Fast Resilience Assessment of Integrated Electricity-Gas Systems Considering Non-Convex Gas Flow Dynamics

This paper proposes a novel spatial branch-and-cut approach for fast resilience analysis of integrated electricity-gas systems. The proposed method makes two improvements to the traditional branch-and-cut algorithms to accelerate the solution process of the scenario-based post-contingency load restoration. First, a new cone-specific spatial branching strategy is developed to handle the non-convex nature of the integrated electricity-gas flow. Compared with the existing branching strategies which can only handle bilinear equality constraints, the scale of branching variables is reduced by two-thirds without affecting the accuracy of solutions. Second, a combinatorial cut generation strategy is proposed to convert the traversal of all failure scenarios into a guided generation of single branch-and-cut search tree. The key idea is to equate the solution process of each failure scenario to a subset of the search tree, so that the load restoration problem only needs to be solved once. The performance of the proposed approach is tested on three integrated electricity-gas systems of different sizes. The comparison against seven alternative solution methods further verifies the superiority of the proposed method in terms of accuracy and efficiency.