Voltage Stability in Three-Phase Unbalanced Networks: A Pseudo-Measurement Modeling Based Dynamic State Estimation Framework

The integration of distributed generation (DG) units, such as photovoltaic systems, poses significant challenges for state estimation and voltage stability analysis in unbalanced three-phase distribution networks. Traditional state estimation methods are often limited by the lack of real-time measurements and the unbalanced nature of distribution networks. To address these issues, this paper proposes a dynamic state estimation framework that utilizes pseudo-measurements derived from a Back Propagation (BP) neural network. The framework is applied to the IEEE 37-node test system, where it performs state estimation at 15-minute intervals over a 24-hour period. The results indicate that nodes close to the DG injection point experience the highest voltage fluctuations, highlighting the impact of DG variability on voltage stability in unbalanced networks. Additionally, the voltage at all nodes tends to peak at specific times following the integration of photovoltaic generation. This study provides valuable insights into the influence of DG variability on voltage stability and demonstrates the effectiveness of using data-driven pseudo-measurements for state estimation in unbalanced distribution networks.