To mitigate power distribution network load superposition caused by heterogeneous private electric vehicle (EV) behaviors, a bi-layer optimal scheduling method integrating battery swapping entitlement profiling with a dynamic incentive “blind boxes” is proposed. Through a two-dimensional profiling system, users are categorized into three groups, and a “charging inertia” theory is employed to quantify the energy leverage effect during replenishment path shifts. A bidirectional dynamic incentive mechanism is designed to guide entitled users toward residential charging via non-linear psychological incentives while attracting paying users during off-peak hours. Based on this, a bi-layer model is established to maximize operator profits and minimize network power loss and voltage deviations. It is demonstrated through simulation using real-world data from Shanghai that user response thresholds are accurately identified by the strategy. Operator profits are increased, peak-to-valley differences are reduced, and nodal voltage fluctuations are significantly stabilized without compromising user experience.