Interval characterization for active distribution networks with high penetration of PV based on dynamic decoupled mapping

High penetration of photovoltaic (PV) exacerbates the operational uncertainty of active distribution networks (ADNs). It is essential to accurately extract the nonlinear relationship between system state and PV uncertainties. Traditional affine-based methods reduce the computational complexity of uncertainty quantification, which may struggle to ensure sufficient accuracy. In this paper, an interval characterization method is proposed based on dynamic decoupled mapping. First, by treating the nodal net power with PV integration as an uncertain variable, a decoupled mapping paradigm (DMP) is constructed to accurately extract the relationship between the nodal voltage/line power and the uncertain variable. To further improve computational speed, a stepwise optimization and screening strategy is designed without compromising accuracy. Then, a dynamic characterization and adaptation mechanism is developed to express nodal net power intervals. Based on the decoupled mapping paradigm, the intervals of nodal voltage/line power are further characterized with the varying bounds of nodal net power intervals. Finally, the effectiveness of DMP is validated on a series of standard power systems. It is shown that the proposed interval characterization method can effectively improve the mapping accuracy while ensuring computational speed. It facilitates the identification of operational risks of ADNs with high PV penetration.