Fast and Robust State Estimation for Active Distribution Networks Considering Measurement Data Fusion and Network Topology Changes
Dai Wan,Miao Zhao,Guidong He,Liang Che,Qi Guo,Qianfan Zhou,Wan,Zhao,He,Che,Guo,Zhou
DOI: https://doi.org/10.3390/su151813800
IF: 3.9
2023-09-16
Sustainability
Abstract:With the integration of distributed generations (DGs), distribution networks are being transformed into active distribution networks (ADNs). Due to ADNs' complex operational scenarios, massive data, and fast-changing network topologies, traditional state-estimation (SE) methods are inadequate to meet the requirements of computational accuracy, computational speed, and robustness. Aiming at the SE of ADNs, this paper proposes a data-driven and classic-model-integrated SE method, which uses an SE neural network (NN) to perform an initial estimation, and then uses linear SE to refine the estimation. It applies PMU and SCADA data fusion and is robust to noise and ADN topology changes. The simulations on the IEEE standard system verify that the proposed method is superior to traditional SE methods in terms of estimation accuracy, calculation speed, and robustness. This study provides ADNS with a new effective estimation scheme, which is of great significance in the context of promoting the development of renewable energy.
environmental sciences,environmental studies,green & sustainable science & technology