Aiming at the problem of DC voltage control deviation and instability caused by a large-scale renewable energy access VSC–MTDC system, this paper combines voltage margin control and droop control. A strategy for controlling collaborative optimization in a sparsely distributed communication network has been proposed. Firstly, the distributed modeling of the system is carried out by combining MAS technology with small signal modeling. Then, a distributed model predictive controller is designed for a single droop control converter station. On this basis, a distributed cooperative optimization control strategy is proposed. According to the DC voltage deviation, the system adopts different control methods to control the receiving converter station. Finally, based on PSCAD/EMTDC and MATLAB co-simulation platforms, a six-terminal flexible HVDC system is built to verify the effectiveness of the control strategy under different conditions such as input power fluctuation, any converter station out of operation and system communication failure.