A consensus algorithm based decentralized autonomous control of hybrid multi-terminal direct current(MTDC)system is proposed. Consensus algorithm has been widely used in multi-agent systems to obtain global information only communicating with neighbouring agents

which is the foundation of global optimization control. Therefore

each terminal is equipped with an agent to constitute a multi-agent system

and each agent contains two function modules. Global information discovery module obtains total injected power and the number of converters participating optimization control. Optimization control module calculates power and voltage reference based on the discovered global information. First

the consensus algorithm is verified in MATLAB

the results show that the consensus algorithm can guarantee convergence and obtain global information in power fluctuation situation. Then

the power losses optimization of a hybrid MTDC system is used as an example to illustrate the feasible application of consensus algorithm based decentralized autonomous control. The simulation results in PSCAD manifest that the proposed control strategy can be applied in system losses optimization and effectively decrease the hybrid MTDC losses. This work is supported by National Natural Science Foundation of China(No. 51177044)and State Grid Corporation of China(No. XT71-14-042).