Pseudo-Label Generation Method Based on Wind Turbine SCADA Data
Chao Zhang, Wu Yue, Guanghan Zhao · 2021
During the daily operation of the wind turbine, the SCADA system continuously records the various operating parameters of the wind turbine, and has stored a large amount of data for many years. The amount of these data is huge, but most of them are unlabeled data. If you want to use deep learning to study the state of the wind turbine, these data cannot be directly used for analysis. In response to the above problems, this paper proposes a method of pseudo-label generation based on deep neural networks. This method first builds a deep neural network model, secondly uses part of the labeled data to train the built deep neural network model, and finally uses the trained deep neural network model to pseudo-label the unlabeled data, so as to obtain a large number of labeled data, label data set.