With the increasing load and a large number of new energies connected to the grid, the flexible DC interconnection system is widely used. The flexible DC interconnection system based on sagging control strategy can realize automatic regulation of power and voltage according to sagging characteristics, but the transmission capacity of the system is limited by the capacities of different converter stations and requirement of voltage quality. So it is difficult to make full use of the transmission capacity of the system. An adaptive sagging control strategy for capacity optimization of flexible DC interconnection systems is proposed in this paper. By the proposed strategy, a sagging coefficient correction loop is added to gradually adjust the sagging coefficient of the corresponding converter station when the voltage or power is close to the limit. By the correction loop, the power can continue to increase without exceeding the limit. In addition, in order to reduce the influence of droop coefficient change on the traditional control and optimize the energy consumption characteristics of the system, a mathematical model is established with the goal of minimizing the adjustment rate of the power flow and the energy consumption of the power generation of the system. Then the non-dominated sorting in genetic algorithm (NSGA-Ⅱ) is used to solve the model and the optimal correction gradient of each converter station can be obtained. Finally, PLECS is used to build a four-terminal flexible direct interconnection model which verifies the effectiveness and feasibility of the proposed strategy.