With the global acceleration of energy transition， the integration of high-penetration photovoltaic （PV） power presents new challenges to the economic dispatch of power systems. Addressing the multi-dimensional uncertainties in PV output caused by meteorological conditions and geographical distribution， this paper proposes a robust economic dispatch method that accounts for PV forecasting uncertainty. First， based on Chebyshev’s inequality and historical data， a multi-dimensional uncertainty set is constructed considering the spatiotemporal correlation of clustered PV output. Second， a two-stage robust optimization model is established with the objective of minimizing the total cost under the worst-case scenario . Finally， leveraging the model’s characteristics， the strong duality theory and the column-and-constraint generation （C&CG） algorithm are employed to decompose the main problem and the subproblem. The master problem optimizes the unit commitment， while the subproblem identifies the extreme scenarios using strong duality theory. Simulations conducted on a modified IEEE 30-bus power system show that the proposed method effectively mitigates PV forecasting uncertainty while ensuring the economic operation of the system.