Distribution system, generalized stochastic Petri net, grid island, Bottom-up restoration
Abstract
When a major outage has occurred in an area with constant disruptive storms, a coordinated decision to safely restore power in the region is critical in the recovery efforts. However, the certainty of grid status can be hard to predict, especially determining if a communication infrastructure is devastated. In this paper, characterizing the potential restoration likelihood is formulated based on a generalization of the restoration process. Such cluster of restorative methodology is proposed as a decision tool for power service restoration. The milestone captures models of transitions, probabilistic estimation, and statuses for generic power system components that are developed to establish the network behavior. Multiple test scenarios have been developed to validate the proposed methodology. Results indicate that after a catastrophic event, the proposed approach can suggest the steps for the switching sequence of the TD system during the grid recovery process based on the power monitoring devices. This approach can be integrated as a planning tool for grid planning.
