Identification and Prevention of Cascading Failures in Autonomous Microgrid

Abstract

This paper presents a new method to model and deal with the cascading failures in an autonomous microgrid (MG) in order to protect it against risks that may lead to its complete collapse. Although cascading failure is a term mainly associated with power system networks and transmission lines, its extension to MGs as the future structure of power systems is proposed in this paper. An MG operating as an independent entity in autonomous mode is prone to catastrophic failures due to its limited supply resources and power flow paths; therefore, a blackout model based on ac power flow is extracted that is tailored to these specifications. Based on the associated risks presented by the model for any configuration in advance of the anticipated events, remedial actions are proposed to mitigate the vulnerability of the MG by reconfiguration to alleviate the loadings of the heavily loaded lines. A genetic algorithm is used to find the optimum MG's configuration with less vulnerable lines. Critical loading is also determined for each configuration in order to indicate the necessity of taking remedial actions upon approaching to this level. The effectiveness of the proposed method is investigated by using the IEEE 33 bus sample power network.