Large-scale electric vehicles (EVs) disorderly charging will bring strong impact to the grid and threaten the security and economic operation of it. A bi-layer scheduling strategy considering temporal and spatial dimensions is proposed for the charging load optimal scheduling problem in the context of the fast charging mode of EVs. In the upper layer mode, the strategy obtains the optimal allocation of the EVs charging load under temporal domain by using a time-of-use (TOU) electricity price to minimizing operators’ cost. In the lower layer mode, according to the congestion level of charging station, a regional time-of-use (RTOU) charging electricity price is formulated and the EVs’ total charging cost and charging time are minimimum. And the optimization is solved by MCSO (modified cat swarm optimization) containing nonlinear decreasing inertia MR. Finally the results of examples demonstrate effectiveness and applicability of the proposed model. And the strategy can satisfy the interests of power grid, operators and EV users while realizing the friendly access of large-scale EVs fast charging.