A Fault Diagnosis Method for Wind Turbines of New Wind Farm Based on Joint Matching Adaptive Network

Traditional machine learning or deep learning methods rely on large amounts of data for model training. For new wind farms, they often face the challenge of lack of sufficient labeled data. In addition, due to the different data distribution between different wind farms, it is impossible to directly use the wind turbine data of other existing wind farms for construction of fault diagnosis model for wind turbines in new wind farms. Many wind turbine fault diagnosis methods based on machine learning and deep learning are no longer applicable. Therefore, in view of the lack of labeled data in new wind farms, this paper proposed a new domain adaptive method: joint matching adaptive network (JMAN). This method is improved based on the existing joint adaptive network. The experimental results show that the proposed JMAN has higher diagnostic accuracy than common transfer learning methods.