Probabilistic-based identification of coherent generators

Abstract

This paper proposes a new probabilistic identification method of coherent generators using Monte Carlo simulation and graph modeling. The simulation generates operating states defined by component availability, demand and generation. For each state, the electrical condition is assessed using AC power flow and community detection is applied to a graph representation of the system to detect the coherent generators groups. Finally, the probability of occurrence of each coherent generators group is computed. This methodology was tested on the IEEE 118-bus test system. Results shows that the approach is computationally simple and fast, which makes it very appealing for large power systems.