Abstract

Sensitive loads in microgrids should have high uninterruptible power supply reliability. To reach this goal, the precise setting of directional overcurrent relays (DOCRs) has a key role in the protection coordination optimization procedure (PCOP). In the event of a single contingency, PCOP's creative goal is to sectionalize the location of the fault without any defects. However, there is a research gap in the literature about experiencing the cascading events in the protection coordination of DOCRs, which significantly increases the complexity and dimensions of the problem model. In this paper, an innovative method called contingency reduction mechanism (CRM) is introduced to solve PCOP. The main purpose of CRM is to reduce the complexity and dimensions of the PCOP. Next, a new algorithm called clustering-based zero-miscoordination (CBZMC) and the k-means algorithm are used to cluster the selected contingencies by CRM. As well as, the decision variables of DOCRs are achieved by four meta-heuristic optimization algorithms (MOAs) and linear programming. MOAs include Fmincon algorithm (FA), simulated annealing algorithm (SAA), pattern search algorithm (PSA), and genetic algorithm (GA). Finally, the real radial network called TESKO and the two IEEE meshed networks (IEEE 14- and 30-bus systems) are tested as case studies with high penetration of distributed generation. The results indicate the effectiveness of CRM and CBZMC in reducing the size, complexity, and solution time of PCOP with the suitable operation time of DOCRs. The PSA is known as the superior algorithm for the TESKO grid, which has zero-miscoordination. For the IEEE 14- and 30-bus systems, the FA with CBZMC method was recognized as the preferable procedure, and its convergence was fully completed with zero violations in the setting of DOCRs. On the other hand, FA has the most optimal objective function in all networks and solves the problem in the shortest period of time.