Estimate Three-Phase Distribution Line Parameters With Physics-Informed Graphical Learning Method

Accurate estimates of network parameters are essential for modeling, monitoring, and control in power distribution systems. In this paper, we develop a physics-informed graphical learning algorithm to estimate three-phase Our proposed uses only readily available smart meter data the series resistance reactance primary line segments. We first parametric physics-based model replace black-box deep neural networks conventional (GNN). Then derive gradient loss function with respect use stochastic descent (SGD) physical parameters. Prior knowledge is also considered further improve accuracy estimation. Comprehensive numerical study results show that our yields high outperforms existing methods.