Optimal Switch Placement to Improve the Reliability of Distribution Network in Interconnected Network of Microgrids Using a Graph-Based Approach

Ensuring reliable electricity supply to customers is a primary objective of the power distribution system. Over the past years, the reliability assessment has witnessed significant advancements due to the emergence of distributed generation (DG) and the introduction of the microgrid concept. Consequently, it is crucial to examine the reliability of distribution grids by considering microgrids. Hence, this article addresses the problem of determining the best placement for switches in interconnected microgrids. This paper presents an approach for optimal placement of switches in interconnected network of microgrids. This method can be easily adopted and applied to any network of any size. Using a graph-based approach, a method is suggested, which can accelerate the solution of switch placement problem. Any optimization tool can be employed for solving the switch placement problem using this method. To show this feature, in this paper, accelerated particle swarm optimization (APSO) and genetic algorithm (GA) are utilized to solve this problem. The objective is to locate automatic switches in the network of interconnected microgrids. The article presents simulation results for a network of interconnected microgrids to demonstrate the effectiveness of the proposed method and assess the impact of microgrid interconnections on the reliability of distribution networks. The simulation results confirm the efficiency of the employed process and highlight the positive effect of microgrid interconnections in enhancing the reliability of distribution networks. For effective demonstration of the influence of connecting microgrids and the existence of DGs, the test system is simulated across distinct scenarios. To provide a thorough comparison, this method's advantages are discussed from financial and reliability perspectives. Moreover, to show the scalability and proficiency of this method, it is applied on a modified IEEE 123-node test system. The results are presented and discussed to validate the performance of the method. The results showed that using this approach can reduce the runtime of the solution process by roughly 20%.