The uncertainty of hydro, wind resources and load will affect the economics and robustness of microgrid configuration, this paper proposes a bi-level optimal planning and design model for grid-connected hydro-wind-diesel microgrid, considering uncertainty of renewable energy and load. The up-level optimization is a microgrid configuration model. The low-level optimization is a microgrid economic dispatch model considering the uncertainty of hydro-wind power and load. Based on the robust linear optimization theory proposed by Seng-Cheol Kang(SCK-type robust linear optimization), the uncertainty model is transformed into a deterministic model. Then, combined with the fast non-dominated sorting genetic algorithm with elite strategy(NSGA-II) and mixed integer linear programming (MILP), the multi-objective optimal configuration problem of microgrid is solved. Finally, a distribution system in Shaoguan city is taken as a simulation example. The results show that the model not only improves the robustness of the microgrid, but also ensures the economics and reliability of the configuration.