A nonlinear binary programming model for fault section estimation in electric power systems

This paper proposes a method to solve the fault section estimation (FSE) problem in power systems by developing a nonlinear binary programming model using the AMPL mathematical programming language. This model is developed based on the minimum set covering problem that contemplates the philosophy and logic of the protection system employed by electrical power companies and associates the alarms of the protective relays reported by the SCADA system with the expected states of the protective relays' functions. The proposed method was evaluated using a benchmark test system found in the related literature and the results show its efficiency, correct diagnosis, computational robustness, with the advantage of not needing to calibrate any control parameter or training in order to solve the fault section estimation problem in power systems.