The Research on Bi-Level Programming for Distribution Network Energy Storage in Remote Areas

To tackle the challenges of poor power quality and high line losses in remote mountainous areas due to long distribution lines and heavy loads, we propose a distributed energy storage optimization method based on genetic algorithms. Considering the geographic constraints of these regions and the ongoing development of electricity infrastructure, the method maintains grid safety and cost-effectiveness through a two-tier model for optimizing energy storage placement and capacity. The upper layer focuses on minimizing the installation costs of energy storage devices by optimizing their locations and capacities, while the lower layer aims to minimize distribution system operation and maintenance costs by reducing network losses and voltage deviations. An adaptive genetic algorithm is used for the upper layer, and a solver is employed for the lower layer. A simulation on a remote mountainous distribution system demonstrates that, compared to centralized energy storage, the optimized configuration of distributed energy storage improves both the economic efficiency and the stability and reliability of the system.