Overcurrent relay coordination by evolutionary programming

This paper presents a novel optimization technique based on evolutionary programming for overcurrent (OC) relay coordination in ring fed distribution networks. Since the problem has multi-optimum points, ordinary mathematical based optimization techniques will fail. Evolutionary programming is a stochastic multi-point searching optimization algorithm and is capable of escape from local optimum problems. When used in OC relay coordination, the chance of obtaining global optimum relay settings is thus drastically increased. The developed technique will make sure that the relay grading pairs will operate in correct grading order. It can also handle the resultant relay operation due to the changes of fault current distribution as a result of circuit breaker operation while the fault still exists. The relay grading requirements such as changes of fault level and system configuration, are reflected in the optimization process as constraint checking. The difficulties in the application of evolutionary programming in a highly constrained system will also be discussed. (C) 2000 Elsevier Science S.A. All rights reserved.