Backtracking based algorithm in hierarchical time-extended Petri net model for power system restoration

The research on power system restoration (PSR) is attracting more and more attention. During the process of system restoration, a wide variety of control actions have to be performed on time within the operating constraints. In this paper, the actions and their relationship are modeled as a controlled hierarchical time-extended Petri net (H-EPN). In order to find a feasible sequence of actions quickly and automatically and to minimize the loss of the blackout, an algorithm based on best-first search algorithm is proposed. Different rules are marked with assessed weights by analytical hierarchy process (AHP). In addition to the classical heuristic rules, the backtracking search algorithm is employed to optimize the final solution. The result indicates that common heuristic rules cannot always find the optimal solution. The backtracking search algorithm is necessary when various heuristic rules are in collision or even with constraints. Test results on practical system are given to demonstrate the validity of the algorithm