Distributed cooperative scheduling for distribution network and smart charging hubs driven by unbalanced distribution locational marginal price

The low-carbon energy transition introduces new challenges to the operation of power distribution networks (PDN). In particular, the growing proliferation of single-phase loads and distributed renewable energies exacerbates three-phase imbalances within the PDN. The smart charging hub, as a low-carbon unit for transportation electrification, integrating photovoltaic (PV) generation, energy storage (ES), and electric vehicle (EV) charging infrastructure, processes phase-balancing capabilities by regulating the three-phase power. However, existing distribution markets lack effective mechanisms to incentivize the participation of PV-ES-EV integrated smart charging hubs (PEV-Hubs) in mitigating imbalances. This paper proposes a three-phase distribution locational marginal price (DLMP)-driven distributed optimization framework that coordinates PDN and PEV-Hubs to actively mitigate phase imbalances. A bi-level optimization model is formulated to capture the interaction between the distribution system operator (DSO) and PEV-Hubs. At the lower level, a three-phase distribution market clearing model is established based on the branch power flow model with voltage imbalance constraints, which can generate unbalanced DLMPs. At the upper level, PEV-Hubs optimize phase-to-phase power allocation to maximize revenue by taking unbalanced DLMP from the lower level as inputs. To solve the bi-level model, the Karush-Kuhn-Tucker (KKT) conditions and strong duality theory are employed to reformulate the bi-level problem into a single-level problem. A novel distributed solution method based on dual problem decomposition is then proposed for privacy protection. Simulation results on two different-scale systems confirm that the proposed method effectively limits the voltage imbalance in the PDN and enhances economic benefits for both the DSO and PEV-Hubs. Compared to the traditional iterative algorithms, the proposed distributed solution method exhibits better convergence and solution optimality.