Optimal reconfiguration of radial distribution networks with renewable energy resources by considering configuration shift steps

The optimal reconfiguration of distribution networks (DNs) is a critical strategy to improve the operational efficiency of modern power systems considering the increasing integration of renewable energy resources. Existing approaches, including heuristic algorithms and Mixed-Integer Linear Programming (MILP) techniques have demonstrated their relevance in addressing important objectives for DNs, such as loss minimization, feeder load balancing, and reliability maximization. However, these methods have some limitations, including single-time formulation, absence of explicit modeling of the transition steps during the shift period from the initial configuration to the optimal one, and the approximated representation of zero-impedance branches, which are necessary for accurate system modeling. This study introduces a novel MILP-based formulation for the multi-period optimal reconfiguration of DNs based on the knowledge of multi-temporal generation and load profiles. The model also considers practical operational constraints, including Rapid Voltage Change (RVC) limits and realistic switching sequences. In particular, the maneuvering process is explicitly modeled, enabling the identification of feasible transition paths while avoiding intermediate configurations that would violate network constraints. Moreover, an efficient formulation to include breakers and bus-ties as zero-impedance branches is developed, which improves the accuracy of the electrical model and ensures a more realistic representation of the DN topology. Simulation results on the IEEE 12-bus and IEEE 69-bus systems confirm the effectiveness of the approach: objective functions are close to ideal values even with a limited number of switching actions, while computation times remain below practical operational limits for the 69-bus test case, demonstrating the method's scalability and suitability for real-time operation.