This report analyses the ’10 of 10’ baselining methodology used in the AEMO-ARENA Short Notice Reliability and Emergency Reserve Trader (SN RERT) trial.
In 2017, ARENA and AEMO entered into a Memorandum of Understanding to jointly develop ‘proof of concept’ projects that support the integration of renewable energy into the energy market, while maintaining system reliability and security. As part of this initiative, a three-year Demand Response (DR) Short Notice Reliability and Emergency Reserve Trader (SN RERT) trial (RERT Trial) was developed to provide evidence to inform DR’s role in maintaining grid security and reliability.
A key characteristic of DR, as compared to energy consumption, is that it cannot be measured directly. It is estimated by comparing actual consumption with a prediction of what would have occurred if the request for DR had not been made. There are several approaches for generating this prediction, of which baselining – using the history of the site’s demand – is the most common.
This report analyses the relative accuracy, bias and precision of the ‘10 of 10’ baselining methodology used to predict the actual metered load for participants in the RERT Trial. Put simply, the ‘10 of 10’ methodology uses the consumption of the 10 most recent qualifying days to construct a baseline.