Fault location in distribution system with load uncertainty analysis

Accurate fault location in distribution network of power system is an essential technology, yet facing more challenges than that in transmission systems. This study develops novel methods for feeder fault location using current/voltage sensors sparsely deployed in the network, in addition to the voltages and currents measured in the substation. Depending on the available sensor locations relative to the fault and the lateral conditions, the algorithm searches every possible path and calculates the fault distance and fault resistance by reducing the circuit to one of the two scenarios. Like all impedance based methods for feeder fault location, the proposed method relies on load values. However, the load values are difficult to obtain and time varying. In this paper, two load estimation methods, based on optimization and regression, respectively, are developed. The impact of load uncertainty on the proposed method is also studied. Simulation experiments show that the proposed fault location method is robust to load uncertainty and achieves high accuracy in various network and fault conditions.