ϑ Abstract--Para-consistent Logic is a non-classic logic whose foundations allow for the contradictions treatment without invalidating the conclusions. Several researches have been demonstrating that through algorithms named PANs -Paraconsistent Analysis Nodes is possible to establish equations determined through signals representing restrictions, risks and configurations of electric power systems networks. In this paper we presented a model with this logic where each electric power grid component is modeled to create a Net of Analysis capable to offer a risk degree for analysis and diagnosis. In that way the model offers a diagnosis and the optimal restorative strategy proposition to the electrical power after an outage. The Paraconsistent logic type used for the models is the Annotated Paraconsistent Logic (APL). The APL utilization brings certain advantages in comparison with the classic logic because allow to manipulate contradictory signals, and like this presenting a faster and reliable action to decision in some situations where the information can be vague, ambiguous or inconsistent. This method is being validated through off-line tests applied at the Eletropaulo-Eletricidade de São Paulo, subtransmission power system. Eletropaulo is an electric power distribution utility of São Paulo state, in Brazil.