To improve the operational efficiency of wind turbines and optimize the operation and maintenance costs of wind farms， this paper combines time-domain feature index analysis with multi-sensor information fusion technology to propose a wind turbine gearbox state monitoring method based on grey wolf optimization （GWO） algorithm-support vector machine （SVM）. Firstly， different time-domain statistical eigenvalues representing vibration energy are calculated， and parallel stacking is used for feature level and data level fusion to obtain an information fusion matrix. Secondly， on this basis， establish a fault diagnosis classification model based on GWO-SVM. Finally， the proposed method is validated and analyzed using the measured data of the gearbox collected from the QPZZ-Ⅱ rotating machinery vibration test. The results show that this method is significantly better than other traditional methods， and its classification and diagnostic accuracy demonstrate significant advantages.