Abstract: With numerous distributed renewable energy resources connected to the active distribution network, there exist more and more couplings between the transmission and distribution (T-D) networks. Therefore, the uncertainty of renewable energy outputs and the mass communications caused by the centralized dispatching has brought about great challenges to the coordinated dispatching of (T-D) networks. Based on the requirements of low-carbon power development and the characteristics of low-carbon and peak load regulation in the carbon capture power plants, a decentralized coordination two-stage robust optimal scheduling model for (T-D) networks based on the improved alternating direction method of multipliers (ADMM) is proposed. Firstly, a convex polyhedron uncertainty set is used to describe the uncertainty of wind power output, and a two-stage robust optimal scheduling model of the (T-D) coordination with the carbon capture is constructed. Furthermore, the (T-D) networks are decoupled and iteratively solved by an iterative loop method (ILM-ADMM) and the column and constraint generation algorithm (C & CG). Finally, simulation analysis is carried out on two examples with different scales. The results show that the proposed method reduces the calculation scale and improves the solving efficiency. It realizes the complete decentralized autonomy of (T-D) networks, and effectively alleviates the peak load regulation pressure caused by the renewable energy.