Optimal energy management at solar-integrated hubs for e-taxis with battery swapping and fast charging under power network constraints

An integrated energy management approach is designed and presented for e-taxis at a multi-functional hub, which provides both battery swapping and fast charging services. This approach illustrates the collaboration among key actors, including the e-taxi fleet operator, solar-integrated hub operator, distribution system operator, and energy supplier. This approach comprises two essential models: the e-taxi visits model and the bidirectional charging optimization model for the multi-functional hub. To tackle the stochasticity of taxi arrivals and departures at the hub, a Monte Carlo simulation approach is employed to derive probability distribution functions. Meanwhile, deep reinforcement learning methods are utilized to optimize bidirectional charging while considering power network constraints. Furthermore, the optimization model integrates both contracted solar power through a power purchase agreement and dynamic pricing tariffs. This ensures that the charging demands at the multi-functional hub are always met. The viability of the proposed energy management approach is evaluated using real-world datasets of taxis, and CIGRE medium voltage (MV) benchmark network is used. The results evidently validate the effectiveness of the proposed approach in optimizing bidirectional charging at a multi-functional hub.