Segmentation of distribution grids: prescribing automatic reclosers as a prescriptive maintenance strategy

The increasing demand for reliable energy supply presents power distribution companies with the challenge of building resilient grids capable of minimizing interruptions. With the growing severity of extreme weather events, adopting advanced technological solutions is essential to protect grid infrastructures. This paper, part of the R&D project "Analytical and Computational Management Center for TOTEX 4.0," regulated by Brazilian Electricity Regulatory Agency (ANEEL), employs Artificial Intelligence and Data Hyperautomation to optimize asset management in power distribution networks. It focuses on the strategic prescription of automatic reclosers to segment large customer blocks and implement cascading automatic transfer logics. Using Mixed-Integer Programming (MIP), the system evaluates millions of scenarios to determine optimal recloser placement, delivering significant improvements in System Average Interruption Frequency Index (SAIFI) and System Average Interruption Duration Index (SAIDI), with measurable operational and financial impacts. Results achieved between 2020 and 2023 highlight a 31% reduction in SAIFI and an 11% reduction in SAIDI in two regions, outperforming the previous decade’s advancements. This study underscores how advanced algorithms and hyperautomation ensure operational efficiency, adaptability, and enhanced resilience in power distribution grids.