A New Probabilistic Framework for EV Charging Station Planning in Distribution Systems Considering Spatio-temporal Uncertainties

In the light of increased electric vehicle (EV) integration into smart grids, impact assessment studies and planning tools have gained a significant interest. If not properly planned, EV charging stations may cause excessive power losses and voltage violations. Existing model-based approaches to quantity the impact of EV charging station on distribution systems are limited to deterministic analysis that does not accurately capture temporal uncertainties. On the other hand, meta-heuristic approximations tend to be relatively complex with no performance guarantees. Therefore, the main goal of this paper is to develop a simple model-based probabilistic EV charging station planning strategy that accounts for spatio-temporal uncertainties. First, the developed probabilistic approach computes the probability distribution of change in power losses and voltage deviation due to EV charging station integration. Then, nodes with least contribution to losses and voltage deviations are identified and used as candidate nodes for charging station placement. Finally, a quadratic optimization framework is formulated to determine the optimal size of charging stations that meets EV demand and satisfies system constraints. The proposed approach is validated via simulation on the IEEE 123-node test system. Results show that the proposed approach significantly minimizes power losses and voltage deviations.