Toward Detection of Distribution System Faulted Line Sections in Real Time: A Mixed Integer Linear Programming Approach

With smart fault indicators (SFIs) being widely deployed in distribution networks, availability of SFIs statuses and their current measurement provides an opportunity to locate distribution system faulted line sections in real time. However, the integrity of SFIs data together with complex fault scenarios poses a challenge for fault diagnosis, especially with the presence of distributed generators (DGs). To solve this issue, this paper proposes a new method based on mixed integer linear programming (MILP) for identification of the faulted line sections using incomplete and incorrect SFIs statuses. By minimizing the discrepancy between the expected SFI statuses and the available evidence at the distribution operating center, the proposed approach is able to handle complex fault scenarios with multiple faults and failures or malfunctions of SFIs. When DGs are present, the current measurement of SFIs and their statuses are jointly used. The result is a robust algorithm for detection of faulted line sections of distribution systems with DGs in real time. In addition, a new technique is proposed to convert nonlinear logical constraints into a linear combination of decision variables. The resulting linearity of the developed optimization model ensures its optimality. Simulation results based on a utility feeder demonstrate the effectiveness of the proposed approach.