A Mixed Formulation of Matrix Completion in Low Voltage Distribution Network State Estimation

Various distributed energy resources are integrated into distribution networks causing operational problems. Before identifying these problems and taking action, a crucial step is to identify the network states. However, sufficient measurements or full observability are difficult to achieve in low-voltage distribution networks; state estimation technologies are developed to evaluate network states based on partial measurements for system operators to take the exact actions in distribution networks. This work proposes a formulation of state estimation that exploits all available information in low observability networks and explores the effects of sensor positioning and network observability on the accuracy of state estimation through an empirical analysis. The investigation is designed to numerically simulate real-world low-voltage networks with high-penetration electric vehicles. The assessment provides results with estimated voltages and currents.