Data-Driven Approach to form Energy Resilient Smart Microgrids with Identification of Vulnerable Nodes in Active Electrical Distribution Network

We propose a methodology for identifying the optimal DERs allocation with vulnerable node identification into consideration in active electrical distribution network and named those nodes as critical nodes. Power variation in these critical nodes would significantly affect the operation of other linked nodes, thus these nodes are suitable and considered optimal for DERs placement. We demonstrated our method evaluation in a standard IEEE-123 test feeder system. Initially, we partitioned the distribution system into optimal microgrid networks using graph theory. The partitioning was validated using graph neural network architecture for suitable formation of the microgrids. Further, using an effective measurable causality analysis like granger causality, we identified critical nodes in the partitioned microgrid and placement of DERs on these nodes resulted in enhanced network reliability and resiliency. Further, to validate the system performance and energy resiliency, we computed percolation threshold for the microgrid network that indicates the system resiliency after incorporating DERs at those critical nodes. This proposed methodology for the first ensures effective microgrid partitioning, identification of critical nodes, optimal DERs allocation and system resiliency evaluation through data driven analysis approach in a distribution network.