Abstract
The large-scale deployment of smart meters has led to significant amount of electricity demand data available, driving it into the realm of Big Data. It is a major challenge to exploit this Big Data in order to characterise electricity use patterns and to support demand response policies. In this paper, we perform a featured-based cluster analysis on nine building archetypes (hospitals, schools, offices, hotels, flats, houses etc.) to identify electricity use patterns. Then, four metrics are developed, which are entropy, load curviness, peak intensity and index of hourly ramp rates, to measure these archetypes’ suitability to be involved in demand response schemes. A significant difference in electricity use patterns between the archetypes is found, as well as among the seasons and days of the week. We present a number of metrics for each archetype to establish which type of archetype should be prioritised for demand response programmes in terms of peak management, ramp rates as well as demand flexibility. A key finding of our study is that households offer more demand flexibility than the non-domestic sector and should therefore be incentivized to participate in dynamic electricity tariffs.
Power Precise Load Shedding Control Strategy Considering User Demand Response Clustering