Abstract Network reconfiguration (NR) serves as an effective mechanism for optimizing electricity delivery by modifying the network topology through sectionalizing and tie-switches. Dynamic reconfiguration is gaining prominence with the commercial use of remotely controlled line switches; however, unbalanced power distribution networks (UPDNs) with diverse customer categories present a significant challenge in identifying optimal configurations that simultaneously minimize losses and mitigate voltage unbalances across multiple supply phases, particularly when accounting for hourly and seasonal demand variability prevalent in the Indian electricity system. This study proposes a comprehensive, hourly- and seasonally-adaptive network reconfiguration strategy for three-phase UPDNs that remains largely unexplored in the prior literature. The approach employs probabilistic load modeling to capture demand uncertainty and extrapolate seasonal daily variations across residential, commercial, and industrial consumer categories. The network reconfiguration problem is formulated as a multi-objective optimization framework that explicitly minimizes active power loss and voltage unbalance simultaneously, solved using a Grey Wolf Optimizer (GWO) algorithm that efficiently navigates the large combinatorial solution space of hourly switching configurations while maintaining radial network topology. The proposed methodology is validated on a modified IEEE 34-bus unbalanced radial power delivery system (URPDS) representative of the Kakdwip region, West Bengal, India. Simulation results demonstrate that the GWO-based strategy achieves significant improvements in both power loss minimization and voltage unbalance mitigation across all four seasons compared to the base network configuration and Particle Swarm Optimization (PSO)-based alternatives, confirming the potential of the proposed approach for practical deployment in modern smart distribution systems operating under highly variable and unbalanced loading conditions.
Maji et al. (Mon,) studied this question.