Component-based modeling and simulation of an electrical substation for dependability study and replacement optimization

In this paper we propose a multi-component modeling approach of an electrical substation for its dependability evaluation. Several failure modes that can occur at different levels of the substation, i.e. components, protective systems, feeders, bus bars and customers, need to be taken into account with their dependences. Petri nets have proved to be a powerful modeling tool, because they permit to account, in the same modeling framework, for both the deterministic and stochastic processes defining the system evolution. A Monte Carlo simulation, based on a Petri net model of the whole system, is performed to assess various key performance indicators, such as the total operational cost, the expected number of failures and the mean availability of components and subsystems; these reliability indices can be used by decision-makers (i.e. the power system operator) in a substation dependability assessment and in asset management studies. In this paper, as illustrative example of our approach, the replacement policy of some sub-station components is investigated.