Hierarchical Clustering Cuckoo Search Optimization Implemented in Optimal Setting of Directional Overcurrent Relays

Abstract

This study has proposed a hierarchical clustering cuckoo search optimization (HCSO) to solve the optimal setting problems of directional overcurrent relays (DOCRs) in the power system. Different from the randomly generated stepsize θ in traditional CSO, stepsize θ in HCSO is updated by a hierarchical clustering mechanism, which can balance the searching priority of exploration and exploitation for each solution. The superiority of HCSO is evaluated by solving the DOCRs coordination problem with the 3-bus, 8-bus, and 9-bus system, presented by linear programming (LP), nonlinear programming (NLP), and mixed integer nonlinear programming (MINLP) forms. Results show that under the same parameter setting conditions, the minimum value of the objective function by HCSO has reduced by 10.93 % , 2.93 % , and 0.46 % , respectively, compared to that of CSO in the experimented cases 2, 3, and 4. It verifies that HCSO is more competitive than CSO in solving the DOCRs optimal setting problem irrespective of the DOCRs number and formulation type.