<![CDATA[This study addresses a large-scale Multi-Objective Covering Problem (MOCP) in urban electricity distribution, involving 416.472 customers and 732 substations. The main objective is to optimize the allocation of customers to substations, taking into account trade-offs between three objectives: minimizing average service distance, balancing load, and minimizing substation investment costs. To achieve this, we propose a hybrid framework that combines a constrained fuzzy C-means algorithm for customer assignment, NSGA-II algorithm for multi-objective optimization, and K-means clustering for facility expansion. The results show a trade-off between the objectives, with the composite evaluation identifying k=7 as the best compromise. Statistical validation confirms the significance of these results. Significantly, the proposed framework can generate scalable solutions for MOCP in the real world. By integrating construction cost data and spatial data, the model shows a 15% reduction in substation investment costs and an 8% increase in service reliability. This research provides practical insights for distribution network planners, by offering a data-driven approach to determine the optimal number and strategic placement of additional substations, considering trade-offs between cost, performance, and spatial constraints.]]>
Copyrights © 2026
