data-driven polyhedron-based robust scheduling method for AC/DC hybrid distribution networks is proposed to address the lack of consideration of uncertain parameter correlation in existing robust scheduling methods. First, a traditional polyhedral set of distributed photovoltaic output is constructed, and a correlation envelope graph is driven by historical data, and the correlation polyhedral set model is established by bending the polyhedral set boundary. Second, based on this, a data-driven polyhedral set model is established to address the issues of poor robustness and high conservatism in correlated polyhedral sets. Then, a robust scheduling model for AC/DC hybrid distribution networks based on data-driven polyhedral sets is established, and a column and constraint generation (CCG) algorithm is used to analyze the robust scheduling model. Finally, the simulation results of the improved IEEE33 node system show that the proposed scheduling method can reduce the conservatism of the optimization results and improve their robustness, proving the effectiveness of the method.