Adaptive Topology Process Automation for Islanding Detection and Restoration

Active distribution power grid transitions pose challenges for monitoring and controlling operations, especially with limited system data. To address this, Micro-Phasor Measurement Units (μPMUs) have been implemented in distribution providing synchronized measurements that assist in the real-time recognition of states. However, there are inconsistencies between switching operations and measurement changes in topology identification during grid operation, particularly on distribution energy sources. Island topologies are regarded as unbalanced in this context because islanding detection and restoration on grids are based on steady states with acyclic and unidirectional energy flows in graph analysis. To avoid the information mismatch, the proposed model structures measurements and switch operations. The algorithm mimics the recognition process of grid operators controlling section switches for line energization and de-energization. The automated model proposed in this study evaluated various aspects of adaptive state recognition, including islanding detection, acyclic energization, path evaluation, and real-time optimization through graph manipulation using μPMUs.