An Interactive Partitioning Algorithm-Based Electrical Power Crisis Management for Service Restoration with Existing Black-Start Resources Considering Load Priority

Abstract

This article explores a new clustering-based solution for power systems that would be restored by post-outage power supply. The suggested solution has divided the overall network into separate clusters in such a way that each cluster has at least one ‘black-start unit,’ which can restart its own generation. The proposed algorithm manages a variety of restrictions when creating clusters, such as black-start generation capability, supply–demand balancing, and load priority, and attempts to minimize the active and reactive power difference between black-start unit and cluster loads as well as minimize power losses. This paper focuses on the standby available sources in a network to act like ‘black-start units,’ while their neighboring high-priority loads as defined by the supply provider were called cluster participants. K-means neighbor clustering algorithm was used to form the clusters, and, before validating each cluster, the steady-state voltage stability of the clusters was tested using equivalent network L-index value analysis technique. Two separate case studies were performed on the modified distribution IEEE 33-bus system and New England electrical transmission system with 39 buses to demonstrate the efficacy of the proposed algorithm. The results of the simulation show that each cluster was consistently able to maintain reasonable voltage stability with minimum power imbalance between available BSU and loads.