Abstract: In order to accurately describe the impact of the fluctuation and randomness of new energy output power on the operation of integrated energy complementary microgrid systems, a data-driven robust optimization method for multi energy microgrids is proposed. Firstly, based on the traditional interval set, the uncertain parameters of new energy output are modeled using a polyhedral set. Then, an ellipsoidal uncertainty set is established using historical data of new energy output with spatiotemporal correlation. By connecting high-dimensional ellipsoidal vertices, a data-driven convex hull polyhedron set is established. Then, the uncertain parameters are better enveloped by scaling the convex hull set. A data-driven robust optimization model for multi energy complementary microgrids is further established, and the column and constraint generation algorithm(C&CG) is used to solve the model. Finally, simulation comparisons are conducted through examples, and the results show that the data-driven robust optimization method for multi energy complementary microgrids can reduce conservatism and improve the robustness of optimization results, demonstrating the effectiveness of the proposed method.