Robust Enhancement of ADN Topology Against Load and Local Power Supply Uncertainties

Abstract

Active distribution networks (ADNs) constitute a vital component of modern power systems, integrating multiple distributed generation (DG) units. Although DG integration improves efficiency by decreasing power losses, its deployment is frequently limited by technical constraints and financial considerations. Traditionally, network reconfiguration has been applied to loss optimization in distribution grids; however, in radial networks, its effectiveness is restricted because of the limited flexibility of power flow paths. Consequently, the combined application of DG placement and reconfiguration offers a more effective strategy for reducing active losses in contemporary distribution systems. During such optimization, it is crucial to include uncertainties in load and renewable generation to ensure realistic and stable results. In this context, the present study introduces a robust optimization framework designed to maintain reliable performance despite variations in generation and consumption. The proposed method ensures that optimal DG allocation and network configuration remain stable even under moderate fluctuations in system conditions. To evaluate its performance, both robust and deterministic formulations were run for a 70-bus system, and the obtained outcomes were compared with those reported in a reference study. The outcomes prove that the proposed approach significantly reduces daily energy losses under normal and uncertain operating conditions compared with the reference method.