Search

Over the last decade, the electricity sector has seen a significant increase in the number of residential battery systems, as well as increasing interest in using them to reduce demand during network peaks. Although there is an abundance of literature assessing this ability using modelled residential batteries, there is a lack of detailed assessment using deployed residential batteries. This paper analyses 1-min resolution data from 15 non-coordinated residential batteries deployed in Australia across 6 network peak demand periods. A novel metric was used to quantify errors in BESS load-following, which occurred when the batteries did not completely mitigate grid import and export even when they had sufficient energy capacity and rated power. On average the 15 batteries discharged around 25% of their rated power during network demand peaks, whereas those that load-followed discharged around 40%. Despite the small sample size, these results suggest that the outcomes from modelled batteries represent the ideal upper bound and the actual performance of some batteries is likely to be lower. This there is a need for more research into the actual operation of deployed batteries, and what this means for the current modelled findings regarding their ability to reduce demand during network peaks. • Analysis of operation of residential batteries (BESS) was performed using 1-min data. • Examined 90 cases of response from uncoordinated BESS during demand peaks. • In 44 cases, the uncoordinated BESS reduced household grid import on average by 90%. • Those BESS discharged on average 40%–50% of their rated power during the peaks. • Coordinating residential BESS may potentially double their demand peak reduction. [ABSTRACT FROM AUTHOR]