Hybrid optimization algorithm applied to adaptive protection in distribution systems with distributed generation

Abstract

The expansion of distributed generation (DG) to supply the energy demand is a reality in the modern electricity sector. However, it may cause changes in the load current and short-circuit current levels of feeders and transmission lines, among other undesirable impacts, which impair the coordination of protection schemes that were formerly designed for a radial configuration. This work proposes a method for optimizing the coordination of overcurrent relays based on computational intelligence techniques applied to adaptive protection in the context of DG. The proposed hybrid algorithm uses fuzzy logic to control the setting currents and genetic algorithms (GAs) to obtain the time dials and curves of the relay. In order to validate the proposed technique, a case study based on a 13-bus test distribution system is analyzed to compare the proposed method with two other protection methods widely known in the literature. ATPDraw software is used for the system simulation and MATLAB software allows implementing the protection algorithms. By using the proposed solution, it is possible to obtain a protection system with improved response with respect to the power system dynamics, thus enabling improvements in sensitivity and minimizing the trip times of the relays without violating the selectivity criteria.