TY - JOUR
T1 - Multiple home-to-home energy transactions for peak load shaving
AU - Mahmud, Khizir
AU - Nizami, Mohammad Sohrab Hasan
AU - Ravishankar, Jayashri
AU - Hossain, M. Jahangir
AU - Siano, Pierluigi
PY - 2020
Y1 - 2020
N2 - This article proposes a new technique to manage the domestic peak load demand through peer-to-peer energy transaction among multiple homes. In this process, the houses willing to sell energy are identified as the Parent, and the houses that require energy are identified as a Child. The parents having energy resources such as photovoltaics, battery storage and electric vehicles will utilize their resources to meet their peak power demand and sell the extra energy to a child. A mixed integer linear programming optimization is used to find the parent-child matching based on their energy availability, power demand, and distances. After selecting the parent-child match, the power demand of a child is forecasted using two different techniques, i.e., autoregressive moving average and artificial neural networks, to identify to child's need in a day ahead of the actual operation. The proposed algorithm calculates the available energy of a parent to sell in real-time and the required energy of a child in a day-ahead, while ensuring to minimize the peak load demand. The proposed method, as confirmed by the presented analysis using data of a real Australian power distribution network, is able to significantly minimize the peak load demand, which in-turn is expected to minimize the electricity costs. The method also facilitates two agreed prosumers to transact energy between themselves without the involvement of a third party.
AB - This article proposes a new technique to manage the domestic peak load demand through peer-to-peer energy transaction among multiple homes. In this process, the houses willing to sell energy are identified as the Parent, and the houses that require energy are identified as a Child. The parents having energy resources such as photovoltaics, battery storage and electric vehicles will utilize their resources to meet their peak power demand and sell the extra energy to a child. A mixed integer linear programming optimization is used to find the parent-child matching based on their energy availability, power demand, and distances. After selecting the parent-child match, the power demand of a child is forecasted using two different techniques, i.e., autoregressive moving average and artificial neural networks, to identify to child's need in a day ahead of the actual operation. The proposed algorithm calculates the available energy of a parent to sell in real-time and the required energy of a child in a day-ahead, while ensuring to minimize the peak load demand. The proposed method, as confirmed by the presented analysis using data of a real Australian power distribution network, is able to significantly minimize the peak load demand, which in-turn is expected to minimize the electricity costs. The method also facilitates two agreed prosumers to transact energy between themselves without the involvement of a third party.
UR - http://www.scopus.com/inward/record.url?scp=85082102816&partnerID=8YFLogxK
U2 - 10.1109/TIA.2020.2964593
DO - 10.1109/TIA.2020.2964593
M3 - Article
AN - SCOPUS:85082102816
SN - 0093-9994
VL - 56
SP - 1074
EP - 1085
JO - IEEE Transactions on Industry Applications
JF - IEEE Transactions on Industry Applications
IS - 2
ER -