Abstract: Wideband oscillation emergency control is one of the “bottom line” measures to ensure the safe operation of wind power systems. It is difficult to establish an accurate mathematical model to assess this risk, due to the complex and variable nature of wind power systems. To address this issue, a new emergency control strategy has been proposed that is based on data analysis. Firstly, a wideband phasor monitoring algorithm has been developed to detect and assess wideband oscillation. Secondly, the largest Lyapunov exponent（LLE） is used to measure the dynamic evolution trend of wideband oscillation and identify its source. The oscillation source is identified by analyzing the negative resistance characteristic of the wind turbine. Based on the identification results, an online risk assessment method for wideband oscillation has been proposed and applied to the emergency control strategy. The effectiveness and reliability of the proposed method have been verified using a real-time digital simulator（RTDS） hardware-in-the-loop platform.