Aggregated peak-load analysis and demand response potential of a residential building in Osaka, Japan

Global climate change has expedited the growth of renewable generation worldwide, particularly in Japan, which aims to attain carbon neutrality by 2050. Rapid increases in rooftop photovoltaic power and grid constraints in Japan have highlighted the need for efficient demand-management strategies for residential sectors. Under this background, identification of flexible consumers can be decisive for planning demand response (DR) programmes. This study analysed longitudinal data on electricity demand from 2013 to 2014 in a residential building of Osaka Prefecture, Japan. Score analyses quantified individual DR potentials of air-conditioning (AC) loads, a promising DR resource for domestic consumers. Next, the k-medoids clustering approach classified target households and quantile regression models identified household clusters with a high DR potential. Behavioural traits such as the frequency, peak-hour propensity, and grid adversity of AC use depicted the higher AC consumption, larger dwelling size, and daytime occupancy as influential characteristics to increase the potency of DR programmes. Thus, scrutinising the consumers’ attributes and AC-usage patterns, this study revealed valuable implications for ramping up DR schemes and contributed to a sustainable energy future.