Equity-Aware Planning of Urban Ultra-Fast Electric Vehicle Charging Infrastructure with Spatial Graph Learning and Grid Constraints

Urban ultra-fast-charging planning must connect equitable user service with physically deliverable charging power. This study formulates a heterogeneous graph linking demand zones, roads, candidate stations, and distribution buses, together with an infrastructure model for station locations, port-level power tiers, interconnection, battery storage, photovoltaics, and network reinforcement. Charging occupancy is defined by a state-of-charge (SOC) integral with vehicle, local-temperature, cable-current, power-sharing, and conversion-efficiency limits. Minimum completion probability, group-service gaps, and conditional value at risk represent charging equity. The urban throughput and service metrics are illustrative and require recalculation with the integral charging model before quantitative planning use. For an assumed 80 kWh battery, separate SOC-only integration from 10% to 90% gives 19.10 min occupancy under the specified nonlinear curve and 14.31 min at constant power, with thermal limits nonbinding in both cases. An academic five-bus teaching-feeder demonstration gives a critical-bus peak-load change from 2.807 to 2.213 MW and a minimum-voltage change from 0.9612 to 0.9683 per unit (p.u.) under storage dispatch. These component calculations do not establish summer thermal performance, metropolitan grid reliability, or project financial feasibility. The contribution is a consistent methodological formulation; city-scale optimization and learning performance require empirical evaluation.