Hamid Salarvand, Meysam Doostizadeh, Farhad Namdari,
Volume 18, Issue 4 (12-2022)
Abstract
Owing to the portability and flexibility of mobile energy storage systems (MESSs), they seem to be a promising solution to improve the resilience of the distribution system (DS). So, this paper presents a rolling optimization mechanism for dispatching MESSs and other resources in microgrids in case of a natural disaster occurrence. The proposed mechanism aims to minimize the total system cost based on the updated information of the status of the DS and transportation network (TN). In addition, the characteristics of the protection system in DS (i.e., relays with fixed protection settings), the constraints related to the protection coordination are examined under pre- and post-event conditions. The coordinated scheduling at each time step is formulated as a two-stage stochastic mixed-integer linear program (MILP) with temporal-spatial and operation constraints. The proposed model is carried out on the Sioux Falls TN and the IEEE 33-bus test system. The results demonstrate the effectiveness of MESS mobility in enhancing DS resilience due to the coordination of mobile and stationary resources.
Mahdi Arabsadegh, Aref Doroudi,
Volume 21, Issue 4 (11-2025)
Abstract
This paper presents an advanced methodology for post-storm power system restoration. A real-time Condition Index (CI)-based classification scheme is introduced to categorize circuit breakers into high-reliability (Type A) and moderate-reliability (Type B) groups. Leveraging this classification, a genetic algorithm (GA) optimizes microgrid configurations to maximize power restoration probabilities by explicitly modeling the stochastic failure risks associated with circuit breakers under severe weather conditions. The approach was validated on the IEEE 118-bus system with five critical breakers deactivated due to storm conditions. The GA achieved a 92.5% load restoration after 200 iterations, surpassing a baseline Monte Carlo simulation that attained 85.2%. Computational efficiency was significantly improved, reducing execution time to approximately 15 minutes compared to 60 minutes for traditional methods, with enhanced accuracy indicated by a 1.8% error margin versus 7.5%. Key contributions include utilizing live CI data for dynamic breaker classification, which resulted in a 20% reduction in computational time, and demonstrating scalability and effectiveness on large-scale test systems such as the 118-bus network. The methodology's performance decreases to 78.3% load restoration when more than 14 breakers are compromised. Future research will focus on integrating detailed storm modeling—including wind speed profiles—and incorporating renewable energy resources to enhance grid resilience.
