Fast and Robust State Estimation for Active Distribution Networks Considering Measurement Data Fusion and Network Topology Changes

With the integration of distributed generations (DGs), distribution networks are being transformed into active (ADNs). Due to ADNs‘ complex operational scenarios, massive data, and fast-changing network topologies, traditional state-estimation (SE) methods inadequate meet requirements computational accuracy, speed, robustness. Aiming at SE ADNs, this paper proposes a data-driven classic-model-integrated method, which uses an neural (NN) perform initial estimation, then linear refine estimation. It applies PMU SCADA data fusion is robust noise ADN topology changes. The simulations on IEEE standard system verify that proposed method superior in terms estimation calculation This study provides ADNS with new effective scheme, great significance context promoting development renewable energy.