Optimal Restoration of Distribution Systems Considering Temporary Closed-Loop Operation

• Published in: IEEE systems journal Vol. 15; no. 4; pp. 5483 - 5494
• Main Authors: Vargas, Renzo, Macedo, Leonardo H., Home-Ortiz, Juan M., Romero, Ruben
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.12.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Closed loop systems; Closed-loop topology operation; Distributed generation; distributed generation (DG); Distributed power generation; distribution systems optimization; Linear programming; Mathematical model; Mathematical models; Mixed integer; mixed-integer second-order cone programming (MISOCP); Network topologies; Network topology; Reactive power; Restoration; restoration problem; Substations; Switches; Switching
• ISSN: 1932-8184; 1937-9234
• DOI: 10.1109/JSYST.2021.3073941

This article presents a new mathematical model to solve the restoration problem in balanced distribution systems with distributed generators (DGs) considering closed-loop topology operation during the restorative state. The restorative state is comprised of the interval of time since the permanent fault has been isolated until the time at which the faulted zone is repaired. During this interval of time, switching operations are performed to minimize the negative effects resulting from the occurrence of a permanent fault in the network. In this way, the two main objective functions of the restoration problem are to minimize the amount of load curtailment in the restored system and to minimize the number of switching operations. Conventionally, the network topology is maintained radial throughout the restorative state. In this article, the possibility of forming loops is considered for improving both objective functions. As such, a new mixed-integer second-order cone programming model is proposed, considering the temporary formation of operational loops in the restorative state, and both connected and islanded operation of the DGs. Several tests are carried out using a 53-node test system and a 2313-node system for single and multiple fault scenarios. The results obtained with the proposed model outperform the solutions achieved when only open-loop configurations are considered for the restoration problem. Moreover, it is verified that the islanded operation of the DGs provides more flexibility to the network, allowing more load to be restored.