Abstract: In recent years, large-scale access of renewable energy sources such as wind power causes frequent occurrence of subsynchronous oscillation accidents. Given the existing monitoring system is difficult to meet the monitoring requirements for the subsynchronous oscillation, many places in China are actively developing a new oscillation monitoring system. Considering the limitation of time and cost, it is the primary problem to determine the critical monitoring nodes of the system and select the number of configuration economically and reasonably. The nodal critical index of each node in the oscillation monitoring system is defined from the perspectives of oscillation monitoring and source location, and an optimal placement method considering nodal critical index is proposed. In this method, the nodal critical index is established by synthesizing the nodal observability and damping influential degree, and then the optimal placement model is constructed by taking the probability of N-1 contingency and zero injection nodes into account. Finally, the placement scheme is solved by integer linear programming. The feasibility and economy of the proposed algorithm are verified by comparing the different placement schemes of the ERCOT wind power system in Texas,USA, the modified New England 39-bus system, and an actual grid-connected wind power system in Guyuan, Hebei Province,China.