This research project considers the theoretical analysis and computational implementation of a metaheuristic called variable neighborhood search, applied to the problem of reconfiguration of primary feeders in distribution systems of electric power with variable demand, i.e., considering various demand levels during the operation. Thus, we intend to develop various forms of the variable neighborhood search algorithm (VND, RVNS, GVNS) applied to the problem of reconfiguration of feeders in distribution systems. The strategy should also develop techniques for reducing the size of neighborhood. As part of the research project, one must also implement a specialized radial load flow algorithm to find losses of each bid solution chosen by the metaheuristic and also to verify if the proposed solution is feasible from an operational standpoint, i.e., there are no violations of current in feeders or limits of voltage in buses of the system. With the developed algorithm, we intend to test large systems existing in the literature.
News published in Agência FAPESP Newsletter about the scholarship: