Abstract: The generator is a crucial component of the wind turbine. However, the failure probability of the generator is high and maintenance is difficult due to its complicated internal structure and harsh operating environment. To solve this problem, we propose a health assessment method for the wind-turbine generator based on supervisory control and data acquisition (SCADA) data. First, we identify variables related to the operating status of the generator and redundant variables based on expert experience and correlation analysis of the state variables. On this basis, we select some reasonable state parameters. Then, using historical data from normal operation, we establish a health benchmark model based on the Gaussian mixture model (GMM). Finally, to evaluate the health status of a generator, we design a health degradation index (HDI) based on the Mahalanobis distance. We verify the effectiveness of our proposed method, and we apply it to 2016 SCADA data from a wind farm of the Shanghai Electric Wind Power Group Co., Ltd. The test results show that the proposed method can accurately track changes in the generator operating status and facilitate early fault identification. In addition, the proposed method is universal in its application.