Location of outages in distribution systems based on statistical hypotheses testing

Quick restoration of power following an outage has been a challenge for power distribution engineers. Searchers have drawn attention to this problem by proposing methods for location of outages in distribution systems. The method presented in this paper for outage location is based on statistical hypotheses testing. The input data consists of time-of-outage measurements obtained from customer-ends. The effect of various levels of noise in the measurements on performance of the method is studied. A procedure for discarding outliers in the measurements is included. A theoretical procedure to determine the maximum noise allowed in the measurements is presented. Performance of the method at different locations in a test system is also studied.