Overcurrent relay coordination by Differential Evolution algorithm

In this paper, three modified Differential Evolution algorithms based on local neighborhood search (LNS) is used to calculate the optimal relay settings of directional overcurrent relays in power systems. The process of LNS is done by exploring the vicinity of the particle having the best fitness function value (global best particle) with the help of mutation operators based on different probability distributions viz. Laplace, Cauchy and Gaussian. The proposed schemes named as LMDE, CMDE and GMDE help in enhancing the exploration and exploitation capabilities of DE without imposing any serious threat on the number of function evaluations. Two models are considered namely IEEE 3-bus model and IEEE 4-bus model to check the efficiency of the algorithms proposed in this paper. The results obtained by all the DE algorithms are compared with the basic Differential Evolution (DE) and eleven other algorithms available in the literature; the numerical results show that the proposed algorithms are able to provide superior results in comparison to the other algorithms.