimators to optimize the operation of a battery storage system in the management of a micro-grid facing charges computed according to several CP programs.