Optimal coordination of overcurrent relays using gravitational search algorithm with DG penetration

Abstract

Modern distribution systems consist of various distributed generators (DG) to make reliable power system. In this DG integrated distribution systems, coordination of overcurrent relays is a big challenge for protection engineers. At addition of DG, distribution system experiences change in the short circuit level of the system and thus earlier relay settings causes mal operation of relays. Nowadays, various programming optimization techniques are frequently used to find optimal relay settings of overcurrent (OC) relays. This paper presents a comparative study of Particle Swarm Optimization (PSO) and Gravitational Search Algorithm for the coordination of overcurrent relay for a system containing DG. A proper combination of primary and backup relay is selected to avoid mal operation of relays and unwanted outages when DG is penetrated. A four bus radial system is simulated in PSCAD/EMTDC platform and programming is done using MATLAB software.

A Modified Gravitational Search Algorithm for Function Optimization

“…. (19) The gravitational force and repulsive force acting on particle i in dimension d are calculated by the following equations, respectively:…”

A new iterative linear programming approach to find optimal protective relay settings

“…In Reference 27, a transient stability index is included as a constraint in the OcR‐CP model including DG, and the OcR‐CP is solved by using ant lion optimizer (ALO). Several other heuristic optimization based techniques like adaptive modified firefly algorithm (AFA), 28 Gravitational Search Algorithm (GSA), 29,30 oppositional Jaya algorithm (OJaya), 31 modified water cycle algorithm (MWCA) 32 have also been suggested in the literature. Though these works have more or less provided better optimal solutions as compared with respective past literature, each of these methods are non‐traditional and, being random search based techniques, they do not guarantee optimal solution for each and every trail run.…”

“…Heuristic optimization techniques, 17‐31 sometimes do not fetch the global optimum and do not converge to feasible solutions because there is no specific algorithm exist that can achieve the best solution for a particular optimization problem. This disadvantage can be overcome by applying hybrid optimization techniques which combine both heuristic and conventional optimization methods.…”