Computationally Efficient Dynamic Simulation of Unbalanced Distribution Systems with PV Inverters

Distribution systems (DSs) are experiencing growing integration of Photovoltaic (PV) inverters to meet the increased load demands. PV inverters can create significant dynamics on DS and it is crucial to determine the system dynamics on all three phases as the DSs are unbalanced in nature. Current approaches for solving the dynamics of PV-integrated DSs are reliant on off-the-shelf solvers which are computationally inefficient, making them unsuitable for real-time applications. In this context, this paper presents a numerical approach for efficiently solving the dynamics of unbalanced DSs with PV inverters operating in Volt-VAr or Volt-Watt modes. The solve time of such simulations largely depends on the underlying network power flow model and the algorithm. As nonlinear solvers are iterative and computationally expensive, this work adopts a linear distribution three-phase power flow (LinDist3Flow) model to speed up the computation of dynamic analyses. Case studies are conducted on a European low-voltage feeder with five PV inverters and a 2520-node feeder with 558 PV inverters. The proposed method accurately captures the dynamics of network voltage as well as the active and reactive power of PV inverters, with minimal root mean square errors. Notably, the results show that the proposed linear approach is over twice as fast as nonlinear power flow-based dynamic models, offering a computationally efficient solution for analyzing the dynamics of unbalanced DSs.