A Non-Convex and Non-Iterative Approach for Fast Vulnerability Analysis of Smart Grids

The computational burden of vulnerability analysis of power networks depends heavily on the management of non-convexities and N-k contingencies. In this paper, a novel spatial branch-and-cut is proposed for the fast screening of vulnerable components. To address the non-convexities in the form of conic equalities, a conic-specific spatial branching strategy is designed to prune the solutions which violate non-convex constraints. To accelerate the process of contingency screening, the modeling of post-contingency load restoration for each contingency is unified and explored through combinatorial cuts. The core idea of the proposed assessment approach is to manage both non-convexities and contingencies in a single spatial branch-and-cut search tree through callback functions, so that the screening for all possible contingencies can be conducted only once. The performance of the vulnerability analysis approach is evaluated on two testing systems. Numerical results show the effectiveness and efficiency of the proposed method.