Two-stage Solving Algorithm for Online Decision-making on Multi-period Distribution Network Restoration

If the distribution network loses the support of upper-level grids after an power outage, multiple local sources can be coordinated to restore critical loads, decreasing the losses and enhancing the resilience. To realize optimized allocation of limited generation resources, the restoration problem in this case can be formulated as a multi-period restoration problem. However, due to the integer variables and nonconvex power flow constraints, it is difficult to solve the multi-period restoration problem efficiently so as to satisfy the online decision-making requirement. This paper proposes a two-stage solving method to efficiently solve the multi-period restoration problem. In the first stage, a maximum spanning tree heuristic (MST) based algorithm is proposed to determine the radial topology. In the second stage, an iterative algorithm based on integer variable relaxation is proposed to determine the load status. Finally, the modified 62-bus distribution system is used to demonstrate the proposed algorithms of the two-stage solving method. The results show that the proposed method is effective and superior, and the proposed algorithm is promising in online application to assist the restoration decision of operators. © 2021 Automation of Electric Power Systems Press.