Bi-Level Optimal Scheduling for Distribution Network with Electric Vehicles Considering Demand Response and Energy Storage Regulation

The large-scale uncoordinated integration of electric vehicles (EVs) into distribution networks exacerbates the load peak-valley difference and impairs voltage stability, posing significant challenges to the safe and economical operation of distribution networks. This paper proposes a bi-level optimal scheduling method incorporating demand response (DR) and energy storage regulation. A spatiotemporal EV load model is constructed using Quasi-Monte Carlo (QMC) simulation. By integrating a time-of-use (TOU) price-based DR model and an energy storage regulation strategy, a bi-level optimization model is established: the upper level minimizes the total operating cost of the distribution network operator (DNO) by optimizing electricity pricing and energy storage system (ESS) charge-discharge strategies; the lower level minimizes EV users' charging costs by optimizing their charging behaviors. The Dream Optimization Algorithm (DOA), utilizing a "memory, forgetting-supplementation, and dream sharing" strategy, solves this high-dimensional, constrained problem. Comparisons show the method reduces DNO total operating cost by 3.84% (vs. TOU-only pricing) and user charging costs by 4.1% (vs. no optimization), while enhancing renewable energy accommodation, demonstrating its feasibility.