A Bilevel Intelligent Dispatch Method for Distribution Networks with Wind, PV, Diesel, and Storage

As microgrids are increasingly integrated into distribution networks, a transition to bidirectional power flow is underway. To address the challenges of high dimensionality, solving complexity, and low precision found in current intelligent dispatching methods for such networks, this paper presents a bilevel intelligent dispatch model for distribution systems that incorporate wind, photovoltaic, diesel, and battery energy storage systems (WT-PV-DE-BES). The upper-level model for the distribution network is formulated as a multi-criteria scheduling problem, aiming to simultaneously minimize operating costs, network losses, and the voltage stability index, which is tackled by an improved Non-dominated Sorting Genetic Algorithm III (NSGA-III). The lower-level model for the microgrid focuses on minimizing its own operational costs and is solved using an enhanced particle swarm optimization (PSO) algorithm. The effectiveness of the proposed approach is validated via case studies. The results demonstrate that this method successfully balances economic efficiency with operational security, enabling mutual benefits for multiple stakeholders through deep coordination within the bilevel framework and showing significant practical potential.