Improved Fault Location Using Fault Indicators and Loads Probabilistic Model

Nowadays, smart grid implementation is a new and challenging issue. One of the important parameters in the smart grids is their fast and accurate fault locator capability. This capability reduces power outage time, which the repair teams of the power distribution companies mostly lose to find the fault location in a traditional way. The fault allocation in the distribution networks, due to their topological nature (high complexity, intermediate loads, load uncertainty, no homogeneity of lines, etc.), is a very difficult and different from transmission networks. In this research, fault indicator information is used to deal with multiple result problem introduced by fault locator in distribution networks.  In this paper, a forward-backward sweep based method is introduced to find fault location. There are some uncertain parameter in fault location because of the error on the load estimation. There is correlation between these parameters due to similarity of the environmental conditions and customer behavior. Therefore, two methods (Monte Carlo and two-point estimation) are used to model probabilistic parameters on fault location. Finally, the 11-bus test network and the actual 306-bus Sharfabad feeder network in Kerman are utilized to evaluate the proposed method.