TY - GEN
T1 - Exploring the Potential of Residential Energy Sharing Communities
T2 - 2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024
AU - Panwar, Dheerendra
AU - Dodda, Suresh
AU - Mahadeva, Rajesh
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - The potential of residential energy sharing communities - especially those with photovoltaic and battery storage systems - to facilitate the transition to decentralised, sustainable renewable energy is explored in this study. The study simulates the profitability of 1,000 communities and looks at the relationship between household load profiles and whether or not they are suitable for energy sharing using actual load profile data from USA. To concisely define the high-dimensional load profiles, a comprehensive collection of electricity parameters was built from both a thorough analysis of the literature and new creations. The inability of Machine Learning (ML) models to predict community revenues effectively, even with efforts in hyperparameteroptimisation and different training dataset configurations, led to recommendations for future research topics. Therefore, a thorough examination of the load profiles of both low- and high-performing communities, theories about the favourable traits of residential load profiles for energy sharing were developed. These theories highlighted the intricate relationship between household energy habits and community energy dynamics.
AB - The potential of residential energy sharing communities - especially those with photovoltaic and battery storage systems - to facilitate the transition to decentralised, sustainable renewable energy is explored in this study. The study simulates the profitability of 1,000 communities and looks at the relationship between household load profiles and whether or not they are suitable for energy sharing using actual load profile data from USA. To concisely define the high-dimensional load profiles, a comprehensive collection of electricity parameters was built from both a thorough analysis of the literature and new creations. The inability of Machine Learning (ML) models to predict community revenues effectively, even with efforts in hyperparameteroptimisation and different training dataset configurations, led to recommendations for future research topics. Therefore, a thorough examination of the load profiles of both low- and high-performing communities, theories about the favourable traits of residential load profiles for energy sharing were developed. These theories highlighted the intricate relationship between household energy habits and community energy dynamics.
UR - https://www.scopus.com/pages/publications/85211189876
UR - https://www.scopus.com/pages/publications/85211189876#tab=citedBy
U2 - 10.1109/ICEECT61758.2024.10739007
DO - 10.1109/ICEECT61758.2024.10739007
M3 - Conference contribution
AN - SCOPUS:85211189876
T3 - 2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024
BT - 2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 29 August 2024 through 31 August 2024
ER -