.png){kind=icon}
.png){kind=icon}
Return
Online calculation of distribution network power loss considering the absence of real-time node data
.png){kind=icon}
PRE
AI
M
X
P
H
Y
DOI：10.1016/j.epsr.2026.113028.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In Distribution Networks (DN), missing node data makes it hard to calculate network losses. This paper proposes a prediction model based on Edge Aggregated Graph Attention Network (EGAT), which utilizes power, voltage, and network loss measurements to calculate network losses in real-time when node data is missing. The model first conducts offline training using historical data from all accessible nodes in DN. Then, the differential evolution algorithm (DE) is employed to dynamically fill in missing data using collected real-time nodal data, ultimately achieving optimization of network losses. Simulation tests on an improved IEEE-33 node system reveal that with 15% and 30% data loss, this approach achieves high accuracy in data filling and low network loss calculation errors.
Keywords:
Distribution network loss
Online calculation
Data missing
Edge graph attention
Differential evolution algorithm
Journal
IF:
4.2
Papers:
1.1W
Citations:
2.2W
Organization
Cited Papers