Abstract
As large-scale distributed generation and electric vehicles are promoted, it enhances the randomness and uncertainty of distribution network. It challenges state estimation for distribution networks, which commonly have limited measurements. To improve distribution network state perception, it is critical to utilize limited measurement and distribution network topology reconstruction to achieve the overall distribution network situation awareness. This paper proposes a distribution network measurement super-resolution model based on graph convolution neural network (GCN). The limited measurements from distribution network and multiple topologies were used as training sets. IEEE 33-nodes distribution network was used to verify the proposed method and the testing results are compared and analyzed. The testing results demonstrate the proposed GCN model can achieve distribution network situation awareness with limited measurements. It shows that the model has a good performance while the distribution network topology changes.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call