With the rapid growth of wind power-photovoltaic grid-connected penetration, the uncertainty of the power system increases dramatically. Aiming at the impact of large-scale grid-connected wind power-photovoltaic on power grid planning, this paper constructs a two-stage robust planning model for the wind power-photovoltaic-thermal power-pumped storage-battery hybrid system, and introduces uncertain adjustment parameters to reduce the conservatism of the model. In the first stage, the capacity configuration of the hybrid system is aiming at minimizing the sum of investment cost and operation and maintenance cost. In the second stage, under a given capacity configuration, the optimal scheduling scheme is determined by constructing an uncertain set of wind power-photovoltaic output, aiming at minimizing the sum of environmental cost and cost of wind power-photovoltaic abandonment. According to the characteristics of the robust planning model, the column and constraint generation algorithm(C&CG) and strong pairwise theory are used to decompose the original problem into a major problem and sub-problem with mixed-integer linear characteristics to solve. The improved IEEE30-node system is followed to analyze the economics of the planning scheme under different conservative degrees, and the validity of the proposed model is verified.