A Novel Two-Stage Partitioning Based Reconfiguration Method for Active Distribution Networks

This article presents a novel two-stage partitioning based network reconfiguration methodology for active distribution networks (ADNs) operating in grid-connected or islanded modes of operation. Firstly, the entire ADN is partitioned into sub-networks (SNs) using the concept of hierarchical spectral clustering (HSC) and network equivalencing. Then, the optimization of the entire network is completed in two stages. The first stage is responsible for obtaining the reconfiguration of the independent SNs. In the second stage, the entire network reconfiguration is obtained, given the switch states from the prior stage. The optimization problem is solved using modified particle swarm optimization (MPSO). Following any contingency, the ADN operates in islanded mode and the survival of the created islands is examined using the proposed adaptive load flow (ALF). Simulation results of the proposed methodology are then compared with the centralized approach. The analysis on the IEEE 33-Bus and the IEEE 118-Bus distribution network integrated with distributed generators (DGs) shows that the proposed approach significantly reduces the computation time, while the solution remains close to the one from centralized approach. The developed reconfiguration methodology is also validated in real-time using real-time digital simulator (RTDS) and dSPACE1104 (R&D controller).