A probability model for grid faults using incomplete information

Utilities face the challenge of responding to power outages due to storms and ice damage, but most power grids are not equipped with sensors to pinpoint the precise location of fault causing the outage. Instead, utilities have to depend primarily on phone calls (trouble calls) from customers who have lost power. This paper presents an information model of the grid in the presence of outages; the developed model is used to estimate the probability of power line faults causing the outages. However, the computational complexity of the problem grows exponentially with the number of power lines in the grid. Thus, several methods are proposed for handling the combinatorial growth of events and the behavior is demonstrated using the data of a real-power grid. Performance results show that power line fault detection can be achieved with high accuracy, even with a very low percentage of customers calling to report an outage.