Enhancing electrical distribution network performance amidst rising electric vehicle integration
M. A. Ebrahim, Esraa E. Ahmed, M.M. Salama, M.M.R. Ahmed
Abstract
The rapid growth in electric vehicle adoption presents both operational challenges and modernization opportunities for electrical distribution networks. As electric vehicles become more prevalent, their inherently unpredictable charging behavior can introduce issues such as voltage instability, feeder congestion, and elevated power losses, which threaten the overall reliability and efficiency of the grid. To address these emerging challenges, this study introduces an advanced optimization framework that leverages two recent metaheuristic algorithms: Sand Cat Swarm Optimization and Mexican Axolotl Optimization. The proposed approach focuses on identifying the optimal locations for distributed generation units and electric vehicle charging stations in distribution networks. A weighted multi-objective function is employed to minimize total active power losses, enhance voltage profiles, and reduce operational costs associated with network reinforcement. The methodology is validated on IEEE 14-bus, 33-bus, and 200-bus test systems to demonstrate scalability and robustness. Simulation results show that in the single distributed generation placement scenario, power losses are reduced by up to 73 % with Mexican Axolotl Optimization and 67 % with Sand Cat Swarm Optimization. In the multi-unit configuration, power losses decreased further by up to 76 % and 84 %, respectively. Across all test cases, voltage magnitudes at system buses remain within the standard operational range of 0.95–1.05 per unit, confirming effective voltage regulation. An additional uncertainty analysis highlighted the impact of renewable variability and load fluctuations on system stability. • Novel bio-inspired algorithms applied for distributed generation units and electric vehicle charging stations co-optimization. • Significant reduction in power losses and improved voltage stability achieved. • Proposed methods outperform classical and existing metaheuristic techniques. • Robustness validated under multiple uncertainty and real-world scenarios. • Practical framework for sustainable distribution network operation developed.