Cross-layer coordinated optimization of integrated ultra-fast charging stations considering station-power module-charging port coupling

The rapid increase in electric vehicle charging power enabled by 800 V platforms and high-rate batteries has pushed the instantaneous load of integrated ultra-fast charging stations to the megawatt level. Under high-concurrency conditions, stochastic vehicle arrivals and heterogeneous charging stages further introduce significant power fluctuations, posing major operational challenges for integrated ultra-fast charging stations. Conventional fixed power allocation strategies fail to exploit the inherent flexibility of modular charging architectures, resulting in low resource utilization and constrained operational performance. To address these issues, this paper proposes a cross-layer operational optimization method based on a modular power pool, enabling coordinated optimization of power module-to-charging port connectivity and station-level operation. The dynamic coupling between power modules and charging ports is explicitly modeled as decision variables, allowing adaptive power sharing among multiple charging demands. Based on battery characteristic data, SOC-dependent charging power boundaries are generated to characterize the nonlinear variation in charging capability with SOC and are incorporated into the optimization model as time-dependent feasible limits. Case studies demonstrate that the proposed method improves the average utilization of power modules from 11.36% to 28.06%, while reducing the grid-side peak power by 24.5% under charging demand satisfaction. These results highlight that power module-level reconfigurability is a key enabler for enhancing system flexibility, providing a scalable solution to improve both the economic performance and grid-friendliness of integrated ultra-fast charging stations.