Comparative study of power grid centrality measures using complex network framework

New closeness and betweenness based centrality measures have been evaluated in this paper. Power grid is modeled as a directed graph. The graph is analyzed in terms of complex network theory to identify influential nodes which control power flow pattern throughout the whole grid and as a result can create cascade if removed unintentionally or targetedly. Various measures of impacts have been analyzed to show that power grid has scale-free network characteristics, i.e., it is very much vulnerable to targeted node removal. Measures of impacts include characteristic path length, connectivity loss and blackout size. Rank similarity analysis have been carried out to show that nominal condition of power system gives critical nodes which remain critical with changes in system operating conditions as well.