User-Transformer Connectivity Relationship Identification Based on Knowledge-Driven Approaches

Accurate user-transformer connectivity relationship (UTCR) plays a key role in fine management of low-voltage distribution network (LVDN) i.e., load expansion, line loss management, and electrical service restoration after outage. Limited data low discriminability noise increase the difficulty to identify UTCR for existing analytics methods. To overcome these hurdles, this paper proposes novel algorithm which combining preprocessing with multi-dimensional priori knowledge based on voltage characteristics LVDN. Firstly, prior related are refined account correlation users at different locations provide theoretical foundation. Then, Z-score principal component analysis combined standardize extract features from original magnify differences between reduce impact noise. Further, basis characteristics, knowledge-driven identification model is proposed wrong their real UTCR. Finally, performance verified simulated LVNDs. The comparison method other published methods number components accuracy also investigated. results indicate that achieves higher recognition than data.