Augmenting Path Max Flow Min Cut Algorithm
The augmenting path max flow – min cut algorithm is used to identify the minimum number of branches that need to be opened or removed from the system in order to isolate the Facility (power system device) from an External region.
The application consists of three stages:
-
Convert the electric network to a graph structure
In this stage, each branch of the system is converted to an undirected arc with graph flow capacity equal to one. Thus, only one unit of graph flow can be sent through a branch. The Facility buses and the External buses are converted to Facility and External supernodes, respectively. This effectively reduces the problem to finding the min cut between these two supernodes.
The number of nodes and capacities of the Facility and the External region are also computed during this stage. -
Find the Max Flow using the Augmenting Path Algorithm
This is an iterative process. At each step, the algorithm finds a new augmenting path from the Facility to the External region and augments the graph flow along this path in one unit. Consequently, the branches in the path won’t be available for flow augmentation in the next step. Each new path is determined using a shortest path routine.
The algorithm stops when no augmenting path from the Facility to the External region can be found. The number of units transferred from the Facility to the External region (path augmentations) reaches the number of branches in a certain cut. Note that the number of branches in the min cut can not exceed the capacity of either the Facility or the External region. -
Determine the branches in the min cut
The identification of the branches that constitute the min cut consists in tracking down labels in the buses and branches used during each path augmentation.