A Compressive Sensing Approach for Fault Location in Distribution Grid Branches

The accurate location of faults through smart algorithms is an essential requirement for grid operators to apply automation schemes aimed at promptly isolating the fault and at restoring the power supply in non-faulty segments of the network. In the distribution system, this task is particularly challenging mainly due to the large size of the network and the low number of measurement devices deployed on the field. In such a scenario, ad-hoc techniques need to be used to have fault location algorithms able to provide reliable results based on the limited amount of available information. This work presents a fault location algorithm based on the use of compressive sensing techniques. Differently from the previous literature on this topic, this paper proposes a procedure that allows the location of faults occurring also at the branches of the distribution grid, and not only on its nodes. The paper describes the mathematical model behind the conceived algorithm and presents tests on a sample distribution grid to show the potential fault location performance.