Unlocking Grid Hosting Capacity for EV Chargers via Non-Firm Connection: A Methodology for Power Limit Determination

The rapid growth of the electric vehicle (EV) fleet places increasing strain on existing power distribution networks, often requiring costly infrastructure upgrades to accommodate rising charging demand. Non-firm connection, which allows temporary power limitation of EV charging stations (EVCSs), represents a promising alternative to defer or avoid such investments. However, its widespread adoption is hindered by the lack of quantitative methodologies for determining the allowable connection capacity while ensuring satisfactory charging quality. This paper presents a novel analytical methodology for assessing the permissible power of homogeneous groups of EVCSs under both static and dynamic non-firm connection schemes. It is emphasized that the analytical expressions provide a conservative planning-oriented equivalent for dynamic non-firm connection capacity sizing rather than a real-time dynamic control algorithm. The proposed approach accounts for key influencing factors, including charger occupancy patterns, demand factors, fast-charging power curves, and the reserve capacity of the power grid. A closed-form expression is derived for the energy delivery coefficient, whereas the allowable power-equivalent number of simultaneously loaded chargers is determined from a nonlinear self-consistent equation solved iteratively. The methodology is validated through extensive Monte Carlo simulations over a 5-year period using real-world charging session statistics and EV fleet composition data. The results demonstrate high accuracy, with mean absolute percentage error (MAPE) below 5% for both slow and fast chargers. It is shown that fast chargers generally allow 21.8% lower connection capacity compared with slow chargers, while the idle time of EVs at slow charging stations can compensate for this difference. Furthermore, the shape of the fast-charging power curve significantly affects the allowable capacity, with variations of up to 25.8% between different EV battery charging profiles. The proposed methodology provides a rational basis for implementing non-firm connections, enabling grid operators and infrastructure planners to determine optimal connection capacities without compromising the quality of EV charging service.