In a grid-connected wind farm based on permanent magnet synchronous generators (PMSGs)

the wind speed and the number of operating PMSGs are the two most important influencing factors along with the stochastic nature of sub-synchronous oscillation (SSO) from the point view of the farm. This paper proposes a method of unstable SSO risk evaluation for grid-connected PMSG-based wind farms based on the sequential Monte Carlo simulation (SMCS). The determination of critical wind speed (CWS) of SSO and the sequential simulation strategy of wind speed states and PMSG states in a wind farm at the same wind speed (S-WF)

as well as in a wind farm at different wind speeds (D-WF)

are studied. Five indices evaluating the expectation

duration

frequency and energy loss of SSO risk are proposed. Moreover

a strategy to reduce SSO risk by adjusting the cut-in wind speed is discussed. The effectiveness of the discussed issues in this paper are proved by the case studies of a 750-PMSG wind farm based on the actual wind speed data collected.